diff options
144 files changed, 6419 insertions, 1917 deletions
diff --git a/Kconfig.host b/Kconfig.host index d763d89269..2ee71578f3 100644 --- a/Kconfig.host +++ b/Kconfig.host @@ -46,3 +46,6 @@ config FUZZ config VFIO_USER_SERVER_ALLOWED bool imply VFIO_USER_SERVER + +config HV_BALLOON_POSSIBLE + bool diff --git a/MAINTAINERS b/MAINTAINERS index 1de7f381e8..59b92ee640 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2656,6 +2656,14 @@ F: hw/usb/canokey.c F: hw/usb/canokey.h F: docs/system/devices/canokey.rst +Hyper-V Dynamic Memory Protocol +M: Maciej S. Szmigiero <maciej.szmigiero@oracle.com> +S: Supported +F: hw/hyperv/hv-balloon*.c +F: hw/hyperv/hv-balloon*.h +F: include/hw/hyperv/dynmem-proto.h +F: include/hw/hyperv/hv-balloon.h + Subsystems ---------- Overall Audio backends diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index b8c5e345b8..f35c5f359b 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -321,21 +321,6 @@ static void flush_all_helper(CPUState *src, run_on_cpu_func fn, } } -void tlb_flush_counts(size_t *pfull, size_t *ppart, size_t *pelide) -{ - CPUState *cpu; - size_t full = 0, part = 0, elide = 0; - - CPU_FOREACH(cpu) { - full += qatomic_read(&cpu->neg.tlb.c.full_flush_count); - part += qatomic_read(&cpu->neg.tlb.c.part_flush_count); - elide += qatomic_read(&cpu->neg.tlb.c.elide_flush_count); - } - *pfull = full; - *ppart = part; - *pelide = elide; -} - static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data) { uint16_t asked = data.host_int; @@ -2706,7 +2691,7 @@ static uint64_t do_st16_leN(CPUState *cpu, MMULookupPageData *p, case MO_ATOM_WITHIN16_PAIR: /* Since size > 8, this is the half that must be atomic. */ - if (!HAVE_ATOMIC128_RW) { + if (!HAVE_CMPXCHG128) { cpu_loop_exit_atomic(cpu, ra); } return store_whole_le16(p->haddr, p->size, val_le); diff --git a/accel/tcg/internal-common.h b/accel/tcg/internal-common.h index 3b2277e6e9..edefd0dcb7 100644 --- a/accel/tcg/internal-common.h +++ b/accel/tcg/internal-common.h @@ -14,8 +14,6 @@ extern int64_t max_delay; extern int64_t max_advance; -void dump_exec_info(GString *buf); - /* * Return true if CS is not running in parallel with other cpus, either * because there are no other cpus or we are within an exclusive context. diff --git a/accel/tcg/ldst_atomicity.c.inc b/accel/tcg/ldst_atomicity.c.inc index 1cf5b92166..33a04dec52 100644 --- a/accel/tcg/ldst_atomicity.c.inc +++ b/accel/tcg/ldst_atomicity.c.inc @@ -825,7 +825,7 @@ static uint64_t store_whole_le16(void *pv, int size, Int128 val_le) int sh = o * 8; Int128 m, v; - qemu_build_assert(HAVE_ATOMIC128_RW); + qemu_build_assert(HAVE_CMPXCHG128); /* Like MAKE_64BIT_MASK(0, sz), but larger. */ if (sz <= 64) { @@ -887,7 +887,7 @@ static void store_atom_2(CPUState *cpu, uintptr_t ra, return; } } else if ((pi & 15) == 7) { - if (HAVE_ATOMIC128_RW) { + if (HAVE_CMPXCHG128) { Int128 v = int128_lshift(int128_make64(val), 56); Int128 m = int128_lshift(int128_make64(0xffff), 56); store_atom_insert_al16(pv - 7, v, m); @@ -956,7 +956,7 @@ static void store_atom_4(CPUState *cpu, uintptr_t ra, return; } } else { - if (HAVE_ATOMIC128_RW) { + if (HAVE_CMPXCHG128) { store_whole_le16(pv, 4, int128_make64(cpu_to_le32(val))); return; } @@ -1021,7 +1021,7 @@ static void store_atom_8(CPUState *cpu, uintptr_t ra, } break; case MO_64: - if (HAVE_ATOMIC128_RW) { + if (HAVE_CMPXCHG128) { store_whole_le16(pv, 8, int128_make64(cpu_to_le64(val))); return; } @@ -1076,7 +1076,7 @@ static void store_atom_16(CPUState *cpu, uintptr_t ra, } break; case -MO_64: - if (HAVE_ATOMIC128_RW) { + if (HAVE_CMPXCHG128) { uint64_t val_le; int s2 = pi & 15; int s1 = 16 - s2; @@ -1103,10 +1103,6 @@ static void store_atom_16(CPUState *cpu, uintptr_t ra, } break; case MO_128: - if (HAVE_ATOMIC128_RW) { - atomic16_set(pv, val); - return; - } break; default: g_assert_not_reached(); diff --git a/accel/tcg/monitor.c b/accel/tcg/monitor.c index caf1189e0b..093efe9714 100644 --- a/accel/tcg/monitor.c +++ b/accel/tcg/monitor.c @@ -8,6 +8,7 @@ #include "qemu/osdep.h" #include "qemu/accel.h" +#include "qemu/qht.h" #include "qapi/error.h" #include "qapi/type-helpers.h" #include "qapi/qapi-commands-machine.h" @@ -17,6 +18,7 @@ #include "sysemu/tcg.h" #include "tcg/tcg.h" #include "internal-common.h" +#include "tb-context.h" static void dump_drift_info(GString *buf) @@ -50,6 +52,153 @@ static void dump_accel_info(GString *buf) one_insn_per_tb ? "on" : "off"); } +static void print_qht_statistics(struct qht_stats hst, GString *buf) +{ + uint32_t hgram_opts; + size_t hgram_bins; + char *hgram; + + if (!hst.head_buckets) { + return; + } + g_string_append_printf(buf, "TB hash buckets %zu/%zu " + "(%0.2f%% head buckets used)\n", + hst.used_head_buckets, hst.head_buckets, + (double)hst.used_head_buckets / + hst.head_buckets * 100); + + hgram_opts = QDIST_PR_BORDER | QDIST_PR_LABELS; + hgram_opts |= QDIST_PR_100X | QDIST_PR_PERCENT; + if (qdist_xmax(&hst.occupancy) - qdist_xmin(&hst.occupancy) == 1) { + hgram_opts |= QDIST_PR_NODECIMAL; + } + hgram = qdist_pr(&hst.occupancy, 10, hgram_opts); + g_string_append_printf(buf, "TB hash occupancy %0.2f%% avg chain occ. " + "Histogram: %s\n", + qdist_avg(&hst.occupancy) * 100, hgram); + g_free(hgram); + + hgram_opts = QDIST_PR_BORDER | QDIST_PR_LABELS; + hgram_bins = qdist_xmax(&hst.chain) - qdist_xmin(&hst.chain); + if (hgram_bins > 10) { + hgram_bins = 10; + } else { + hgram_bins = 0; + hgram_opts |= QDIST_PR_NODECIMAL | QDIST_PR_NOBINRANGE; + } + hgram = qdist_pr(&hst.chain, hgram_bins, hgram_opts); + g_string_append_printf(buf, "TB hash avg chain %0.3f buckets. " + "Histogram: %s\n", + qdist_avg(&hst.chain), hgram); + g_free(hgram); +} + +struct tb_tree_stats { + size_t nb_tbs; + size_t host_size; + size_t target_size; + size_t max_target_size; + size_t direct_jmp_count; + size_t direct_jmp2_count; + size_t cross_page; +}; + +static gboolean tb_tree_stats_iter(gpointer key, gpointer value, gpointer data) +{ + const TranslationBlock *tb = value; + struct tb_tree_stats *tst = data; + + tst->nb_tbs++; + tst->host_size += tb->tc.size; + tst->target_size += tb->size; + if (tb->size > tst->max_target_size) { + tst->max_target_size = tb->size; + } + if (tb->page_addr[1] != -1) { + tst->cross_page++; + } + if (tb->jmp_reset_offset[0] != TB_JMP_OFFSET_INVALID) { + tst->direct_jmp_count++; + if (tb->jmp_reset_offset[1] != TB_JMP_OFFSET_INVALID) { + tst->direct_jmp2_count++; + } + } + return false; +} + +static void tlb_flush_counts(size_t *pfull, size_t *ppart, size_t *pelide) +{ + CPUState *cpu; + size_t full = 0, part = 0, elide = 0; + + CPU_FOREACH(cpu) { + full += qatomic_read(&cpu->neg.tlb.c.full_flush_count); + part += qatomic_read(&cpu->neg.tlb.c.part_flush_count); + elide += qatomic_read(&cpu->neg.tlb.c.elide_flush_count); + } + *pfull = full; + *ppart = part; + *pelide = elide; +} + +static void tcg_dump_info(GString *buf) +{ + g_string_append_printf(buf, "[TCG profiler not compiled]\n"); +} + +static void dump_exec_info(GString *buf) +{ + struct tb_tree_stats tst = {}; + struct qht_stats hst; + size_t nb_tbs, flush_full, flush_part, flush_elide; + + tcg_tb_foreach(tb_tree_stats_iter, &tst); + nb_tbs = tst.nb_tbs; + /* XXX: avoid using doubles ? */ + g_string_append_printf(buf, "Translation buffer state:\n"); + /* + * Report total code size including the padding and TB structs; + * otherwise users might think "-accel tcg,tb-size" is not honoured. + * For avg host size we use the precise numbers from tb_tree_stats though. + */ + g_string_append_printf(buf, "gen code size %zu/%zu\n", + tcg_code_size(), tcg_code_capacity()); + g_string_append_printf(buf, "TB count %zu\n", nb_tbs); + g_string_append_printf(buf, "TB avg target size %zu max=%zu bytes\n", + nb_tbs ? tst.target_size / nb_tbs : 0, + tst.max_target_size); + g_string_append_printf(buf, "TB avg host size %zu bytes " + "(expansion ratio: %0.1f)\n", + nb_tbs ? tst.host_size / nb_tbs : 0, + tst.target_size ? + (double)tst.host_size / tst.target_size : 0); + g_string_append_printf(buf, "cross page TB count %zu (%zu%%)\n", + tst.cross_page, + nb_tbs ? (tst.cross_page * 100) / nb_tbs : 0); + g_string_append_printf(buf, "direct jump count %zu (%zu%%) " + "(2 jumps=%zu %zu%%)\n", + tst.direct_jmp_count, + nb_tbs ? (tst.direct_jmp_count * 100) / nb_tbs : 0, + tst.direct_jmp2_count, + nb_tbs ? (tst.direct_jmp2_count * 100) / nb_tbs : 0); + + qht_statistics_init(&tb_ctx.htable, &hst); + print_qht_statistics(hst, buf); + qht_statistics_destroy(&hst); + + g_string_append_printf(buf, "\nStatistics:\n"); + g_string_append_printf(buf, "TB flush count %u\n", + qatomic_read(&tb_ctx.tb_flush_count)); + g_string_append_printf(buf, "TB invalidate count %u\n", + qatomic_read(&tb_ctx.tb_phys_invalidate_count)); + + tlb_flush_counts(&flush_full, &flush_part, &flush_elide); + g_string_append_printf(buf, "TLB full flushes %zu\n", flush_full); + g_string_append_printf(buf, "TLB partial flushes %zu\n", flush_part); + g_string_append_printf(buf, "TLB elided flushes %zu\n", flush_elide); + tcg_dump_info(buf); +} + HumanReadableText *qmp_x_query_jit(Error **errp) { g_autoptr(GString) buf = g_string_new(""); @@ -66,6 +215,11 @@ HumanReadableText *qmp_x_query_jit(Error **errp) return human_readable_text_from_str(buf); } +static void tcg_dump_op_count(GString *buf) +{ + g_string_append_printf(buf, "[TCG profiler not compiled]\n"); +} + HumanReadableText *qmp_x_query_opcount(Error **errp) { g_autoptr(GString) buf = g_string_new(""); diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index 8cb6ad3511..e579b0891d 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -645,133 +645,6 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) cpu_loop_exit_noexc(cpu); } -static void print_qht_statistics(struct qht_stats hst, GString *buf) -{ - uint32_t hgram_opts; - size_t hgram_bins; - char *hgram; - - if (!hst.head_buckets) { - return; - } - g_string_append_printf(buf, "TB hash buckets %zu/%zu " - "(%0.2f%% head buckets used)\n", - hst.used_head_buckets, hst.head_buckets, - (double)hst.used_head_buckets / - hst.head_buckets * 100); - - hgram_opts = QDIST_PR_BORDER | QDIST_PR_LABELS; - hgram_opts |= QDIST_PR_100X | QDIST_PR_PERCENT; - if (qdist_xmax(&hst.occupancy) - qdist_xmin(&hst.occupancy) == 1) { - hgram_opts |= QDIST_PR_NODECIMAL; - } - hgram = qdist_pr(&hst.occupancy, 10, hgram_opts); - g_string_append_printf(buf, "TB hash occupancy %0.2f%% avg chain occ. " - "Histogram: %s\n", - qdist_avg(&hst.occupancy) * 100, hgram); - g_free(hgram); - - hgram_opts = QDIST_PR_BORDER | QDIST_PR_LABELS; - hgram_bins = qdist_xmax(&hst.chain) - qdist_xmin(&hst.chain); - if (hgram_bins > 10) { - hgram_bins = 10; - } else { - hgram_bins = 0; - hgram_opts |= QDIST_PR_NODECIMAL | QDIST_PR_NOBINRANGE; - } - hgram = qdist_pr(&hst.chain, hgram_bins, hgram_opts); - g_string_append_printf(buf, "TB hash avg chain %0.3f buckets. " - "Histogram: %s\n", - qdist_avg(&hst.chain), hgram); - g_free(hgram); -} - -struct tb_tree_stats { - size_t nb_tbs; - size_t host_size; - size_t target_size; - size_t max_target_size; - size_t direct_jmp_count; - size_t direct_jmp2_count; - size_t cross_page; -}; - -static gboolean tb_tree_stats_iter(gpointer key, gpointer value, gpointer data) -{ - const TranslationBlock *tb = value; - struct tb_tree_stats *tst = data; - - tst->nb_tbs++; - tst->host_size += tb->tc.size; - tst->target_size += tb->size; - if (tb->size > tst->max_target_size) { - tst->max_target_size = tb->size; - } - if (tb_page_addr1(tb) != -1) { - tst->cross_page++; - } - if (tb->jmp_reset_offset[0] != TB_JMP_OFFSET_INVALID) { - tst->direct_jmp_count++; - if (tb->jmp_reset_offset[1] != TB_JMP_OFFSET_INVALID) { - tst->direct_jmp2_count++; - } - } - return false; -} - -void dump_exec_info(GString *buf) -{ - struct tb_tree_stats tst = {}; - struct qht_stats hst; - size_t nb_tbs, flush_full, flush_part, flush_elide; - - tcg_tb_foreach(tb_tree_stats_iter, &tst); - nb_tbs = tst.nb_tbs; - /* XXX: avoid using doubles ? */ - g_string_append_printf(buf, "Translation buffer state:\n"); - /* - * Report total code size including the padding and TB structs; - * otherwise users might think "-accel tcg,tb-size" is not honoured. - * For avg host size we use the precise numbers from tb_tree_stats though. - */ - g_string_append_printf(buf, "gen code size %zu/%zu\n", - tcg_code_size(), tcg_code_capacity()); - g_string_append_printf(buf, "TB count %zu\n", nb_tbs); - g_string_append_printf(buf, "TB avg target size %zu max=%zu bytes\n", - nb_tbs ? tst.target_size / nb_tbs : 0, - tst.max_target_size); - g_string_append_printf(buf, "TB avg host size %zu bytes " - "(expansion ratio: %0.1f)\n", - nb_tbs ? tst.host_size / nb_tbs : 0, - tst.target_size ? - (double)tst.host_size / tst.target_size : 0); - g_string_append_printf(buf, "cross page TB count %zu (%zu%%)\n", - tst.cross_page, - nb_tbs ? (tst.cross_page * 100) / nb_tbs : 0); - g_string_append_printf(buf, "direct jump count %zu (%zu%%) " - "(2 jumps=%zu %zu%%)\n", - tst.direct_jmp_count, - nb_tbs ? (tst.direct_jmp_count * 100) / nb_tbs : 0, - tst.direct_jmp2_count, - nb_tbs ? (tst.direct_jmp2_count * 100) / nb_tbs : 0); - - qht_statistics_init(&tb_ctx.htable, &hst); - print_qht_statistics(hst, buf); - qht_statistics_destroy(&hst); - - g_string_append_printf(buf, "\nStatistics:\n"); - g_string_append_printf(buf, "TB flush count %u\n", - qatomic_read(&tb_ctx.tb_flush_count)); - g_string_append_printf(buf, "TB invalidate count %u\n", - qatomic_read(&tb_ctx.tb_phys_invalidate_count)); - - tlb_flush_counts(&flush_full, &flush_part, &flush_elide); - g_string_append_printf(buf, "TLB full flushes %zu\n", flush_full); - g_string_append_printf(buf, "TLB partial flushes %zu\n", flush_part); - g_string_append_printf(buf, "TLB elided flushes %zu\n", flush_elide); - tcg_dump_info(buf); -} - #else /* CONFIG_USER_ONLY */ void cpu_interrupt(CPUState *cpu, int mask) diff --git a/block/file-posix.c b/block/file-posix.c index 50e2b20d5c..b862406c71 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -160,7 +160,6 @@ typedef struct BDRVRawState { bool has_write_zeroes:1; bool use_linux_aio:1; bool use_linux_io_uring:1; - int64_t *offset; /* offset of zone append operation */ int page_cache_inconsistent; /* errno from fdatasync failure */ bool has_fallocate; bool needs_alignment; @@ -2445,12 +2444,13 @@ static bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov) return true; } -static int coroutine_fn raw_co_prw(BlockDriverState *bs, uint64_t offset, +static int coroutine_fn raw_co_prw(BlockDriverState *bs, int64_t *offset_ptr, uint64_t bytes, QEMUIOVector *qiov, int type) { BDRVRawState *s = bs->opaque; RawPosixAIOData acb; int ret; + uint64_t offset = *offset_ptr; if (fd_open(bs) < 0) return -EIO; @@ -2513,8 +2513,8 @@ out: uint64_t *wp = &wps->wp[offset / bs->bl.zone_size]; if (!BDRV_ZT_IS_CONV(*wp)) { if (type & QEMU_AIO_ZONE_APPEND) { - *s->offset = *wp; - trace_zbd_zone_append_complete(bs, *s->offset + *offset_ptr = *wp; + trace_zbd_zone_append_complete(bs, *offset_ptr >> BDRV_SECTOR_BITS); } /* Advance the wp if needed */ @@ -2523,7 +2523,10 @@ out: } } } else { - update_zones_wp(bs, s->fd, 0, 1); + /* + * write and append write are not allowed to cross zone boundaries + */ + update_zones_wp(bs, s->fd, offset, 1); } qemu_co_mutex_unlock(&wps->colock); @@ -2536,14 +2539,14 @@ static int coroutine_fn raw_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags) { - return raw_co_prw(bs, offset, bytes, qiov, QEMU_AIO_READ); + return raw_co_prw(bs, &offset, bytes, qiov, QEMU_AIO_READ); } static int coroutine_fn raw_co_pwritev(BlockDriverState *bs, int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags) { - return raw_co_prw(bs, offset, bytes, qiov, QEMU_AIO_WRITE); + return raw_co_prw(bs, &offset, bytes, qiov, QEMU_AIO_WRITE); } static int coroutine_fn raw_co_flush_to_disk(BlockDriverState *bs) @@ -3470,7 +3473,7 @@ static int coroutine_fn raw_co_zone_mgmt(BlockDriverState *bs, BlockZoneOp op, len >> BDRV_SECTOR_BITS); ret = raw_thread_pool_submit(handle_aiocb_zone_mgmt, &acb); if (ret != 0) { - update_zones_wp(bs, s->fd, offset, i); + update_zones_wp(bs, s->fd, offset, nrz); error_report("ioctl %s failed %d", op_name, ret); return ret; } @@ -3506,8 +3509,6 @@ static int coroutine_fn raw_co_zone_append(BlockDriverState *bs, int64_t zone_size_mask = bs->bl.zone_size - 1; int64_t iov_len = 0; int64_t len = 0; - BDRVRawState *s = bs->opaque; - s->offset = offset; if (*offset & zone_size_mask) { error_report("sector offset %" PRId64 " is not aligned to zone size " @@ -3528,7 +3529,7 @@ static int coroutine_fn raw_co_zone_append(BlockDriverState *bs, } trace_zbd_zone_append(bs, *offset >> BDRV_SECTOR_BITS); - return raw_co_prw(bs, *offset, len, qiov, QEMU_AIO_ZONE_APPEND); + return raw_co_prw(bs, offset, len, qiov, QEMU_AIO_ZONE_APPEND); } #endif diff --git a/block/nvme.c b/block/nvme.c index 96b3f8f2fa..0a0a0a6b36 100644 --- a/block/nvme.c +++ b/block/nvme.c @@ -417,9 +417,10 @@ static bool nvme_process_completion(NVMeQueuePair *q) q->cq_phase = !q->cq_phase; } cid = le16_to_cpu(c->cid); - if (cid == 0 || cid > NVME_QUEUE_SIZE) { - warn_report("NVMe: Unexpected CID in completion queue: %"PRIu32", " - "queue size: %u", cid, NVME_QUEUE_SIZE); + if (cid == 0 || cid > NVME_NUM_REQS) { + warn_report("NVMe: Unexpected CID in completion queue: %" PRIu32 + ", should be within: 1..%u inclusively", cid, + NVME_NUM_REQS); continue; } trace_nvme_complete_command(s, q->index, cid); diff --git a/block/parallels-ext.c b/block/parallels-ext.c index 8a109f005a..4d8ecf5047 100644 --- a/block/parallels-ext.c +++ b/block/parallels-ext.c @@ -130,7 +130,7 @@ static BdrvDirtyBitmap *parallels_load_bitmap(BlockDriverState *bs, g_autofree uint64_t *l1_table = NULL; BdrvDirtyBitmap *bitmap; QemuUUID uuid; - char uuidstr[UUID_FMT_LEN + 1]; + char uuidstr[UUID_STR_LEN]; int i; if (data_size < sizeof(bf)) { diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index 904f00d1b3..5af439bd11 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -1983,7 +1983,7 @@ discard_in_l2_slice(BlockDriverState *bs, uint64_t offset, uint64_t nb_clusters, /* If we keep the reference, pass on the discard still */ bdrv_pdiscard(s->data_file, old_l2_entry & L2E_OFFSET_MASK, s->cluster_size); - } + } } qcow2_cache_put(s->l2_table_cache, (void **) &l2_slice); @@ -2061,9 +2061,15 @@ zero_in_l2_slice(BlockDriverState *bs, uint64_t offset, QCow2ClusterType type = qcow2_get_cluster_type(bs, old_l2_entry); bool unmap = (type == QCOW2_CLUSTER_COMPRESSED) || ((flags & BDRV_REQ_MAY_UNMAP) && qcow2_cluster_is_allocated(type)); - uint64_t new_l2_entry = unmap ? 0 : old_l2_entry; + bool keep_reference = + (s->discard_no_unref && type != QCOW2_CLUSTER_COMPRESSED); + uint64_t new_l2_entry = old_l2_entry; uint64_t new_l2_bitmap = old_l2_bitmap; + if (unmap && !keep_reference) { + new_l2_entry = 0; + } + if (has_subclusters(s)) { new_l2_bitmap = QCOW_L2_BITMAP_ALL_ZEROES; } else { @@ -2081,9 +2087,17 @@ zero_in_l2_slice(BlockDriverState *bs, uint64_t offset, set_l2_bitmap(s, l2_slice, l2_index + i, new_l2_bitmap); } - /* Then decrease the refcount */ if (unmap) { - qcow2_free_any_cluster(bs, old_l2_entry, QCOW2_DISCARD_REQUEST); + if (!keep_reference) { + /* Then decrease the refcount */ + qcow2_free_any_cluster(bs, old_l2_entry, QCOW2_DISCARD_REQUEST); + } else if (s->discard_passthrough[QCOW2_DISCARD_REQUEST] && + (type == QCOW2_CLUSTER_NORMAL || + type == QCOW2_CLUSTER_ZERO_ALLOC)) { + /* If we keep the reference, pass on the discard still */ + bdrv_pdiscard(s->data_file, old_l2_entry & L2E_OFFSET_MASK, + s->cluster_size); + } } } diff --git a/block/vdi.c b/block/vdi.c index c647d72895..7cfd12b50d 100644 --- a/block/vdi.c +++ b/block/vdi.c @@ -239,7 +239,7 @@ static void vdi_header_to_le(VdiHeader *header) static void vdi_header_print(VdiHeader *header) { - char uuidstr[37]; + char uuidstr[UUID_STR_LEN]; QemuUUID uuid; logout("text %s", header->text); logout("signature 0x%08x\n", header->signature); diff --git a/docs/devel/index-api.rst b/docs/devel/index-api.rst index 539ad29c21..fe01b2b488 100644 --- a/docs/devel/index-api.rst +++ b/docs/devel/index-api.rst @@ -11,6 +11,7 @@ generated from in-code annotations to function prototypes. loads-stores memory modules + pci qom-api qdev-api ui diff --git a/docs/devel/pci.rst b/docs/devel/pci.rst new file mode 100644 index 0000000000..68739334f3 --- /dev/null +++ b/docs/devel/pci.rst @@ -0,0 +1,8 @@ +============= +PCI subsystem +============= + +API Reference +------------- + +.. kernel-doc:: include/hw/pci/pci.h diff --git a/docs/system/arm/vexpress.rst b/docs/system/arm/vexpress.rst index 3e3839e923..38f29c73e7 100644 --- a/docs/system/arm/vexpress.rst +++ b/docs/system/arm/vexpress.rst @@ -58,6 +58,9 @@ Other differences between the hardware and the QEMU model: ``vexpress-a15``, and have IRQs from 40 upwards. If a dtb is provided on the command line then QEMU will edit it to include suitable entries describing these transports for the guest. +- QEMU does not currently support either dynamic or static remapping + of the area of memory at address 0: it is always mapped to alias + the first flash bank Booting a Linux kernel ---------------------- diff --git a/host/include/loongarch64/host/atomic128-ldst.h b/host/include/loongarch64/host/atomic128-ldst.h new file mode 100644 index 0000000000..9a4a8f8b9e --- /dev/null +++ b/host/include/loongarch64/host/atomic128-ldst.h @@ -0,0 +1,52 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * Load/store for 128-bit atomic operations, LoongArch version. + * + * See docs/devel/atomics.rst for discussion about the guarantees each + * atomic primitive is meant to provide. + */ + +#ifndef LOONGARCH_ATOMIC128_LDST_H +#define LOONGARCH_ATOMIC128_LDST_H + +#include "host/cpuinfo.h" +#include "tcg/debug-assert.h" + +#define HAVE_ATOMIC128_RO likely(cpuinfo & CPUINFO_LSX) +#define HAVE_ATOMIC128_RW HAVE_ATOMIC128_RO + +/* + * As of gcc 13 and clang 16, there is no compiler support for LSX at all. + * Use inline assembly throughout. + */ + +static inline Int128 atomic16_read_ro(const Int128 *ptr) +{ + uint64_t l, h; + + tcg_debug_assert(HAVE_ATOMIC128_RO); + asm("vld $vr0, %2, 0\n\t" + "vpickve2gr.d %0, $vr0, 0\n\t" + "vpickve2gr.d %1, $vr0, 1" + : "=r"(l), "=r"(h) : "r"(ptr), "m"(*ptr) : "f0"); + + return int128_make128(l, h); +} + +static inline Int128 atomic16_read_rw(Int128 *ptr) +{ + return atomic16_read_ro(ptr); +} + +static inline void atomic16_set(Int128 *ptr, Int128 val) +{ + uint64_t l = int128_getlo(val), h = int128_gethi(val); + + tcg_debug_assert(HAVE_ATOMIC128_RW); + asm("vinsgr2vr.d $vr0, %1, 0\n\t" + "vinsgr2vr.d $vr0, %2, 1\n\t" + "vst $vr0, %3, 0" + : "=m"(*ptr) : "r"(l), "r"(h), "r"(ptr) : "f0"); +} + +#endif /* LOONGARCH_ATOMIC128_LDST_H */ diff --git a/host/include/loongarch64/host/cpuinfo.h b/host/include/loongarch64/host/cpuinfo.h new file mode 100644 index 0000000000..fab664a10b --- /dev/null +++ b/host/include/loongarch64/host/cpuinfo.h @@ -0,0 +1,21 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * Host specific cpu identification for LoongArch + */ + +#ifndef HOST_CPUINFO_H +#define HOST_CPUINFO_H + +#define CPUINFO_ALWAYS (1u << 0) /* so cpuinfo is nonzero */ +#define CPUINFO_LSX (1u << 1) + +/* Initialized with a constructor. */ +extern unsigned cpuinfo; + +/* + * We cannot rely on constructor ordering, so other constructors must + * use the function interface rather than the variable above. + */ +unsigned cpuinfo_init(void); + +#endif /* HOST_CPUINFO_H */ diff --git a/host/include/loongarch64/host/load-extract-al16-al8.h b/host/include/loongarch64/host/load-extract-al16-al8.h new file mode 100644 index 0000000000..d1fb59d8af --- /dev/null +++ b/host/include/loongarch64/host/load-extract-al16-al8.h @@ -0,0 +1,39 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * Atomic extract 64 from 128-bit, LoongArch version. + * + * Copyright (C) 2023 Linaro, Ltd. + */ + +#ifndef LOONGARCH_LOAD_EXTRACT_AL16_AL8_H +#define LOONGARCH_LOAD_EXTRACT_AL16_AL8_H + +#include "host/cpuinfo.h" +#include "tcg/debug-assert.h" + +/** + * load_atom_extract_al16_or_al8: + * @pv: host address + * @s: object size in bytes, @s <= 8. + * + * Load @s bytes from @pv, when pv % s != 0. If [p, p+s-1] does not + * cross an 16-byte boundary then the access must be 16-byte atomic, + * otherwise the access must be 8-byte atomic. + */ +static inline uint64_t load_atom_extract_al16_or_al8(void *pv, int s) +{ + uintptr_t pi = (uintptr_t)pv; + Int128 *ptr_align = (Int128 *)(pi & ~7); + int shr = (pi & 7) * 8; + uint64_t l, h; + + tcg_debug_assert(HAVE_ATOMIC128_RO); + asm("vld $vr0, %2, 0\n\t" + "vpickve2gr.d %0, $vr0, 0\n\t" + "vpickve2gr.d %1, $vr0, 1" + : "=r"(l), "=r"(h) : "r"(ptr_align), "m"(*ptr_align) : "f0"); + + return (l >> shr) | (h << (-shr & 63)); +} + +#endif /* LOONGARCH_LOAD_EXTRACT_AL16_AL8_H */ diff --git a/host/include/loongarch64/host/store-insert-al16.h b/host/include/loongarch64/host/store-insert-al16.h new file mode 100644 index 0000000000..919fd8d744 --- /dev/null +++ b/host/include/loongarch64/host/store-insert-al16.h @@ -0,0 +1,12 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * Atomic store insert into 128-bit, LoongArch version. + */ + +#ifndef LOONGARCH_STORE_INSERT_AL16_H +#define LOONGARCH_STORE_INSERT_AL16_H + +void store_atom_insert_al16(Int128 *ps, Int128 val, Int128 msk) + QEMU_ERROR("unsupported atomic"); + +#endif /* LOONGARCH_STORE_INSERT_AL16_H */ diff --git a/hw/alpha/typhoon.c b/hw/alpha/typhoon.c index 49a80550c5..e8711ae16a 100644 --- a/hw/alpha/typhoon.c +++ b/hw/alpha/typhoon.c @@ -738,6 +738,10 @@ static AddressSpace *typhoon_pci_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &s->pchip.iommu_as; } +static const PCIIOMMUOps typhoon_iommu_ops = { + .get_address_space = typhoon_pci_dma_iommu, +}; + static void typhoon_set_irq(void *opaque, int irq, int level) { TyphoonState *s = opaque; @@ -897,7 +901,7 @@ PCIBus *typhoon_init(MemoryRegion *ram, qemu_irq *p_isa_irq, "iommu-typhoon", UINT64_MAX); address_space_init(&s->pchip.iommu_as, MEMORY_REGION(&s->pchip.iommu), "pchip0-pci"); - pci_setup_iommu(b, typhoon_pci_dma_iommu, s); + pci_setup_iommu(b, &typhoon_iommu_ops, s); /* Pchip0 PCI special/interrupt acknowledge, 0x801.F800.0000, 64MB. */ memory_region_init_io(&s->pchip.reg_iack, OBJECT(s), &alpha_pci_iack_ops, diff --git a/hw/arm/smmu-common.c b/hw/arm/smmu-common.c index f35ae9aa22..9a8ac45431 100644 --- a/hw/arm/smmu-common.c +++ b/hw/arm/smmu-common.c @@ -605,6 +605,10 @@ static AddressSpace *smmu_find_add_as(PCIBus *bus, void *opaque, int devfn) return &sdev->as; } +static const PCIIOMMUOps smmu_ops = { + .get_address_space = smmu_find_add_as, +}; + IOMMUMemoryRegion *smmu_iommu_mr(SMMUState *s, uint32_t sid) { uint8_t bus_n, devfn; @@ -661,7 +665,7 @@ static void smmu_base_realize(DeviceState *dev, Error **errp) s->smmu_pcibus_by_busptr = g_hash_table_new(NULL, NULL); if (s->primary_bus) { - pci_setup_iommu(s->primary_bus, smmu_find_add_as, s); + pci_setup_iommu(s->primary_bus, &smmu_ops, s); } else { error_setg(errp, "SMMU is not attached to any PCI bus!"); } diff --git a/hw/arm/vexpress.c b/hw/arm/vexpress.c index 8ff37f52ca..c08ea34e92 100644 --- a/hw/arm/vexpress.c +++ b/hw/arm/vexpress.c @@ -177,7 +177,6 @@ struct VexpressMachineState { MemoryRegion vram; MemoryRegion sram; MemoryRegion flashalias; - MemoryRegion lowram; MemoryRegion a15sram; bool secure; bool virt; @@ -276,7 +275,6 @@ static void a9_daughterboard_init(VexpressMachineState *vms, { MachineState *machine = MACHINE(vms); MemoryRegion *sysmem = get_system_memory(); - ram_addr_t low_ram_size; if (ram_size > 0x40000000) { /* 1GB is the maximum the address space permits */ @@ -284,17 +282,11 @@ static void a9_daughterboard_init(VexpressMachineState *vms, exit(1); } - low_ram_size = ram_size; - if (low_ram_size > 0x4000000) { - low_ram_size = 0x4000000; - } - /* RAM is from 0x60000000 upwards. The bottom 64MB of the + /* + * RAM is from 0x60000000 upwards. The bottom 64MB of the * address space should in theory be remappable to various - * things including ROM or RAM; we always map the RAM there. + * things including ROM or RAM; we always map the flash there. */ - memory_region_init_alias(&vms->lowram, NULL, "vexpress.lowmem", - machine->ram, 0, low_ram_size); - memory_region_add_subregion(sysmem, 0x0, &vms->lowram); memory_region_add_subregion(sysmem, 0x60000000, machine->ram); /* 0x1e000000 A9MPCore (SCU) private memory region */ diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c index 9ce136cd88..8bc35a483c 100644 --- a/hw/arm/virt-acpi-build.c +++ b/hw/arm/virt-acpi-build.c @@ -482,7 +482,7 @@ build_spcr(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms) build_append_int_noprefix(table_data, 3, 1); /* ARM PL011 UART */ build_append_int_noprefix(table_data, 0, 3); /* Reserved */ /* Base Address */ - build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 8, 0, 1, + build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 32, 0, 3, vms->memmap[VIRT_UART].base); /* Interrupt Type */ build_append_int_noprefix(table_data, @@ -673,7 +673,7 @@ build_dbg2(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms) build_append_int_noprefix(table_data, 34, 2); /* BaseAddressRegister[] */ - build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 8, 0, 1, + build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 32, 0, 3, vms->memmap[VIRT_UART].base); /* AddressSize[] */ diff --git a/hw/arm/virt.c b/hw/arm/virt.c index 92085d2d8f..0a16ab3095 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -631,7 +631,8 @@ static void fdt_add_pmu_nodes(const VirtMachineState *vms) qemu_fdt_setprop(ms->fdt, "/pmu", "compatible", compat, sizeof(compat)); qemu_fdt_setprop_cells(ms->fdt, "/pmu", "interrupts", - GIC_FDT_IRQ_TYPE_PPI, VIRTUAL_PMU_IRQ, irqflags); + GIC_FDT_IRQ_TYPE_PPI, + INTID_TO_PPI(VIRTUAL_PMU_IRQ), irqflags); } } diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c index a07cd7eb5d..bfa53960c3 100644 --- a/hw/block/xen-block.c +++ b/hw/block/xen-block.c @@ -115,9 +115,13 @@ static void xen_block_connect(XenDevice *xendev, Error **errp) return; } - if (xen_device_frontend_scanf(xendev, "protocol", "%ms", - &str) != 1) { - protocol = BLKIF_PROTOCOL_NATIVE; + if (xen_device_frontend_scanf(xendev, "protocol", "%ms", &str) != 1) { + /* x86 defaults to the 32-bit protocol even for 64-bit guests. */ + if (object_dynamic_cast(OBJECT(qdev_get_machine()), "x86-machine")) { + protocol = BLKIF_PROTOCOL_X86_32; + } else { + protocol = BLKIF_PROTOCOL_NATIVE; + } } else { if (strcmp(str, XEN_IO_PROTO_ABI_X86_32) == 0) { protocol = BLKIF_PROTOCOL_X86_32; diff --git a/hw/core/loader.c b/hw/core/loader.c index 4dd5a71fb7..b7bb44b7f7 100644 --- a/hw/core/loader.c +++ b/hw/core/loader.c @@ -558,7 +558,7 @@ static void zfree(void *x, void *addr) ssize_t gunzip(void *dst, size_t dstlen, uint8_t *src, size_t srclen) { - z_stream s; + z_stream s = {}; ssize_t dstbytes; int r, i, flags; diff --git a/hw/core/machine-hmp-cmds.c b/hw/core/machine-hmp-cmds.c index 9a4b59c6f2..a6ff6a4875 100644 --- a/hw/core/machine-hmp-cmds.c +++ b/hw/core/machine-hmp-cmds.c @@ -253,6 +253,7 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict) MemoryDeviceInfo *value; PCDIMMDeviceInfo *di; SgxEPCDeviceInfo *se; + HvBalloonDeviceInfo *hi; for (info = info_list; info; info = info->next) { value = info->value; @@ -310,6 +311,20 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict) monitor_printf(mon, " node: %" PRId64 "\n", se->node); monitor_printf(mon, " memdev: %s\n", se->memdev); break; + case MEMORY_DEVICE_INFO_KIND_HV_BALLOON: + hi = value->u.hv_balloon.data; + monitor_printf(mon, "Memory device [%s]: \"%s\"\n", + MemoryDeviceInfoKind_str(value->type), + hi->id ? hi->id : ""); + if (hi->has_memaddr) { + monitor_printf(mon, " memaddr: 0x%" PRIx64 "\n", + hi->memaddr); + } + monitor_printf(mon, " max-size: %" PRIu64 "\n", hi->max_size); + if (hi->memdev) { + monitor_printf(mon, " memdev: %s\n", hi->memdev); + } + break; default: g_assert_not_reached(); } diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c index 2f1dbb3fd7..b46d16cd2c 100644 --- a/hw/core/qdev-properties-system.c +++ b/hw/core/qdev-properties-system.c @@ -705,7 +705,7 @@ static void get_reserved_region(Object *obj, Visitor *v, const char *name, int rc; rc = snprintf(buffer, sizeof(buffer), "0x%"PRIx64":0x%"PRIx64":%u", - rr->low, rr->high, rr->type); + range_lob(&rr->range), range_upb(&rr->range), rr->type); assert(rc < sizeof(buffer)); visit_type_str(v, name, &p, errp); @@ -717,6 +717,7 @@ static void set_reserved_region(Object *obj, Visitor *v, const char *name, Property *prop = opaque; ReservedRegion *rr = object_field_prop_ptr(obj, prop); const char *endptr; + uint64_t lob, upb; char *str; int ret; @@ -724,7 +725,7 @@ static void set_reserved_region(Object *obj, Visitor *v, const char *name, return; } - ret = qemu_strtou64(str, &endptr, 16, &rr->low); + ret = qemu_strtou64(str, &endptr, 16, &lob); if (ret) { error_setg(errp, "start address of '%s'" " must be a hexadecimal integer", name); @@ -734,7 +735,7 @@ static void set_reserved_region(Object *obj, Visitor *v, const char *name, goto separator_error; } - ret = qemu_strtou64(endptr + 1, &endptr, 16, &rr->high); + ret = qemu_strtou64(endptr + 1, &endptr, 16, &upb); if (ret) { error_setg(errp, "end address of '%s'" " must be a hexadecimal integer", name); @@ -744,6 +745,8 @@ static void set_reserved_region(Object *obj, Visitor *v, const char *name, goto separator_error; } + range_set_bounds(&rr->range, lob, upb); + ret = qemu_strtoui(endptr + 1, &endptr, 10, &rr->type); if (ret) { error_setg(errp, "type of '%s'" @@ -1111,7 +1114,7 @@ static void get_uuid(Object *obj, Visitor *v, const char *name, void *opaque, { Property *prop = opaque; QemuUUID *uuid = object_field_prop_ptr(obj, prop); - char buffer[UUID_FMT_LEN + 1]; + char buffer[UUID_STR_LEN]; char *p = buffer; qemu_uuid_unparse(uuid, buffer); diff --git a/hw/display/ati.c b/hw/display/ati.c index 6e38e00502..9a87a5504a 100644 --- a/hw/display/ati.c +++ b/hw/display/ati.c @@ -319,11 +319,13 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) case DAC_CNTL: val = s->regs.dac_cntl; break; - case GPIO_VGA_DDC: - val = s->regs.gpio_vga_ddc; + case GPIO_VGA_DDC ... GPIO_VGA_DDC + 3: + val = ati_reg_read_offs(s->regs.gpio_vga_ddc, + addr - GPIO_VGA_DDC, size); break; - case GPIO_DVI_DDC: - val = s->regs.gpio_dvi_ddc; + case GPIO_DVI_DDC ... GPIO_DVI_DDC + 3: + val = ati_reg_read_offs(s->regs.gpio_dvi_ddc, + addr - GPIO_DVI_DDC, size); break; case GPIO_MONID ... GPIO_MONID + 3: val = ati_reg_read_offs(s->regs.gpio_monid, @@ -337,6 +339,9 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) case PALETTE_DATA: val = vga_ioport_read(&s->vga, VGA_PEL_D); break; + case PALETTE_30_DATA: + val = s->regs.palette[vga_ioport_read(&s->vga, VGA_PEL_IR)]; + break; case CNFG_CNTL: val = s->regs.config_cntl; break; @@ -349,14 +354,17 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) PCI_BASE_ADDRESS_0, size) & 0xfffffff0; break; case CONFIG_APER_SIZE: - val = s->vga.vram_size; + val = s->vga.vram_size / 2; break; case CONFIG_REG_1_BASE: val = pci_default_read_config(&s->dev, PCI_BASE_ADDRESS_2, size) & 0xfffffff0; break; case CONFIG_REG_APER_SIZE: - val = memory_region_size(&s->mm); + val = memory_region_size(&s->mm) / 2; + break; + case HOST_PATH_CNTL: + val = BIT(23); /* Radeon HDP_APER_CNTL */ break; case MC_STATUS: val = 5; @@ -612,29 +620,34 @@ static void ati_mm_write(void *opaque, hwaddr addr, s->regs.dac_cntl = data & 0xffffe3ff; s->vga.dac_8bit = !!(data & DAC_8BIT_EN); break; - case GPIO_VGA_DDC: + /* + * GPIO regs for DDC access. Because some drivers access these via + * multiple byte writes we have to be careful when we send bits to + * avoid spurious changes in bitbang_i2c state. Only do it when either + * the enable bits are changed or output bits changed while enabled. + */ + case GPIO_VGA_DDC ... GPIO_VGA_DDC + 3: if (s->dev_id != PCI_DEVICE_ID_ATI_RAGE128_PF) { /* FIXME: Maybe add a property to select VGA or DVI port? */ } break; - case GPIO_DVI_DDC: + case GPIO_DVI_DDC ... GPIO_DVI_DDC + 3: if (s->dev_id != PCI_DEVICE_ID_ATI_RAGE128_PF) { - s->regs.gpio_dvi_ddc = ati_i2c(&s->bbi2c, data, 0); + ati_reg_write_offs(&s->regs.gpio_dvi_ddc, + addr - GPIO_DVI_DDC, data, size); + if ((addr <= GPIO_DVI_DDC + 2 && addr + size > GPIO_DVI_DDC + 2) || + (addr == GPIO_DVI_DDC && (s->regs.gpio_dvi_ddc & 0x30000))) { + s->regs.gpio_dvi_ddc = ati_i2c(&s->bbi2c, + s->regs.gpio_dvi_ddc, 0); + } } break; case GPIO_MONID ... GPIO_MONID + 3: /* FIXME What does Radeon have here? */ if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) { + /* Rage128p accesses DDC via MONID(1-2) with additional mask bit */ ati_reg_write_offs(&s->regs.gpio_monid, addr - GPIO_MONID, data, size); - /* - * Rage128p accesses DDC used to get EDID via these bits. - * Because some drivers access this via multiple byte writes - * we have to be careful when we send bits to avoid spurious - * changes in bitbang_i2c state. So only do it when mask is set - * and either the enable bits are changed or output bits changed - * while enabled. - */ if ((s->regs.gpio_monid & BIT(25)) && ((addr <= GPIO_MONID + 2 && addr + size > GPIO_MONID + 2) || (addr == GPIO_MONID && (s->regs.gpio_monid & 0x60000)))) { @@ -663,6 +676,12 @@ static void ati_mm_write(void *opaque, hwaddr addr, data >>= 8; vga_ioport_write(&s->vga, VGA_PEL_D, data & 0xff); break; + case PALETTE_30_DATA: + s->regs.palette[vga_ioport_read(&s->vga, VGA_PEL_IW)] = data; + vga_ioport_write(&s->vga, VGA_PEL_D, (data >> 22) & 0xff); + vga_ioport_write(&s->vga, VGA_PEL_D, (data >> 12) & 0xff); + vga_ioport_write(&s->vga, VGA_PEL_D, (data >> 2) & 0xff); + break; case CNFG_CNTL: s->regs.config_cntl = data; break; @@ -1014,6 +1033,7 @@ static Property ati_vga_properties[] = { DEFINE_PROP_UINT16("x-device-id", ATIVGAState, dev_id, PCI_DEVICE_ID_ATI_RAGE128_PF), DEFINE_PROP_BOOL("guest_hwcursor", ATIVGAState, cursor_guest_mode, false), + DEFINE_PROP_UINT8("x-pixman", ATIVGAState, use_pixman, 3), DEFINE_PROP_END_OF_LIST() }; @@ -1035,11 +1055,18 @@ static void ati_vga_class_init(ObjectClass *klass, void *data) k->exit = ati_vga_exit; } +static void ati_vga_init(Object *o) +{ + object_property_set_description(o, "x-pixman", "Use pixman for: " + "1: fill, 2: blit"); +} + static const TypeInfo ati_vga_info = { .name = TYPE_ATI_VGA, .parent = TYPE_PCI_DEVICE, .instance_size = sizeof(ATIVGAState), .class_init = ati_vga_class_init, + .instance_init = ati_vga_init, .interfaces = (InterfaceInfo[]) { { INTERFACE_CONVENTIONAL_PCI_DEVICE }, { }, diff --git a/hw/display/ati_2d.c b/hw/display/ati_2d.c index 7d786653e8..0e6b8e4367 100644 --- a/hw/display/ati_2d.c +++ b/hw/display/ati_2d.c @@ -92,6 +92,7 @@ void ati_2d_blt(ATIVGAState *s) switch (s->regs.dp_mix & GMC_ROP3_MASK) { case ROP3_SRCCOPY: { + bool fallback = false; unsigned src_x = (s->regs.dp_cntl & DST_X_LEFT_TO_RIGHT ? s->regs.src_x : s->regs.src_x + 1 - s->regs.dst_width); unsigned src_y = (s->regs.dp_cntl & DST_Y_TOP_TO_BOTTOM ? @@ -122,27 +123,50 @@ void ati_2d_blt(ATIVGAState *s) src_bits, dst_bits, src_stride, dst_stride, bpp, bpp, src_x, src_y, dst_x, dst_y, s->regs.dst_width, s->regs.dst_height); - if (s->regs.dp_cntl & DST_X_LEFT_TO_RIGHT && + if ((s->use_pixman & BIT(1)) && + s->regs.dp_cntl & DST_X_LEFT_TO_RIGHT && s->regs.dp_cntl & DST_Y_TOP_TO_BOTTOM) { - pixman_blt((uint32_t *)src_bits, (uint32_t *)dst_bits, - src_stride, dst_stride, bpp, bpp, - src_x, src_y, dst_x, dst_y, - s->regs.dst_width, s->regs.dst_height); - } else { + fallback = !pixman_blt((uint32_t *)src_bits, (uint32_t *)dst_bits, + src_stride, dst_stride, bpp, bpp, + src_x, src_y, dst_x, dst_y, + s->regs.dst_width, s->regs.dst_height); + } else if (s->use_pixman & BIT(1)) { /* FIXME: We only really need a temporary if src and dst overlap */ int llb = s->regs.dst_width * (bpp / 8); int tmp_stride = DIV_ROUND_UP(llb, sizeof(uint32_t)); uint32_t *tmp = g_malloc(tmp_stride * sizeof(uint32_t) * s->regs.dst_height); - pixman_blt((uint32_t *)src_bits, tmp, - src_stride, tmp_stride, bpp, bpp, - src_x, src_y, 0, 0, - s->regs.dst_width, s->regs.dst_height); - pixman_blt(tmp, (uint32_t *)dst_bits, - tmp_stride, dst_stride, bpp, bpp, - 0, 0, dst_x, dst_y, - s->regs.dst_width, s->regs.dst_height); + fallback = !pixman_blt((uint32_t *)src_bits, tmp, + src_stride, tmp_stride, bpp, bpp, + src_x, src_y, 0, 0, + s->regs.dst_width, s->regs.dst_height); + if (!fallback) { + fallback = !pixman_blt(tmp, (uint32_t *)dst_bits, + tmp_stride, dst_stride, bpp, bpp, + 0, 0, dst_x, dst_y, + s->regs.dst_width, s->regs.dst_height); + } g_free(tmp); + } else { + fallback = true; + } + if (fallback) { + unsigned int y, i, j, bypp = bpp / 8; + unsigned int src_pitch = src_stride * sizeof(uint32_t); + unsigned int dst_pitch = dst_stride * sizeof(uint32_t); + + for (y = 0; y < s->regs.dst_height; y++) { + i = dst_x * bypp; + j = src_x * bypp; + if (s->regs.dp_cntl & DST_Y_TOP_TO_BOTTOM) { + i += (dst_y + y) * dst_pitch; + j += (src_y + y) * src_pitch; + } else { + i += (dst_y + s->regs.dst_height - 1 - y) * dst_pitch; + j += (src_y + s->regs.dst_height - 1 - y) * src_pitch; + } + memmove(&dst_bits[i], &src_bits[j], s->regs.dst_width * bypp); + } } if (dst_bits >= s->vga.vram_ptr + s->vga.vbe_start_addr && dst_bits < s->vga.vram_ptr + s->vga.vbe_start_addr + @@ -180,14 +204,21 @@ void ati_2d_blt(ATIVGAState *s) dst_stride /= sizeof(uint32_t); DPRINTF("pixman_fill(%p, %d, %d, %d, %d, %d, %d, %x)\n", - dst_bits, dst_stride, bpp, - dst_x, dst_y, - s->regs.dst_width, s->regs.dst_height, - filler); - pixman_fill((uint32_t *)dst_bits, dst_stride, bpp, - dst_x, dst_y, - s->regs.dst_width, s->regs.dst_height, - filler); + dst_bits, dst_stride, bpp, dst_x, dst_y, + s->regs.dst_width, s->regs.dst_height, filler); + if (!(s->use_pixman & BIT(0)) || + !pixman_fill((uint32_t *)dst_bits, dst_stride, bpp, dst_x, dst_y, + s->regs.dst_width, s->regs.dst_height, filler)) { + /* fallback when pixman failed or we don't want to call it */ + unsigned int x, y, i, bypp = bpp / 8; + unsigned int dst_pitch = dst_stride * sizeof(uint32_t); + for (y = 0; y < s->regs.dst_height; y++) { + i = dst_x * bypp + (dst_y + y) * dst_pitch; + for (x = 0; x < s->regs.dst_width; x++, i += bypp) { + stn_he_p(&dst_bits[i], bypp, filler); + } + } + } if (dst_bits >= s->vga.vram_ptr + s->vga.vbe_start_addr && dst_bits < s->vga.vram_ptr + s->vga.vbe_start_addr + s->vga.vbe_regs[VBE_DISPI_INDEX_YRES] * s->vga.vbe_line_offset) { diff --git a/hw/display/ati_dbg.c b/hw/display/ati_dbg.c index bd0ecd48c7..3ffa7f35df 100644 --- a/hw/display/ati_dbg.c +++ b/hw/display/ati_dbg.c @@ -30,6 +30,7 @@ static struct ati_regdesc ati_reg_names[] = { {"AMCGPIO_EN_MIR", 0x00a8}, {"PALETTE_INDEX", 0x00b0}, {"PALETTE_DATA", 0x00b4}, + {"PALETTE_30_DATA", 0x00b8}, {"CNFG_CNTL", 0x00e0}, {"GEN_RESET_CNTL", 0x00f0}, {"CNFG_MEMSIZE", 0x00f8}, @@ -38,6 +39,7 @@ static struct ati_regdesc ati_reg_names[] = { {"CONFIG_APER_SIZE", 0x0108}, {"CONFIG_REG_1_BASE", 0x010c}, {"CONFIG_REG_APER_SIZE", 0x0110}, + {"HOST_PATH_CNTL", 0x0130}, {"MEM_CNTL", 0x0140}, {"MC_FB_LOCATION", 0x0148}, {"MC_AGP_LOCATION", 0x014C}, diff --git a/hw/display/ati_int.h b/hw/display/ati_int.h index e8d3c7af75..f5a47b82b0 100644 --- a/hw/display/ati_int.h +++ b/hw/display/ati_int.h @@ -44,6 +44,7 @@ typedef struct ATIVGARegs { uint32_t gpio_dvi_ddc; uint32_t gpio_monid; uint32_t config_cntl; + uint32_t palette[256]; uint32_t crtc_h_total_disp; uint32_t crtc_h_sync_strt_wid; uint32_t crtc_v_total_disp; @@ -89,6 +90,7 @@ struct ATIVGAState { char *model; uint16_t dev_id; uint8_t mode; + uint8_t use_pixman; bool cursor_guest_mode; uint16_t cursor_size; uint32_t cursor_offset; diff --git a/hw/display/ati_regs.h b/hw/display/ati_regs.h index d6282b2ef2..d7127748ff 100644 --- a/hw/display/ati_regs.h +++ b/hw/display/ati_regs.h @@ -48,6 +48,7 @@ #define AMCGPIO_EN_MIR 0x00a8 #define PALETTE_INDEX 0x00b0 #define PALETTE_DATA 0x00b4 +#define PALETTE_30_DATA 0x00b8 #define CNFG_CNTL 0x00e0 #define GEN_RESET_CNTL 0x00f0 #define CNFG_MEMSIZE 0x00f8 @@ -56,6 +57,7 @@ #define CONFIG_APER_SIZE 0x0108 #define CONFIG_REG_1_BASE 0x010c #define CONFIG_REG_APER_SIZE 0x0110 +#define HOST_PATH_CNTL 0x0130 #define MEM_CNTL 0x0140 #define MC_FB_LOCATION 0x0148 #define MC_AGP_LOCATION 0x014C diff --git a/hw/display/macfb.c b/hw/display/macfb.c index 2f8e016566..d61541ccb5 100644 --- a/hw/display/macfb.c +++ b/hw/display/macfb.c @@ -36,8 +36,8 @@ #define DAFB_INTR_MASK 0x104 #define DAFB_INTR_STAT 0x108 #define DAFB_INTR_CLEAR 0x10c -#define DAFB_RESET 0x200 -#define DAFB_LUT 0x213 +#define DAFB_LUT_INDEX 0x200 +#define DAFB_LUT 0x210 #define DAFB_INTR_VBL 0x4 @@ -537,6 +537,11 @@ static uint64_t macfb_ctrl_read(void *opaque, case DAFB_MODE_SENSE: val = macfb_sense_read(s); break; + case DAFB_LUT ... DAFB_LUT + 3: + val = s->color_palette[s->palette_current]; + s->palette_current = (s->palette_current + 1) % + ARRAY_SIZE(s->color_palette); + break; default: if (addr < MACFB_CTRL_TOPADDR) { val = s->regs[addr >> 2]; @@ -583,13 +588,11 @@ static void macfb_ctrl_write(void *opaque, s->regs[DAFB_INTR_STAT >> 2] &= ~DAFB_INTR_VBL; macfb_update_irq(s); break; - case DAFB_RESET: - s->palette_current = 0; - s->regs[DAFB_INTR_STAT >> 2] &= ~DAFB_INTR_VBL; - macfb_update_irq(s); + case DAFB_LUT_INDEX: + s->palette_current = (val & 0xff) * 3; break; - case DAFB_LUT: - s->color_palette[s->palette_current] = val; + case DAFB_LUT ... DAFB_LUT + 3: + s->color_palette[s->palette_current] = val & 0xff; s->palette_current = (s->palette_current + 1) % ARRAY_SIZE(s->color_palette); if (s->palette_current % 3) { diff --git a/hw/display/virtio-gpu-pci-rutabaga.c b/hw/display/virtio-gpu-pci-rutabaga.c index c96729e198..abbb898c65 100644 --- a/hw/display/virtio-gpu-pci-rutabaga.c +++ b/hw/display/virtio-gpu-pci-rutabaga.c @@ -36,6 +36,7 @@ static const TypeInfo virtio_gpu_rutabaga_pci_info[] = { .instance_init = virtio_gpu_rutabaga_initfn, .interfaces = (InterfaceInfo[]) { { INTERFACE_CONVENTIONAL_PCI_DEVICE }, + { }, } }, }; diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c index 4265316cbb..2707bceea8 100644 --- a/hw/display/virtio-gpu.c +++ b/hw/display/virtio-gpu.c @@ -1213,6 +1213,9 @@ static int virtio_gpu_save(QEMUFile *f, void *opaque, size_t size, assert(QTAILQ_EMPTY(&g->cmdq)); QTAILQ_FOREACH(res, &g->reslist, next) { + if (res->blob_size) { + continue; + } qemu_put_be32(f, res->resource_id); qemu_put_be32(f, res->width); qemu_put_be32(f, res->height); @@ -1230,12 +1233,40 @@ static int virtio_gpu_save(QEMUFile *f, void *opaque, size_t size, return vmstate_save_state(f, &vmstate_virtio_gpu_scanouts, g, NULL); } +static bool virtio_gpu_load_restore_mapping(VirtIOGPU *g, + struct virtio_gpu_simple_resource *res) +{ + int i; + + for (i = 0; i < res->iov_cnt; i++) { + hwaddr len = res->iov[i].iov_len; + res->iov[i].iov_base = + dma_memory_map(VIRTIO_DEVICE(g)->dma_as, res->addrs[i], &len, + DMA_DIRECTION_TO_DEVICE, MEMTXATTRS_UNSPECIFIED); + + if (!res->iov[i].iov_base || len != res->iov[i].iov_len) { + /* Clean up the half-a-mapping we just created... */ + if (res->iov[i].iov_base) { + dma_memory_unmap(VIRTIO_DEVICE(g)->dma_as, res->iov[i].iov_base, + len, DMA_DIRECTION_TO_DEVICE, 0); + } + /* ...and the mappings for previous loop iterations */ + res->iov_cnt = i; + virtio_gpu_cleanup_mapping(g, res); + return false; + } + } + + QTAILQ_INSERT_HEAD(&g->reslist, res, next); + g->hostmem += res->hostmem; + return true; +} + static int virtio_gpu_load(QEMUFile *f, void *opaque, size_t size, const VMStateField *field) { VirtIOGPU *g = opaque; struct virtio_gpu_simple_resource *res; - struct virtio_gpu_scanout *scanout; uint32_t resource_id, pformat; void *bits = NULL; int i; @@ -1294,40 +1325,96 @@ static int virtio_gpu_load(QEMUFile *f, void *opaque, size_t size, qemu_get_buffer(f, (void *)pixman_image_get_data(res->image), pixman_image_get_stride(res->image) * res->height); - /* restore mapping */ - for (i = 0; i < res->iov_cnt; i++) { - hwaddr len = res->iov[i].iov_len; - res->iov[i].iov_base = - dma_memory_map(VIRTIO_DEVICE(g)->dma_as, res->addrs[i], &len, - DMA_DIRECTION_TO_DEVICE, - MEMTXATTRS_UNSPECIFIED); - - if (!res->iov[i].iov_base || len != res->iov[i].iov_len) { - /* Clean up the half-a-mapping we just created... */ - if (res->iov[i].iov_base) { - dma_memory_unmap(VIRTIO_DEVICE(g)->dma_as, - res->iov[i].iov_base, - len, - DMA_DIRECTION_TO_DEVICE, - 0); - } - /* ...and the mappings for previous loop iterations */ - res->iov_cnt = i; - virtio_gpu_cleanup_mapping(g, res); - pixman_image_unref(res->image); - g_free(res); - return -EINVAL; - } + if (!virtio_gpu_load_restore_mapping(g, res)) { + pixman_image_unref(res->image); + g_free(res); + return -EINVAL; } - QTAILQ_INSERT_HEAD(&g->reslist, res, next); - g->hostmem += res->hostmem; - resource_id = qemu_get_be32(f); } /* load & apply scanout state */ vmstate_load_state(f, &vmstate_virtio_gpu_scanouts, g, 1); + + return 0; +} + +static int virtio_gpu_blob_save(QEMUFile *f, void *opaque, size_t size, + const VMStateField *field, JSONWriter *vmdesc) +{ + VirtIOGPU *g = opaque; + struct virtio_gpu_simple_resource *res; + int i; + + /* in 2d mode we should never find unprocessed commands here */ + assert(QTAILQ_EMPTY(&g->cmdq)); + + QTAILQ_FOREACH(res, &g->reslist, next) { + if (!res->blob_size) { + continue; + } + qemu_put_be32(f, res->resource_id); + qemu_put_be32(f, res->blob_size); + qemu_put_be32(f, res->iov_cnt); + for (i = 0; i < res->iov_cnt; i++) { + qemu_put_be64(f, res->addrs[i]); + qemu_put_be32(f, res->iov[i].iov_len); + } + } + qemu_put_be32(f, 0); /* end of list */ + + return 0; +} + +static int virtio_gpu_blob_load(QEMUFile *f, void *opaque, size_t size, + const VMStateField *field) +{ + VirtIOGPU *g = opaque; + struct virtio_gpu_simple_resource *res; + uint32_t resource_id; + int i; + + resource_id = qemu_get_be32(f); + while (resource_id != 0) { + res = virtio_gpu_find_resource(g, resource_id); + if (res) { + return -EINVAL; + } + + res = g_new0(struct virtio_gpu_simple_resource, 1); + res->resource_id = resource_id; + res->blob_size = qemu_get_be32(f); + res->iov_cnt = qemu_get_be32(f); + res->addrs = g_new(uint64_t, res->iov_cnt); + res->iov = g_new(struct iovec, res->iov_cnt); + + /* read data */ + for (i = 0; i < res->iov_cnt; i++) { + res->addrs[i] = qemu_get_be64(f); + res->iov[i].iov_len = qemu_get_be32(f); + } + + if (!virtio_gpu_load_restore_mapping(g, res)) { + g_free(res); + return -EINVAL; + } + + virtio_gpu_init_udmabuf(res); + + resource_id = qemu_get_be32(f); + } + + return 0; +} + +static int virtio_gpu_post_load(void *opaque, int version_id) +{ + VirtIOGPU *g = opaque; + struct virtio_gpu_scanout *scanout; + struct virtio_gpu_simple_resource *res; + int i; + for (i = 0; i < g->parent_obj.conf.max_outputs; i++) { /* FIXME: should take scanout.r.{x,y} into account */ scanout = &g->parent_obj.scanout[i]; @@ -1475,6 +1562,32 @@ virtio_gpu_set_config(VirtIODevice *vdev, const uint8_t *config) } } +static bool virtio_gpu_blob_state_needed(void *opaque) +{ + VirtIOGPU *g = VIRTIO_GPU(opaque); + + return virtio_gpu_blob_enabled(g->parent_obj.conf); +} + +const VMStateDescription vmstate_virtio_gpu_blob_state = { + .name = "virtio-gpu/blob", + .minimum_version_id = VIRTIO_GPU_VM_VERSION, + .version_id = VIRTIO_GPU_VM_VERSION, + .needed = virtio_gpu_blob_state_needed, + .fields = (const VMStateField[]){ + { + .name = "virtio-gpu/blob", + .info = &(const VMStateInfo) { + .name = "blob", + .get = virtio_gpu_blob_load, + .put = virtio_gpu_blob_save, + }, + .flags = VMS_SINGLE, + } /* device */, + VMSTATE_END_OF_LIST() + }, +}; + /* * For historical reasons virtio_gpu does not adhere to virtio migration * scheme as described in doc/virtio-migration.txt, in a sense that no @@ -1500,6 +1613,11 @@ static const VMStateDescription vmstate_virtio_gpu = { } /* device */, VMSTATE_END_OF_LIST() }, + .subsections = (const VMStateDescription * []) { + &vmstate_virtio_gpu_blob_state, + NULL + }, + .post_load = virtio_gpu_post_load, }; static Property virtio_gpu_properties[] = { diff --git a/hw/hyperv/Kconfig b/hw/hyperv/Kconfig index fcf65903bd..41dd827c84 100644 --- a/hw/hyperv/Kconfig +++ b/hw/hyperv/Kconfig @@ -16,3 +16,13 @@ config SYNDBG bool default y depends on VMBUS + +config HV_BALLOON_SUPPORTED + bool + +config HV_BALLOON + bool + default y + depends on VMBUS + depends on HV_BALLOON_POSSIBLE + depends on HV_BALLOON_SUPPORTED diff --git a/hw/hyperv/hv-balloon-internal.h b/hw/hyperv/hv-balloon-internal.h new file mode 100644 index 0000000000..164c2e5825 --- /dev/null +++ b/hw/hyperv/hv-balloon-internal.h @@ -0,0 +1,33 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_HYPERV_HV_BALLOON_INTERNAL_H +#define HW_HYPERV_HV_BALLOON_INTERNAL_H + +#include "qemu/osdep.h" + +#define HV_BALLOON_PFN_SHIFT 12 +#define HV_BALLOON_PAGE_SIZE (1 << HV_BALLOON_PFN_SHIFT) + +#define SUM_OVERFLOW_U64(in1, in2) ((in1) > UINT64_MAX - (in2)) +#define SUM_SATURATE_U64(in1, in2) \ + ({ \ + uint64_t _in1 = (in1), _in2 = (in2); \ + uint64_t _result; \ + \ + if (!SUM_OVERFLOW_U64(_in1, _in2)) { \ + _result = _in1 + _in2; \ + } else { \ + _result = UINT64_MAX; \ + } \ + \ + _result; \ + }) + +#endif diff --git a/hw/hyperv/hv-balloon-our_range_memslots.c b/hw/hyperv/hv-balloon-our_range_memslots.c new file mode 100644 index 0000000000..99bae870f3 --- /dev/null +++ b/hw/hyperv/hv-balloon-our_range_memslots.c @@ -0,0 +1,201 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "hv-balloon-internal.h" +#include "hv-balloon-our_range_memslots.h" +#include "trace.h" + +/* OurRange */ +static void our_range_init(OurRange *our_range, uint64_t start, uint64_t count) +{ + assert(count <= UINT64_MAX - start); + our_range->range.start = start; + our_range->range.count = count; + + hvb_page_range_tree_init(&our_range->removed_guest); + hvb_page_range_tree_init(&our_range->removed_both); + + /* mark the whole range as unused but for potential use */ + our_range->added = 0; + our_range->unusable_tail = 0; +} + +static void our_range_destroy(OurRange *our_range) +{ + hvb_page_range_tree_destroy(&our_range->removed_guest); + hvb_page_range_tree_destroy(&our_range->removed_both); +} + +void hvb_our_range_clear_removed_trees(OurRange *our_range) +{ + hvb_page_range_tree_destroy(&our_range->removed_guest); + hvb_page_range_tree_destroy(&our_range->removed_both); + hvb_page_range_tree_init(&our_range->removed_guest); + hvb_page_range_tree_init(&our_range->removed_both); +} + +void hvb_our_range_mark_added(OurRange *our_range, uint64_t additional_size) +{ + assert(additional_size <= UINT64_MAX - our_range->added); + + our_range->added += additional_size; + + assert(our_range->added <= UINT64_MAX - our_range->unusable_tail); + assert(our_range->added + our_range->unusable_tail <= + our_range->range.count); +} + +/* OurRangeMemslots */ +static void our_range_memslots_init_slots(OurRangeMemslots *our_range, + MemoryRegion *backing_mr, + Object *memslot_owner) +{ + OurRangeMemslotsSlots *memslots = &our_range->slots; + unsigned int idx; + uint64_t memslot_offset; + + assert(memslots->count > 0); + memslots->slots = g_new0(MemoryRegion, memslots->count); + + /* Initialize our memslots, but don't map them yet. */ + assert(memslots->size_each > 0); + for (idx = 0, memslot_offset = 0; idx < memslots->count; + idx++, memslot_offset += memslots->size_each) { + uint64_t memslot_size; + g_autofree char *name = NULL; + + /* The size of the last memslot might be smaller. */ + if (idx == memslots->count - 1) { + uint64_t region_size; + + assert(our_range->mr); + region_size = memory_region_size(our_range->mr); + memslot_size = region_size - memslot_offset; + } else { + memslot_size = memslots->size_each; + } + + name = g_strdup_printf("memslot-%u", idx); + memory_region_init_alias(&memslots->slots[idx], memslot_owner, name, + backing_mr, memslot_offset, memslot_size); + /* + * We want to be able to atomically and efficiently activate/deactivate + * individual memslots without affecting adjacent memslots in memory + * notifiers. + */ + memory_region_set_unmergeable(&memslots->slots[idx], true); + } + + memslots->mapped_count = 0; +} + +OurRangeMemslots *hvb_our_range_memslots_new(uint64_t addr, + MemoryRegion *parent_mr, + MemoryRegion *backing_mr, + Object *memslot_owner, + unsigned int memslot_count, + uint64_t memslot_size) +{ + OurRangeMemslots *our_range; + + our_range = g_malloc(sizeof(*our_range)); + our_range_init(&our_range->range, + addr / HV_BALLOON_PAGE_SIZE, + memory_region_size(parent_mr) / HV_BALLOON_PAGE_SIZE); + our_range->slots.size_each = memslot_size; + our_range->slots.count = memslot_count; + our_range->mr = parent_mr; + our_range_memslots_init_slots(our_range, backing_mr, memslot_owner); + + return our_range; +} + +static void our_range_memslots_free_memslots(OurRangeMemslots *our_range) +{ + OurRangeMemslotsSlots *memslots = &our_range->slots; + unsigned int idx; + uint64_t offset; + + memory_region_transaction_begin(); + for (idx = 0, offset = 0; idx < memslots->mapped_count; + idx++, offset += memslots->size_each) { + trace_hv_balloon_unmap_slot(idx, memslots->count, offset); + assert(memory_region_is_mapped(&memslots->slots[idx])); + memory_region_del_subregion(our_range->mr, &memslots->slots[idx]); + } + memory_region_transaction_commit(); + + for (idx = 0; idx < memslots->count; idx++) { + object_unparent(OBJECT(&memslots->slots[idx])); + } + + g_clear_pointer(&our_range->slots.slots, g_free); +} + +void hvb_our_range_memslots_free(OurRangeMemslots *our_range) +{ + OurRangeMemslotsSlots *memslots = &our_range->slots; + MemoryRegion *hostmem_mr; + RAMBlock *rb; + + assert(our_range->slots.count > 0); + assert(our_range->slots.slots); + + hostmem_mr = memslots->slots[0].alias; + rb = hostmem_mr->ram_block; + ram_block_discard_range(rb, 0, qemu_ram_get_used_length(rb)); + + our_range_memslots_free_memslots(our_range); + our_range_destroy(&our_range->range); + g_free(our_range); +} + +void hvb_our_range_memslots_ensure_mapped_additional(OurRangeMemslots *our_range, + uint64_t additional_map_size) +{ + OurRangeMemslotsSlots *memslots = &our_range->slots; + uint64_t total_map_size; + unsigned int idx; + uint64_t offset; + + total_map_size = (our_range->range.added + additional_map_size) * + HV_BALLOON_PAGE_SIZE; + idx = memslots->mapped_count; + assert(memslots->size_each > 0); + offset = idx * memslots->size_each; + + /* + * Activate all memslots covered by the newly added region in a single + * transaction. + */ + memory_region_transaction_begin(); + for ( ; idx < memslots->count; + idx++, offset += memslots->size_each) { + /* + * If this memslot starts beyond or at the end of the range to map so + * does every next one. + */ + if (offset >= total_map_size) { + break; + } + + /* + * Instead of enabling/disabling memslot, we add/remove them. This + * should make address space updates faster, because we don't have to + * loop over many disabled subregions. + */ + trace_hv_balloon_map_slot(idx, memslots->count, offset); + assert(!memory_region_is_mapped(&memslots->slots[idx])); + memory_region_add_subregion(our_range->mr, offset, + &memslots->slots[idx]); + + memslots->mapped_count++; + } + memory_region_transaction_commit(); +} diff --git a/hw/hyperv/hv-balloon-our_range_memslots.h b/hw/hyperv/hv-balloon-our_range_memslots.h new file mode 100644 index 0000000000..b6f592d34b --- /dev/null +++ b/hw/hyperv/hv-balloon-our_range_memslots.h @@ -0,0 +1,110 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_HYPERV_HV_BALLOON_OUR_RANGE_MEMSLOTS_H +#define HW_HYPERV_HV_BALLOON_OUR_RANGE_MEMSLOTS_H + +#include "qemu/osdep.h" + +#include "exec/memory.h" +#include "qom/object.h" +#include "hv-balloon-page_range_tree.h" + +/* OurRange */ +#define OUR_RANGE(ptr) ((OurRange *)(ptr)) + +/* "our range" means the memory range owned by this driver (for hot-adding) */ +typedef struct OurRange { + PageRange range; + + /* How many pages were hot-added to the guest */ + uint64_t added; + + /* Pages at the end not currently usable */ + uint64_t unusable_tail; + + /* Memory removed from the guest */ + PageRangeTree removed_guest, removed_both; +} OurRange; + +static inline uint64_t our_range_get_remaining_start(OurRange *our_range) +{ + return our_range->range.start + our_range->added; +} + +static inline uint64_t our_range_get_remaining_size(OurRange *our_range) +{ + return our_range->range.count - our_range->added - our_range->unusable_tail; +} + +void hvb_our_range_mark_added(OurRange *our_range, uint64_t additional_size); + +static inline void our_range_mark_remaining_unusable(OurRange *our_range) +{ + our_range->unusable_tail = our_range->range.count - our_range->added; +} + +static inline PageRangeTree our_range_get_removed_tree(OurRange *our_range, + bool both) +{ + if (both) { + return our_range->removed_both; + } else { + return our_range->removed_guest; + } +} + +static inline bool our_range_is_removed_tree_empty(OurRange *our_range, + bool both) +{ + if (both) { + return page_range_tree_is_empty(our_range->removed_both); + } else { + return page_range_tree_is_empty(our_range->removed_guest); + } +} + +void hvb_our_range_clear_removed_trees(OurRange *our_range); + +/* OurRangeMemslots */ +typedef struct OurRangeMemslotsSlots { + /* Nominal size of each memslot (the last one might be smaller) */ + uint64_t size_each; + + /* Slots array and its element count */ + MemoryRegion *slots; + unsigned int count; + + /* How many slots are currently mapped */ + unsigned int mapped_count; +} OurRangeMemslotsSlots; + +typedef struct OurRangeMemslots { + OurRange range; + + /* Memslots covering our range */ + OurRangeMemslotsSlots slots; + + MemoryRegion *mr; +} OurRangeMemslots; + +OurRangeMemslots *hvb_our_range_memslots_new(uint64_t addr, + MemoryRegion *parent_mr, + MemoryRegion *backing_mr, + Object *memslot_owner, + unsigned int memslot_count, + uint64_t memslot_size); +void hvb_our_range_memslots_free(OurRangeMemslots *our_range); + +G_DEFINE_AUTOPTR_CLEANUP_FUNC(OurRangeMemslots, hvb_our_range_memslots_free) + +void hvb_our_range_memslots_ensure_mapped_additional(OurRangeMemslots *our_range, + uint64_t additional_map_size); + +#endif diff --git a/hw/hyperv/hv-balloon-page_range_tree.c b/hw/hyperv/hv-balloon-page_range_tree.c new file mode 100644 index 0000000000..e178d8b413 --- /dev/null +++ b/hw/hyperv/hv-balloon-page_range_tree.c @@ -0,0 +1,228 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "hv-balloon-internal.h" +#include "hv-balloon-page_range_tree.h" + +/* + * temporarily avoid warnings about enhanced GTree API usage requiring a + * too recent Glib version until GLIB_VERSION_MAX_ALLOWED finally reaches + * the Glib version with this API + */ +#pragma GCC diagnostic ignored "-Wdeprecated-declarations" + +/* PageRangeTree */ +static gint page_range_tree_key_compare(gconstpointer leftp, + gconstpointer rightp, + gpointer user_data) +{ + const uint64_t *left = leftp, *right = rightp; + + if (*left < *right) { + return -1; + } else if (*left > *right) { + return 1; + } else { /* *left == *right */ + return 0; + } +} + +static GTreeNode *page_range_tree_insert_new(PageRangeTree tree, + uint64_t start, uint64_t count) +{ + uint64_t *key = g_malloc(sizeof(*key)); + PageRange *range = g_malloc(sizeof(*range)); + + assert(count > 0); + + *key = range->start = start; + range->count = count; + + return g_tree_insert_node(tree.t, key, range); +} + +void hvb_page_range_tree_insert(PageRangeTree tree, + uint64_t start, uint64_t count, + uint64_t *dupcount) +{ + GTreeNode *node; + bool joinable; + uint64_t intersection; + PageRange *range; + + assert(!SUM_OVERFLOW_U64(start, count)); + if (count == 0) { + return; + } + + node = g_tree_upper_bound(tree.t, &start); + if (node) { + node = g_tree_node_previous(node); + } else { + node = g_tree_node_last(tree.t); + } + + if (node) { + range = g_tree_node_value(node); + assert(range); + intersection = page_range_intersection_size(range, start, count); + joinable = page_range_joinable_right(range, start, count); + } + + if (!node || + (!intersection && !joinable)) { + /* + * !node case: the tree is empty or the very first node in the tree + * already has a higher key (the start of its range). + * the other case: there is a gap in the tree between the new range + * and the previous one. + * anyway, let's just insert the new range into the tree. + */ + node = page_range_tree_insert_new(tree, start, count); + assert(node); + range = g_tree_node_value(node); + assert(range); + } else { + /* + * the previous range in the tree either partially covers the new + * range or ends just at its beginning - extend it + */ + if (dupcount) { + *dupcount += intersection; + } + + count += start - range->start; + range->count = MAX(range->count, count); + } + + /* check next nodes for possible merging */ + for (node = g_tree_node_next(node); node; ) { + PageRange *rangecur; + + rangecur = g_tree_node_value(node); + assert(rangecur); + + intersection = page_range_intersection_size(rangecur, + range->start, range->count); + joinable = page_range_joinable_left(rangecur, + range->start, range->count); + if (!intersection && !joinable) { + /* the current node is disjoint */ + break; + } + + if (dupcount) { + *dupcount += intersection; + } + + count = rangecur->count + (rangecur->start - range->start); + range->count = MAX(range->count, count); + + /* the current node was merged in, remove it */ + start = rangecur->start; + node = g_tree_node_next(node); + /* no hinted removal in GTree... */ + g_tree_remove(tree.t, &start); + } +} + +bool hvb_page_range_tree_pop(PageRangeTree tree, PageRange *out, + uint64_t maxcount) +{ + GTreeNode *node; + PageRange *range; + + node = g_tree_node_last(tree.t); + if (!node) { + return false; + } + + range = g_tree_node_value(node); + assert(range); + + out->start = range->start; + + /* can't modify range->start as it is the node key */ + if (range->count > maxcount) { + out->start += range->count - maxcount; + out->count = maxcount; + range->count -= maxcount; + } else { + out->count = range->count; + /* no hinted removal in GTree... */ + g_tree_remove(tree.t, &out->start); + } + + return true; +} + +bool hvb_page_range_tree_intree_any(PageRangeTree tree, + uint64_t start, uint64_t count) +{ + GTreeNode *node; + + if (count == 0) { + return false; + } + + /* find the first node that can possibly intersect our range */ + node = g_tree_upper_bound(tree.t, &start); + if (node) { + /* + * a NULL node below means that the very first node in the tree + * already has a higher key (the start of its range). + */ + node = g_tree_node_previous(node); + } else { + /* a NULL node below means that the tree is empty */ + node = g_tree_node_last(tree.t); + } + /* node range start <= range start */ + + if (!node) { + /* node range start > range start */ + node = g_tree_node_first(tree.t); + } + + for ( ; node; node = g_tree_node_next(node)) { + PageRange *range = g_tree_node_value(node); + + assert(range); + /* + * if this node starts beyond or at the end of our range so does + * every next one + */ + if (range->start >= start + count) { + break; + } + + if (page_range_intersection_size(range, start, count) > 0) { + return true; + } + } + + return false; +} + +void hvb_page_range_tree_init(PageRangeTree *tree) +{ + tree->t = g_tree_new_full(page_range_tree_key_compare, NULL, + g_free, g_free); +} + +void hvb_page_range_tree_destroy(PageRangeTree *tree) +{ + /* g_tree_destroy() is not NULL-safe */ + if (!tree->t) { + return; + } + + g_tree_destroy(tree->t); + tree->t = NULL; +} diff --git a/hw/hyperv/hv-balloon-page_range_tree.h b/hw/hyperv/hv-balloon-page_range_tree.h new file mode 100644 index 0000000000..07a9ae0da6 --- /dev/null +++ b/hw/hyperv/hv-balloon-page_range_tree.h @@ -0,0 +1,118 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_HYPERV_HV_BALLOON_PAGE_RANGE_TREE_H +#define HW_HYPERV_HV_BALLOON_PAGE_RANGE_TREE_H + +#include "qemu/osdep.h" + +/* PageRange */ +typedef struct PageRange { + uint64_t start; + uint64_t count; +} PageRange; + +/* return just the part of range before (start) */ +static inline void page_range_part_before(const PageRange *range, + uint64_t start, PageRange *out) +{ + uint64_t endr = range->start + range->count; + uint64_t end = MIN(endr, start); + + out->start = range->start; + if (end > out->start) { + out->count = end - out->start; + } else { + out->count = 0; + } +} + +/* return just the part of range after (start, count) */ +static inline void page_range_part_after(const PageRange *range, + uint64_t start, uint64_t count, + PageRange *out) +{ + uint64_t end = range->start + range->count; + uint64_t ends = start + count; + + out->start = MAX(range->start, ends); + if (end > out->start) { + out->count = end - out->start; + } else { + out->count = 0; + } +} + +static inline void page_range_intersect(const PageRange *range, + uint64_t start, uint64_t count, + PageRange *out) +{ + uint64_t end1 = range->start + range->count; + uint64_t end2 = start + count; + uint64_t end = MIN(end1, end2); + + out->start = MAX(range->start, start); + out->count = out->start < end ? end - out->start : 0; +} + +static inline uint64_t page_range_intersection_size(const PageRange *range, + uint64_t start, uint64_t count) +{ + PageRange trange; + + page_range_intersect(range, start, count, &trange); + return trange.count; +} + +static inline bool page_range_joinable_left(const PageRange *range, + uint64_t start, uint64_t count) +{ + return start + count == range->start; +} + +static inline bool page_range_joinable_right(const PageRange *range, + uint64_t start, uint64_t count) +{ + return range->start + range->count == start; +} + +static inline bool page_range_joinable(const PageRange *range, + uint64_t start, uint64_t count) +{ + return page_range_joinable_left(range, start, count) || + page_range_joinable_right(range, start, count); +} + +/* PageRangeTree */ +/* type safety */ +typedef struct PageRangeTree { + GTree *t; +} PageRangeTree; + +static inline bool page_range_tree_is_empty(PageRangeTree tree) +{ + guint nnodes = g_tree_nnodes(tree.t); + + return nnodes == 0; +} + +void hvb_page_range_tree_init(PageRangeTree *tree); +void hvb_page_range_tree_destroy(PageRangeTree *tree); + +bool hvb_page_range_tree_intree_any(PageRangeTree tree, + uint64_t start, uint64_t count); + +bool hvb_page_range_tree_pop(PageRangeTree tree, PageRange *out, + uint64_t maxcount); + +void hvb_page_range_tree_insert(PageRangeTree tree, + uint64_t start, uint64_t count, + uint64_t *dupcount); + +#endif diff --git a/hw/hyperv/hv-balloon-stub.c b/hw/hyperv/hv-balloon-stub.c new file mode 100644 index 0000000000..a47412d4a8 --- /dev/null +++ b/hw/hyperv/hv-balloon-stub.c @@ -0,0 +1,19 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "qapi/qapi-commands-machine.h" +#include "qapi/qapi-types-machine.h" + +HvBalloonInfo *qmp_query_hv_balloon_status_report(Error **errp) +{ + error_setg(errp, "hv-balloon device not enabled in this build"); + return NULL; +} diff --git a/hw/hyperv/hv-balloon.c b/hw/hyperv/hv-balloon.c new file mode 100644 index 0000000000..66f297c1d7 --- /dev/null +++ b/hw/hyperv/hv-balloon.c @@ -0,0 +1,1769 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "hv-balloon-internal.h" + +#include "exec/address-spaces.h" +#include "exec/cpu-common.h" +#include "exec/ramblock.h" +#include "hw/boards.h" +#include "hw/hyperv/dynmem-proto.h" +#include "hw/hyperv/hv-balloon.h" +#include "hw/hyperv/vmbus.h" +#include "hw/mem/memory-device.h" +#include "hw/mem/pc-dimm.h" +#include "hw/qdev-core.h" +#include "hw/qdev-properties.h" +#include "monitor/qdev.h" +#include "qapi/error.h" +#include "qapi/qapi-commands-machine.h" +#include "qapi/qapi-events-machine.h" +#include "qapi/qapi-types-machine.h" +#include "qapi/qmp/qdict.h" +#include "qapi/visitor.h" +#include "qemu/error-report.h" +#include "qemu/module.h" +#include "qemu/units.h" +#include "qemu/timer.h" +#include "sysemu/balloon.h" +#include "sysemu/hostmem.h" +#include "sysemu/reset.h" +#include "hv-balloon-our_range_memslots.h" +#include "hv-balloon-page_range_tree.h" +#include "trace.h" + +#define HV_BALLOON_ADDR_PROP "addr" +#define HV_BALLOON_MEMDEV_PROP "memdev" +#define HV_BALLOON_GUID "525074DC-8985-46e2-8057-A307DC18A502" + +/* + * Some Windows versions (at least Server 2019) will crash with various + * error codes when receiving DM protocol requests (at least + * DM_MEM_HOT_ADD_REQUEST) immediately after boot. + * + * It looks like Hyper-V from Server 2016 uses a 50-second after-boot + * delay, probably to workaround this issue, so we'll use this value, too. + */ +#define HV_BALLOON_POST_INIT_WAIT (50 * 1000) + +#define HV_BALLOON_HA_CHUNK_SIZE (2 * GiB) +#define HV_BALLOON_HA_CHUNK_PAGES (HV_BALLOON_HA_CHUNK_SIZE / HV_BALLOON_PAGE_SIZE) + +#define HV_BALLOON_HA_MEMSLOT_SIZE_ALIGN (128 * MiB) + +#define HV_BALLOON_HR_CHUNK_PAGES 585728 +/* + * ^ that's the maximum number of pages + * that Windows returns in one hot remove response + * + * If the number requested is too high Windows will no longer honor + * these requests + */ + +struct HvBalloonClass { + VMBusDeviceClass parent_class; +} HvBalloonClass; + +typedef enum State { + /* not a real state */ + S_NO_CHANGE = 0, + + S_WAIT_RESET, + S_POST_RESET_CLOSED, + + /* init flow */ + S_VERSION, + S_CAPS, + S_POST_INIT_WAIT, + + S_IDLE, + + /* balloon op flow */ + S_BALLOON_POSTING, + S_BALLOON_RB_WAIT, + S_BALLOON_REPLY_WAIT, + + /* unballoon + hot add ops flow */ + S_UNBALLOON_POSTING, + S_UNBALLOON_RB_WAIT, + S_UNBALLOON_REPLY_WAIT, + S_HOT_ADD_SETUP, + S_HOT_ADD_RB_WAIT, + S_HOT_ADD_POSTING, + S_HOT_ADD_REPLY_WAIT, +} State; + +typedef struct StateDesc { + State state; + const char *desc; +} StateDesc; + +typedef struct HvBalloon { + VMBusDevice parent; + State state; + + union dm_version version; + union dm_caps caps; + + QEMUTimer post_init_timer; + + unsigned int trans_id; + + struct { + bool enabled; + bool received; + uint64_t committed; + uint64_t available; + } status_report; + + /* Guest target size */ + uint64_t target; + bool target_changed; + + /* Current (un)balloon / hot-add operation parameters */ + union { + uint64_t balloon_diff; + + struct { + uint64_t unballoon_diff; + uint64_t hot_add_diff; + }; + + struct { + PageRange hot_add_range; + uint64_t ha_current_count; + }; + }; + + OurRangeMemslots *our_range; + + /* Count of memslots covering our memory */ + unsigned int memslot_count; + + /* Nominal size of each memslot (the last one might be smaller) */ + uint64_t memslot_size; + + /* Non-ours removed memory */ + PageRangeTree removed_guest, removed_both; + + /* Grand totals of removed memory (both ours and non-ours) */ + uint64_t removed_guest_ctr, removed_both_ctr; + + /* MEMORY_DEVICE props */ + uint64_t addr; + HostMemoryBackend *hostmem; + MemoryRegion *mr; +} HvBalloon; + +OBJECT_DEFINE_TYPE_WITH_INTERFACES(HvBalloon, hv_balloon, HV_BALLOON, VMBUS_DEVICE, \ + { TYPE_MEMORY_DEVICE }, { }) + +#define HV_BALLOON_SET_STATE(hvb, news) \ + do { \ + assert(news != S_NO_CHANGE); \ + hv_balloon_state_set(hvb, news, # news); \ + } while (0) + +#define HV_BALLOON_STATE_DESC_SET(stdesc, news) \ + _hv_balloon_state_desc_set(stdesc, news, # news) + +#define HV_BALLOON_STATE_DESC_INIT \ + { \ + .state = S_NO_CHANGE, \ + } + +typedef struct HvBalloonReq { + VMBusChanReq vmreq; +} HvBalloonReq; + +/* total our memory includes parts currently removed from the guest */ +static uint64_t hv_balloon_total_our_ram(HvBalloon *balloon) +{ + if (!balloon->our_range) { + return 0; + } + + return balloon->our_range->range.added; +} + +/* TODO: unify the code below with virtio-balloon and cache the value */ +static int build_dimm_list(Object *obj, void *opaque) +{ + GSList **list = opaque; + + if (object_dynamic_cast(obj, TYPE_PC_DIMM)) { + DeviceState *dev = DEVICE(obj); + if (dev->realized) { /* only realized DIMMs matter */ + *list = g_slist_prepend(*list, dev); + } + } + + object_child_foreach(obj, build_dimm_list, opaque); + return 0; +} + +static ram_addr_t get_current_ram_size(void) +{ + GSList *list = NULL, *item; + ram_addr_t size = current_machine->ram_size; + + build_dimm_list(qdev_get_machine(), &list); + for (item = list; item; item = g_slist_next(item)) { + Object *obj = OBJECT(item->data); + if (!strcmp(object_get_typename(obj), TYPE_PC_DIMM)) + size += object_property_get_int(obj, PC_DIMM_SIZE_PROP, + &error_abort); + } + g_slist_free(list); + + return size; +} + +/* total RAM includes memory currently removed from the guest */ +static uint64_t hv_balloon_total_ram(HvBalloon *balloon) +{ + ram_addr_t ram_size = get_current_ram_size(); + uint64_t ram_size_pages = ram_size >> HV_BALLOON_PFN_SHIFT; + uint64_t our_ram_size_pages = hv_balloon_total_our_ram(balloon); + + assert(ram_size_pages > 0); + + return SUM_SATURATE_U64(ram_size_pages, our_ram_size_pages); +} + +/* + * calculating the total RAM size is a slow operation, + * avoid it as much as possible + */ +static uint64_t hv_balloon_total_removed_rs(HvBalloon *balloon, + uint64_t ram_size_pages) +{ + uint64_t total_removed; + + total_removed = SUM_SATURATE_U64(balloon->removed_guest_ctr, + balloon->removed_both_ctr); + + /* possible if guest returns pages outside actual RAM */ + if (total_removed > ram_size_pages) { + total_removed = ram_size_pages; + } + + return total_removed; +} + +/* Returns whether the state has actually changed */ +static bool hv_balloon_state_set(HvBalloon *balloon, + State newst, const char *newststr) +{ + if (newst == S_NO_CHANGE || balloon->state == newst) { + return false; + } + + balloon->state = newst; + trace_hv_balloon_state_change(newststr); + return true; +} + +static void _hv_balloon_state_desc_set(StateDesc *stdesc, + State newst, const char *newststr) +{ + /* state setting is only permitted on a freshly init desc */ + assert(stdesc->state == S_NO_CHANGE); + + assert(newst != S_NO_CHANGE); + + stdesc->state = newst; + stdesc->desc = newststr; +} + +static VMBusChannel *hv_balloon_get_channel_maybe(HvBalloon *balloon) +{ + return vmbus_device_channel(&balloon->parent, 0); +} + +static VMBusChannel *hv_balloon_get_channel(HvBalloon *balloon) +{ + VMBusChannel *chan; + + chan = hv_balloon_get_channel_maybe(balloon); + assert(chan != NULL); + return chan; +} + +static ssize_t hv_balloon_send_packet(VMBusChannel *chan, + struct dm_message *msg) +{ + int ret; + + ret = vmbus_channel_reserve(chan, 0, msg->hdr.size); + if (ret < 0) { + return ret; + } + + return vmbus_channel_send(chan, VMBUS_PACKET_DATA_INBAND, + NULL, 0, msg, msg->hdr.size, false, + msg->hdr.trans_id); +} + +static bool hv_balloon_unballoon_get_source(HvBalloon *balloon, + PageRangeTree *dtree, + uint64_t **dctr, + bool *is_our_range) +{ + OurRange *our_range = OUR_RANGE(balloon->our_range); + + /* Try the boot memory first */ + if (g_tree_nnodes(balloon->removed_guest.t) > 0) { + *dtree = balloon->removed_guest; + *dctr = &balloon->removed_guest_ctr; + *is_our_range = false; + } else if (g_tree_nnodes(balloon->removed_both.t) > 0) { + *dtree = balloon->removed_both; + *dctr = &balloon->removed_both_ctr; + *is_our_range = false; + } else if (!our_range) { + return false; + } else if (!our_range_is_removed_tree_empty(our_range, false)) { + *dtree = our_range_get_removed_tree(our_range, false); + *dctr = &balloon->removed_guest_ctr; + *is_our_range = true; + } else if (!our_range_is_removed_tree_empty(our_range, true)) { + *dtree = our_range_get_removed_tree(our_range, true); + *dctr = &balloon->removed_both_ctr; + *is_our_range = true; + } else { + return false; + } + + return true; +} + +static void hv_balloon_unballoon_rb_wait(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan = hv_balloon_get_channel(balloon); + struct dm_unballoon_request *ur; + size_t ur_size = sizeof(*ur) + sizeof(ur->range_array[0]); + + assert(balloon->state == S_UNBALLOON_RB_WAIT); + + if (vmbus_channel_reserve(chan, 0, ur_size) < 0) { + return; + } + + HV_BALLOON_STATE_DESC_SET(stdesc, S_UNBALLOON_POSTING); +} + +static void hv_balloon_unballoon_posting(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan = hv_balloon_get_channel(balloon); + PageRangeTree dtree; + uint64_t *dctr; + bool our_range; + struct dm_unballoon_request *ur; + size_t ur_size = sizeof(*ur) + sizeof(ur->range_array[0]); + PageRange range; + bool bret; + ssize_t ret; + + assert(balloon->state == S_UNBALLOON_POSTING); + assert(balloon->unballoon_diff > 0); + + if (!hv_balloon_unballoon_get_source(balloon, &dtree, &dctr, &our_range)) { + error_report("trying to unballoon but nothing seems to be ballooned"); + /* + * there is little we can do as we might have already + * sent the guest a partial request we can't cancel + */ + return; + } + + assert(balloon->our_range || !our_range); + assert(dtree.t); + assert(dctr); + + ur = alloca(ur_size); + memset(ur, 0, ur_size); + ur->hdr.type = DM_UNBALLOON_REQUEST; + ur->hdr.size = ur_size; + ur->hdr.trans_id = balloon->trans_id; + + bret = hvb_page_range_tree_pop(dtree, &range, MIN(balloon->unballoon_diff, + HV_BALLOON_HA_CHUNK_PAGES)); + assert(bret); + /* TODO: madvise? */ + + *dctr -= range.count; + balloon->unballoon_diff -= range.count; + + ur->range_count = 1; + ur->range_array[0].finfo.start_page = range.start; + ur->range_array[0].finfo.page_cnt = range.count; + ur->more_pages = balloon->unballoon_diff > 0; + + trace_hv_balloon_outgoing_unballoon(ur->hdr.trans_id, + range.count, range.start, + balloon->unballoon_diff); + + if (ur->more_pages) { + HV_BALLOON_STATE_DESC_SET(stdesc, S_UNBALLOON_RB_WAIT); + } else { + HV_BALLOON_STATE_DESC_SET(stdesc, S_UNBALLOON_REPLY_WAIT); + } + + ret = vmbus_channel_send(chan, VMBUS_PACKET_DATA_INBAND, + NULL, 0, ur, ur_size, false, + ur->hdr.trans_id); + if (ret <= 0) { + error_report("error %zd when posting unballoon msg, expect problems", + ret); + } +} + +static bool hv_balloon_our_range_ensure(HvBalloon *balloon) +{ + uint64_t align; + MemoryRegion *hostmem_mr; + g_autoptr(OurRangeMemslots) our_range_memslots = NULL; + OurRange *our_range; + + if (balloon->our_range) { + return true; + } + + if (!balloon->hostmem) { + return false; + } + + align = (1 << balloon->caps.cap_bits.hot_add_alignment) * MiB; + assert(QEMU_IS_ALIGNED(balloon->addr, align)); + + hostmem_mr = host_memory_backend_get_memory(balloon->hostmem); + + our_range_memslots = hvb_our_range_memslots_new(balloon->addr, + balloon->mr, hostmem_mr, + OBJECT(balloon), + balloon->memslot_count, + balloon->memslot_size); + our_range = OUR_RANGE(our_range_memslots); + + if (hvb_page_range_tree_intree_any(balloon->removed_guest, + our_range->range.start, + our_range->range.count) || + hvb_page_range_tree_intree_any(balloon->removed_both, + our_range->range.start, + our_range->range.count)) { + error_report("some parts of the memory backend were already returned by the guest. this should not happen, please reboot the guest and try again"); + return false; + } + + trace_hv_balloon_our_range_add(our_range->range.count, + our_range->range.start); + + balloon->our_range = g_steal_pointer(&our_range_memslots); + return true; +} + +static void hv_balloon_hot_add_setup(HvBalloon *balloon, StateDesc *stdesc) +{ + /* need to make copy since it is in union with hot_add_range */ + uint64_t hot_add_diff = balloon->hot_add_diff; + PageRange *hot_add_range = &balloon->hot_add_range; + uint64_t align, our_range_remaining; + OurRange *our_range; + + assert(balloon->state == S_HOT_ADD_SETUP); + assert(hot_add_diff > 0); + + if (!hv_balloon_our_range_ensure(balloon)) { + goto ret_idle; + } + + our_range = OUR_RANGE(balloon->our_range); + + align = (1 << balloon->caps.cap_bits.hot_add_alignment) * + (MiB / HV_BALLOON_PAGE_SIZE); + + /* Absolute GPA in pages */ + hot_add_range->start = our_range_get_remaining_start(our_range); + assert(QEMU_IS_ALIGNED(hot_add_range->start, align)); + + our_range_remaining = our_range_get_remaining_size(our_range); + hot_add_range->count = MIN(our_range_remaining, hot_add_diff); + hot_add_range->count = QEMU_ALIGN_DOWN(hot_add_range->count, align); + if (hot_add_range->count == 0) { + goto ret_idle; + } + + hvb_our_range_memslots_ensure_mapped_additional(balloon->our_range, + hot_add_range->count); + + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_RB_WAIT); + return; + +ret_idle: + HV_BALLOON_STATE_DESC_SET(stdesc, S_IDLE); +} + +static void hv_balloon_hot_add_rb_wait(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan = hv_balloon_get_channel(balloon); + struct dm_hot_add *ha; + size_t ha_size = sizeof(*ha) + sizeof(ha->range); + + assert(balloon->state == S_HOT_ADD_RB_WAIT); + + if (vmbus_channel_reserve(chan, 0, ha_size) < 0) { + return; + } + + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_POSTING); +} + +static void hv_balloon_hot_add_posting(HvBalloon *balloon, StateDesc *stdesc) +{ + PageRange *hot_add_range = &balloon->hot_add_range; + uint64_t *current_count = &balloon->ha_current_count; + VMBusChannel *chan = hv_balloon_get_channel(balloon); + struct dm_hot_add *ha; + size_t ha_size = sizeof(*ha) + sizeof(ha->range); + union dm_mem_page_range *ha_region; + uint64_t align, chunk_max_size; + ssize_t ret; + + assert(balloon->state == S_HOT_ADD_POSTING); + assert(hot_add_range->count > 0); + + align = (1 << balloon->caps.cap_bits.hot_add_alignment) * + (MiB / HV_BALLOON_PAGE_SIZE); + if (align >= HV_BALLOON_HA_CHUNK_PAGES) { + /* + * If the required alignment is higher than the chunk size we let it + * override that size. + */ + chunk_max_size = align; + } else { + chunk_max_size = QEMU_ALIGN_DOWN(HV_BALLOON_HA_CHUNK_PAGES, align); + } + + /* + * hot_add_range->count starts aligned in hv_balloon_hot_add_setup(), + * then it is either reduced by subtracting aligned current_count or + * further hot-adds are prevented by marking the whole remaining our range + * as unusable in hv_balloon_handle_hot_add_response(). + */ + *current_count = MIN(hot_add_range->count, chunk_max_size); + + ha = alloca(ha_size); + ha_region = &(&ha->range)[1]; + memset(ha, 0, ha_size); + ha->hdr.type = DM_MEM_HOT_ADD_REQUEST; + ha->hdr.size = ha_size; + ha->hdr.trans_id = balloon->trans_id; + + ha->range.finfo.start_page = hot_add_range->start; + ha->range.finfo.page_cnt = *current_count; + ha_region->finfo.start_page = hot_add_range->start; + ha_region->finfo.page_cnt = ha->range.finfo.page_cnt; + + trace_hv_balloon_outgoing_hot_add(ha->hdr.trans_id, + *current_count, hot_add_range->start); + + ret = vmbus_channel_send(chan, VMBUS_PACKET_DATA_INBAND, + NULL, 0, ha, ha_size, false, + ha->hdr.trans_id); + if (ret <= 0) { + error_report("error %zd when posting hot add msg, expect problems", + ret); + } + + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_REPLY_WAIT); +} + +static void hv_balloon_balloon_rb_wait(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan = hv_balloon_get_channel(balloon); + size_t bl_size = sizeof(struct dm_balloon); + + assert(balloon->state == S_BALLOON_RB_WAIT); + + if (vmbus_channel_reserve(chan, 0, bl_size) < 0) { + return; + } + + HV_BALLOON_STATE_DESC_SET(stdesc, S_BALLOON_POSTING); +} + +static void hv_balloon_balloon_posting(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan = hv_balloon_get_channel(balloon); + struct dm_balloon bl; + size_t bl_size = sizeof(bl); + ssize_t ret; + + assert(balloon->state == S_BALLOON_POSTING); + assert(balloon->balloon_diff > 0); + + memset(&bl, 0, sizeof(bl)); + bl.hdr.type = DM_BALLOON_REQUEST; + bl.hdr.size = bl_size; + bl.hdr.trans_id = balloon->trans_id; + bl.num_pages = MIN(balloon->balloon_diff, HV_BALLOON_HR_CHUNK_PAGES); + + trace_hv_balloon_outgoing_balloon(bl.hdr.trans_id, bl.num_pages, + balloon->balloon_diff); + + ret = vmbus_channel_send(chan, VMBUS_PACKET_DATA_INBAND, + NULL, 0, &bl, bl_size, false, + bl.hdr.trans_id); + if (ret <= 0) { + error_report("error %zd when posting balloon msg, expect problems", + ret); + } + + HV_BALLOON_STATE_DESC_SET(stdesc, S_BALLOON_REPLY_WAIT); +} + +static void hv_balloon_idle_state_process_target(HvBalloon *balloon, + StateDesc *stdesc) +{ + bool can_balloon = balloon->caps.cap_bits.balloon; + uint64_t ram_size_pages, total_removed; + + ram_size_pages = hv_balloon_total_ram(balloon); + total_removed = hv_balloon_total_removed_rs(balloon, ram_size_pages); + + /* + * we need to cache the values computed from the balloon target value when + * starting the adjustment procedure in case someone changes the target when + * the procedure is in progress + */ + if (balloon->target > ram_size_pages - total_removed) { + bool can_hot_add = balloon->caps.cap_bits.hot_add; + uint64_t target_diff = balloon->target - + (ram_size_pages - total_removed); + + balloon->unballoon_diff = MIN(target_diff, total_removed); + + if (can_hot_add) { + balloon->hot_add_diff = target_diff - balloon->unballoon_diff; + } else { + balloon->hot_add_diff = 0; + } + + if (balloon->unballoon_diff > 0) { + assert(can_balloon); + HV_BALLOON_STATE_DESC_SET(stdesc, S_UNBALLOON_RB_WAIT); + } else if (balloon->hot_add_diff > 0) { + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_SETUP); + } + } else if (can_balloon && + balloon->target < ram_size_pages - total_removed) { + balloon->balloon_diff = ram_size_pages - total_removed - + balloon->target; + HV_BALLOON_STATE_DESC_SET(stdesc, S_BALLOON_RB_WAIT); + } +} + +static void hv_balloon_idle_state(HvBalloon *balloon, + StateDesc *stdesc) +{ + assert(balloon->state == S_IDLE); + + if (balloon->target_changed) { + balloon->target_changed = false; + hv_balloon_idle_state_process_target(balloon, stdesc); + return; + } +} + +static const struct { + void (*handler)(HvBalloon *balloon, StateDesc *stdesc); +} state_handlers[] = { + [S_IDLE].handler = hv_balloon_idle_state, + [S_BALLOON_POSTING].handler = hv_balloon_balloon_posting, + [S_BALLOON_RB_WAIT].handler = hv_balloon_balloon_rb_wait, + [S_UNBALLOON_POSTING].handler = hv_balloon_unballoon_posting, + [S_UNBALLOON_RB_WAIT].handler = hv_balloon_unballoon_rb_wait, + [S_HOT_ADD_SETUP].handler = hv_balloon_hot_add_setup, + [S_HOT_ADD_RB_WAIT].handler = hv_balloon_hot_add_rb_wait, + [S_HOT_ADD_POSTING].handler = hv_balloon_hot_add_posting, +}; + +static void hv_balloon_handle_state(HvBalloon *balloon, StateDesc *stdesc) +{ + if (balloon->state >= ARRAY_SIZE(state_handlers) || + !state_handlers[balloon->state].handler) { + return; + } + + state_handlers[balloon->state].handler(balloon, stdesc); +} + +static void hv_balloon_remove_response_insert_range(PageRangeTree tree, + const PageRange *range, + uint64_t *ctr1, + uint64_t *ctr2, + uint64_t *ctr3) +{ + uint64_t dupcount, effcount; + + if (range->count == 0) { + return; + } + + dupcount = 0; + hvb_page_range_tree_insert(tree, range->start, range->count, &dupcount); + + assert(dupcount <= range->count); + effcount = range->count - dupcount; + + *ctr1 += effcount; + *ctr2 += effcount; + if (ctr3) { + *ctr3 += effcount; + } +} + +static void hv_balloon_remove_response_handle_range(HvBalloon *balloon, + PageRange *range, + bool both, + uint64_t *removedctr) +{ + OurRange *our_range = OUR_RANGE(balloon->our_range); + PageRangeTree globaltree = + both ? balloon->removed_both : balloon->removed_guest; + uint64_t *globalctr = + both ? &balloon->removed_both_ctr : &balloon->removed_guest_ctr; + PageRange rangeeff; + + if (range->count == 0) { + return; + } + + trace_hv_balloon_remove_response(range->count, range->start, both); + + if (our_range) { + /* Includes the not-yet-hot-added and unusable parts. */ + rangeeff = our_range->range; + } else { + rangeeff.start = rangeeff.count = 0; + } + + if (page_range_intersection_size(range, rangeeff.start, rangeeff.count) > 0) { + PageRangeTree ourtree = our_range_get_removed_tree(our_range, both); + PageRange rangehole, rangecommon; + uint64_t ourremoved = 0; + + /* process the hole before our range, if it exists */ + page_range_part_before(range, rangeeff.start, &rangehole); + hv_balloon_remove_response_insert_range(globaltree, &rangehole, + globalctr, removedctr, NULL); + if (rangehole.count > 0) { + trace_hv_balloon_remove_response_hole(rangehole.count, + rangehole.start, + range->count, range->start, + rangeeff.start, both); + } + + /* process our part */ + page_range_intersect(range, rangeeff.start, rangeeff.count, + &rangecommon); + hv_balloon_remove_response_insert_range(ourtree, &rangecommon, + globalctr, removedctr, + &ourremoved); + if (rangecommon.count > 0) { + trace_hv_balloon_remove_response_common(rangecommon.count, + rangecommon.start, + range->count, range->start, + rangeeff.count, + rangeeff.start, ourremoved, + both); + } + + /* calculate what's left after our range */ + rangecommon = *range; + page_range_part_after(&rangecommon, rangeeff.start, rangeeff.count, + range); + } + + /* process the remainder of the range that lies after our range */ + if (range->count > 0) { + hv_balloon_remove_response_insert_range(globaltree, range, + globalctr, removedctr, NULL); + trace_hv_balloon_remove_response_remainder(range->count, range->start, + both); + range->count = 0; + } +} + +static void hv_balloon_remove_response_handle_pages(HvBalloon *balloon, + PageRange *range, + uint64_t start, + uint64_t count, + bool both, + uint64_t *removedctr) +{ + assert(count > 0); + + /* + * if there is an existing range that the new range can't be joined to + * dump it into tree(s) + */ + if (range->count > 0 && !page_range_joinable(range, start, count)) { + hv_balloon_remove_response_handle_range(balloon, range, both, + removedctr); + } + + if (range->count == 0) { + range->start = start; + range->count = count; + } else if (page_range_joinable_left(range, start, count)) { + range->start = start; + range->count += count; + } else { /* page_range_joinable_right() */ + range->count += count; + } +} + +static gboolean hv_balloon_handle_remove_host_addr_node(gpointer key, + gpointer value, + gpointer data) +{ + PageRange *range = value; + uint64_t pageoff; + + for (pageoff = 0; pageoff < range->count; ) { + uint64_t addr_64 = (range->start + pageoff) * HV_BALLOON_PAGE_SIZE; + void *addr; + RAMBlock *rb; + ram_addr_t rb_offset; + size_t rb_page_size; + size_t discard_size; + + assert(addr_64 <= UINTPTR_MAX); + addr = (void *)((uintptr_t)addr_64); + rb = qemu_ram_block_from_host(addr, false, &rb_offset); + rb_page_size = qemu_ram_pagesize(rb); + + if (rb_page_size != HV_BALLOON_PAGE_SIZE) { + /* TODO: these should end in "removed_guest" */ + warn_report("guest reported removed page backed by unsupported page size %zu", + rb_page_size); + pageoff++; + continue; + } + + discard_size = MIN(range->count - pageoff, + (rb->max_length - rb_offset) / + HV_BALLOON_PAGE_SIZE); + discard_size = MAX(discard_size, 1); + + if (ram_block_discard_range(rb, rb_offset, discard_size * + HV_BALLOON_PAGE_SIZE) != 0) { + warn_report("guest reported removed page failed discard"); + } + + pageoff += discard_size; + } + + return false; +} + +static void hv_balloon_handle_remove_host_addr_tree(PageRangeTree tree) +{ + g_tree_foreach(tree.t, hv_balloon_handle_remove_host_addr_node, NULL); +} + +static int hv_balloon_handle_remove_section(PageRangeTree tree, + const MemoryRegionSection *section, + uint64_t count) +{ + void *addr = memory_region_get_ram_ptr(section->mr) + + section->offset_within_region; + uint64_t addr_page; + + assert(count > 0); + + if ((uintptr_t)addr % HV_BALLOON_PAGE_SIZE) { + warn_report("guest reported removed pages at an unaligned host addr %p", + addr); + return -EINVAL; + } + + addr_page = (uintptr_t)addr / HV_BALLOON_PAGE_SIZE; + hvb_page_range_tree_insert(tree, addr_page, count, NULL); + + return 0; +} + +static void hv_balloon_handle_remove_ranges(HvBalloon *balloon, + union dm_mem_page_range ranges[], + uint32_t count) +{ + uint64_t removedcnt; + PageRangeTree removed_host_addr; + PageRange range_guest, range_both; + + hvb_page_range_tree_init(&removed_host_addr); + range_guest.count = range_both.count = removedcnt = 0; + for (unsigned int ctr = 0; ctr < count; ctr++) { + union dm_mem_page_range *mr = &ranges[ctr]; + hwaddr pa; + MemoryRegionSection section; + + for (unsigned int offset = 0; offset < mr->finfo.page_cnt; ) { + int ret; + uint64_t pageno = mr->finfo.start_page + offset; + uint64_t pagecnt = 1; + + pa = (hwaddr)pageno << HV_BALLOON_PFN_SHIFT; + section = memory_region_find(get_system_memory(), pa, + (mr->finfo.page_cnt - offset) * + HV_BALLOON_PAGE_SIZE); + if (!section.mr) { + warn_report("guest reported removed page %"PRIu64" not found in RAM", + pageno); + ret = -EINVAL; + goto finish_page; + } + + pagecnt = int128_get64(section.size) / HV_BALLOON_PAGE_SIZE; + if (pagecnt <= 0) { + warn_report("guest reported removed page %"PRIu64" in a section smaller than page size", + pageno); + pagecnt = 1; /* skip the whole page */ + ret = -EINVAL; + goto finish_page; + } + + if (!memory_region_is_ram(section.mr) || + memory_region_is_rom(section.mr) || + memory_region_is_romd(section.mr)) { + warn_report("guest reported removed page %"PRIu64" in a section that is not an ordinary RAM", + pageno); + ret = -EINVAL; + goto finish_page; + } + + ret = hv_balloon_handle_remove_section(removed_host_addr, §ion, + pagecnt); + + finish_page: + if (ret == 0) { + hv_balloon_remove_response_handle_pages(balloon, + &range_both, + pageno, pagecnt, + true, &removedcnt); + } else { + hv_balloon_remove_response_handle_pages(balloon, + &range_guest, + pageno, pagecnt, + false, &removedcnt); + } + + if (section.mr) { + memory_region_unref(section.mr); + } + + offset += pagecnt; + } + } + + hv_balloon_remove_response_handle_range(balloon, &range_both, true, + &removedcnt); + hv_balloon_remove_response_handle_range(balloon, &range_guest, false, + &removedcnt); + + hv_balloon_handle_remove_host_addr_tree(removed_host_addr); + hvb_page_range_tree_destroy(&removed_host_addr); + + if (removedcnt > balloon->balloon_diff) { + warn_report("guest reported more pages removed than currently pending (%"PRIu64" vs %"PRIu64")", + removedcnt, balloon->balloon_diff); + balloon->balloon_diff = 0; + } else { + balloon->balloon_diff -= removedcnt; + } +} + +static bool hv_balloon_handle_msg_size(HvBalloonReq *req, size_t minsize, + const char *msgname) +{ + VMBusChanReq *vmreq = &req->vmreq; + uint32_t msglen = vmreq->msglen; + + if (msglen >= minsize) { + return true; + } + + warn_report("%s message too short (%u vs %zu), ignoring", msgname, + (unsigned int)msglen, minsize); + return false; +} + +static void hv_balloon_handle_version_request(HvBalloon *balloon, + HvBalloonReq *req, + StateDesc *stdesc) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_version_request *msgVr = vmreq->msg; + struct dm_version_response respVr; + + if (balloon->state != S_VERSION) { + warn_report("unexpected DM_VERSION_REQUEST in %d state", + balloon->state); + return; + } + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgVr), + "DM_VERSION_REQUEST")) { + return; + } + + trace_hv_balloon_incoming_version(msgVr->version.major_version, + msgVr->version.minor_version); + + memset(&respVr, 0, sizeof(respVr)); + respVr.hdr.type = DM_VERSION_RESPONSE; + respVr.hdr.size = sizeof(respVr); + respVr.hdr.trans_id = msgVr->hdr.trans_id; + respVr.is_accepted = msgVr->version.version >= DYNMEM_PROTOCOL_VERSION_1 && + msgVr->version.version <= DYNMEM_PROTOCOL_VERSION_3; + + hv_balloon_send_packet(vmreq->chan, (struct dm_message *)&respVr); + + if (respVr.is_accepted) { + HV_BALLOON_STATE_DESC_SET(stdesc, S_CAPS); + } +} + +static void hv_balloon_handle_caps_report(HvBalloon *balloon, + HvBalloonReq *req, + StateDesc *stdesc) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_capabilities *msgCap = vmreq->msg; + struct dm_capabilities_resp_msg respCap; + + if (balloon->state != S_CAPS) { + warn_report("unexpected DM_CAPABILITIES_REPORT in %d state", + balloon->state); + return; + } + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgCap), + "DM_CAPABILITIES_REPORT")) { + return; + } + + trace_hv_balloon_incoming_caps(msgCap->caps.caps); + balloon->caps = msgCap->caps; + + memset(&respCap, 0, sizeof(respCap)); + respCap.hdr.type = DM_CAPABILITIES_RESPONSE; + respCap.hdr.size = sizeof(respCap); + respCap.hdr.trans_id = msgCap->hdr.trans_id; + respCap.is_accepted = 1; + respCap.hot_remove = 1; + respCap.suppress_pressure_reports = !balloon->status_report.enabled; + hv_balloon_send_packet(vmreq->chan, (struct dm_message *)&respCap); + + timer_mod(&balloon->post_init_timer, + qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + + HV_BALLOON_POST_INIT_WAIT); + + HV_BALLOON_STATE_DESC_SET(stdesc, S_POST_INIT_WAIT); +} + +static void hv_balloon_handle_status_report(HvBalloon *balloon, + HvBalloonReq *req) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_status *msgStatus = vmreq->msg; + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgStatus), + "DM_STATUS_REPORT")) { + return; + } + + if (!balloon->status_report.enabled) { + return; + } + + balloon->status_report.committed = msgStatus->num_committed; + balloon->status_report.committed *= HV_BALLOON_PAGE_SIZE; + balloon->status_report.available = msgStatus->num_avail; + balloon->status_report.available *= HV_BALLOON_PAGE_SIZE; + balloon->status_report.received = true; + + qapi_event_send_hv_balloon_status_report(balloon->status_report.committed, + balloon->status_report.available); +} + +HvBalloonInfo *qmp_query_hv_balloon_status_report(Error **errp) +{ + HvBalloon *balloon; + HvBalloonInfo *info; + + balloon = HV_BALLOON(object_resolve_path_type("", TYPE_HV_BALLOON, NULL)); + if (!balloon) { + error_setg(errp, "no %s device present", TYPE_HV_BALLOON); + return NULL; + } + + if (!balloon->status_report.enabled) { + error_setg(errp, "guest memory status reporting not enabled"); + return NULL; + } + + if (!balloon->status_report.received) { + error_setg(errp, "no guest memory status report received yet"); + return NULL; + } + + info = g_malloc0(sizeof(*info)); + info->committed = balloon->status_report.committed; + info->available = balloon->status_report.available; + return info; +} + +static void hv_balloon_handle_unballoon_response(HvBalloon *balloon, + HvBalloonReq *req, + StateDesc *stdesc) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_unballoon_response *msgUrR = vmreq->msg; + + if (balloon->state != S_UNBALLOON_REPLY_WAIT) { + warn_report("unexpected DM_UNBALLOON_RESPONSE in %d state", + balloon->state); + return; + } + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgUrR), + "DM_UNBALLOON_RESPONSE")) + return; + + trace_hv_balloon_incoming_unballoon(msgUrR->hdr.trans_id); + + balloon->trans_id++; + + if (balloon->hot_add_diff > 0) { + bool can_hot_add = balloon->caps.cap_bits.hot_add; + + assert(can_hot_add); + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_SETUP); + } else { + HV_BALLOON_STATE_DESC_SET(stdesc, S_IDLE); + } +} + +static void hv_balloon_handle_hot_add_response(HvBalloon *balloon, + HvBalloonReq *req, + StateDesc *stdesc) +{ + PageRange *hot_add_range = &balloon->hot_add_range; + VMBusChanReq *vmreq = &req->vmreq; + struct dm_hot_add_response *msgHaR = vmreq->msg; + OurRange *our_range; + + if (balloon->state != S_HOT_ADD_REPLY_WAIT) { + warn_report("unexpected DM_HOT_ADD_RESPONSE in %d state", + balloon->state); + return; + } + + assert(balloon->our_range); + our_range = OUR_RANGE(balloon->our_range); + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgHaR), + "DM_HOT_ADD_RESPONSE")) + return; + + trace_hv_balloon_incoming_hot_add(msgHaR->hdr.trans_id, msgHaR->result, + msgHaR->page_count); + + balloon->trans_id++; + + if (msgHaR->result) { + if (msgHaR->page_count > balloon->ha_current_count) { + warn_report("DM_HOT_ADD_RESPONSE page count higher than requested (%"PRIu32" vs %"PRIu64")", + msgHaR->page_count, balloon->ha_current_count); + msgHaR->page_count = balloon->ha_current_count; + } + + hvb_our_range_mark_added(our_range, msgHaR->page_count); + hot_add_range->start += msgHaR->page_count; + hot_add_range->count -= msgHaR->page_count; + } + + if (!msgHaR->result || msgHaR->page_count < balloon->ha_current_count) { + /* + * the current planned range was only partially hot-added, take note + * how much of it remains and don't attempt any further hot adds + */ + our_range_mark_remaining_unusable(our_range); + + goto ret_idle; + } + + /* any pages remaining to hot-add in our range? */ + if (hot_add_range->count > 0) { + HV_BALLOON_STATE_DESC_SET(stdesc, S_HOT_ADD_RB_WAIT); + return; + } + +ret_idle: + HV_BALLOON_STATE_DESC_SET(stdesc, S_IDLE); +} + +static void hv_balloon_handle_balloon_response(HvBalloon *balloon, + HvBalloonReq *req, + StateDesc *stdesc) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_balloon_response *msgBR = vmreq->msg; + + if (balloon->state != S_BALLOON_REPLY_WAIT) { + warn_report("unexpected DM_BALLOON_RESPONSE in %d state", + balloon->state); + return; + } + + if (!hv_balloon_handle_msg_size(req, sizeof(*msgBR), + "DM_BALLOON_RESPONSE")) + return; + + trace_hv_balloon_incoming_balloon(msgBR->hdr.trans_id, msgBR->range_count, + msgBR->more_pages); + + if (vmreq->msglen < sizeof(*msgBR) + + (uint64_t)sizeof(msgBR->range_array[0]) * msgBR->range_count) { + warn_report("DM_BALLOON_RESPONSE too short for the range count"); + return; + } + + if (msgBR->range_count == 0) { + /* The guest is already at its minimum size */ + balloon->balloon_diff = 0; + goto ret_end_trans; + } else { + hv_balloon_handle_remove_ranges(balloon, + msgBR->range_array, + msgBR->range_count); + } + + /* More responses expected? */ + if (msgBR->more_pages) { + return; + } + +ret_end_trans: + balloon->trans_id++; + + if (balloon->balloon_diff > 0) { + HV_BALLOON_STATE_DESC_SET(stdesc, S_BALLOON_RB_WAIT); + } else { + HV_BALLOON_STATE_DESC_SET(stdesc, S_IDLE); + } +} + +static void hv_balloon_handle_packet(HvBalloon *balloon, HvBalloonReq *req, + StateDesc *stdesc) +{ + VMBusChanReq *vmreq = &req->vmreq; + struct dm_message *msg = vmreq->msg; + + if (vmreq->msglen < sizeof(msg->hdr)) { + return; + } + + switch (msg->hdr.type) { + case DM_VERSION_REQUEST: + hv_balloon_handle_version_request(balloon, req, stdesc); + break; + + case DM_CAPABILITIES_REPORT: + hv_balloon_handle_caps_report(balloon, req, stdesc); + break; + + case DM_STATUS_REPORT: + hv_balloon_handle_status_report(balloon, req); + break; + + case DM_MEM_HOT_ADD_RESPONSE: + hv_balloon_handle_hot_add_response(balloon, req, stdesc); + break; + + case DM_UNBALLOON_RESPONSE: + hv_balloon_handle_unballoon_response(balloon, req, stdesc); + break; + + case DM_BALLOON_RESPONSE: + hv_balloon_handle_balloon_response(balloon, req, stdesc); + break; + + default: + warn_report("unknown DM message %u", msg->hdr.type); + break; + } +} + +static bool hv_balloon_recv_channel(HvBalloon *balloon, StateDesc *stdesc) +{ + VMBusChannel *chan; + HvBalloonReq *req; + + if (balloon->state == S_WAIT_RESET || + balloon->state == S_POST_RESET_CLOSED) { + return false; + } + + chan = hv_balloon_get_channel(balloon); + if (vmbus_channel_recv_start(chan)) { + return false; + } + + while ((req = vmbus_channel_recv_peek(chan, sizeof(*req)))) { + hv_balloon_handle_packet(balloon, req, stdesc); + vmbus_free_req(req); + vmbus_channel_recv_pop(chan); + + if (stdesc->state != S_NO_CHANGE) { + break; + } + } + + return vmbus_channel_recv_done(chan) > 0; +} + +/* old state handler -> new state transition (potential) */ +static bool hv_balloon_event_loop_state(HvBalloon *balloon) +{ + StateDesc state_new = HV_BALLOON_STATE_DESC_INIT; + + hv_balloon_handle_state(balloon, &state_new); + return hv_balloon_state_set(balloon, state_new.state, state_new.desc); +} + +/* VMBus message -> new state transition (potential) */ +static bool hv_balloon_event_loop_recv(HvBalloon *balloon) +{ + StateDesc state_new = HV_BALLOON_STATE_DESC_INIT; + bool any_recv, state_changed; + + any_recv = hv_balloon_recv_channel(balloon, &state_new); + state_changed = hv_balloon_state_set(balloon, + state_new.state, state_new.desc); + + return state_changed || any_recv; +} + +static void hv_balloon_event_loop(HvBalloon *balloon) +{ + bool state_repeat, recv_repeat; + + do { + state_repeat = hv_balloon_event_loop_state(balloon); + recv_repeat = hv_balloon_event_loop_recv(balloon); + } while (state_repeat || recv_repeat); +} + +static void hv_balloon_vmdev_chan_notify(VMBusChannel *chan) +{ + HvBalloon *balloon = HV_BALLOON(vmbus_channel_device(chan)); + + hv_balloon_event_loop(balloon); +} + +static void hv_balloon_stat(void *opaque, BalloonInfo *info) +{ + HvBalloon *balloon = opaque; + info->actual = (hv_balloon_total_ram(balloon) - balloon->removed_both_ctr) + << HV_BALLOON_PFN_SHIFT; +} + +static void hv_balloon_to_target(void *opaque, ram_addr_t target) +{ + HvBalloon *balloon = opaque; + uint64_t target_pages = target >> HV_BALLOON_PFN_SHIFT; + + if (!target_pages) { + return; + } + + /* + * always set target_changed, even with unchanged target, as the user + * might be asking us to try again reaching it + */ + balloon->target = target_pages; + balloon->target_changed = true; + + hv_balloon_event_loop(balloon); +} + +static int hv_balloon_vmdev_open_channel(VMBusChannel *chan) +{ + HvBalloon *balloon = HV_BALLOON(vmbus_channel_device(chan)); + + if (balloon->state != S_POST_RESET_CLOSED) { + warn_report("guest trying to open a DM channel in invalid %d state", + balloon->state); + return -EINVAL; + } + + HV_BALLOON_SET_STATE(balloon, S_VERSION); + hv_balloon_event_loop(balloon); + + return 0; +} + +static void hv_balloon_vmdev_close_channel(VMBusChannel *chan) +{ + HvBalloon *balloon = HV_BALLOON(vmbus_channel_device(chan)); + + timer_del(&balloon->post_init_timer); + + /* Don't report stale data */ + balloon->status_report.received = false; + + HV_BALLOON_SET_STATE(balloon, S_WAIT_RESET); + hv_balloon_event_loop(balloon); +} + +static void hv_balloon_post_init_timer(void *opaque) +{ + HvBalloon *balloon = opaque; + + if (balloon->state != S_POST_INIT_WAIT) { + return; + } + + HV_BALLOON_SET_STATE(balloon, S_IDLE); + hv_balloon_event_loop(balloon); +} + +static void hv_balloon_system_reset_unrealize_common(HvBalloon *balloon) +{ + g_clear_pointer(&balloon->our_range, hvb_our_range_memslots_free); +} + +static void hv_balloon_system_reset(void *opaque) +{ + HvBalloon *balloon = HV_BALLOON(opaque); + + hv_balloon_system_reset_unrealize_common(balloon); +} + +static void hv_balloon_ensure_mr(HvBalloon *balloon) +{ + MemoryRegion *hostmem_mr; + + assert(balloon->hostmem); + + if (balloon->mr) { + return; + } + + hostmem_mr = host_memory_backend_get_memory(balloon->hostmem); + + balloon->mr = g_new0(MemoryRegion, 1); + memory_region_init(balloon->mr, OBJECT(balloon), TYPE_HV_BALLOON, + memory_region_size(hostmem_mr)); + + /* + * The VM can indicate an alignment up to 32 GiB. Memory device core can + * usually only handle/guarantee 1 GiB alignment. The user will have to + * specify a larger maxmem eventually. + * + * The memory device core will warn the user in case maxmem might have to be + * increased and will fail plugging the device if there is not sufficient + * space after alignment. + * + * TODO: we could do the alignment ourselves in a slightly bigger region. + * But this feels better, although the warning might be annoying. Maybe + * we can optimize that in the future (e.g., with such a device on the + * cmdline place/size the device memory region differently. + */ + balloon->mr->align = MAX(32 * GiB, memory_region_get_alignment(hostmem_mr)); +} + +static void hv_balloon_free_mr(HvBalloon *balloon) +{ + if (!balloon->mr) { + return; + } + + object_unparent(OBJECT(balloon->mr)); + g_clear_pointer(&balloon->mr, g_free); +} + +static void hv_balloon_vmdev_realize(VMBusDevice *vdev, Error **errp) +{ + ERRP_GUARD(); + HvBalloon *balloon = HV_BALLOON(vdev); + int ret; + + balloon->state = S_WAIT_RESET; + + ret = qemu_add_balloon_handler(hv_balloon_to_target, hv_balloon_stat, + balloon); + if (ret < 0) { + /* This also protects against having multiple hv-balloon instances */ + error_setg(errp, "Only one balloon device is supported"); + return; + } + + if (balloon->hostmem) { + if (host_memory_backend_is_mapped(balloon->hostmem)) { + Object *obj = OBJECT(balloon->hostmem); + + error_setg(errp, "'%s' property specifies a busy memdev: %s", + HV_BALLOON_MEMDEV_PROP, + object_get_canonical_path_component(obj)); + goto out_balloon_handler; + } + + hv_balloon_ensure_mr(balloon); + + /* This is rather unlikely to happen, but let's still check for it. */ + if (!QEMU_IS_ALIGNED(memory_region_size(balloon->mr), + HV_BALLOON_PAGE_SIZE)) { + error_setg(errp, "'%s' property memdev size has to be a multiple of 0x%" PRIx64, + HV_BALLOON_MEMDEV_PROP, (uint64_t)HV_BALLOON_PAGE_SIZE); + goto out_balloon_handler; + } + + host_memory_backend_set_mapped(balloon->hostmem, true); + vmstate_register_ram(host_memory_backend_get_memory(balloon->hostmem), + DEVICE(balloon)); + } else if (balloon->addr) { + error_setg(errp, "'%s' property must not be set without a memdev", + HV_BALLOON_MEMDEV_PROP); + goto out_balloon_handler; + } + + timer_init_ms(&balloon->post_init_timer, QEMU_CLOCK_VIRTUAL, + hv_balloon_post_init_timer, balloon); + + qemu_register_reset(hv_balloon_system_reset, balloon); + + return; + +out_balloon_handler: + qemu_remove_balloon_handler(balloon); +} + +/* + * VMBus device reset has to be implemented in case the guest decides to + * disconnect and reconnect to the VMBus without rebooting the whole system. + * + * However, the hot-added memory can't be removed here as Windows keeps on using + * it until the system is restarted, even after disconnecting from the VMBus. + */ +static void hv_balloon_vmdev_reset(VMBusDevice *vdev) +{ + HvBalloon *balloon = HV_BALLOON(vdev); + + if (balloon->state == S_POST_RESET_CLOSED) { + return; + } + + if (balloon->our_range) { + hvb_our_range_clear_removed_trees(OUR_RANGE(balloon->our_range)); + } + + hvb_page_range_tree_destroy(&balloon->removed_guest); + hvb_page_range_tree_destroy(&balloon->removed_both); + hvb_page_range_tree_init(&balloon->removed_guest); + hvb_page_range_tree_init(&balloon->removed_both); + + balloon->trans_id = 0; + balloon->removed_guest_ctr = 0; + balloon->removed_both_ctr = 0; + + HV_BALLOON_SET_STATE(balloon, S_POST_RESET_CLOSED); + hv_balloon_event_loop(balloon); +} + +/* + * Clean up things that were (possibly) allocated pre-realization, for example + * from memory_device_pre_plug(), so we don't leak them if the device don't + * actually get realized in the end. + */ +static void hv_balloon_unrealize_finalize_common(HvBalloon *balloon) +{ + hv_balloon_free_mr(balloon); + balloon->addr = 0; + + balloon->memslot_count = 0; +} + +static void hv_balloon_vmdev_unrealize(VMBusDevice *vdev) +{ + HvBalloon *balloon = HV_BALLOON(vdev); + + qemu_unregister_reset(hv_balloon_system_reset, balloon); + + hv_balloon_system_reset_unrealize_common(balloon); + + qemu_remove_balloon_handler(balloon); + + if (balloon->hostmem) { + vmstate_unregister_ram(host_memory_backend_get_memory(balloon->hostmem), + DEVICE(balloon)); + host_memory_backend_set_mapped(balloon->hostmem, false); + } + + hvb_page_range_tree_destroy(&balloon->removed_guest); + hvb_page_range_tree_destroy(&balloon->removed_both); + + hv_balloon_unrealize_finalize_common(balloon); +} + +static uint64_t hv_balloon_md_get_addr(const MemoryDeviceState *md) +{ + return object_property_get_uint(OBJECT(md), HV_BALLOON_ADDR_PROP, + &error_abort); +} + +static void hv_balloon_md_set_addr(MemoryDeviceState *md, uint64_t addr, + Error **errp) +{ + object_property_set_uint(OBJECT(md), HV_BALLOON_ADDR_PROP, addr, errp); +} + +static MemoryRegion *hv_balloon_md_get_memory_region(MemoryDeviceState *md, + Error **errp) +{ + HvBalloon *balloon = HV_BALLOON(md); + + if (!balloon->hostmem) { + return NULL; + } + + hv_balloon_ensure_mr(balloon); + + return balloon->mr; +} + +static void hv_balloon_md_fill_device_info(const MemoryDeviceState *md, + MemoryDeviceInfo *info) +{ + HvBalloonDeviceInfo *hi = g_new0(HvBalloonDeviceInfo, 1); + const HvBalloon *balloon = HV_BALLOON(md); + DeviceState *dev = DEVICE(md); + + if (dev->id) { + hi->id = g_strdup(dev->id); + } + + if (balloon->hostmem) { + hi->memdev = object_get_canonical_path(OBJECT(balloon->hostmem)); + hi->memaddr = balloon->addr; + hi->has_memaddr = true; + hi->max_size = memory_region_size(balloon->mr); + /* TODO: expose current provided size or something else? */ + } else { + hi->max_size = 0; + } + + info->u.hv_balloon.data = hi; + info->type = MEMORY_DEVICE_INFO_KIND_HV_BALLOON; +} + +static void hv_balloon_decide_memslots(MemoryDeviceState *md, + unsigned int limit) +{ + HvBalloon *balloon = HV_BALLOON(md); + MemoryRegion *hostmem_mr; + uint64_t region_size, memslot_size, memslots; + + /* We're called exactly once, before realizing the device. */ + assert(!balloon->memslot_count); + + /* We should not be called if we don't have a memory backend */ + assert(balloon->hostmem); + + hostmem_mr = host_memory_backend_get_memory(balloon->hostmem); + region_size = memory_region_size(hostmem_mr); + + assert(region_size > 0); + memslot_size = QEMU_ALIGN_UP(region_size / limit, + HV_BALLOON_HA_MEMSLOT_SIZE_ALIGN); + memslots = QEMU_ALIGN_UP(region_size, memslot_size) / memslot_size; + + if (memslots > 1) { + balloon->memslot_size = memslot_size; + } else { + balloon->memslot_size = region_size; + } + + assert(memslots <= UINT_MAX); + balloon->memslot_count = memslots; +} + +static unsigned int hv_balloon_get_memslots(MemoryDeviceState *md) +{ + const HvBalloon *balloon = HV_BALLOON(md); + + /* We're called after setting the suggested limit. */ + assert(balloon->memslot_count > 0); + + return balloon->memslot_count; +} + +static void hv_balloon_init(Object *obj) +{ +} + +static void hv_balloon_finalize(Object *obj) +{ + HvBalloon *balloon = HV_BALLOON(obj); + + hv_balloon_unrealize_finalize_common(balloon); +} + +static Property hv_balloon_properties[] = { + DEFINE_PROP_BOOL("status-report", HvBalloon, + status_report.enabled, false), + + /* MEMORY_DEVICE props */ + DEFINE_PROP_LINK(HV_BALLOON_MEMDEV_PROP, HvBalloon, hostmem, + TYPE_MEMORY_BACKEND, HostMemoryBackend *), + DEFINE_PROP_UINT64(HV_BALLOON_ADDR_PROP, HvBalloon, addr, 0), + + DEFINE_PROP_END_OF_LIST(), +}; + +static void hv_balloon_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + VMBusDeviceClass *vdc = VMBUS_DEVICE_CLASS(klass); + MemoryDeviceClass *mdc = MEMORY_DEVICE_CLASS(klass); + + device_class_set_props(dc, hv_balloon_properties); + qemu_uuid_parse(HV_BALLOON_GUID, &vdc->classid); + set_bit(DEVICE_CATEGORY_MISC, dc->categories); + + vdc->vmdev_realize = hv_balloon_vmdev_realize; + vdc->vmdev_unrealize = hv_balloon_vmdev_unrealize; + vdc->vmdev_reset = hv_balloon_vmdev_reset; + vdc->open_channel = hv_balloon_vmdev_open_channel; + vdc->close_channel = hv_balloon_vmdev_close_channel; + vdc->chan_notify_cb = hv_balloon_vmdev_chan_notify; + + mdc->get_addr = hv_balloon_md_get_addr; + mdc->set_addr = hv_balloon_md_set_addr; + mdc->get_plugged_size = memory_device_get_region_size; + mdc->get_memory_region = hv_balloon_md_get_memory_region; + mdc->decide_memslots = hv_balloon_decide_memslots; + mdc->get_memslots = hv_balloon_get_memslots; + mdc->fill_device_info = hv_balloon_md_fill_device_info; +} diff --git a/hw/hyperv/meson.build b/hw/hyperv/meson.build index b43f119ea5..d3d2668c71 100644 --- a/hw/hyperv/meson.build +++ b/hw/hyperv/meson.build @@ -2,3 +2,4 @@ specific_ss.add(when: 'CONFIG_HYPERV', if_true: files('hyperv.c')) specific_ss.add(when: 'CONFIG_HYPERV_TESTDEV', if_true: files('hyperv_testdev.c')) specific_ss.add(when: 'CONFIG_VMBUS', if_true: files('vmbus.c')) specific_ss.add(when: 'CONFIG_SYNDBG', if_true: files('syndbg.c')) +specific_ss.add(when: 'CONFIG_HV_BALLOON', if_true: files('hv-balloon.c', 'hv-balloon-page_range_tree.c', 'hv-balloon-our_range_memslots.c'), if_false: files('hv-balloon-stub.c')) diff --git a/hw/hyperv/trace-events b/hw/hyperv/trace-events index b4c35ca8e3..7963c215b1 100644 --- a/hw/hyperv/trace-events +++ b/hw/hyperv/trace-events @@ -16,3 +16,21 @@ vmbus_gpadl_torndown(uint32_t gpadl_id) "gpadl #%d" vmbus_open_channel(uint32_t chan_id, uint32_t gpadl_id, uint32_t target_vp) "channel #%d gpadl #%d target vp %d" vmbus_channel_open(uint32_t chan_id, uint32_t status) "channel #%d status %d" vmbus_close_channel(uint32_t chan_id) "channel #%d" + +# hv-balloon +hv_balloon_state_change(const char *tostr) "-> %s" +hv_balloon_incoming_version(uint16_t major, uint16_t minor) "incoming proto version %u.%u" +hv_balloon_incoming_caps(uint32_t caps) "incoming caps 0x%x" +hv_balloon_outgoing_unballoon(uint32_t trans_id, uint64_t count, uint64_t start, uint64_t rempages) "posting unballoon %"PRIu32" for %"PRIu64" @ 0x%"PRIx64", remaining %"PRIu64 +hv_balloon_incoming_unballoon(uint32_t trans_id) "incoming unballoon response %"PRIu32 +hv_balloon_outgoing_hot_add(uint32_t trans_id, uint64_t count, uint64_t start) "posting hot add %"PRIu32" for %"PRIu64" @ 0x%"PRIx64 +hv_balloon_incoming_hot_add(uint32_t trans_id, uint32_t result, uint32_t count) "incoming hot add response %"PRIu32", result %"PRIu32", count %"PRIu32 +hv_balloon_outgoing_balloon(uint32_t trans_id, uint64_t count, uint64_t rempages) "posting balloon %"PRIu32" for %"PRIu64", remaining %"PRIu64 +hv_balloon_incoming_balloon(uint32_t trans_id, uint32_t range_count, uint32_t more_pages) "incoming balloon response %"PRIu32", ranges %"PRIu32", more %"PRIu32 +hv_balloon_our_range_add(uint64_t count, uint64_t start) "adding our range %"PRIu64" @ 0x%"PRIx64 +hv_balloon_remove_response(uint64_t count, uint64_t start, unsigned int both) "processing remove response range %"PRIu64" @ 0x%"PRIx64", both %u" +hv_balloon_remove_response_hole(uint64_t counthole, uint64_t starthole, uint64_t countrange, uint64_t startrange, uint64_t starthpr, unsigned int both) "response range hole %"PRIu64" @ 0x%"PRIx64" from range %"PRIu64" @ 0x%"PRIx64", before our start 0x%"PRIx64", both %u" +hv_balloon_remove_response_common(uint64_t countcommon, uint64_t startcommon, uint64_t countrange, uint64_t startrange, uint64_t counthpr, uint64_t starthpr, uint64_t removed, unsigned int both) "response common range %"PRIu64" @ 0x%"PRIx64" from range %"PRIu64" @ 0x%"PRIx64" with our %"PRIu64" @ 0x%"PRIx64", removed %"PRIu64", both %u" +hv_balloon_remove_response_remainder(uint64_t count, uint64_t start, unsigned int both) "remove response remaining range %"PRIu64" @ 0x%"PRIx64", both %u" +hv_balloon_map_slot(unsigned int idx, unsigned int total_slots, uint64_t offset) "mapping memslot %u / %u @ 0x%"PRIx64 +hv_balloon_unmap_slot(unsigned int idx, unsigned int total_slots, uint64_t offset) "unmapping memslot %u / %u @ 0x%"PRIx64 diff --git a/hw/hyperv/vmbus.c b/hw/hyperv/vmbus.c index 271289f902..c64eaa5a46 100644 --- a/hw/hyperv/vmbus.c +++ b/hw/hyperv/vmbus.c @@ -2271,7 +2271,7 @@ static void vmbus_dev_realize(DeviceState *dev, Error **errp) VMBus *vmbus = VMBUS(qdev_get_parent_bus(dev)); BusChild *child; Error *err = NULL; - char idstr[UUID_FMT_LEN + 1]; + char idstr[UUID_STR_LEN]; assert(!qemu_uuid_is_null(&vdev->instanceid)); @@ -2467,7 +2467,7 @@ static char *vmbus_get_dev_path(DeviceState *dev) static char *vmbus_get_fw_dev_path(DeviceState *dev) { VMBusDevice *vdev = VMBUS_DEVICE(dev); - char uuid[UUID_FMT_LEN + 1]; + char uuid[UUID_STR_LEN]; qemu_uuid_unparse(&vdev->instanceid, uuid); return g_strdup_printf("%s@%s", qdev_fw_name(dev), uuid); diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig index 94772c726b..55850791df 100644 --- a/hw/i386/Kconfig +++ b/hw/i386/Kconfig @@ -45,6 +45,7 @@ config PC select ACPI_VMGENID select VIRTIO_PMEM_SUPPORTED select VIRTIO_MEM_SUPPORTED + select HV_BALLOON_SUPPORTED config PC_PCI bool diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c index 7965415b47..4203144da9 100644 --- a/hw/i386/amd_iommu.c +++ b/hw/i386/amd_iommu.c @@ -1450,6 +1450,10 @@ static AddressSpace *amdvi_host_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &iommu_as[devfn]->as; } +static const PCIIOMMUOps amdvi_iommu_ops = { + .get_address_space = amdvi_host_dma_iommu, +}; + static const MemoryRegionOps mmio_mem_ops = { .read = amdvi_mmio_read, .write = amdvi_mmio_write, @@ -1581,7 +1585,7 @@ static void amdvi_sysbus_realize(DeviceState *dev, Error **errp) AMDVI_MMIO_SIZE); memory_region_add_subregion(get_system_memory(), AMDVI_BASE_ADDR, &s->mmio); - pci_setup_iommu(bus, amdvi_host_dma_iommu, s); + pci_setup_iommu(bus, &amdvi_iommu_ops, s); amdvi_init(s); } diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c index 1c6c18622f..5085a6fee3 100644 --- a/hw/i386/intel_iommu.c +++ b/hw/i386/intel_iommu.c @@ -1045,18 +1045,35 @@ static dma_addr_t vtd_get_iova_pgtbl_base(IntelIOMMUState *s, * Rsvd field masks for spte: * vtd_spte_rsvd 4k pages * vtd_spte_rsvd_large large pages + * + * We support only 3-level and 4-level page tables (see vtd_init() which + * sets only VTD_CAP_SAGAW_39bit and maybe VTD_CAP_SAGAW_48bit bits in s->cap). */ -static uint64_t vtd_spte_rsvd[5]; -static uint64_t vtd_spte_rsvd_large[5]; +#define VTD_SPTE_RSVD_LEN 5 +static uint64_t vtd_spte_rsvd[VTD_SPTE_RSVD_LEN]; +static uint64_t vtd_spte_rsvd_large[VTD_SPTE_RSVD_LEN]; static bool vtd_slpte_nonzero_rsvd(uint64_t slpte, uint32_t level) { - uint64_t rsvd_mask = vtd_spte_rsvd[level]; + uint64_t rsvd_mask; + + /* + * We should have caught a guest-mis-programmed level earlier, + * via vtd_is_level_supported. + */ + assert(level < VTD_SPTE_RSVD_LEN); + /* + * Zero level doesn't exist. The smallest level is VTD_SL_PT_LEVEL=1 and + * checked by vtd_is_last_slpte(). + */ + assert(level); if ((level == VTD_SL_PD_LEVEL || level == VTD_SL_PDP_LEVEL) && (slpte & VTD_SL_PT_PAGE_SIZE_MASK)) { /* large page */ rsvd_mask = vtd_spte_rsvd_large[level]; + } else { + rsvd_mask = vtd_spte_rsvd[level]; } return slpte & rsvd_mask; @@ -4088,6 +4105,10 @@ static AddressSpace *vtd_host_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &vtd_as->as; } +static PCIIOMMUOps vtd_iommu_ops = { + .get_address_space = vtd_host_dma_iommu, +}; + static bool vtd_decide_config(IntelIOMMUState *s, Error **errp) { X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(s); @@ -4210,7 +4231,7 @@ static void vtd_realize(DeviceState *dev, Error **errp) s->vtd_address_spaces = g_hash_table_new_full(vtd_as_hash, vtd_as_equal, g_free, g_free); vtd_init(s); - pci_setup_iommu(bus, vtd_host_dma_iommu, dev); + pci_setup_iommu(bus, &vtd_iommu_ops, dev); /* Pseudo address space under root PCI bus. */ x86ms->ioapic_as = vtd_host_dma_iommu(bus, s, Q35_PSEUDO_DEVFN_IOAPIC); qemu_add_machine_init_done_notifier(&vtd_machine_done_notify); diff --git a/hw/i386/kvm/xen_evtchn.c b/hw/i386/kvm/xen_evtchn.c index a731738411..b2b4be9983 100644 --- a/hw/i386/kvm/xen_evtchn.c +++ b/hw/i386/kvm/xen_evtchn.c @@ -490,6 +490,12 @@ int xen_evtchn_set_callback_param(uint64_t param) break; } + /* If the guest has set a per-vCPU callback vector, prefer that. */ + if (gsi && kvm_xen_has_vcpu_callback_vector()) { + in_kernel = kvm_xen_has_cap(EVTCHN_SEND); + gsi = 0; + } + if (!ret) { /* If vector delivery was turned *off* then tell the kernel */ if ((s->callback_param >> CALLBACK_VIA_TYPE_SHIFT) == @@ -1129,6 +1135,7 @@ int xen_evtchn_reset_op(struct evtchn_reset *reset) return -ESRCH; } + QEMU_IOTHREAD_LOCK_GUARD(); return xen_evtchn_soft_reset(); } diff --git a/hw/i386/kvm/xen_gnttab.c b/hw/i386/kvm/xen_gnttab.c index 21c30e3659..839ec920a1 100644 --- a/hw/i386/kvm/xen_gnttab.c +++ b/hw/i386/kvm/xen_gnttab.c @@ -541,7 +541,5 @@ int xen_gnttab_reset(void) s->entries.v1[GNTTAB_RESERVED_XENSTORE].flags = GTF_permit_access; s->entries.v1[GNTTAB_RESERVED_XENSTORE].frame = XEN_SPECIAL_PFN(XENSTORE); - memset(s->map_track, 0, s->max_frames * ENTRIES_PER_FRAME_V1); - return 0; } diff --git a/hw/i386/kvm/xen_xenstore.c b/hw/i386/kvm/xen_xenstore.c index 660d0b72f9..8e716a7009 100644 --- a/hw/i386/kvm/xen_xenstore.c +++ b/hw/i386/kvm/xen_xenstore.c @@ -1357,10 +1357,12 @@ static void fire_watch_cb(void *opaque, const char *path, const char *token) } else { deliver_watch(s, path, token); /* - * If the message was queued because there was already ring activity, - * no need to wake the guest. But if not, we need to send the evtchn. + * Attempt to queue the message into the actual ring, and send + * the event channel notification if any bytes are copied. */ - xen_be_evtchn_notify(s->eh, s->be_port); + if (s->rsp_pending && put_rsp(s) > 0) { + xen_be_evtchn_notify(s->eh, s->be_port); + } } } diff --git a/hw/i386/pc.c b/hw/i386/pc.c index 6031234a73..1aef21aa2c 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -27,6 +27,7 @@ #include "hw/i386/pc.h" #include "hw/char/serial.h" #include "hw/char/parallel.h" +#include "hw/hyperv/hv-balloon.h" #include "hw/i386/fw_cfg.h" #include "hw/i386/vmport.h" #include "sysemu/cpus.h" @@ -57,6 +58,7 @@ #include "hw/i386/kvm/xen_evtchn.h" #include "hw/i386/kvm/xen_gnttab.h" #include "hw/i386/kvm/xen_xenstore.h" +#include "hw/mem/memory-device.h" #include "e820_memory_layout.h" #include "trace.h" #include CONFIG_DEVICES @@ -1422,6 +1424,21 @@ static void pc_memory_unplug(HotplugHandler *hotplug_dev, error_propagate(errp, local_err); } +static void pc_hv_balloon_pre_plug(HotplugHandler *hotplug_dev, + DeviceState *dev, Error **errp) +{ + /* The vmbus handler has no hotplug handler; we should never end up here. */ + g_assert(!dev->hotplugged); + memory_device_pre_plug(MEMORY_DEVICE(dev), MACHINE(hotplug_dev), NULL, + errp); +} + +static void pc_hv_balloon_plug(HotplugHandler *hotplug_dev, + DeviceState *dev, Error **errp) +{ + memory_device_plug(MEMORY_DEVICE(dev), MACHINE(hotplug_dev)); +} + static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev, Error **errp) { @@ -1452,6 +1469,8 @@ static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev, return; } pcms->iommu = dev; + } else if (object_dynamic_cast(OBJECT(dev), TYPE_HV_BALLOON)) { + pc_hv_balloon_pre_plug(hotplug_dev, dev, errp); } } @@ -1464,6 +1483,8 @@ static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev, x86_cpu_plug(hotplug_dev, dev, errp); } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_PCI)) { virtio_md_pci_plug(VIRTIO_MD_PCI(dev), MACHINE(hotplug_dev), errp); + } else if (object_dynamic_cast(OBJECT(dev), TYPE_HV_BALLOON)) { + pc_hv_balloon_plug(hotplug_dev, dev, errp); } } @@ -1505,6 +1526,7 @@ static HotplugHandler *pc_get_hotplug_handler(MachineState *machine, object_dynamic_cast(OBJECT(dev), TYPE_CPU) || object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_PCI) || object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) || + object_dynamic_cast(OBJECT(dev), TYPE_HV_BALLOON) || object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE)) { return HOTPLUG_HANDLER(machine); } diff --git a/hw/mem/memory-device.c b/hw/mem/memory-device.c index ae38f48f16..e0704b8dc3 100644 --- a/hw/mem/memory-device.c +++ b/hw/mem/memory-device.c @@ -20,6 +20,22 @@ #include "exec/address-spaces.h" #include "trace.h" +static bool memory_device_is_empty(const MemoryDeviceState *md) +{ + const MemoryDeviceClass *mdc = MEMORY_DEVICE_GET_CLASS(md); + Error *local_err = NULL; + MemoryRegion *mr; + + /* dropping const here is fine as we don't touch the memory region */ + mr = mdc->get_memory_region((MemoryDeviceState *)md, &local_err); + if (local_err) { + /* Not empty, we'll report errors later when ontaining the MR again. */ + error_free(local_err); + return false; + } + return !mr; +} + static gint memory_device_addr_sort(gconstpointer a, gconstpointer b) { const MemoryDeviceState *md_a = MEMORY_DEVICE(a); @@ -220,12 +236,6 @@ static uint64_t memory_device_get_free_addr(MachineState *ms, return 0; } - if (!QEMU_IS_ALIGNED(size, align)) { - error_setg(errp, "backend memory size must be multiple of 0x%" - PRIx64, align); - return 0; - } - if (hint) { if (range_init(&new, *hint, size) || !range_contains_range(&as, &new)) { error_setg(errp, "can't add memory device [0x%" PRIx64 ":0x%" PRIx64 @@ -249,6 +259,10 @@ static uint64_t memory_device_get_free_addr(MachineState *ms, uint64_t next_addr; Range tmp; + if (memory_device_is_empty(md)) { + continue; + } + range_init_nofail(&tmp, mdc->get_addr(md), memory_device_get_region_size(md, &error_abort)); @@ -292,6 +306,7 @@ MemoryDeviceInfoList *qmp_memory_device_list(void) const MemoryDeviceClass *mdc = MEMORY_DEVICE_GET_CLASS(item->data); MemoryDeviceInfo *info = g_new0(MemoryDeviceInfo, 1); + /* Let's query infotmation even for empty memory devices. */ mdc->fill_device_info(md, info); QAPI_LIST_APPEND(tail, info); @@ -311,7 +326,7 @@ static int memory_device_plugged_size(Object *obj, void *opaque) const MemoryDeviceState *md = MEMORY_DEVICE(obj); const MemoryDeviceClass *mdc = MEMORY_DEVICE_GET_CLASS(obj); - if (dev->realized) { + if (dev->realized && !memory_device_is_empty(md)) { *size += mdc->get_plugged_size(md, &error_abort); } } @@ -337,6 +352,11 @@ void memory_device_pre_plug(MemoryDeviceState *md, MachineState *ms, uint64_t addr, align = 0; MemoryRegion *mr; + /* We support empty memory devices even without device memory. */ + if (memory_device_is_empty(md)) { + return; + } + if (!ms->device_memory) { error_setg(errp, "the configuration is not prepared for memory devices" " (e.g., for memory hotplug), consider specifying the" @@ -380,10 +400,17 @@ out: void memory_device_plug(MemoryDeviceState *md, MachineState *ms) { const MemoryDeviceClass *mdc = MEMORY_DEVICE_GET_CLASS(md); - const unsigned int memslots = memory_device_get_memslots(md); - const uint64_t addr = mdc->get_addr(md); + unsigned int memslots; + uint64_t addr; MemoryRegion *mr; + if (memory_device_is_empty(md)) { + return; + } + + memslots = memory_device_get_memslots(md); + addr = mdc->get_addr(md); + /* * We expect that a previous call to memory_device_pre_plug() succeeded, so * it can't fail at this point. @@ -408,6 +435,10 @@ void memory_device_unplug(MemoryDeviceState *md, MachineState *ms) const unsigned int memslots = memory_device_get_memslots(md); MemoryRegion *mr; + if (memory_device_is_empty(md)) { + return; + } + /* * We expect that a previous call to memory_device_pre_plug() succeeded, so * it can't fail at this point. diff --git a/hw/pci-host/astro.c b/hw/pci-host/astro.c index 4b2d7caf2d..84e0ca14ac 100644 --- a/hw/pci-host/astro.c +++ b/hw/pci-host/astro.c @@ -345,6 +345,10 @@ static AddressSpace *elroy_pcihost_set_iommu(PCIBus *bus, void *opaque, return &s->astro->iommu_as; } +static const PCIIOMMUOps elroy_pcihost_iommu_ops = { + .get_address_space = elroy_pcihost_set_iommu, +}; + /* * Encoding in IOSAPIC: * base_addr == 0xfffa0000, we want to get 0xa0ff0000. @@ -834,7 +838,7 @@ static void astro_realize(DeviceState *obj, Error **errp) &elroy->pci_io); /* Host memory as seen from the PCI side, via the IOMMU. */ - pci_setup_iommu(PCI_HOST_BRIDGE(elroy)->bus, elroy_pcihost_set_iommu, + pci_setup_iommu(PCI_HOST_BRIDGE(elroy)->bus, &elroy_pcihost_iommu_ops, elroy); } } diff --git a/hw/pci-host/designware.c b/hw/pci-host/designware.c index 6f5442f108..f477f97847 100644 --- a/hw/pci-host/designware.c +++ b/hw/pci-host/designware.c @@ -663,6 +663,10 @@ static AddressSpace *designware_pcie_host_set_iommu(PCIBus *bus, void *opaque, return &s->pci.address_space; } +static const PCIIOMMUOps designware_iommu_ops = { + .get_address_space = designware_pcie_host_set_iommu, +}; + static void designware_pcie_host_realize(DeviceState *dev, Error **errp) { PCIHostState *pci = PCI_HOST_BRIDGE(dev); @@ -705,7 +709,7 @@ static void designware_pcie_host_realize(DeviceState *dev, Error **errp) address_space_init(&s->pci.address_space, &s->pci.address_space_root, "pcie-bus-address-space"); - pci_setup_iommu(pci->bus, designware_pcie_host_set_iommu, s); + pci_setup_iommu(pci->bus, &designware_iommu_ops, s); qdev_realize(DEVICE(&s->root), BUS(pci->bus), &error_fatal); } diff --git a/hw/pci-host/dino.c b/hw/pci-host/dino.c index 82503229fa..5b0947a16c 100644 --- a/hw/pci-host/dino.c +++ b/hw/pci-host/dino.c @@ -354,6 +354,10 @@ static AddressSpace *dino_pcihost_set_iommu(PCIBus *bus, void *opaque, return &s->bm_as; } +static const PCIIOMMUOps dino_iommu_ops = { + .get_address_space = dino_pcihost_set_iommu, +}; + /* * Dino interrupts are connected as shown on Page 78, Table 23 * (Little-endian bit numbers) @@ -481,7 +485,7 @@ static void dino_pcihost_init(Object *obj) g_free(name); } - pci_setup_iommu(phb->bus, dino_pcihost_set_iommu, s); + pci_setup_iommu(phb->bus, &dino_iommu_ops, s); sysbus_init_mmio(sbd, &s->this_mem); diff --git a/hw/pci-host/pnv_phb3.c b/hw/pci-host/pnv_phb3.c index c5e58f4086..2a74dbe45f 100644 --- a/hw/pci-host/pnv_phb3.c +++ b/hw/pci-host/pnv_phb3.c @@ -968,6 +968,10 @@ static AddressSpace *pnv_phb3_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &ds->dma_as; } +static PCIIOMMUOps pnv_phb3_iommu_ops = { + .get_address_space = pnv_phb3_dma_iommu, +}; + static void pnv_phb3_instance_init(Object *obj) { PnvPHB3 *phb = PNV_PHB3(obj); @@ -1012,7 +1016,7 @@ void pnv_phb3_bus_init(DeviceState *dev, PnvPHB3 *phb) object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id, &error_abort); - pci_setup_iommu(pci->bus, pnv_phb3_dma_iommu, phb); + pci_setup_iommu(pci->bus, &pnv_phb3_iommu_ops, phb); } static void pnv_phb3_realize(DeviceState *dev, Error **errp) diff --git a/hw/pci-host/pnv_phb4.c b/hw/pci-host/pnv_phb4.c index 29cb11a5d9..37c7afc18c 100644 --- a/hw/pci-host/pnv_phb4.c +++ b/hw/pci-host/pnv_phb4.c @@ -1518,6 +1518,10 @@ static void pnv_phb4_xscom_realize(PnvPHB4 *phb) &phb->phb_regs_mr); } +static PCIIOMMUOps pnv_phb4_iommu_ops = { + .get_address_space = pnv_phb4_dma_iommu, +}; + static void pnv_phb4_instance_init(Object *obj) { PnvPHB4 *phb = PNV_PHB4(obj); @@ -1557,7 +1561,7 @@ void pnv_phb4_bus_init(DeviceState *dev, PnvPHB4 *phb) object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id, &error_abort); - pci_setup_iommu(pci->bus, pnv_phb4_dma_iommu, phb); + pci_setup_iommu(pci->bus, &pnv_phb4_iommu_ops, phb); pci->bus->flags |= PCI_BUS_EXTENDED_CONFIG_SPACE; } diff --git a/hw/pci-host/ppce500.c b/hw/pci-host/ppce500.c index 38814247f2..453a4e6ed3 100644 --- a/hw/pci-host/ppce500.c +++ b/hw/pci-host/ppce500.c @@ -435,6 +435,10 @@ static AddressSpace *e500_pcihost_set_iommu(PCIBus *bus, void *opaque, return &s->bm_as; } +static const PCIIOMMUOps ppce500_iommu_ops = { + .get_address_space = e500_pcihost_set_iommu, +}; + static void e500_pcihost_realize(DeviceState *dev, Error **errp) { SysBusDevice *sbd = SYS_BUS_DEVICE(dev); @@ -469,7 +473,7 @@ static void e500_pcihost_realize(DeviceState *dev, Error **errp) memory_region_init(&s->bm, OBJECT(s), "bm-e500", UINT64_MAX); memory_region_add_subregion(&s->bm, 0x0, &s->busmem); address_space_init(&s->bm_as, &s->bm, "pci-bm"); - pci_setup_iommu(b, e500_pcihost_set_iommu, s); + pci_setup_iommu(b, &ppce500_iommu_ops, s); pci_create_simple(b, 0, "e500-host-bridge"); diff --git a/hw/pci-host/raven.c b/hw/pci-host/raven.c index 9a11ac4b2b..86c3a49087 100644 --- a/hw/pci-host/raven.c +++ b/hw/pci-host/raven.c @@ -223,6 +223,10 @@ static AddressSpace *raven_pcihost_set_iommu(PCIBus *bus, void *opaque, return &s->bm_as; } +static const PCIIOMMUOps raven_iommu_ops = { + .get_address_space = raven_pcihost_set_iommu, +}; + static void raven_change_gpio(void *opaque, int n, int level) { PREPPCIState *s = opaque; @@ -320,7 +324,7 @@ static void raven_pcihost_initfn(Object *obj) memory_region_add_subregion(&s->bm, 0 , &s->bm_pci_memory_alias); memory_region_add_subregion(&s->bm, 0x80000000, &s->bm_ram_alias); address_space_init(&s->bm_as, &s->bm, "raven-bm"); - pci_setup_iommu(&s->pci_bus, raven_pcihost_set_iommu, s); + pci_setup_iommu(&s->pci_bus, &raven_iommu_ops, s); h->bus = &s->pci_bus; diff --git a/hw/pci-host/sabre.c b/hw/pci-host/sabre.c index dcb2e230b6..d0851b48b0 100644 --- a/hw/pci-host/sabre.c +++ b/hw/pci-host/sabre.c @@ -112,6 +112,10 @@ static AddressSpace *sabre_pci_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &is->iommu_as; } +static const PCIIOMMUOps sabre_iommu_ops = { + .get_address_space = sabre_pci_dma_iommu, +}; + static void sabre_config_write(void *opaque, hwaddr addr, uint64_t val, unsigned size) { @@ -384,7 +388,7 @@ static void sabre_realize(DeviceState *dev, Error **errp) /* IOMMU */ memory_region_add_subregion_overlap(&s->sabre_config, 0x200, sysbus_mmio_get_region(SYS_BUS_DEVICE(s->iommu), 0), 1); - pci_setup_iommu(phb->bus, sabre_pci_dma_iommu, s->iommu); + pci_setup_iommu(phb->bus, &sabre_iommu_ops, s->iommu); /* APB secondary busses */ pci_dev = pci_new_multifunction(PCI_DEVFN(1, 0), TYPE_SIMBA_PCI_BRIDGE); diff --git a/hw/pci/pci.c b/hw/pci/pci.c index 885c04b6f5..c49417abb2 100644 --- a/hw/pci/pci.c +++ b/hw/pci/pci.c @@ -2678,7 +2678,7 @@ AddressSpace *pci_device_iommu_address_space(PCIDevice *dev) PCIBus *iommu_bus = bus; uint8_t devfn = dev->devfn; - while (iommu_bus && !iommu_bus->iommu_fn && iommu_bus->parent_dev) { + while (iommu_bus && !iommu_bus->iommu_ops && iommu_bus->parent_dev) { PCIBus *parent_bus = pci_get_bus(iommu_bus->parent_dev); /* @@ -2717,15 +2717,23 @@ AddressSpace *pci_device_iommu_address_space(PCIDevice *dev) iommu_bus = parent_bus; } - if (!pci_bus_bypass_iommu(bus) && iommu_bus && iommu_bus->iommu_fn) { - return iommu_bus->iommu_fn(bus, iommu_bus->iommu_opaque, devfn); + if (!pci_bus_bypass_iommu(bus) && iommu_bus->iommu_ops) { + return iommu_bus->iommu_ops->get_address_space(bus, + iommu_bus->iommu_opaque, devfn); } return &address_space_memory; } -void pci_setup_iommu(PCIBus *bus, PCIIOMMUFunc fn, void *opaque) +void pci_setup_iommu(PCIBus *bus, const PCIIOMMUOps *ops, void *opaque) { - bus->iommu_fn = fn; + /* + * If called, pci_setup_iommu() should provide a minimum set of + * useful callbacks for the bus. + */ + assert(ops); + assert(ops->get_address_space); + + bus->iommu_ops = ops; bus->iommu_opaque = opaque; } diff --git a/hw/ppc/ppc440_pcix.c b/hw/ppc/ppc440_pcix.c index 672090de94..df4ee374d0 100644 --- a/hw/ppc/ppc440_pcix.c +++ b/hw/ppc/ppc440_pcix.c @@ -449,6 +449,10 @@ static AddressSpace *ppc440_pcix_set_iommu(PCIBus *b, void *opaque, int devfn) return &s->bm_as; } +static const PCIIOMMUOps ppc440_iommu_ops = { + .get_address_space = ppc440_pcix_set_iommu, +}; + /* * Some guests on sam460ex write all kinds of garbage here such as * missing enable bit and low bits set and still expect this to work @@ -503,7 +507,7 @@ static void ppc440_pcix_realize(DeviceState *dev, Error **errp) memory_region_init(&s->bm, OBJECT(s), "bm-ppc440-pcix", UINT64_MAX); memory_region_add_subregion(&s->bm, 0x0, &s->busmem); address_space_init(&s->bm_as, &s->bm, "pci-bm"); - pci_setup_iommu(h->bus, ppc440_pcix_set_iommu, s); + pci_setup_iommu(h->bus, &ppc440_iommu_ops, s); memory_region_init(&s->container, OBJECT(s), "pci-container", PCI_ALL_SIZE); memory_region_init_io(&h->conf_mem, OBJECT(s), &ppc440_pcix_host_conf_ops, diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c index 370c5a90f2..a27024e45a 100644 --- a/hw/ppc/spapr_pci.c +++ b/hw/ppc/spapr_pci.c @@ -780,6 +780,10 @@ static AddressSpace *spapr_pci_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &phb->iommu_as; } +static const PCIIOMMUOps spapr_iommu_ops = { + .get_address_space = spapr_pci_dma_iommu, +}; + static char *spapr_phb_vfio_get_loc_code(SpaprPhbState *sphb, PCIDevice *pdev) { g_autofree char *path = NULL; @@ -1978,7 +1982,7 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp) memory_region_add_subregion(&sphb->iommu_root, SPAPR_PCI_MSI_WINDOW, &sphb->msiwindow); - pci_setup_iommu(bus, spapr_pci_dma_iommu, sphb); + pci_setup_iommu(bus, &spapr_iommu_ops, sphb); pci_bus_set_route_irq_fn(bus, spapr_route_intx_pin_to_irq); diff --git a/hw/ppc/spapr_pci_vfio.c b/hw/ppc/spapr_pci_vfio.c index 9016720547..f283f7e38d 100644 --- a/hw/ppc/spapr_pci_vfio.c +++ b/hw/ppc/spapr_pci_vfio.c @@ -18,14 +18,112 @@ */ #include "qemu/osdep.h" +#include <sys/ioctl.h> #include <linux/vfio.h> #include "hw/ppc/spapr.h" #include "hw/pci-host/spapr.h" #include "hw/pci/msix.h" #include "hw/pci/pci_device.h" -#include "hw/vfio/vfio.h" +#include "hw/vfio/vfio-common.h" #include "qemu/error-report.h" +/* + * Interfaces for IBM EEH (Enhanced Error Handling) + */ +static bool vfio_eeh_container_ok(VFIOContainer *container) +{ + /* + * As of 2016-03-04 (linux-4.5) the host kernel EEH/VFIO + * implementation is broken if there are multiple groups in a + * container. The hardware works in units of Partitionable + * Endpoints (== IOMMU groups) and the EEH operations naively + * iterate across all groups in the container, without any logic + * to make sure the groups have their state synchronized. For + * certain operations (ENABLE) that might be ok, until an error + * occurs, but for others (GET_STATE) it's clearly broken. + */ + + /* + * XXX Once fixed kernels exist, test for them here + */ + + if (QLIST_EMPTY(&container->group_list)) { + return false; + } + + if (QLIST_NEXT(QLIST_FIRST(&container->group_list), container_next)) { + return false; + } + + return true; +} + +static int vfio_eeh_container_op(VFIOContainer *container, uint32_t op) +{ + struct vfio_eeh_pe_op pe_op = { + .argsz = sizeof(pe_op), + .op = op, + }; + int ret; + + if (!vfio_eeh_container_ok(container)) { + error_report("vfio/eeh: EEH_PE_OP 0x%x: " + "kernel requires a container with exactly one group", op); + return -EPERM; + } + + ret = ioctl(container->fd, VFIO_EEH_PE_OP, &pe_op); + if (ret < 0) { + error_report("vfio/eeh: EEH_PE_OP 0x%x failed: %m", op); + return -errno; + } + + return ret; +} + +static VFIOContainer *vfio_eeh_as_container(AddressSpace *as) +{ + VFIOAddressSpace *space = vfio_get_address_space(as); + VFIOContainer *container = NULL; + + if (QLIST_EMPTY(&space->containers)) { + /* No containers to act on */ + goto out; + } + + container = QLIST_FIRST(&space->containers); + + if (QLIST_NEXT(container, next)) { + /* + * We don't yet have logic to synchronize EEH state across + * multiple containers + */ + container = NULL; + goto out; + } + +out: + vfio_put_address_space(space); + return container; +} + +static bool vfio_eeh_as_ok(AddressSpace *as) +{ + VFIOContainer *container = vfio_eeh_as_container(as); + + return (container != NULL) && vfio_eeh_container_ok(container); +} + +static int vfio_eeh_as_op(AddressSpace *as, uint32_t op) +{ + VFIOContainer *container = vfio_eeh_as_container(as); + + if (!container) { + return -ENODEV; + } + return vfio_eeh_container_op(container, op); +} + bool spapr_phb_eeh_available(SpaprPhbState *sphb) { return vfio_eeh_as_ok(&sphb->iommu_as); diff --git a/hw/remote/iommu.c b/hw/remote/iommu.c index 1391dd712c..7c56aad0fc 100644 --- a/hw/remote/iommu.c +++ b/hw/remote/iommu.c @@ -100,6 +100,10 @@ static void remote_iommu_finalize(Object *obj) iommu->elem_by_devfn = NULL; } +static const PCIIOMMUOps remote_iommu_ops = { + .get_address_space = remote_iommu_find_add_as, +}; + void remote_iommu_setup(PCIBus *pci_bus) { RemoteIommu *iommu = NULL; @@ -108,7 +112,7 @@ void remote_iommu_setup(PCIBus *pci_bus) iommu = REMOTE_IOMMU(object_new(TYPE_REMOTE_IOMMU)); - pci_setup_iommu(pci_bus, remote_iommu_find_add_as, iommu); + pci_setup_iommu(pci_bus, &remote_iommu_ops, iommu); object_property_add_child(OBJECT(pci_bus), "remote-iommu", OBJECT(iommu)); diff --git a/hw/rtc/mc146818rtc.c b/hw/rtc/mc146818rtc.c index c27c362db9..2d391a8396 100644 --- a/hw/rtc/mc146818rtc.c +++ b/hw/rtc/mc146818rtc.c @@ -599,7 +599,7 @@ static void rtc_get_time(MC146818RtcState *s, struct tm *tm) static void rtc_set_time(MC146818RtcState *s) { - struct tm tm; + struct tm tm = {}; g_autofree const char *qom_path = object_get_canonical_path(OBJECT(s)); rtc_get_time(s, &tm); diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c index 2ca36f9f3b..347580ebac 100644 --- a/hw/s390x/s390-pci-bus.c +++ b/hw/s390x/s390-pci-bus.c @@ -652,6 +652,10 @@ static AddressSpace *s390_pci_dma_iommu(PCIBus *bus, void *opaque, int devfn) return &iommu->as; } +static const PCIIOMMUOps s390_iommu_ops = { + .get_address_space = s390_pci_dma_iommu, +}; + static uint8_t set_ind_atomic(uint64_t ind_loc, uint8_t to_be_set) { uint8_t expected, actual; @@ -839,7 +843,7 @@ static void s390_pcihost_realize(DeviceState *dev, Error **errp) b = pci_register_root_bus(dev, NULL, s390_pci_set_irq, s390_pci_map_irq, NULL, get_system_memory(), get_system_io(), 0, 64, TYPE_PCI_BUS); - pci_setup_iommu(b, s390_pci_dma_iommu, s); + pci_setup_iommu(b, &s390_iommu_ops, s); bus = BUS(b); qbus_set_hotplug_handler(bus, OBJECT(dev)); @@ -1058,7 +1062,7 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev, pdev = PCI_DEVICE(dev); pci_bridge_map_irq(pb, dev->id, s390_pci_map_irq); - pci_setup_iommu(&pb->sec_bus, s390_pci_dma_iommu, s); + pci_setup_iommu(&pb->sec_bus, &s390_iommu_ops, s); qbus_set_hotplug_handler(BUS(&pb->sec_bus), OBJECT(s)); diff --git a/hw/vfio/ap.c b/hw/vfio/ap.c index 5f257bffb9..bbf69ff55a 100644 --- a/hw/vfio/ap.c +++ b/hw/vfio/ap.c @@ -14,7 +14,6 @@ #include <linux/vfio.h> #include <sys/ioctl.h> #include "qapi/error.h" -#include "hw/vfio/vfio.h" #include "hw/vfio/vfio-common.h" #include "hw/s390x/ap-device.h" #include "qemu/error-report.h" diff --git a/hw/vfio/ccw.c b/hw/vfio/ccw.c index 6623ae237b..d857bb8d0f 100644 --- a/hw/vfio/ccw.c +++ b/hw/vfio/ccw.c @@ -20,7 +20,6 @@ #include <sys/ioctl.h> #include "qapi/error.h" -#include "hw/vfio/vfio.h" #include "hw/vfio/vfio-common.h" #include "hw/s390x/s390-ccw.h" #include "hw/s390x/vfio-ccw.h" diff --git a/hw/vfio/common.c b/hw/vfio/common.c index d806057b40..e70fdf5e0c 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -26,7 +26,6 @@ #include <linux/vfio.h> #include "hw/vfio/vfio-common.h" -#include "hw/vfio/vfio.h" #include "hw/vfio/pci.h" #include "exec/address-spaces.h" #include "exec/memory.h" @@ -246,44 +245,6 @@ bool vfio_devices_all_running_and_mig_active(VFIOContainer *container) return true; } -void vfio_host_win_add(VFIOContainer *container, hwaddr min_iova, - hwaddr max_iova, uint64_t iova_pgsizes) -{ - VFIOHostDMAWindow *hostwin; - - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (ranges_overlap(hostwin->min_iova, - hostwin->max_iova - hostwin->min_iova + 1, - min_iova, - max_iova - min_iova + 1)) { - hw_error("%s: Overlapped IOMMU are not enabled", __func__); - } - } - - hostwin = g_malloc0(sizeof(*hostwin)); - - hostwin->min_iova = min_iova; - hostwin->max_iova = max_iova; - hostwin->iova_pgsizes = iova_pgsizes; - QLIST_INSERT_HEAD(&container->hostwin_list, hostwin, hostwin_next); -} - -int vfio_host_win_del(VFIOContainer *container, - hwaddr min_iova, hwaddr max_iova) -{ - VFIOHostDMAWindow *hostwin; - - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (hostwin->min_iova == min_iova && hostwin->max_iova == max_iova) { - QLIST_REMOVE(hostwin, hostwin_next); - g_free(hostwin); - return 0; - } - } - - return -1; -} - static bool vfio_listener_skipped_section(MemoryRegionSection *section) { return (!memory_region_is_ram(section->mr) && @@ -532,22 +493,6 @@ static void vfio_unregister_ram_discard_listener(VFIOContainer *container, g_free(vrdl); } -static VFIOHostDMAWindow *vfio_find_hostwin(VFIOContainer *container, - hwaddr iova, hwaddr end) -{ - VFIOHostDMAWindow *hostwin; - bool hostwin_found = false; - - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (hostwin->min_iova <= iova && end <= hostwin->max_iova) { - hostwin_found = true; - break; - } - } - - return hostwin_found ? hostwin : NULL; -} - static bool vfio_known_safe_misalignment(MemoryRegionSection *section) { MemoryRegion *mr = section->mr; @@ -626,7 +571,6 @@ static void vfio_listener_region_add(MemoryListener *listener, Int128 llend, llsize; void *vaddr; int ret; - VFIOHostDMAWindow *hostwin; Error *err = NULL; if (!vfio_listener_valid_section(section, "region_add")) { @@ -648,13 +592,6 @@ static void vfio_listener_region_add(MemoryListener *listener, goto fail; } - hostwin = vfio_find_hostwin(container, iova, end); - if (!hostwin) { - error_setg(&err, "Container %p can't map guest IOVA region" - " 0x%"HWADDR_PRIx"..0x%"HWADDR_PRIx, container, iova, end); - goto fail; - } - memory_region_ref(section->mr); if (memory_region_is_iommu(section->mr)) { @@ -693,6 +630,15 @@ static void vfio_listener_region_add(MemoryListener *listener, goto fail; } + if (container->iova_ranges) { + ret = memory_region_iommu_set_iova_ranges(giommu->iommu_mr, + container->iova_ranges, &err); + if (ret) { + g_free(giommu); + goto fail; + } + } + ret = memory_region_register_iommu_notifier(section->mr, &giommu->n, &err); if (ret) { @@ -726,7 +672,7 @@ static void vfio_listener_region_add(MemoryListener *listener, llsize = int128_sub(llend, int128_make64(iova)); if (memory_region_is_ram_device(section->mr)) { - hwaddr pgmask = (1ULL << ctz64(hostwin->iova_pgsizes)) - 1; + hwaddr pgmask = (1ULL << ctz64(container->pgsizes)) - 1; if ((iova & pgmask) || (int128_get64(llsize) & pgmask)) { trace_vfio_listener_region_add_no_dma_map( @@ -825,12 +771,8 @@ static void vfio_listener_region_del(MemoryListener *listener, if (memory_region_is_ram_device(section->mr)) { hwaddr pgmask; - VFIOHostDMAWindow *hostwin; - - hostwin = vfio_find_hostwin(container, iova, end); - assert(hostwin); /* or region_add() would have failed */ - pgmask = (1ULL << ctz64(hostwin->iova_pgsizes)) - 1; + pgmask = (1ULL << ctz64(container->pgsizes)) - 1; try_unmap = !((iova & pgmask) || (int128_get64(llsize) & pgmask)); } else if (memory_region_has_ram_discard_manager(section->mr)) { vfio_unregister_ram_discard_listener(container, section); diff --git a/hw/vfio/container.c b/hw/vfio/container.c index adc467210f..242010036a 100644 --- a/hw/vfio/container.c +++ b/hw/vfio/container.c @@ -20,20 +20,15 @@ #include "qemu/osdep.h" #include <sys/ioctl.h> -#ifdef CONFIG_KVM -#include <linux/kvm.h> -#endif #include <linux/vfio.h> #include "hw/vfio/vfio-common.h" -#include "hw/vfio/vfio.h" #include "exec/address-spaces.h" #include "exec/memory.h" #include "exec/ram_addr.h" #include "hw/hw.h" #include "qemu/error-report.h" #include "qemu/range.h" -#include "sysemu/kvm.h" #include "sysemu/reset.h" #include "trace.h" #include "qapi/error.h" @@ -205,92 +200,6 @@ int vfio_dma_map(VFIOContainer *container, hwaddr iova, return -errno; } -int vfio_container_add_section_window(VFIOContainer *container, - MemoryRegionSection *section, - Error **errp) -{ - VFIOHostDMAWindow *hostwin; - hwaddr pgsize = 0; - int ret; - - if (container->iommu_type != VFIO_SPAPR_TCE_v2_IOMMU) { - return 0; - } - - /* For now intersections are not allowed, we may relax this later */ - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (ranges_overlap(hostwin->min_iova, - hostwin->max_iova - hostwin->min_iova + 1, - section->offset_within_address_space, - int128_get64(section->size))) { - error_setg(errp, - "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existing" - "host DMA window [0x%"PRIx64",0x%"PRIx64"]", - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, - hostwin->min_iova, hostwin->max_iova); - return -EINVAL; - } - } - - ret = vfio_spapr_create_window(container, section, &pgsize); - if (ret) { - error_setg_errno(errp, -ret, "Failed to create SPAPR window"); - return ret; - } - - vfio_host_win_add(container, section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, pgsize); -#ifdef CONFIG_KVM - if (kvm_enabled()) { - VFIOGroup *group; - IOMMUMemoryRegion *iommu_mr = IOMMU_MEMORY_REGION(section->mr); - struct kvm_vfio_spapr_tce param; - struct kvm_device_attr attr = { - .group = KVM_DEV_VFIO_GROUP, - .attr = KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, - .addr = (uint64_t)(unsigned long)¶m, - }; - - if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_FD, - ¶m.tablefd)) { - QLIST_FOREACH(group, &container->group_list, container_next) { - param.groupfd = group->fd; - if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) { - error_setg_errno(errp, errno, - "vfio: failed GROUP_SET_SPAPR_TCE for " - "KVM VFIO device %d and group fd %d", - param.tablefd, param.groupfd); - return -errno; - } - trace_vfio_spapr_group_attach(param.groupfd, param.tablefd); - } - } - } -#endif - return 0; -} - -void vfio_container_del_section_window(VFIOContainer *container, - MemoryRegionSection *section) -{ - if (container->iommu_type != VFIO_SPAPR_TCE_v2_IOMMU) { - return; - } - - vfio_spapr_remove_window(container, - section->offset_within_address_space); - if (vfio_host_win_del(container, - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1) < 0) { - hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, - __func__, section->offset_within_address_space); - } -} - int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start) { int ret; @@ -355,14 +264,6 @@ int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, return ret; } -static void vfio_listener_release(VFIOContainer *container) -{ - memory_listener_unregister(&container->listener); - if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU) { - memory_listener_unregister(&container->prereg_listener); - } -} - static struct vfio_info_cap_header * vfio_get_iommu_type1_info_cap(struct vfio_iommu_type1_info *info, uint16_t id) { @@ -382,7 +283,7 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info, /* If the capability cannot be found, assume no DMA limiting */ hdr = vfio_get_iommu_type1_info_cap(info, VFIO_IOMMU_TYPE1_INFO_DMA_AVAIL); - if (hdr == NULL) { + if (!hdr) { return false; } @@ -394,6 +295,32 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info, return true; } +static bool vfio_get_info_iova_range(struct vfio_iommu_type1_info *info, + VFIOContainer *container) +{ + struct vfio_info_cap_header *hdr; + struct vfio_iommu_type1_info_cap_iova_range *cap; + + hdr = vfio_get_iommu_type1_info_cap(info, + VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE); + if (!hdr) { + return false; + } + + cap = (void *)hdr; + + for (int i = 0; i < cap->nr_iovas; i++) { + Range *range = g_new(Range, 1); + + range_set_bounds(range, cap->iova_ranges[i].start, + cap->iova_ranges[i].end); + container->iova_ranges = + range_list_insert(container->iova_ranges, range); + } + + return true; +} + static void vfio_kvm_device_add_group(VFIOGroup *group) { Error *err = NULL; @@ -535,6 +462,12 @@ static void vfio_get_iommu_info_migration(VFIOContainer *container, } } +static void vfio_free_container(VFIOContainer *container) +{ + g_list_free_full(container->iova_ranges, g_free); + g_free(container); +} + static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, Error **errp) { @@ -616,8 +549,8 @@ static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, container->error = NULL; container->dirty_pages_supported = false; container->dma_max_mappings = 0; + container->iova_ranges = NULL; QLIST_INIT(&container->giommu_list); - QLIST_INIT(&container->hostwin_list); QLIST_INIT(&container->vrdl_list); ret = vfio_init_container(container, group->fd, errp); @@ -652,84 +585,21 @@ static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, if (!vfio_get_info_dma_avail(info, &container->dma_max_mappings)) { container->dma_max_mappings = 65535; } - vfio_get_iommu_info_migration(container, info); - g_free(info); - /* - * FIXME: We should parse VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE - * information to get the actual window extent rather than assume - * a 64-bit IOVA address space. - */ - vfio_host_win_add(container, 0, (hwaddr)-1, container->pgsizes); + vfio_get_info_iova_range(info, container); + vfio_get_iommu_info_migration(container, info); + g_free(info); break; } case VFIO_SPAPR_TCE_v2_IOMMU: case VFIO_SPAPR_TCE_IOMMU: { - struct vfio_iommu_spapr_tce_info info; - bool v2 = container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU; - - /* - * The host kernel code implementing VFIO_IOMMU_DISABLE is called - * when container fd is closed so we do not call it explicitly - * in this file. - */ - if (!v2) { - ret = ioctl(fd, VFIO_IOMMU_ENABLE); - if (ret) { - error_setg_errno(errp, errno, "failed to enable container"); - ret = -errno; - goto enable_discards_exit; - } - } else { - container->prereg_listener = vfio_prereg_listener; - - memory_listener_register(&container->prereg_listener, - &address_space_memory); - if (container->error) { - memory_listener_unregister(&container->prereg_listener); - ret = -1; - error_propagate_prepend(errp, container->error, - "RAM memory listener initialization failed: "); - goto enable_discards_exit; - } - } - - info.argsz = sizeof(info); - ret = ioctl(fd, VFIO_IOMMU_SPAPR_TCE_GET_INFO, &info); + ret = vfio_spapr_container_init(container, errp); if (ret) { - error_setg_errno(errp, errno, - "VFIO_IOMMU_SPAPR_TCE_GET_INFO failed"); - ret = -errno; - if (v2) { - memory_listener_unregister(&container->prereg_listener); - } goto enable_discards_exit; } - - if (v2) { - container->pgsizes = info.ddw.pgsizes; - /* - * There is a default window in just created container. - * To make region_add/del simpler, we better remove this - * window now and let those iommu_listener callbacks - * create/remove them when needed. - */ - ret = vfio_spapr_remove_window(container, info.dma32_window_start); - if (ret) { - error_setg_errno(errp, -ret, - "failed to remove existing window"); - goto enable_discards_exit; - } - } else { - /* The default table uses 4K pages */ - container->pgsizes = 0x1000; - vfio_host_win_add(container, info.dma32_window_start, - info.dma32_window_start + - info.dma32_window_size - 1, - 0x1000); - } + break; } } @@ -759,13 +629,17 @@ listener_release_exit: QLIST_REMOVE(group, container_next); QLIST_REMOVE(container, next); vfio_kvm_device_del_group(group); - vfio_listener_release(container); + memory_listener_unregister(&container->listener); + if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU || + container->iommu_type == VFIO_SPAPR_TCE_IOMMU) { + vfio_spapr_container_deinit(container); + } enable_discards_exit: vfio_ram_block_discard_disable(container, false); free_container_exit: - g_free(container); + vfio_free_container(container); close_fd_exit: close(fd); @@ -789,7 +663,11 @@ static void vfio_disconnect_container(VFIOGroup *group) * group. */ if (QLIST_EMPTY(&container->group_list)) { - vfio_listener_release(container); + memory_listener_unregister(&container->listener); + if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU || + container->iommu_type == VFIO_SPAPR_TCE_IOMMU) { + vfio_spapr_container_deinit(container); + } } if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, &container->fd)) { @@ -800,7 +678,6 @@ static void vfio_disconnect_container(VFIOGroup *group) if (QLIST_EMPTY(&container->group_list)) { VFIOAddressSpace *space = container->space; VFIOGuestIOMMU *giommu, *tmp; - VFIOHostDMAWindow *hostwin, *next; QLIST_REMOVE(container, next); @@ -811,15 +688,9 @@ static void vfio_disconnect_container(VFIOGroup *group) g_free(giommu); } - QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, - next) { - QLIST_REMOVE(hostwin, hostwin_next); - g_free(hostwin); - } - trace_vfio_disconnect_container(container->fd); close(container->fd); - g_free(container); + vfio_free_container(container); vfio_put_address_space(space); } @@ -975,103 +846,6 @@ static void vfio_put_base_device(VFIODevice *vbasedev) close(vbasedev->fd); } -/* - * Interfaces for IBM EEH (Enhanced Error Handling) - */ -static bool vfio_eeh_container_ok(VFIOContainer *container) -{ - /* - * As of 2016-03-04 (linux-4.5) the host kernel EEH/VFIO - * implementation is broken if there are multiple groups in a - * container. The hardware works in units of Partitionable - * Endpoints (== IOMMU groups) and the EEH operations naively - * iterate across all groups in the container, without any logic - * to make sure the groups have their state synchronized. For - * certain operations (ENABLE) that might be ok, until an error - * occurs, but for others (GET_STATE) it's clearly broken. - */ - - /* - * XXX Once fixed kernels exist, test for them here - */ - - if (QLIST_EMPTY(&container->group_list)) { - return false; - } - - if (QLIST_NEXT(QLIST_FIRST(&container->group_list), container_next)) { - return false; - } - - return true; -} - -static int vfio_eeh_container_op(VFIOContainer *container, uint32_t op) -{ - struct vfio_eeh_pe_op pe_op = { - .argsz = sizeof(pe_op), - .op = op, - }; - int ret; - - if (!vfio_eeh_container_ok(container)) { - error_report("vfio/eeh: EEH_PE_OP 0x%x: " - "kernel requires a container with exactly one group", op); - return -EPERM; - } - - ret = ioctl(container->fd, VFIO_EEH_PE_OP, &pe_op); - if (ret < 0) { - error_report("vfio/eeh: EEH_PE_OP 0x%x failed: %m", op); - return -errno; - } - - return ret; -} - -static VFIOContainer *vfio_eeh_as_container(AddressSpace *as) -{ - VFIOAddressSpace *space = vfio_get_address_space(as); - VFIOContainer *container = NULL; - - if (QLIST_EMPTY(&space->containers)) { - /* No containers to act on */ - goto out; - } - - container = QLIST_FIRST(&space->containers); - - if (QLIST_NEXT(container, next)) { - /* - * We don't yet have logic to synchronize EEH state across - * multiple containers - */ - container = NULL; - goto out; - } - -out: - vfio_put_address_space(space); - return container; -} - -bool vfio_eeh_as_ok(AddressSpace *as) -{ - VFIOContainer *container = vfio_eeh_as_container(as); - - return (container != NULL) && vfio_eeh_container_ok(container); -} - -int vfio_eeh_as_op(AddressSpace *as, uint32_t op) -{ - VFIOContainer *container = vfio_eeh_as_container(as); - - if (!container) { - return -ENODEV; - } - return vfio_eeh_container_op(container, op); -} - static int vfio_device_groupid(VFIODevice *vbasedev, Error **errp) { char *tmp, group_path[PATH_MAX], *group_name; diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c index 7e5da21b31..168847e7c5 100644 --- a/hw/vfio/helpers.c +++ b/hw/vfio/helpers.c @@ -23,7 +23,6 @@ #include <sys/ioctl.h> #include "hw/vfio/vfio-common.h" -#include "hw/vfio/vfio.h" #include "hw/hw.h" #include "trace.h" #include "qapi/error.h" diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index b27011cee7..c62c02f7b6 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -3081,7 +3081,7 @@ static void vfio_realize(PCIDevice *pdev, Error **errp) struct stat st; int i, ret; bool is_mdev; - char uuid[UUID_FMT_LEN]; + char uuid[UUID_STR_LEN]; char *name; if (!vbasedev->sysfsdev) { diff --git a/hw/vfio/spapr.c b/hw/vfio/spapr.c index 9ec1e95f6d..83da2f7ec2 100644 --- a/hw/vfio/spapr.c +++ b/hw/vfio/spapr.c @@ -11,6 +11,11 @@ #include "qemu/osdep.h" #include <sys/ioctl.h> #include <linux/vfio.h> +#ifdef CONFIG_KVM +#include <linux/kvm.h> +#endif +#include "sysemu/kvm.h" +#include "exec/address-spaces.h" #include "hw/vfio/vfio-common.h" #include "hw/hw.h" @@ -135,15 +140,90 @@ static void vfio_prereg_listener_region_del(MemoryListener *listener, trace_vfio_prereg_unregister(reg.vaddr, reg.size, ret ? -errno : 0); } -const MemoryListener vfio_prereg_listener = { +static const MemoryListener vfio_prereg_listener = { .name = "vfio-pre-reg", .region_add = vfio_prereg_listener_region_add, .region_del = vfio_prereg_listener_region_del, }; -int vfio_spapr_create_window(VFIOContainer *container, - MemoryRegionSection *section, - hwaddr *pgsize) +static void vfio_host_win_add(VFIOContainer *container, hwaddr min_iova, + hwaddr max_iova, uint64_t iova_pgsizes) +{ + VFIOHostDMAWindow *hostwin; + + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (ranges_overlap(hostwin->min_iova, + hostwin->max_iova - hostwin->min_iova + 1, + min_iova, + max_iova - min_iova + 1)) { + hw_error("%s: Overlapped IOMMU are not enabled", __func__); + } + } + + hostwin = g_malloc0(sizeof(*hostwin)); + + hostwin->min_iova = min_iova; + hostwin->max_iova = max_iova; + hostwin->iova_pgsizes = iova_pgsizes; + QLIST_INSERT_HEAD(&container->hostwin_list, hostwin, hostwin_next); +} + +static int vfio_host_win_del(VFIOContainer *container, + hwaddr min_iova, hwaddr max_iova) +{ + VFIOHostDMAWindow *hostwin; + + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (hostwin->min_iova == min_iova && hostwin->max_iova == max_iova) { + QLIST_REMOVE(hostwin, hostwin_next); + g_free(hostwin); + return 0; + } + } + + return -1; +} + +static VFIOHostDMAWindow *vfio_find_hostwin(VFIOContainer *container, + hwaddr iova, hwaddr end) +{ + VFIOHostDMAWindow *hostwin; + bool hostwin_found = false; + + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (hostwin->min_iova <= iova && end <= hostwin->max_iova) { + hostwin_found = true; + break; + } + } + + return hostwin_found ? hostwin : NULL; +} + +static int vfio_spapr_remove_window(VFIOContainer *container, + hwaddr offset_within_address_space) +{ + struct vfio_iommu_spapr_tce_remove remove = { + .argsz = sizeof(remove), + .start_addr = offset_within_address_space, + }; + int ret; + + ret = ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_REMOVE, &remove); + if (ret) { + error_report("Failed to remove window at %"PRIx64, + (uint64_t)remove.start_addr); + return -errno; + } + + trace_vfio_spapr_remove_window(offset_within_address_space); + + return 0; +} + +static int vfio_spapr_create_window(VFIOContainer *container, + MemoryRegionSection *section, + hwaddr *pgsize) { int ret = 0; IOMMUMemoryRegion *iommu_mr = IOMMU_MEMORY_REGION(section->mr); @@ -233,23 +313,195 @@ int vfio_spapr_create_window(VFIOContainer *container, return 0; } -int vfio_spapr_remove_window(VFIOContainer *container, - hwaddr offset_within_address_space) +int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp) { - struct vfio_iommu_spapr_tce_remove remove = { - .argsz = sizeof(remove), - .start_addr = offset_within_address_space, - }; + VFIOHostDMAWindow *hostwin; + hwaddr pgsize = 0; int ret; - ret = ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_REMOVE, &remove); + /* + * VFIO_SPAPR_TCE_IOMMU supports a single host window between + * [dma32_window_start, dma32_window_size), we need to ensure + * the section fall in this range. + */ + if (container->iommu_type == VFIO_SPAPR_TCE_IOMMU) { + hwaddr iova, end; + + iova = section->offset_within_address_space; + end = iova + int128_get64(section->size) - 1; + + if (!vfio_find_hostwin(container, iova, end)) { + error_setg(errp, "Container %p can't map guest IOVA region" + " 0x%"HWADDR_PRIx"..0x%"HWADDR_PRIx, container, + iova, end); + return -EINVAL; + } + return 0; + } + + if (container->iommu_type != VFIO_SPAPR_TCE_v2_IOMMU) { + return 0; + } + + /* For now intersections are not allowed, we may relax this later */ + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (ranges_overlap(hostwin->min_iova, + hostwin->max_iova - hostwin->min_iova + 1, + section->offset_within_address_space, + int128_get64(section->size))) { + error_setg(errp, + "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existing" + "host DMA window [0x%"PRIx64",0x%"PRIx64"]", + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, + hostwin->min_iova, hostwin->max_iova); + return -EINVAL; + } + } + + ret = vfio_spapr_create_window(container, section, &pgsize); if (ret) { - error_report("Failed to remove window at %"PRIx64, - (uint64_t)remove.start_addr); - return -errno; + error_setg_errno(errp, -ret, "Failed to create SPAPR window"); + return ret; } - trace_vfio_spapr_remove_window(offset_within_address_space); + vfio_host_win_add(container, section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, pgsize); +#ifdef CONFIG_KVM + if (kvm_enabled()) { + VFIOGroup *group; + IOMMUMemoryRegion *iommu_mr = IOMMU_MEMORY_REGION(section->mr); + struct kvm_vfio_spapr_tce param; + struct kvm_device_attr attr = { + .group = KVM_DEV_VFIO_GROUP, + .attr = KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, + .addr = (uint64_t)(unsigned long)¶m, + }; + + if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_FD, + ¶m.tablefd)) { + QLIST_FOREACH(group, &container->group_list, container_next) { + param.groupfd = group->fd; + if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) { + error_setg_errno(errp, errno, + "vfio: failed GROUP_SET_SPAPR_TCE for " + "KVM VFIO device %d and group fd %d", + param.tablefd, param.groupfd); + return -errno; + } + trace_vfio_spapr_group_attach(param.groupfd, param.tablefd); + } + } + } +#endif + return 0; +} + +void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section) +{ + if (container->iommu_type != VFIO_SPAPR_TCE_v2_IOMMU) { + return; + } + + vfio_spapr_remove_window(container, + section->offset_within_address_space); + if (vfio_host_win_del(container, + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1) < 0) { + hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, + __func__, section->offset_within_address_space); + } +} + +int vfio_spapr_container_init(VFIOContainer *container, Error **errp) +{ + struct vfio_iommu_spapr_tce_info info; + bool v2 = container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU; + int ret, fd = container->fd; + + QLIST_INIT(&container->hostwin_list); + + /* + * The host kernel code implementing VFIO_IOMMU_DISABLE is called + * when container fd is closed so we do not call it explicitly + * in this file. + */ + if (!v2) { + ret = ioctl(fd, VFIO_IOMMU_ENABLE); + if (ret) { + error_setg_errno(errp, errno, "failed to enable container"); + return -errno; + } + } else { + container->prereg_listener = vfio_prereg_listener; + + memory_listener_register(&container->prereg_listener, + &address_space_memory); + if (container->error) { + ret = -1; + error_propagate_prepend(errp, container->error, + "RAM memory listener initialization failed: "); + goto listener_unregister_exit; + } + } + + info.argsz = sizeof(info); + ret = ioctl(fd, VFIO_IOMMU_SPAPR_TCE_GET_INFO, &info); + if (ret) { + error_setg_errno(errp, errno, + "VFIO_IOMMU_SPAPR_TCE_GET_INFO failed"); + ret = -errno; + goto listener_unregister_exit; + } + + if (v2) { + container->pgsizes = info.ddw.pgsizes; + /* + * There is a default window in just created container. + * To make region_add/del simpler, we better remove this + * window now and let those iommu_listener callbacks + * create/remove them when needed. + */ + ret = vfio_spapr_remove_window(container, info.dma32_window_start); + if (ret) { + error_setg_errno(errp, -ret, + "failed to remove existing window"); + goto listener_unregister_exit; + } + } else { + /* The default table uses 4K pages */ + container->pgsizes = 0x1000; + vfio_host_win_add(container, info.dma32_window_start, + info.dma32_window_start + + info.dma32_window_size - 1, + 0x1000); + } return 0; + +listener_unregister_exit: + if (v2) { + memory_listener_unregister(&container->prereg_listener); + } + return ret; +} + +void vfio_spapr_container_deinit(VFIOContainer *container) +{ + VFIOHostDMAWindow *hostwin, *next; + + if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU) { + memory_listener_unregister(&container->prereg_listener); + } + QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, + next) { + QLIST_REMOVE(hostwin, hostwin_next); + g_free(hostwin); + } } diff --git a/hw/virtio/trace-events b/hw/virtio/trace-events index 0af7a2886c..637cac4edf 100644 --- a/hw/virtio/trace-events +++ b/hw/virtio/trace-events @@ -135,6 +135,7 @@ virtio_iommu_notify_flag_add(const char *name) "add notifier to mr %s" virtio_iommu_notify_flag_del(const char *name) "del notifier from mr %s" virtio_iommu_switch_address_space(uint8_t bus, uint8_t slot, uint8_t fn, bool on) "Device %02x:%02x.%x switching address space (iommu enabled=%d)" virtio_iommu_freeze_granule(uint64_t page_size_mask) "granule set to 0x%"PRIx64 +virtio_iommu_host_resv_regions(const char *name, uint32_t index, uint64_t lob, uint64_t upb) "mr=%s host-resv-reg[%d] = [0x%"PRIx64",0x%"PRIx64"]" # virtio-mem.c virtio_mem_send_response(uint16_t type) "type=%" PRIu16 diff --git a/hw/virtio/virtio-iommu-pci.c b/hw/virtio/virtio-iommu-pci.c index 7ef2f9dcdb..9459fbf6ed 100644 --- a/hw/virtio/virtio-iommu-pci.c +++ b/hw/virtio/virtio-iommu-pci.c @@ -37,7 +37,7 @@ struct VirtIOIOMMUPCI { static Property virtio_iommu_pci_properties[] = { DEFINE_PROP_UINT32("class", VirtIOPCIProxy, class_code, 0), DEFINE_PROP_ARRAY("reserved-regions", VirtIOIOMMUPCI, - vdev.nb_reserved_regions, vdev.reserved_regions, + vdev.nr_prop_resv_regions, vdev.prop_resv_regions, qdev_prop_reserved_region, ReservedRegion), DEFINE_PROP_END_OF_LIST(), }; @@ -54,9 +54,9 @@ static void virtio_iommu_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp) "for the virtio-iommu-pci device"); return; } - for (int i = 0; i < s->nb_reserved_regions; i++) { - if (s->reserved_regions[i].type != VIRTIO_IOMMU_RESV_MEM_T_RESERVED && - s->reserved_regions[i].type != VIRTIO_IOMMU_RESV_MEM_T_MSI) { + for (int i = 0; i < s->nr_prop_resv_regions; i++) { + if (s->prop_resv_regions[i].type != VIRTIO_IOMMU_RESV_MEM_T_RESERVED && + s->prop_resv_regions[i].type != VIRTIO_IOMMU_RESV_MEM_T_MSI) { error_setg(errp, "reserved region %d has an invalid type", i); error_append_hint(errp, "Valid values are 0 and 1\n"); return; diff --git a/hw/virtio/virtio-iommu.c b/hw/virtio/virtio-iommu.c index be51635895..89fb5767d1 100644 --- a/hw/virtio/virtio-iommu.c +++ b/hw/virtio/virtio-iommu.c @@ -20,12 +20,15 @@ #include "qemu/osdep.h" #include "qemu/log.h" #include "qemu/iov.h" +#include "qemu/range.h" +#include "qemu/reserved-region.h" #include "exec/target_page.h" #include "hw/qdev-properties.h" #include "hw/virtio/virtio.h" #include "sysemu/kvm.h" #include "sysemu/reset.h" #include "sysemu/sysemu.h" +#include "qemu/reserved-region.h" #include "qapi/error.h" #include "qemu/error-report.h" #include "trace.h" @@ -378,6 +381,19 @@ static void virtio_iommu_put_domain(gpointer data) g_free(domain); } +static void add_prop_resv_regions(IOMMUDevice *sdev) +{ + VirtIOIOMMU *s = sdev->viommu; + int i; + + for (i = 0; i < s->nr_prop_resv_regions; i++) { + ReservedRegion *reg = g_new0(ReservedRegion, 1); + + *reg = s->prop_resv_regions[i]; + sdev->resv_regions = resv_region_list_insert(sdev->resv_regions, reg); + } +} + static AddressSpace *virtio_iommu_find_add_as(PCIBus *bus, void *opaque, int devfn) { @@ -408,6 +424,7 @@ static AddressSpace *virtio_iommu_find_add_as(PCIBus *bus, void *opaque, memory_region_init(&sdev->root, OBJECT(s), name, UINT64_MAX); address_space_init(&sdev->as, &sdev->root, TYPE_VIRTIO_IOMMU); + add_prop_resv_regions(sdev); /* * Build the IOMMU disabled container with aliases to the @@ -444,6 +461,10 @@ static AddressSpace *virtio_iommu_find_add_as(PCIBus *bus, void *opaque, return &sdev->as; } +static const PCIIOMMUOps virtio_iommu_ops = { + .get_address_space = virtio_iommu_find_add_as, +}; + static int virtio_iommu_attach(VirtIOIOMMU *s, struct virtio_iommu_req_attach *req) { @@ -624,29 +645,30 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s, return ret; } -static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t ep, +static ssize_t virtio_iommu_fill_resv_mem_prop(IOMMUDevice *sdev, uint32_t ep, uint8_t *buf, size_t free) { struct virtio_iommu_probe_resv_mem prop = {}; size_t size = sizeof(prop), length = size - sizeof(prop.head), total; - int i; - - total = size * s->nb_reserved_regions; + GList *l; + total = size * g_list_length(sdev->resv_regions); if (total > free) { return -ENOSPC; } - for (i = 0; i < s->nb_reserved_regions; i++) { - unsigned subtype = s->reserved_regions[i].type; + for (l = sdev->resv_regions; l; l = l->next) { + ReservedRegion *reg = l->data; + unsigned subtype = reg->type; + Range *range = ®->range; assert(subtype == VIRTIO_IOMMU_RESV_MEM_T_RESERVED || subtype == VIRTIO_IOMMU_RESV_MEM_T_MSI); prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM); prop.head.length = cpu_to_le16(length); prop.subtype = subtype; - prop.start = cpu_to_le64(s->reserved_regions[i].low); - prop.end = cpu_to_le64(s->reserved_regions[i].high); + prop.start = cpu_to_le64(range_lob(range)); + prop.end = cpu_to_le64(range_upb(range)); memcpy(buf, &prop, size); @@ -666,19 +688,27 @@ static int virtio_iommu_probe(VirtIOIOMMU *s, uint8_t *buf) { uint32_t ep_id = le32_to_cpu(req->endpoint); + IOMMUMemoryRegion *iommu_mr = virtio_iommu_mr(s, ep_id); size_t free = VIOMMU_PROBE_SIZE; + IOMMUDevice *sdev; ssize_t count; - if (!virtio_iommu_mr(s, ep_id)) { + if (!iommu_mr) { return VIRTIO_IOMMU_S_NOENT; } - count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free); + sdev = container_of(iommu_mr, IOMMUDevice, iommu_mr); + if (!sdev) { + return -EINVAL; + } + + count = virtio_iommu_fill_resv_mem_prop(sdev, ep_id, buf, free); if (count < 0) { return VIRTIO_IOMMU_S_INVAL; } buf += count; free -= count; + sdev->probe_done = true; return VIRTIO_IOMMU_S_OK; } @@ -856,7 +886,7 @@ static IOMMUTLBEntry virtio_iommu_translate(IOMMUMemoryRegion *mr, hwaddr addr, bool bypass_allowed; int granule; bool found; - int i; + GList *l; interval.low = addr; interval.high = addr + 1; @@ -894,10 +924,10 @@ static IOMMUTLBEntry virtio_iommu_translate(IOMMUMemoryRegion *mr, hwaddr addr, goto unlock; } - for (i = 0; i < s->nb_reserved_regions; i++) { - ReservedRegion *reg = &s->reserved_regions[i]; + for (l = sdev->resv_regions; l; l = l->next) { + ReservedRegion *reg = l->data; - if (addr >= reg->low && addr <= reg->high) { + if (range_contains(®->range, addr)) { switch (reg->type) { case VIRTIO_IOMMU_RESV_MEM_T_MSI: entry.perm = flag; @@ -1131,6 +1161,106 @@ static int virtio_iommu_set_page_size_mask(IOMMUMemoryRegion *mr, return 0; } +/** + * rebuild_resv_regions: rebuild resv regions with both the + * info of host resv ranges and property set resv ranges + */ +static int rebuild_resv_regions(IOMMUDevice *sdev) +{ + GList *l; + int i = 0; + + /* free the existing list and rebuild it from scratch */ + g_list_free_full(sdev->resv_regions, g_free); + sdev->resv_regions = NULL; + + /* First add host reserved regions if any, all tagged as RESERVED */ + for (l = sdev->host_resv_ranges; l; l = l->next) { + ReservedRegion *reg = g_new0(ReservedRegion, 1); + Range *r = (Range *)l->data; + + reg->type = VIRTIO_IOMMU_RESV_MEM_T_RESERVED; + range_set_bounds(®->range, range_lob(r), range_upb(r)); + sdev->resv_regions = resv_region_list_insert(sdev->resv_regions, reg); + trace_virtio_iommu_host_resv_regions(sdev->iommu_mr.parent_obj.name, i, + range_lob(®->range), + range_upb(®->range)); + i++; + } + /* + * then add higher priority reserved regions set by the machine + * through properties + */ + add_prop_resv_regions(sdev); + return 0; +} + +/** + * virtio_iommu_set_iova_ranges: Conveys the usable IOVA ranges + * + * The function turns those into reserved ranges. Once some + * reserved ranges have been set, new reserved regions cannot be + * added outside of the original ones. + * + * @mr: IOMMU MR + * @iova_ranges: list of usable IOVA ranges + * @errp: error handle + */ +static int virtio_iommu_set_iova_ranges(IOMMUMemoryRegion *mr, + GList *iova_ranges, + Error **errp) +{ + IOMMUDevice *sdev = container_of(mr, IOMMUDevice, iommu_mr); + GList *current_ranges = sdev->host_resv_ranges; + GList *l, *tmp, *new_ranges = NULL; + int ret = -EINVAL; + + /* check that each new resv region is included in an existing one */ + if (sdev->host_resv_ranges) { + range_inverse_array(iova_ranges, + &new_ranges, + 0, UINT64_MAX); + + for (tmp = new_ranges; tmp; tmp = tmp->next) { + Range *newr = (Range *)tmp->data; + bool included = false; + + for (l = current_ranges; l; l = l->next) { + Range * r = (Range *)l->data; + + if (range_contains_range(r, newr)) { + included = true; + break; + } + } + if (!included) { + goto error; + } + } + /* all new reserved ranges are included in existing ones */ + ret = 0; + goto out; + } + + if (sdev->probe_done) { + warn_report("%s: Notified about new host reserved regions after probe", + mr->parent_obj.name); + } + + range_inverse_array(iova_ranges, + &sdev->host_resv_ranges, + 0, UINT64_MAX); + rebuild_resv_regions(sdev); + + return 0; +error: + error_setg(errp, "IOMMU mr=%s Conflicting host reserved ranges set!", + mr->parent_obj.name); +out: + g_list_free_full(new_ranges, g_free); + return ret; +} + static void virtio_iommu_system_reset(void *opaque) { VirtIOIOMMU *s = opaque; @@ -1206,7 +1336,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, Error **errp) s->as_by_busptr = g_hash_table_new_full(NULL, NULL, NULL, g_free); if (s->primary_bus) { - pci_setup_iommu(s->primary_bus, virtio_iommu_find_add_as, s); + pci_setup_iommu(s->primary_bus, &virtio_iommu_ops, s); } else { error_setg(errp, "VIRTIO-IOMMU is not attached to any PCI bus!"); } @@ -1426,6 +1556,7 @@ static void virtio_iommu_memory_region_class_init(ObjectClass *klass, imrc->replay = virtio_iommu_replay; imrc->notify_flag_changed = virtio_iommu_notify_flag_changed; imrc->iommu_set_page_size_mask = virtio_iommu_set_page_size_mask; + imrc->iommu_set_iova_ranges = virtio_iommu_set_iova_ranges; } static const TypeInfo virtio_iommu_info = { diff --git a/hw/virtio/virtio-pmem.c b/hw/virtio/virtio-pmem.c index cc24812d2e..c3512c2dae 100644 --- a/hw/virtio/virtio-pmem.c +++ b/hw/virtio/virtio-pmem.c @@ -147,7 +147,10 @@ static void virtio_pmem_fill_device_info(const VirtIOPMEM *pmem, static MemoryRegion *virtio_pmem_get_memory_region(VirtIOPMEM *pmem, Error **errp) { - assert(pmem->memdev); + if (!pmem->memdev) { + error_setg(errp, "'%s' property must be set", VIRTIO_PMEM_MEMDEV_PROP); + return NULL; + } return &pmem->memdev->mr; } diff --git a/include/exec/cputlb.h b/include/exec/cputlb.h index 19b16e58f8..6da1462c4f 100644 --- a/include/exec/cputlb.h +++ b/include/exec/cputlb.h @@ -26,6 +26,5 @@ /* cputlb.c */ void tlb_protect_code(ram_addr_t ram_addr); void tlb_unprotect_code(ram_addr_t ram_addr); -void tlb_flush_counts(size_t *full, size_t *part, size_t *elide); #endif #endif diff --git a/include/exec/memory.h b/include/exec/memory.h index 9087d02769..831f7c996d 100644 --- a/include/exec/memory.h +++ b/include/exec/memory.h @@ -24,6 +24,7 @@ #include "qemu/bswap.h" #include "qemu/queue.h" #include "qemu/int128.h" +#include "qemu/range.h" #include "qemu/notify.h" #include "qom/object.h" #include "qemu/rcu.h" @@ -79,8 +80,7 @@ extern unsigned int global_dirty_tracking; typedef struct MemoryRegionOps MemoryRegionOps; struct ReservedRegion { - hwaddr low; - hwaddr high; + Range range; unsigned type; }; @@ -527,6 +527,26 @@ struct IOMMUMemoryRegionClass { int (*iommu_set_page_size_mask)(IOMMUMemoryRegion *iommu, uint64_t page_size_mask, Error **errp); + /** + * @iommu_set_iova_ranges: + * + * Propagate information about the usable IOVA ranges for a given IOMMU + * memory region. Used for example to propagate host physical device + * reserved memory region constraints to the virtual IOMMU. + * + * Optional method: if this method is not provided, then the default IOVA + * aperture is used. + * + * @iommu: the IOMMUMemoryRegion + * + * @iova_ranges: list of ordered IOVA ranges (at least one range) + * + * Returns 0 on success, or a negative error. In case of failure, the error + * object must be created. + */ + int (*iommu_set_iova_ranges)(IOMMUMemoryRegion *iommu, + GList *iova_ranges, + Error **errp); }; typedef struct RamDiscardListener RamDiscardListener; @@ -1857,6 +1877,18 @@ int memory_region_iommu_set_page_size_mask(IOMMUMemoryRegion *iommu_mr, Error **errp); /** + * memory_region_iommu_set_iova_ranges - Set the usable IOVA ranges + * for a given IOMMU MR region + * + * @iommu: IOMMU memory region + * @iova_ranges: list of ordered IOVA ranges (at least one range) + * @errp: pointer to Error*, to store an error if it happens. + */ +int memory_region_iommu_set_iova_ranges(IOMMUMemoryRegion *iommu, + GList *iova_ranges, + Error **errp); + +/** * memory_region_name: get a memory region's name * * Returns the string that was used to initialize the memory region. diff --git a/include/hw/hyperv/dynmem-proto.h b/include/hw/hyperv/dynmem-proto.h new file mode 100644 index 0000000000..d0f9090ac4 --- /dev/null +++ b/include/hw/hyperv/dynmem-proto.h @@ -0,0 +1,423 @@ +#ifndef HW_HYPERV_DYNMEM_PROTO_H +#define HW_HYPERV_DYNMEM_PROTO_H + +/* + * Hyper-V Dynamic Memory Protocol definitions + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * Based on drivers/hv/hv_balloon.c from Linux kernel: + * Copyright (c) 2012, Microsoft Corporation. + * + * Author: K. Y. Srinivasan <kys@microsoft.com> + * + * This work is licensed under the terms of the GNU GPL, version 2. + * See the COPYING file in the top-level directory. + */ + +/* + * Protocol versions. The low word is the minor version, the high word the major + * version. + * + * History: + * Initial version 1.0 + * Changed to 0.1 on 2009/03/25 + * Changes to 0.2 on 2009/05/14 + * Changes to 0.3 on 2009/12/03 + * Changed to 1.0 on 2011/04/05 + * Changed to 2.0 on 2019/12/10 + */ + +#define DYNMEM_MAKE_VERSION(Major, Minor) ((uint32_t)(((Major) << 16) | (Minor))) +#define DYNMEM_MAJOR_VERSION(Version) ((uint32_t)(Version) >> 16) +#define DYNMEM_MINOR_VERSION(Version) ((uint32_t)(Version) & 0xff) + +enum { + DYNMEM_PROTOCOL_VERSION_1 = DYNMEM_MAKE_VERSION(0, 3), + DYNMEM_PROTOCOL_VERSION_2 = DYNMEM_MAKE_VERSION(1, 0), + DYNMEM_PROTOCOL_VERSION_3 = DYNMEM_MAKE_VERSION(2, 0), + + DYNMEM_PROTOCOL_VERSION_WIN7 = DYNMEM_PROTOCOL_VERSION_1, + DYNMEM_PROTOCOL_VERSION_WIN8 = DYNMEM_PROTOCOL_VERSION_2, + DYNMEM_PROTOCOL_VERSION_WIN10 = DYNMEM_PROTOCOL_VERSION_3, + + DYNMEM_PROTOCOL_VERSION_CURRENT = DYNMEM_PROTOCOL_VERSION_WIN10 +}; + + + +/* + * Message Types + */ + +enum dm_message_type { + /* + * Version 0.3 + */ + DM_ERROR = 0, + DM_VERSION_REQUEST = 1, + DM_VERSION_RESPONSE = 2, + DM_CAPABILITIES_REPORT = 3, + DM_CAPABILITIES_RESPONSE = 4, + DM_STATUS_REPORT = 5, + DM_BALLOON_REQUEST = 6, + DM_BALLOON_RESPONSE = 7, + DM_UNBALLOON_REQUEST = 8, + DM_UNBALLOON_RESPONSE = 9, + DM_MEM_HOT_ADD_REQUEST = 10, + DM_MEM_HOT_ADD_RESPONSE = 11, + DM_VERSION_03_MAX = 11, + /* + * Version 1.0. + */ + DM_INFO_MESSAGE = 12, + DM_VERSION_1_MAX = 12, + + /* + * Version 2.0 + */ + DM_MEM_HOT_REMOVE_REQUEST = 13, + DM_MEM_HOT_REMOVE_RESPONSE = 14 +}; + + +/* + * Structures defining the dynamic memory management + * protocol. + */ + +union dm_version { + struct { + uint16_t minor_version; + uint16_t major_version; + }; + uint32_t version; +} QEMU_PACKED; + + +union dm_caps { + struct { + uint64_t balloon:1; + uint64_t hot_add:1; + /* + * To support guests that may have alignment + * limitations on hot-add, the guest can specify + * its alignment requirements; a value of n + * represents an alignment of 2^n in mega bytes. + */ + uint64_t hot_add_alignment:4; + uint64_t hot_remove:1; + uint64_t reservedz:57; + } cap_bits; + uint64_t caps; +} QEMU_PACKED; + +union dm_mem_page_range { + struct { + /* + * The PFN number of the first page in the range. + * 40 bits is the architectural limit of a PFN + * number for AMD64. + */ + uint64_t start_page:40; + /* + * The number of pages in the range. + */ + uint64_t page_cnt:24; + } finfo; + uint64_t page_range; +} QEMU_PACKED; + + + +/* + * The header for all dynamic memory messages: + * + * type: Type of the message. + * size: Size of the message in bytes; including the header. + * trans_id: The guest is responsible for manufacturing this ID. + */ + +struct dm_header { + uint16_t type; + uint16_t size; + uint32_t trans_id; +} QEMU_PACKED; + +/* + * A generic message format for dynamic memory. + * Specific message formats are defined later in the file. + */ + +struct dm_message { + struct dm_header hdr; + uint8_t data[]; /* enclosed message */ +} QEMU_PACKED; + + +/* + * Specific message types supporting the dynamic memory protocol. + */ + +/* + * Version negotiation message. Sent from the guest to the host. + * The guest is free to try different versions until the host + * accepts the version. + * + * dm_version: The protocol version requested. + * is_last_attempt: If TRUE, this is the last version guest will request. + * reservedz: Reserved field, set to zero. + */ + +struct dm_version_request { + struct dm_header hdr; + union dm_version version; + uint32_t is_last_attempt:1; + uint32_t reservedz:31; +} QEMU_PACKED; + +/* + * Version response message; Host to Guest and indicates + * if the host has accepted the version sent by the guest. + * + * is_accepted: If TRUE, host has accepted the version and the guest + * should proceed to the next stage of the protocol. FALSE indicates that + * guest should re-try with a different version. + * + * reservedz: Reserved field, set to zero. + */ + +struct dm_version_response { + struct dm_header hdr; + uint64_t is_accepted:1; + uint64_t reservedz:63; +} QEMU_PACKED; + +/* + * Message reporting capabilities. This is sent from the guest to the + * host. + */ + +struct dm_capabilities { + struct dm_header hdr; + union dm_caps caps; + uint64_t min_page_cnt; + uint64_t max_page_number; +} QEMU_PACKED; + +/* + * Response to the capabilities message. This is sent from the host to the + * guest. This message notifies if the host has accepted the guest's + * capabilities. If the host has not accepted, the guest must shutdown + * the service. + * + * is_accepted: Indicates if the host has accepted guest's capabilities. + * reservedz: Must be 0. + */ + +struct dm_capabilities_resp_msg { + struct dm_header hdr; + uint64_t is_accepted:1; + uint64_t hot_remove:1; + uint64_t suppress_pressure_reports:1; + uint64_t reservedz:61; +} QEMU_PACKED; + +/* + * This message is used to report memory pressure from the guest. + * This message is not part of any transaction and there is no + * response to this message. + * + * num_avail: Available memory in pages. + * num_committed: Committed memory in pages. + * page_file_size: The accumulated size of all page files + * in the system in pages. + * zero_free: The nunber of zero and free pages. + * page_file_writes: The writes to the page file in pages. + * io_diff: An indicator of file cache efficiency or page file activity, + * calculated as File Cache Page Fault Count - Page Read Count. + * This value is in pages. + * + * Some of these metrics are Windows specific and fortunately + * the algorithm on the host side that computes the guest memory + * pressure only uses num_committed value. + */ + +struct dm_status { + struct dm_header hdr; + uint64_t num_avail; + uint64_t num_committed; + uint64_t page_file_size; + uint64_t zero_free; + uint32_t page_file_writes; + uint32_t io_diff; +} QEMU_PACKED; + + +/* + * Message to ask the guest to allocate memory - balloon up message. + * This message is sent from the host to the guest. The guest may not be + * able to allocate as much memory as requested. + * + * num_pages: number of pages to allocate. + */ + +struct dm_balloon { + struct dm_header hdr; + uint32_t num_pages; + uint32_t reservedz; +} QEMU_PACKED; + + +/* + * Balloon response message; this message is sent from the guest + * to the host in response to the balloon message. + * + * reservedz: Reserved; must be set to zero. + * more_pages: If FALSE, this is the last message of the transaction. + * if TRUE there will atleast one more message from the guest. + * + * range_count: The number of ranges in the range array. + * + * range_array: An array of page ranges returned to the host. + * + */ + +struct dm_balloon_response { + struct dm_header hdr; + uint32_t reservedz; + uint32_t more_pages:1; + uint32_t range_count:31; + union dm_mem_page_range range_array[]; +} QEMU_PACKED; + +/* + * Un-balloon message; this message is sent from the host + * to the guest to give guest more memory. + * + * more_pages: If FALSE, this is the last message of the transaction. + * if TRUE there will atleast one more message from the guest. + * + * reservedz: Reserved; must be set to zero. + * + * range_count: The number of ranges in the range array. + * + * range_array: An array of page ranges returned to the host. + * + */ + +struct dm_unballoon_request { + struct dm_header hdr; + uint32_t more_pages:1; + uint32_t reservedz:31; + uint32_t range_count; + union dm_mem_page_range range_array[]; +} QEMU_PACKED; + +/* + * Un-balloon response message; this message is sent from the guest + * to the host in response to an unballoon request. + * + */ + +struct dm_unballoon_response { + struct dm_header hdr; +} QEMU_PACKED; + + +/* + * Hot add request message. Message sent from the host to the guest. + * + * mem_range: Memory range to hot add. + * + */ + +struct dm_hot_add { + struct dm_header hdr; + union dm_mem_page_range range; +} QEMU_PACKED; + +/* + * Hot add response message. + * This message is sent by the guest to report the status of a hot add request. + * If page_count is less than the requested page count, then the host should + * assume all further hot add requests will fail, since this indicates that + * the guest has hit an upper physical memory barrier. + * + * Hot adds may also fail due to low resources; in this case, the guest must + * not complete this message until the hot add can succeed, and the host must + * not send a new hot add request until the response is sent. + * If VSC fails to hot add memory DYNMEM_NUMBER_OF_UNSUCCESSFUL_HOTADD_ATTEMPTS + * times it fails the request. + * + * + * page_count: number of pages that were successfully hot added. + * + * result: result of the operation 1: success, 0: failure. + * + */ + +struct dm_hot_add_response { + struct dm_header hdr; + uint32_t page_count; + uint32_t result; +} QEMU_PACKED; + +struct dm_hot_remove { + struct dm_header hdr; + uint32_t virtual_node; + uint32_t page_count; + uint32_t qos_flags; + uint32_t reservedZ; +} QEMU_PACKED; + +struct dm_hot_remove_response { + struct dm_header hdr; + uint32_t result; + uint32_t range_count; + uint64_t more_pages:1; + uint64_t reservedz:63; + union dm_mem_page_range range_array[]; +} QEMU_PACKED; + +#define DM_REMOVE_QOS_LARGE (1 << 0) +#define DM_REMOVE_QOS_LOCAL (1 << 1) +#define DM_REMOVE_QOS_MASK (0x3) + +/* + * Types of information sent from host to the guest. + */ + +enum dm_info_type { + INFO_TYPE_MAX_PAGE_CNT = 0, + MAX_INFO_TYPE +}; + + +/* + * Header for the information message. + */ + +struct dm_info_header { + enum dm_info_type type; + uint32_t data_size; + uint8_t data[]; +} QEMU_PACKED; + +/* + * This message is sent from the host to the guest to pass + * some relevant information (win8 addition). + * + * reserved: no used. + * info_size: size of the information blob. + * info: information blob. + */ + +struct dm_info_msg { + struct dm_header hdr; + uint32_t reserved; + uint32_t info_size; + uint8_t info[]; +}; + +#endif diff --git a/include/hw/hyperv/hv-balloon.h b/include/hw/hyperv/hv-balloon.h new file mode 100644 index 0000000000..c1efe70fc2 --- /dev/null +++ b/include/hw/hyperv/hv-balloon.h @@ -0,0 +1,18 @@ +/* + * QEMU Hyper-V Dynamic Memory Protocol driver + * + * Copyright (C) 2020-2023 Oracle and/or its affiliates. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_HV_BALLOON_H +#define HW_HV_BALLOON_H + +#include "qom/object.h" + +#define TYPE_HV_BALLOON "hv-balloon" +OBJECT_DECLARE_SIMPLE_TYPE(HvBalloon, HV_BALLOON) + +#endif diff --git a/include/hw/mem/memory-device.h b/include/hw/mem/memory-device.h index 3354d6c166..a1d62cc551 100644 --- a/include/hw/mem/memory-device.h +++ b/include/hw/mem/memory-device.h @@ -38,6 +38,10 @@ typedef struct MemoryDeviceState MemoryDeviceState; * address in guest physical memory can either be specified explicitly * or get assigned automatically. * + * Some memory device might not own a memory region in certain device + * configurations. Such devices can logically get (un)plugged, however, + * empty memory devices are mostly ignored by the memory device code. + * * Conceptually, memory devices only span one memory region. If multiple * successive memory regions are used, a covering memory region has to * be provided. Scattered memory regions are not supported for single @@ -91,7 +95,8 @@ struct MemoryDeviceClass { uint64_t (*get_plugged_size)(const MemoryDeviceState *md, Error **errp); /* - * Return the memory region of the memory device. + * Return the memory region of the memory device. If the device is + * completely empty, returns NULL without an error. * * Called when (un)plugging the memory device, to (un)map the * memory region in guest physical memory, but also to detect the diff --git a/include/hw/pci/pci.h b/include/hw/pci/pci.h index ea5aff118b..fa6313aabc 100644 --- a/include/hw/pci/pci.h +++ b/include/hw/pci/pci.h @@ -363,10 +363,42 @@ void pci_bus_get_w64_range(PCIBus *bus, Range *range); void pci_device_deassert_intx(PCIDevice *dev); -typedef AddressSpace *(*PCIIOMMUFunc)(PCIBus *, void *, int); + +/** + * struct PCIIOMMUOps: callbacks structure for specific IOMMU handlers + * of a PCIBus + * + * Allows to modify the behavior of some IOMMU operations of the PCI + * framework for a set of devices on a PCI bus. + */ +typedef struct PCIIOMMUOps { + /** + * @get_address_space: get the address space for a set of devices + * on a PCI bus. + * + * Mandatory callback which returns a pointer to an #AddressSpace + * + * @bus: the #PCIBus being accessed. + * + * @opaque: the data passed to pci_setup_iommu(). + * + * @devfn: device and function number + */ + AddressSpace * (*get_address_space)(PCIBus *bus, void *opaque, int devfn); +} PCIIOMMUOps; AddressSpace *pci_device_iommu_address_space(PCIDevice *dev); -void pci_setup_iommu(PCIBus *bus, PCIIOMMUFunc fn, void *opaque); + +/** + * pci_setup_iommu: Initialize specific IOMMU handlers for a PCIBus + * + * Let PCI host bridges define specific operations. + * + * @bus: the #PCIBus being updated. + * @ops: the #PCIIOMMUOps + * @opaque: passed to callbacks of the @ops structure. + */ +void pci_setup_iommu(PCIBus *bus, const PCIIOMMUOps *ops, void *opaque); pcibus_t pci_bar_address(PCIDevice *d, int reg, uint8_t type, pcibus_t size); diff --git a/include/hw/pci/pci_bus.h b/include/hw/pci/pci_bus.h index 5653175957..2261312546 100644 --- a/include/hw/pci/pci_bus.h +++ b/include/hw/pci/pci_bus.h @@ -33,7 +33,7 @@ enum PCIBusFlags { struct PCIBus { BusState qbus; enum PCIBusFlags flags; - PCIIOMMUFunc iommu_fn; + const PCIIOMMUOps *iommu_ops; void *iommu_opaque; uint8_t devfn_min; uint32_t slot_reserved_mask; diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index 7780b9073a..a4a22accb9 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -99,6 +99,7 @@ typedef struct VFIOContainer { QLIST_HEAD(, VFIORamDiscardListener) vrdl_list; QLIST_ENTRY(VFIOContainer) next; QLIST_HEAD(, VFIODevice) device_list; + GList *iova_ranges; } VFIOContainer; typedef struct VFIOGuestIOMMU { @@ -206,11 +207,6 @@ typedef struct { hwaddr pages; } VFIOBitmap; -void vfio_host_win_add(VFIOContainer *container, - hwaddr min_iova, hwaddr max_iova, - uint64_t iova_pgsizes); -int vfio_host_win_del(VFIOContainer *container, hwaddr min_iova, - hwaddr max_iova); VFIOAddressSpace *vfio_get_address_space(AddressSpace *as); void vfio_put_address_space(VFIOAddressSpace *space); bool vfio_devices_all_running_and_saving(VFIOContainer *container); @@ -224,11 +220,14 @@ int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start); int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, hwaddr iova, hwaddr size); +/* SPAPR specific */ int vfio_container_add_section_window(VFIOContainer *container, MemoryRegionSection *section, Error **errp); void vfio_container_del_section_window(VFIOContainer *container, MemoryRegionSection *section); +int vfio_spapr_container_init(VFIOContainer *container, Error **errp); +void vfio_spapr_container_deinit(VFIOContainer *container); void vfio_disable_irqindex(VFIODevice *vbasedev, int index); void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index); @@ -288,13 +287,6 @@ vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id); struct vfio_info_cap_header * vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id); #endif -extern const MemoryListener vfio_prereg_listener; - -int vfio_spapr_create_window(VFIOContainer *container, - MemoryRegionSection *section, - hwaddr *pgsize); -int vfio_spapr_remove_window(VFIOContainer *container, - hwaddr offset_within_address_space); bool vfio_migration_realize(VFIODevice *vbasedev, Error **errp); void vfio_migration_exit(VFIODevice *vbasedev); diff --git a/include/hw/vfio/vfio.h b/include/hw/vfio/vfio.h deleted file mode 100644 index 86248f5436..0000000000 --- a/include/hw/vfio/vfio.h +++ /dev/null @@ -1,7 +0,0 @@ -#ifndef HW_VFIO_H -#define HW_VFIO_H - -bool vfio_eeh_as_ok(AddressSpace *as); -int vfio_eeh_as_op(AddressSpace *as, uint32_t op); - -#endif diff --git a/include/hw/virtio/virtio-iommu.h b/include/hw/virtio/virtio-iommu.h index a93fc5383e..781ebaea8f 100644 --- a/include/hw/virtio/virtio-iommu.h +++ b/include/hw/virtio/virtio-iommu.h @@ -39,6 +39,9 @@ typedef struct IOMMUDevice { AddressSpace as; MemoryRegion root; /* The root container of the device */ MemoryRegion bypass_mr; /* The alias of shared memory MR */ + GList *resv_regions; + GList *host_resv_ranges; + bool probe_done; } IOMMUDevice; typedef struct IOMMUPciBus { @@ -55,8 +58,8 @@ struct VirtIOIOMMU { GHashTable *as_by_busptr; IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX]; PCIBus *primary_bus; - ReservedRegion *reserved_regions; - uint32_t nb_reserved_regions; + ReservedRegion *prop_resv_regions; + uint32_t nr_prop_resv_regions; GTree *domains; QemuRecMutex mutex; GTree *endpoints; diff --git a/include/qemu/range.h b/include/qemu/range.h index 7e2b1cc447..205e1da76d 100644 --- a/include/qemu/range.h +++ b/include/qemu/range.h @@ -217,6 +217,20 @@ static inline int ranges_overlap(uint64_t first1, uint64_t len1, return !(last2 < first1 || last1 < first2); } +/* + * Return -1 if @a < @b, 1 @a > @b, and 0 if they touch or overlap. + * Both @a and @b must not be empty. + */ +int range_compare(Range *a, Range *b); + GList *range_list_insert(GList *list, Range *data); +/* + * Inverse an array of sorted ranges over the [low, high] span, ie. + * original ranges becomes holes in the newly allocated inv_ranges + */ +void range_inverse_array(GList *in_ranges, + GList **out_ranges, + uint64_t low, uint64_t high); + #endif diff --git a/include/qemu/reserved-region.h b/include/qemu/reserved-region.h new file mode 100644 index 0000000000..8e6f0a97e2 --- /dev/null +++ b/include/qemu/reserved-region.h @@ -0,0 +1,32 @@ +/* + * QEMU ReservedRegion helpers + * + * Copyright (c) 2023 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef QEMU_RESERVED_REGION_H +#define QEMU_RESERVED_REGION_H + +#include "exec/memory.h" + +/* + * Insert a new region into a sorted list of reserved regions. In case + * there is overlap with existing regions, the new added region has + * higher priority and replaces the overlapped segment. + */ +GList *resv_region_list_insert(GList *list, ReservedRegion *reg); + +#endif diff --git a/include/qemu/uuid.h b/include/qemu/uuid.h index e24a1099e4..869f84af09 100644 --- a/include/qemu/uuid.h +++ b/include/qemu/uuid.h @@ -78,9 +78,10 @@ typedef struct { "%02hhx%02hhx-" \ "%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx" -#define UUID_FMT_LEN 36 - #define UUID_NONE "00000000-0000-0000-0000-000000000000" +QEMU_BUILD_BUG_ON(sizeof(UUID_NONE) - 1 != 36); + +#define UUID_STR_LEN sizeof(UUID_NONE) void qemu_uuid_generate(QemuUUID *out); diff --git a/include/sysemu/kvm_xen.h b/include/sysemu/kvm_xen.h index 595abfbe40..961c702c4e 100644 --- a/include/sysemu/kvm_xen.h +++ b/include/sysemu/kvm_xen.h @@ -22,6 +22,7 @@ int kvm_xen_soft_reset(void); uint32_t kvm_xen_get_caps(void); void *kvm_xen_get_vcpu_info_hva(uint32_t vcpu_id); +bool kvm_xen_has_vcpu_callback_vector(void); void kvm_xen_inject_vcpu_callback_vector(uint32_t vcpu_id, int type); void kvm_xen_set_callback_asserted(void); int kvm_xen_set_vcpu_virq(uint32_t vcpu_id, uint16_t virq, uint16_t port); diff --git a/include/tcg/tcg-op-common.h b/include/tcg/tcg-op-common.h index 677aea6dd1..2d932a515e 100644 --- a/include/tcg/tcg-op-common.h +++ b/include/tcg/tcg-op-common.h @@ -12,236 +12,25 @@ #include "exec/helper-proto-common.h" #include "exec/helper-gen-common.h" -/* Basic output routines. Not for general consumption. */ - -void tcg_gen_op1(TCGOpcode, TCGArg); -void tcg_gen_op2(TCGOpcode, TCGArg, TCGArg); -void tcg_gen_op3(TCGOpcode, TCGArg, TCGArg, TCGArg); -void tcg_gen_op4(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg); -void tcg_gen_op5(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg); -void tcg_gen_op6(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg); - -void vec_gen_2(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg); -void vec_gen_3(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg, TCGArg); -void vec_gen_4(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg, TCGArg, TCGArg); - -static inline void tcg_gen_op1_i32(TCGOpcode opc, TCGv_i32 a1) -{ - tcg_gen_op1(opc, tcgv_i32_arg(a1)); -} - -static inline void tcg_gen_op1_i64(TCGOpcode opc, TCGv_i64 a1) -{ - tcg_gen_op1(opc, tcgv_i64_arg(a1)); -} - -static inline void tcg_gen_op1i(TCGOpcode opc, TCGArg a1) -{ - tcg_gen_op1(opc, a1); -} - -static inline void tcg_gen_op2_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2) -{ - tcg_gen_op2(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2)); -} - -static inline void tcg_gen_op2_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2) -{ - tcg_gen_op2(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2)); -} - -static inline void tcg_gen_op2i_i32(TCGOpcode opc, TCGv_i32 a1, TCGArg a2) -{ - tcg_gen_op2(opc, tcgv_i32_arg(a1), a2); -} - -static inline void tcg_gen_op2i_i64(TCGOpcode opc, TCGv_i64 a1, TCGArg a2) -{ - tcg_gen_op2(opc, tcgv_i64_arg(a1), a2); -} - -static inline void tcg_gen_op2ii(TCGOpcode opc, TCGArg a1, TCGArg a2) -{ - tcg_gen_op2(opc, a1, a2); -} - -static inline void tcg_gen_op3_i32(TCGOpcode opc, TCGv_i32 a1, - TCGv_i32 a2, TCGv_i32 a3) -{ - tcg_gen_op3(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), tcgv_i32_arg(a3)); -} - -static inline void tcg_gen_op3_i64(TCGOpcode opc, TCGv_i64 a1, - TCGv_i64 a2, TCGv_i64 a3) -{ - tcg_gen_op3(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), tcgv_i64_arg(a3)); -} - -static inline void tcg_gen_op3i_i32(TCGOpcode opc, TCGv_i32 a1, - TCGv_i32 a2, TCGArg a3) -{ - tcg_gen_op3(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), a3); -} - -static inline void tcg_gen_op3i_i64(TCGOpcode opc, TCGv_i64 a1, - TCGv_i64 a2, TCGArg a3) -{ - tcg_gen_op3(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), a3); -} - -static inline void tcg_gen_ldst_op_i32(TCGOpcode opc, TCGv_i32 val, - TCGv_ptr base, TCGArg offset) -{ - tcg_gen_op3(opc, tcgv_i32_arg(val), tcgv_ptr_arg(base), offset); -} - -static inline void tcg_gen_ldst_op_i64(TCGOpcode opc, TCGv_i64 val, - TCGv_ptr base, TCGArg offset) -{ - tcg_gen_op3(opc, tcgv_i64_arg(val), tcgv_ptr_arg(base), offset); -} - -static inline void tcg_gen_op4_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4) -{ - tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4)); -} - -static inline void tcg_gen_op4_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4) -{ - tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4)); -} - -static inline void tcg_gen_op4i_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGArg a4) -{ - tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), a4); -} - -static inline void tcg_gen_op4i_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGArg a4) -{ - tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), a4); -} - -static inline void tcg_gen_op4ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGArg a3, TCGArg a4) -{ - tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), a3, a4); -} - -static inline void tcg_gen_op4ii_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGArg a3, TCGArg a4) -{ - tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), a3, a4); -} - -static inline void tcg_gen_op5_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4, TCGv_i32 a5) -{ - tcg_gen_op5(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5)); -} - -static inline void tcg_gen_op5_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4, TCGv_i64 a5) -{ - tcg_gen_op5(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5)); -} - -static inline void tcg_gen_op5i_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4, TCGArg a5) -{ - tcg_gen_op5(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4), a5); -} - -static inline void tcg_gen_op5i_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4, TCGArg a5) -{ - tcg_gen_op5(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4), a5); -} - -static inline void tcg_gen_op5ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGArg a4, TCGArg a5) -{ - tcg_gen_op5(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), a4, a5); -} - -static inline void tcg_gen_op5ii_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGArg a4, TCGArg a5) -{ - tcg_gen_op5(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), a4, a5); -} - -static inline void tcg_gen_op6_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4, - TCGv_i32 a5, TCGv_i32 a6) -{ - tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5), - tcgv_i32_arg(a6)); -} - -static inline void tcg_gen_op6_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4, - TCGv_i64 a5, TCGv_i64 a6) -{ - tcg_gen_op6(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5), - tcgv_i64_arg(a6)); -} - -static inline void tcg_gen_op6i_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4, - TCGv_i32 a5, TCGArg a6) -{ - tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5), a6); -} - -static inline void tcg_gen_op6i_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4, - TCGv_i64 a5, TCGArg a6) -{ - tcg_gen_op6(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5), a6); -} - -static inline void tcg_gen_op6ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, - TCGv_i32 a3, TCGv_i32 a4, - TCGArg a5, TCGArg a6) -{ - tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), - tcgv_i32_arg(a3), tcgv_i32_arg(a4), a5, a6); -} - -static inline void tcg_gen_op6ii_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, - TCGv_i64 a3, TCGv_i64 a4, - TCGArg a5, TCGArg a6) -{ - tcg_gen_op6(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), - tcgv_i64_arg(a3), tcgv_i64_arg(a4), a5, a6); -} - +TCGv_i32 tcg_constant_i32(int32_t val); +TCGv_i64 tcg_constant_i64(int64_t val); +TCGv_vec tcg_constant_vec(TCGType type, unsigned vece, int64_t val); +TCGv_vec tcg_constant_vec_matching(TCGv_vec match, unsigned vece, int64_t val); + +TCGv_i32 tcg_temp_new_i32(void); +TCGv_i64 tcg_temp_new_i64(void); +TCGv_ptr tcg_temp_new_ptr(void); +TCGv_i128 tcg_temp_new_i128(void); +TCGv_vec tcg_temp_new_vec(TCGType type); +TCGv_vec tcg_temp_new_vec_matching(TCGv_vec match); + +TCGv_i32 tcg_global_mem_new_i32(TCGv_ptr reg, intptr_t off, const char *name); +TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t off, const char *name); +TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t off, const char *name); /* Generic ops. */ -static inline void gen_set_label(TCGLabel *l) -{ - l->present = 1; - tcg_gen_op1(INDEX_op_set_label, label_arg(l)); -} - +void gen_set_label(TCGLabel *l); void tcg_gen_br(TCGLabel *l); void tcg_gen_mb(TCGBar); @@ -285,16 +74,8 @@ void tcg_gen_goto_tb(unsigned idx); */ void tcg_gen_lookup_and_goto_ptr(void); -static inline void tcg_gen_plugin_cb_start(unsigned from, unsigned type, - unsigned wr) -{ - tcg_gen_op3(INDEX_op_plugin_cb_start, from, type, wr); -} - -static inline void tcg_gen_plugin_cb_end(void) -{ - tcg_emit_op(INDEX_op_plugin_cb_end, 0); -} +void tcg_gen_plugin_cb_start(unsigned from, unsigned type, unsigned wr); +void tcg_gen_plugin_cb_end(void); /* 32 bit ops */ @@ -374,128 +155,30 @@ void tcg_gen_abs_i32(TCGv_i32, TCGv_i32); /* Replicate a value of size @vece from @in to all the lanes in @out */ void tcg_gen_dup_i32(unsigned vece, TCGv_i32 out, TCGv_i32 in); -static inline void tcg_gen_discard_i32(TCGv_i32 arg) -{ - tcg_gen_op1_i32(INDEX_op_discard, arg); -} - -static inline void tcg_gen_mov_i32(TCGv_i32 ret, TCGv_i32 arg) -{ - if (ret != arg) { - tcg_gen_op2_i32(INDEX_op_mov_i32, ret, arg); - } -} - -static inline void tcg_gen_ld8u_i32(TCGv_i32 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_ld8u_i32, ret, arg2, offset); -} - -static inline void tcg_gen_ld8s_i32(TCGv_i32 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_ld8s_i32, ret, arg2, offset); -} - -static inline void tcg_gen_ld16u_i32(TCGv_i32 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_ld16u_i32, ret, arg2, offset); -} - -static inline void tcg_gen_ld16s_i32(TCGv_i32 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_ld16s_i32, ret, arg2, offset); -} - -static inline void tcg_gen_ld_i32(TCGv_i32 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_ld_i32, ret, arg2, offset); -} - -static inline void tcg_gen_st8_i32(TCGv_i32 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_st8_i32, arg1, arg2, offset); -} - -static inline void tcg_gen_st16_i32(TCGv_i32 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_st16_i32, arg1, arg2, offset); -} - -static inline void tcg_gen_st_i32(TCGv_i32 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i32(INDEX_op_st_i32, arg1, arg2, offset); -} - -static inline void tcg_gen_add_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_add_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_sub_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_sub_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_and_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_and_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_or_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_or_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_xor_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_xor_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_shl_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_shl_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_shr_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_shr_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_sar_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_sar_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_mul_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) -{ - tcg_gen_op3_i32(INDEX_op_mul_i32, ret, arg1, arg2); -} - -static inline void tcg_gen_neg_i32(TCGv_i32 ret, TCGv_i32 arg) -{ - if (TCG_TARGET_HAS_neg_i32) { - tcg_gen_op2_i32(INDEX_op_neg_i32, ret, arg); - } else { - tcg_gen_subfi_i32(ret, 0, arg); - } -} - -static inline void tcg_gen_not_i32(TCGv_i32 ret, TCGv_i32 arg) -{ - if (TCG_TARGET_HAS_not_i32) { - tcg_gen_op2_i32(INDEX_op_not_i32, ret, arg); - } else { - tcg_gen_xori_i32(ret, arg, -1); - } -} +void tcg_gen_discard_i32(TCGv_i32 arg); +void tcg_gen_mov_i32(TCGv_i32 ret, TCGv_i32 arg); + +void tcg_gen_ld8u_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_ld8s_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_ld16u_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_ld16s_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_ld_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset); + +void tcg_gen_st8_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_st16_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset); +void tcg_gen_st_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset); + +void tcg_gen_add_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_sub_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_and_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_or_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_xor_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_shl_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_shr_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_sar_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_mul_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_neg_i32(TCGv_i32 ret, TCGv_i32 arg); +void tcg_gen_not_i32(TCGv_i32 ret, TCGv_i32 arg); /* 64 bit ops */ @@ -580,130 +263,6 @@ void tcg_gen_abs_i64(TCGv_i64, TCGv_i64); /* Replicate a value of size @vece from @in to all the lanes in @out */ void tcg_gen_dup_i64(unsigned vece, TCGv_i64 out, TCGv_i64 in); -#if TCG_TARGET_REG_BITS == 64 -static inline void tcg_gen_discard_i64(TCGv_i64 arg) -{ - tcg_gen_op1_i64(INDEX_op_discard, arg); -} - -static inline void tcg_gen_mov_i64(TCGv_i64 ret, TCGv_i64 arg) -{ - if (ret != arg) { - tcg_gen_op2_i64(INDEX_op_mov_i64, ret, arg); - } -} - -static inline void tcg_gen_ld8u_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld8u_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld8s_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld8s_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld16u_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld16u_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld16s_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld16s_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld32u_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld32u_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld32s_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld32s_i64, ret, arg2, offset); -} - -static inline void tcg_gen_ld_i64(TCGv_i64 ret, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_ld_i64, ret, arg2, offset); -} - -static inline void tcg_gen_st8_i64(TCGv_i64 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_st8_i64, arg1, arg2, offset); -} - -static inline void tcg_gen_st16_i64(TCGv_i64 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_st16_i64, arg1, arg2, offset); -} - -static inline void tcg_gen_st32_i64(TCGv_i64 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_st32_i64, arg1, arg2, offset); -} - -static inline void tcg_gen_st_i64(TCGv_i64 arg1, TCGv_ptr arg2, - tcg_target_long offset) -{ - tcg_gen_ldst_op_i64(INDEX_op_st_i64, arg1, arg2, offset); -} - -static inline void tcg_gen_add_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_add_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_sub_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_sub_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_and_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_and_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_or_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_or_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_xor_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_xor_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_shl_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_shl_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_shr_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_shr_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_sar_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_sar_i64, ret, arg1, arg2); -} - -static inline void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) -{ - tcg_gen_op3_i64(INDEX_op_mul_i64, ret, arg1, arg2); -} -#else /* TCG_TARGET_REG_BITS == 32 */ void tcg_gen_st8_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset); void tcg_gen_st16_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset); void tcg_gen_st32_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset); @@ -728,16 +287,8 @@ void tcg_gen_shl_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2); void tcg_gen_shr_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2); void tcg_gen_sar_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2); void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2); -#endif /* TCG_TARGET_REG_BITS */ +void tcg_gen_neg_i64(TCGv_i64 ret, TCGv_i64 arg); -static inline void tcg_gen_neg_i64(TCGv_i64 ret, TCGv_i64 arg) -{ - if (TCG_TARGET_HAS_neg_i64) { - tcg_gen_op2_i64(INDEX_op_neg_i64, ret, arg); - } else { - tcg_gen_subfi_i64(ret, 0, arg); - } -} /* Size changing operations. */ @@ -748,19 +299,17 @@ void tcg_gen_extrl_i64_i32(TCGv_i32 ret, TCGv_i64 arg); void tcg_gen_extrh_i64_i32(TCGv_i32 ret, TCGv_i64 arg); void tcg_gen_extr_i64_i32(TCGv_i32 lo, TCGv_i32 hi, TCGv_i64 arg); void tcg_gen_extr32_i64(TCGv_i64 lo, TCGv_i64 hi, TCGv_i64 arg); +void tcg_gen_concat32_i64(TCGv_i64 ret, TCGv_i64 lo, TCGv_i64 hi); -void tcg_gen_mov_i128(TCGv_i128 dst, TCGv_i128 src); void tcg_gen_extr_i128_i64(TCGv_i64 lo, TCGv_i64 hi, TCGv_i128 arg); void tcg_gen_concat_i64_i128(TCGv_i128 ret, TCGv_i64 lo, TCGv_i64 hi); +/* 128 bit ops */ + +void tcg_gen_mov_i128(TCGv_i128 dst, TCGv_i128 src); void tcg_gen_ld_i128(TCGv_i128 ret, TCGv_ptr base, tcg_target_long offset); void tcg_gen_st_i128(TCGv_i128 val, TCGv_ptr base, tcg_target_long offset); -static inline void tcg_gen_concat32_i64(TCGv_i64 ret, TCGv_i64 lo, TCGv_i64 hi) -{ - tcg_gen_deposit_i64(ret, lo, hi, 32, 32); -} - /* Local load/store bit ops */ void tcg_gen_qemu_ld_i32_chk(TCGv_i32, TCGTemp *, TCGArg, MemOp, TCGType); @@ -926,6 +475,9 @@ void tcg_gen_stl_vec(TCGv_vec r, TCGv_ptr base, TCGArg offset, TCGType t); # define NAT TCGv_i64 #endif +TCGv_ptr tcg_constant_ptr_int(intptr_t x); +#define tcg_constant_ptr(X) tcg_constant_ptr_int((intptr_t)(X)) + static inline void tcg_gen_ld_ptr(TCGv_ptr r, TCGv_ptr a, intptr_t o) { glue(tcg_gen_ld_,PTR)((NAT)r, a, o); diff --git a/include/tcg/tcg-opc.h b/include/tcg/tcg-opc.h index 6eff3d9106..b80227fa1c 100644 --- a/include/tcg/tcg-opc.h +++ b/include/tcg/tcg-opc.h @@ -47,7 +47,7 @@ DEF(mb, 0, 0, 1, 0) DEF(mov_i32, 1, 1, 0, TCG_OPF_NOT_PRESENT) DEF(setcond_i32, 1, 2, 1, 0) DEF(negsetcond_i32, 1, 2, 1, IMPL(TCG_TARGET_HAS_negsetcond_i32)) -DEF(movcond_i32, 1, 4, 1, IMPL(TCG_TARGET_HAS_movcond_i32)) +DEF(movcond_i32, 1, 4, 1, 0) /* load/store */ DEF(ld8u_i32, 1, 1, 1, 0) DEF(ld8s_i32, 1, 1, 1, 0) @@ -100,7 +100,7 @@ DEF(ext16u_i32, 1, 1, 0, IMPL(TCG_TARGET_HAS_ext16u_i32)) DEF(bswap16_i32, 1, 1, 1, IMPL(TCG_TARGET_HAS_bswap16_i32)) DEF(bswap32_i32, 1, 1, 1, IMPL(TCG_TARGET_HAS_bswap32_i32)) DEF(not_i32, 1, 1, 0, IMPL(TCG_TARGET_HAS_not_i32)) -DEF(neg_i32, 1, 1, 0, IMPL(TCG_TARGET_HAS_neg_i32)) +DEF(neg_i32, 1, 1, 0, 0) DEF(andc_i32, 1, 2, 0, IMPL(TCG_TARGET_HAS_andc_i32)) DEF(orc_i32, 1, 2, 0, IMPL(TCG_TARGET_HAS_orc_i32)) DEF(eqv_i32, 1, 2, 0, IMPL(TCG_TARGET_HAS_eqv_i32)) @@ -113,7 +113,7 @@ DEF(ctpop_i32, 1, 1, 0, IMPL(TCG_TARGET_HAS_ctpop_i32)) DEF(mov_i64, 1, 1, 0, TCG_OPF_64BIT | TCG_OPF_NOT_PRESENT) DEF(setcond_i64, 1, 2, 1, IMPL64) DEF(negsetcond_i64, 1, 2, 1, IMPL64 | IMPL(TCG_TARGET_HAS_negsetcond_i64)) -DEF(movcond_i64, 1, 4, 1, IMPL64 | IMPL(TCG_TARGET_HAS_movcond_i64)) +DEF(movcond_i64, 1, 4, 1, IMPL64) /* load/store */ DEF(ld8u_i64, 1, 1, 1, IMPL64) DEF(ld8s_i64, 1, 1, 1, IMPL64) @@ -171,7 +171,7 @@ DEF(bswap16_i64, 1, 1, 1, IMPL64 | IMPL(TCG_TARGET_HAS_bswap16_i64)) DEF(bswap32_i64, 1, 1, 1, IMPL64 | IMPL(TCG_TARGET_HAS_bswap32_i64)) DEF(bswap64_i64, 1, 1, 1, IMPL64 | IMPL(TCG_TARGET_HAS_bswap64_i64)) DEF(not_i64, 1, 1, 0, IMPL64 | IMPL(TCG_TARGET_HAS_not_i64)) -DEF(neg_i64, 1, 1, 0, IMPL64 | IMPL(TCG_TARGET_HAS_neg_i64)) +DEF(neg_i64, 1, 1, 0, IMPL64) DEF(andc_i64, 1, 2, 0, IMPL64 | IMPL(TCG_TARGET_HAS_andc_i64)) DEF(orc_i64, 1, 2, 0, IMPL64 | IMPL(TCG_TARGET_HAS_orc_i64)) DEF(eqv_i64, 1, 2, 0, IMPL64 | IMPL(TCG_TARGET_HAS_eqv_i64)) diff --git a/include/tcg/tcg-temp-internal.h b/include/tcg/tcg-temp-internal.h index dded2917e5..44192c55a9 100644 --- a/include/tcg/tcg-temp-internal.h +++ b/include/tcg/tcg-temp-internal.h @@ -31,53 +31,15 @@ void tcg_temp_free_internal(TCGTemp *); -static inline void tcg_temp_free_i32(TCGv_i32 arg) -{ - tcg_temp_free_internal(tcgv_i32_temp(arg)); -} - -static inline void tcg_temp_free_i64(TCGv_i64 arg) -{ - tcg_temp_free_internal(tcgv_i64_temp(arg)); -} - -static inline void tcg_temp_free_i128(TCGv_i128 arg) -{ - tcg_temp_free_internal(tcgv_i128_temp(arg)); -} - -static inline void tcg_temp_free_ptr(TCGv_ptr arg) -{ - tcg_temp_free_internal(tcgv_ptr_temp(arg)); -} - -static inline void tcg_temp_free_vec(TCGv_vec arg) -{ - tcg_temp_free_internal(tcgv_vec_temp(arg)); -} - -static inline TCGv_i32 tcg_temp_ebb_new_i32(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB); - return temp_tcgv_i32(t); -} - -static inline TCGv_i64 tcg_temp_ebb_new_i64(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB); - return temp_tcgv_i64(t); -} - -static inline TCGv_i128 tcg_temp_ebb_new_i128(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB); - return temp_tcgv_i128(t); -} - -static inline TCGv_ptr tcg_temp_ebb_new_ptr(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB); - return temp_tcgv_ptr(t); -} +void tcg_temp_free_i32(TCGv_i32 arg); +void tcg_temp_free_i64(TCGv_i64 arg); +void tcg_temp_free_i128(TCGv_i128 arg); +void tcg_temp_free_ptr(TCGv_ptr arg); +void tcg_temp_free_vec(TCGv_vec arg); + +TCGv_i32 tcg_temp_ebb_new_i32(void); +TCGv_i64 tcg_temp_ebb_new_i64(void); +TCGv_ptr tcg_temp_ebb_new_ptr(void); +TCGv_i128 tcg_temp_ebb_new_i128(void); #endif /* TCG_TEMP_FREE_H */ diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index a9282cdcc6..daf2a5bf9e 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -82,7 +82,6 @@ typedef uint64_t TCGRegSet; #define TCG_TARGET_HAS_bswap16_i64 0 #define TCG_TARGET_HAS_bswap32_i64 0 #define TCG_TARGET_HAS_bswap64_i64 0 -#define TCG_TARGET_HAS_neg_i64 0 #define TCG_TARGET_HAS_not_i64 0 #define TCG_TARGET_HAS_andc_i64 0 #define TCG_TARGET_HAS_orc_i64 0 @@ -96,7 +95,6 @@ typedef uint64_t TCGRegSet; #define TCG_TARGET_HAS_extract_i64 0 #define TCG_TARGET_HAS_sextract_i64 0 #define TCG_TARGET_HAS_extract2_i64 0 -#define TCG_TARGET_HAS_movcond_i64 0 #define TCG_TARGET_HAS_negsetcond_i64 0 #define TCG_TARGET_HAS_add2_i64 0 #define TCG_TARGET_HAS_sub2_i64 0 @@ -795,60 +793,6 @@ void tb_target_set_jmp_target(const TranslationBlock *, int, void tcg_set_frame(TCGContext *s, TCGReg reg, intptr_t start, intptr_t size); -TCGTemp *tcg_global_mem_new_internal(TCGType, TCGv_ptr, - intptr_t, const char *); -TCGTemp *tcg_temp_new_internal(TCGType, TCGTempKind); -TCGv_vec tcg_temp_new_vec(TCGType type); -TCGv_vec tcg_temp_new_vec_matching(TCGv_vec match); - -static inline TCGv_i32 tcg_global_mem_new_i32(TCGv_ptr reg, intptr_t offset, - const char *name) -{ - TCGTemp *t = tcg_global_mem_new_internal(TCG_TYPE_I32, reg, offset, name); - return temp_tcgv_i32(t); -} - -static inline TCGv_i32 tcg_temp_new_i32(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_TB); - return temp_tcgv_i32(t); -} - -static inline TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t offset, - const char *name) -{ - TCGTemp *t = tcg_global_mem_new_internal(TCG_TYPE_I64, reg, offset, name); - return temp_tcgv_i64(t); -} - -static inline TCGv_i64 tcg_temp_new_i64(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_TB); - return temp_tcgv_i64(t); -} - -static inline TCGv_i128 tcg_temp_new_i128(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_TB); - return temp_tcgv_i128(t); -} - -static inline TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t offset, - const char *name) -{ - TCGTemp *t = tcg_global_mem_new_internal(TCG_TYPE_PTR, reg, offset, name); - return temp_tcgv_ptr(t); -} - -static inline TCGv_ptr tcg_temp_new_ptr(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_TB); - return temp_tcgv_ptr(t); -} - -void tcg_dump_info(GString *buf); -void tcg_dump_op_count(GString *buf); - #define TCG_CT_CONST 1 /* any constant of register size */ typedef struct TCGArgConstraint { @@ -937,32 +881,6 @@ void tcg_remove_ops_after(TCGOp *op); void tcg_optimize(TCGContext *s); -/* - * Locate or create a read-only temporary that is a constant. - * This kind of temporary need not be freed, but for convenience - * will be silently ignored by tcg_temp_free_*. - */ -TCGTemp *tcg_constant_internal(TCGType type, int64_t val); - -static inline TCGv_i32 tcg_constant_i32(int32_t val) -{ - return temp_tcgv_i32(tcg_constant_internal(TCG_TYPE_I32, val)); -} - -static inline TCGv_i64 tcg_constant_i64(int64_t val) -{ - return temp_tcgv_i64(tcg_constant_internal(TCG_TYPE_I64, val)); -} - -TCGv_vec tcg_constant_vec(TCGType type, unsigned vece, int64_t val); -TCGv_vec tcg_constant_vec_matching(TCGv_vec match, unsigned vece, int64_t val); - -#if UINTPTR_MAX == UINT32_MAX -# define tcg_constant_ptr(x) ((TCGv_ptr)tcg_constant_i32((intptr_t)(x))) -#else -# define tcg_constant_ptr(x) ((TCGv_ptr)tcg_constant_i64((intptr_t)(x))) -#endif - TCGLabel *gen_new_label(void); /** diff --git a/io/channel-socket.c b/io/channel-socket.c index 02ffb51e99..3a899b0608 100644 --- a/io/channel-socket.c +++ b/io/channel-socket.c @@ -782,6 +782,11 @@ static int qio_channel_socket_flush(QIOChannel *ioc, "Error not from zero copy"); return -1; } + if (serr->ee_data < serr->ee_info) { + error_setg_errno(errp, serr->ee_origin, + "Wrong notification bounds"); + return -1; + } /* No errors, count successfully finished sendmsg()*/ sioc->zero_copy_sent += serr->ee_data - serr->ee_info + 1; diff --git a/meson.build b/meson.build index dcef8b1e79..51a51075db 100644 --- a/meson.build +++ b/meson.build @@ -1323,6 +1323,30 @@ if not get_option('glusterfs').auto() or have_block endif endif +hv_balloon = false +if get_option('hv_balloon').allowed() and have_system + if cc.links(''' + #include <string.h> + #include <gmodule.h> + int main(void) { + GTree *tree; + + tree = g_tree_new((GCompareFunc)strcmp); + (void)g_tree_node_first(tree); + g_tree_destroy(tree); + return 0; + } + ''', dependencies: glib) + hv_balloon = true + else + if get_option('hv_balloon').enabled() + error('could not enable hv-balloon, update your glib') + else + warning('could not find glib support for hv-balloon, disabling') + endif + endif +endif + libssh = not_found if not get_option('libssh').auto() or have_block libssh = dependency('libssh', version: '>=0.8.7', @@ -2855,7 +2879,8 @@ host_kconfig = \ (targetos == 'linux' ? ['CONFIG_LINUX=y'] : []) + \ (have_pvrdma ? ['CONFIG_PVRDMA=y'] : []) + \ (multiprocess_allowed ? ['CONFIG_MULTIPROCESS_ALLOWED=y'] : []) + \ - (vfio_user_server_allowed ? ['CONFIG_VFIO_USER_SERVER_ALLOWED=y'] : []) + (vfio_user_server_allowed ? ['CONFIG_VFIO_USER_SERVER_ALLOWED=y'] : []) + \ + (hv_balloon ? ['CONFIG_HV_BALLOON_POSSIBLE=y'] : []) ignored = [ 'TARGET_XML_FILES', 'TARGET_ABI_DIR', 'TARGET_ARCH' ] @@ -4321,6 +4346,7 @@ if targetos == 'windows' endif summary_info += {'seccomp support': seccomp} summary_info += {'GlusterFS support': glusterfs} +summary_info += {'hv-balloon support': hv_balloon} summary_info += {'TPM support': have_tpm} summary_info += {'libssh support': libssh} summary_info += {'lzo support': lzo} diff --git a/meson_options.txt b/meson_options.txt index 3c7398f3c6..5c212fcd45 100644 --- a/meson_options.txt +++ b/meson_options.txt @@ -150,6 +150,8 @@ option('gio', type : 'feature', value : 'auto', description: 'use libgio for D-Bus support') option('glusterfs', type : 'feature', value : 'auto', description: 'Glusterfs block device driver') +option('hv_balloon', type : 'feature', value : 'auto', + description: 'hv-balloon driver (requires Glib 2.68+ GTree API)') option('libdw', type : 'feature', value : 'auto', description: 'debuginfo support') option('libiscsi', type : 'feature', value : 'auto', diff --git a/migration/savevm.c b/migration/savevm.c index bc98c2ea6f..eec5503a42 100644 --- a/migration/savevm.c +++ b/migration/savevm.c @@ -471,8 +471,8 @@ static bool vmstate_uuid_needed(void *opaque) static int vmstate_uuid_post_load(void *opaque, int version_id) { SaveState *state = opaque; - char uuid_src[UUID_FMT_LEN + 1]; - char uuid_dst[UUID_FMT_LEN + 1]; + char uuid_src[UUID_STR_LEN]; + char uuid_dst[UUID_STR_LEN]; if (!qemu_uuid_set) { /* diff --git a/monitor/monitor.c b/monitor/monitor.c index 941f87815a..01ede1babd 100644 --- a/monitor/monitor.c +++ b/monitor/monitor.c @@ -315,6 +315,7 @@ static MonitorQAPIEventConf monitor_qapi_event_conf[QAPI_EVENT__MAX] = { [QAPI_EVENT_QUORUM_FAILURE] = { 1000 * SCALE_MS }, [QAPI_EVENT_VSERPORT_CHANGE] = { 1000 * SCALE_MS }, [QAPI_EVENT_MEMORY_DEVICE_SIZE_CHANGE] = { 1000 * SCALE_MS }, + [QAPI_EVENT_HV_BALLOON_STATUS_REPORT] = { 1000 * SCALE_MS }, }; /* diff --git a/qapi/block-core.json b/qapi/block-core.json index 99961256f2..ca390c5700 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -3528,16 +3528,20 @@ # @pass-discard-other: whether discard requests for the data source # should be issued on other occasions where a cluster gets freed # -# @discard-no-unref: when enabled, discards from the guest will not -# cause cluster allocations to be relinquished. This prevents -# qcow2 fragmentation that would be caused by such discards. -# Besides potential performance degradation, such fragmentation -# can lead to increased allocation of clusters past the end of the -# image file, resulting in image files whose file length can grow -# much larger than their guest disk size would suggest. If image -# file length is of concern (e.g. when storing qcow2 images -# directly on block devices), you should consider enabling this -# option. (since 8.1) +# @discard-no-unref: when enabled, data clusters will remain +# preallocated when they are no longer used, e.g. because they are +# discarded or converted to zero clusters. As usual, whether the +# old data is discarded or kept on the protocol level (i.e. in the +# image file) depends on the setting of the pass-discard-request +# option. Keeping the clusters preallocated prevents qcow2 +# fragmentation that would otherwise be caused by freeing and +# re-allocating them later. Besides potential performance +# degradation, such fragmentation can lead to increased allocation +# of clusters past the end of the image file, resulting in image +# files whose file length can grow much larger than their guest disk +# size would suggest. If image file length is of concern (e.g. when +# storing qcow2 images directly on block devices), you should +# consider enabling this option. (since 8.1) # # @overlap-check: which overlap checks to perform for writes to the # image, defaults to 'cached' (since 2.2) diff --git a/qapi/machine.json b/qapi/machine.json index 6c9d2f6dcf..b6d634b30d 100644 --- a/qapi/machine.json +++ b/qapi/machine.json @@ -1138,6 +1138,68 @@ 'data': { 'actual': 'int' } } ## +# @HvBalloonInfo: +# +# hv-balloon guest-provided memory status information. +# +# @committed: the amount of memory in use inside the guest plus the +# amount of the memory unusable inside the guest (ballooned out, +# offline, etc.) +# +# @available: the amount of the memory inside the guest available for +# new allocations ("free") +# +# Since: 8.2 +## +{ 'struct': 'HvBalloonInfo', + 'data': { 'committed': 'size', 'available': 'size' } } + +## +# @query-hv-balloon-status-report: +# +# Returns the hv-balloon driver data contained in the last received "STATUS" +# message from the guest. +# +# Returns: +# - @HvBalloonInfo on success +# - If no hv-balloon device is present, guest memory status reporting +# is not enabled or no guest memory status report received yet, +# GenericError +# +# Since: 8.2 +# +# Example: +# +# -> { "execute": "query-hv-balloon-status-report" } +# <- { "return": { +# "committed": 816640000, +# "available": 3333054464 +# } +# } +## +{ 'command': 'query-hv-balloon-status-report', 'returns': 'HvBalloonInfo' } + +## +# @HV_BALLOON_STATUS_REPORT: +# +# Emitted when the hv-balloon driver receives a "STATUS" message from +# the guest. +# +# Note: this event is rate-limited. +# +# Since: 8.2 +# +# Example: +# +# <- { "event": "HV_BALLOON_STATUS_REPORT", +# "data": { "committed": 816640000, "available": 3333054464 }, +# "timestamp": { "seconds": 1600295492, "microseconds": 661044 } } +# +## +{ 'event': 'HV_BALLOON_STATUS_REPORT', + 'data': 'HvBalloonInfo' } + +## # @MemoryInfo: # # Actual memory information in bytes. @@ -1290,6 +1352,29 @@ } ## +# @HvBalloonDeviceInfo: +# +# hv-balloon provided memory state information +# +# @id: device's ID +# +# @memaddr: physical address in memory, where device is mapped +# +# @max-size: the maximum size of memory that the device can provide +# +# @memdev: memory backend linked with device +# +# Since: 8.2 +## +{ 'struct': 'HvBalloonDeviceInfo', + 'data': { '*id': 'str', + '*memaddr': 'size', + 'max-size': 'size', + '*memdev': 'str' + } +} + +## # @MemoryDeviceInfoKind: # # @nvdimm: since 2.12 @@ -1300,10 +1385,13 @@ # # @sgx-epc: since 6.2. # +# @hv-balloon: since 8.2. +# # Since: 2.1 ## { 'enum': 'MemoryDeviceInfoKind', - 'data': [ 'dimm', 'nvdimm', 'virtio-pmem', 'virtio-mem', 'sgx-epc' ] } + 'data': [ 'dimm', 'nvdimm', 'virtio-pmem', 'virtio-mem', 'sgx-epc', + 'hv-balloon' ] } ## # @PCDIMMDeviceInfoWrapper: @@ -1338,6 +1426,14 @@ 'data': { 'data': 'SgxEPCDeviceInfo' } } ## +# @HvBalloonDeviceInfoWrapper: +# +# Since: 8.2 +## +{ 'struct': 'HvBalloonDeviceInfoWrapper', + 'data': { 'data': 'HvBalloonDeviceInfo' } } + +## # @MemoryDeviceInfo: # # Union containing information about a memory device @@ -1351,7 +1447,8 @@ 'nvdimm': 'PCDIMMDeviceInfoWrapper', 'virtio-pmem': 'VirtioPMEMDeviceInfoWrapper', 'virtio-mem': 'VirtioMEMDeviceInfoWrapper', - 'sgx-epc': 'SgxEPCDeviceInfoWrapper' + 'sgx-epc': 'SgxEPCDeviceInfoWrapper', + 'hv-balloon': 'HvBalloonDeviceInfoWrapper' } } diff --git a/qemu-options.hx b/qemu-options.hx index e26230bac5..7809036d8c 100644 --- a/qemu-options.hx +++ b/qemu-options.hx @@ -1457,9 +1457,13 @@ SRST (on/off; default: off) ``discard-no-unref`` - When enabled, discards from the guest will not cause cluster - allocations to be relinquished. This prevents qcow2 fragmentation - that would be caused by such discards. Besides potential + When enabled, data clusters will remain preallocated when they are + no longer used, e.g. because they are discarded or converted to + zero clusters. As usual, whether the old data is discarded or kept + on the protocol level (i.e. in the image file) depends on the + setting of the pass-discard-request option. Keeping the clusters + preallocated prevents qcow2 fragmentation that would otherwise be + caused by freeing and re-allocating them later. Besides potential performance degradation, such fragmentation can lead to increased allocation of clusters past the end of the image file, resulting in image files whose file length can grow much larger diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh index 7ca4b77eae..e9d6d39279 100644 --- a/scripts/meson-buildoptions.sh +++ b/scripts/meson-buildoptions.sh @@ -123,6 +123,7 @@ meson_options_help() { printf "%s\n" ' gtk-clipboard clipboard support for the gtk UI (EXPERIMENTAL, MAY HANG)' printf "%s\n" ' guest-agent Build QEMU Guest Agent' printf "%s\n" ' guest-agent-msi Build MSI package for the QEMU Guest Agent' + printf "%s\n" ' hv-balloon hv-balloon driver (requires Glib 2.68+ GTree API)' printf "%s\n" ' hvf HVF acceleration support' printf "%s\n" ' iconv Font glyph conversion support' printf "%s\n" ' jack JACK sound support' @@ -333,6 +334,8 @@ _meson_option_parse() { --disable-guest-agent-msi) printf "%s" -Dguest_agent_msi=disabled ;; --enable-hexagon-idef-parser) printf "%s" -Dhexagon_idef_parser=true ;; --disable-hexagon-idef-parser) printf "%s" -Dhexagon_idef_parser=false ;; + --enable-hv-balloon) printf "%s" -Dhv_balloon=enabled ;; + --disable-hv-balloon) printf "%s" -Dhv_balloon=disabled ;; --enable-hvf) printf "%s" -Dhvf=enabled ;; --disable-hvf) printf "%s" -Dhvf=disabled ;; --iasl=*) quote_sh "-Diasl=$2" ;; diff --git a/system/memory.c b/system/memory.c index 4928f2525d..304fa843ea 100644 --- a/system/memory.c +++ b/system/memory.c @@ -1921,6 +1921,19 @@ int memory_region_iommu_set_page_size_mask(IOMMUMemoryRegion *iommu_mr, return ret; } +int memory_region_iommu_set_iova_ranges(IOMMUMemoryRegion *iommu_mr, + GList *iova_ranges, + Error **errp) +{ + IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr); + int ret = 0; + + if (imrc->iommu_set_iova_ranges) { + ret = imrc->iommu_set_iova_ranges(iommu_mr, iova_ranges, errp); + } + return ret; +} + int memory_region_register_iommu_notifier(MemoryRegion *mr, IOMMUNotifier *n, Error **errp) { diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode index 0cf1147074..8a20dce3c8 100644 --- a/target/arm/tcg/a64.decode +++ b/target/arm/tcg/a64.decode @@ -462,7 +462,7 @@ LDAPR sz:2 111 0 00 1 0 1 11111 1100 00 rn:5 rt:5 # Load/store register (pointer authentication) # LDRA immediate is 10 bits signed and scaled, but the bits aren't all contiguous -%ldra_imm 22:s1 12:9 !function=times_2 +%ldra_imm 22:s1 12:9 !function=times_8 LDRA 11 111 0 00 m:1 . 1 ......... w:1 1 rn:5 rt:5 imm=%ldra_imm diff --git a/target/arm/tcg/translate.h b/target/arm/tcg/translate.h index 9efe00cf6c..3c3bb3431a 100644 --- a/target/arm/tcg/translate.h +++ b/target/arm/tcg/translate.h @@ -205,6 +205,11 @@ static inline int times_4(DisasContext *s, int x) return x * 4; } +static inline int times_8(DisasContext *s, int x) +{ + return x * 8; +} + static inline int times_2_plus_1(DisasContext *s, int x) { return x * 2 + 1; diff --git a/target/i386/kvm/xen-emu.c b/target/i386/kvm/xen-emu.c index 76348f9d5d..75b2c557b9 100644 --- a/target/i386/kvm/xen-emu.c +++ b/target/i386/kvm/xen-emu.c @@ -267,7 +267,6 @@ static bool kvm_xen_hcall_xen_version(struct kvm_xen_exit *exit, X86CPU *cpu, fi.submap |= 1 << XENFEAT_writable_page_tables | 1 << XENFEAT_writable_descriptor_tables | 1 << XENFEAT_auto_translated_physmap | - 1 << XENFEAT_supervisor_mode_kernel | 1 << XENFEAT_hvm_callback_vector | 1 << XENFEAT_hvm_safe_pvclock | 1 << XENFEAT_hvm_pirqs; @@ -307,7 +306,7 @@ static int kvm_xen_set_vcpu_callback_vector(CPUState *cs) trace_kvm_xen_set_vcpu_callback(cs->cpu_index, vector); - return kvm_vcpu_ioctl(cs, KVM_XEN_HVM_SET_ATTR, &xva); + return kvm_vcpu_ioctl(cs, KVM_XEN_VCPU_SET_ATTR, &xva); } static void do_set_vcpu_callback_vector(CPUState *cs, run_on_cpu_data data) @@ -425,6 +424,13 @@ void kvm_xen_set_callback_asserted(void) } } +bool kvm_xen_has_vcpu_callback_vector(void) +{ + CPUState *cs = qemu_get_cpu(0); + + return cs && !!X86_CPU(cs)->env.xen_vcpu_callback_vector; +} + void kvm_xen_inject_vcpu_callback_vector(uint32_t vcpu_id, int type) { CPUState *cs = qemu_get_cpu(vcpu_id); @@ -441,7 +447,8 @@ void kvm_xen_inject_vcpu_callback_vector(uint32_t vcpu_id, int type) * deliver it as an MSI. */ MSIMessage msg = { - .address = APIC_DEFAULT_ADDRESS | X86_CPU(cs)->apic_id, + .address = APIC_DEFAULT_ADDRESS | + (X86_CPU(cs)->apic_id << MSI_ADDR_DEST_ID_SHIFT), .data = vector | (1UL << MSI_DATA_LEVEL_SHIFT), }; kvm_irqchip_send_msi(kvm_state, msg); @@ -850,8 +857,7 @@ static bool kvm_xen_hcall_hvm_op(struct kvm_xen_exit *exit, X86CPU *cpu, int ret = -ENOSYS; switch (cmd) { case HVMOP_set_evtchn_upcall_vector: - ret = kvm_xen_hcall_evtchn_upcall_vector(exit, cpu, - exit->u.hcall.params[0]); + ret = kvm_xen_hcall_evtchn_upcall_vector(exit, cpu, arg); break; case HVMOP_pagetable_dying: diff --git a/tcg/aarch64/tcg-target.h b/tcg/aarch64/tcg-target.h index 98727ea53b..33f15a564a 100644 --- a/tcg/aarch64/tcg-target.h +++ b/tcg/aarch64/tcg-target.h @@ -71,7 +71,6 @@ typedef enum { #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 #define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_rot_i32 1 #define TCG_TARGET_HAS_andc_i32 1 #define TCG_TARGET_HAS_orc_i32 1 @@ -85,7 +84,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i32 1 #define TCG_TARGET_HAS_sextract_i32 1 #define TCG_TARGET_HAS_extract2_i32 1 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_add2_i32 1 #define TCG_TARGET_HAS_sub2_i32 1 @@ -108,7 +106,6 @@ typedef enum { #define TCG_TARGET_HAS_bswap32_i64 1 #define TCG_TARGET_HAS_bswap64_i64 1 #define TCG_TARGET_HAS_not_i64 1 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_rot_i64 1 #define TCG_TARGET_HAS_andc_i64 1 #define TCG_TARGET_HAS_orc_i64 1 @@ -122,7 +119,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i64 1 #define TCG_TARGET_HAS_sextract_i64 1 #define TCG_TARGET_HAS_extract2_i64 1 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_add2_i64 1 #define TCG_TARGET_HAS_sub2_i64 1 diff --git a/tcg/arm/tcg-target.h b/tcg/arm/tcg-target.h index 311a985209..a712cc80ad 100644 --- a/tcg/arm/tcg-target.h +++ b/tcg/arm/tcg-target.h @@ -101,7 +101,6 @@ extern bool use_neon_instructions; #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 #define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_rot_i32 1 #define TCG_TARGET_HAS_andc_i32 1 #define TCG_TARGET_HAS_orc_i32 0 @@ -115,7 +114,6 @@ extern bool use_neon_instructions; #define TCG_TARGET_HAS_extract_i32 use_armv7_instructions #define TCG_TARGET_HAS_sextract_i32 use_armv7_instructions #define TCG_TARGET_HAS_extract2_i32 1 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_mulu2_i32 1 #define TCG_TARGET_HAS_muls2_i32 1 diff --git a/tcg/i386/tcg-target.h b/tcg/i386/tcg-target.h index 8417ea4899..fa34deec47 100644 --- a/tcg/i386/tcg-target.h +++ b/tcg/i386/tcg-target.h @@ -135,7 +135,6 @@ typedef enum { #define TCG_TARGET_HAS_ext16u_i32 1 #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_not_i32 1 #define TCG_TARGET_HAS_andc_i32 have_bmi1 #define TCG_TARGET_HAS_orc_i32 0 @@ -149,7 +148,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i32 1 #define TCG_TARGET_HAS_sextract_i32 1 #define TCG_TARGET_HAS_extract2_i32 1 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_add2_i32 1 #define TCG_TARGET_HAS_sub2_i32 1 @@ -172,7 +170,6 @@ typedef enum { #define TCG_TARGET_HAS_bswap16_i64 1 #define TCG_TARGET_HAS_bswap32_i64 1 #define TCG_TARGET_HAS_bswap64_i64 1 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_not_i64 1 #define TCG_TARGET_HAS_andc_i64 have_bmi1 #define TCG_TARGET_HAS_orc_i64 0 @@ -186,7 +183,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i64 1 #define TCG_TARGET_HAS_sextract_i64 0 #define TCG_TARGET_HAS_extract2_i64 1 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_add2_i64 1 #define TCG_TARGET_HAS_sub2_i64 1 diff --git a/tcg/loongarch64/tcg-target-con-set.h b/tcg/loongarch64/tcg-target-con-set.h index 77d62e38e7..cae6c2aad6 100644 --- a/tcg/loongarch64/tcg-target-con-set.h +++ b/tcg/loongarch64/tcg-target-con-set.h @@ -38,4 +38,4 @@ C_O1_I2(w, w, wM) C_O1_I2(w, w, wA) C_O1_I3(w, w, w, w) C_O1_I4(r, rZ, rJ, rZ, rZ) -C_O2_I1(r, r, r) +C_N2_I1(r, r, r) diff --git a/tcg/loongarch64/tcg-target.c.inc b/tcg/loongarch64/tcg-target.c.inc index ccf133db4b..a588fb3085 100644 --- a/tcg/loongarch64/tcg-target.c.inc +++ b/tcg/loongarch64/tcg-target.c.inc @@ -32,8 +32,6 @@ #include "../tcg-ldst.c.inc" #include <asm/hwcap.h> -bool use_lsx_instructions; - #ifdef CONFIG_DEBUG_TCG static const char * const tcg_target_reg_names[TCG_TARGET_NB_REGS] = { "zero", @@ -1103,13 +1101,18 @@ static void tcg_out_qemu_ldst_i128(TCGContext *s, TCGReg data_lo, TCGReg data_hi } } else { /* Otherwise use a pair of LD/ST. */ - tcg_out_opc_add_d(s, TCG_REG_TMP0, h.base, h.index); + TCGReg base = h.base; + if (h.index != TCG_REG_ZERO) { + base = TCG_REG_TMP0; + tcg_out_opc_add_d(s, base, h.base, h.index); + } if (is_ld) { - tcg_out_opc_ld_d(s, data_lo, TCG_REG_TMP0, 0); - tcg_out_opc_ld_d(s, data_hi, TCG_REG_TMP0, 8); + tcg_debug_assert(base != data_lo); + tcg_out_opc_ld_d(s, data_lo, base, 0); + tcg_out_opc_ld_d(s, data_hi, base, 8); } else { - tcg_out_opc_st_d(s, data_lo, TCG_REG_TMP0, 0); - tcg_out_opc_st_d(s, data_hi, TCG_REG_TMP0, 8); + tcg_out_opc_st_d(s, data_lo, base, 0); + tcg_out_opc_st_d(s, data_hi, base, 8); } } @@ -1438,6 +1441,13 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, } break; + case INDEX_op_neg_i32: + tcg_out_opc_sub_w(s, a0, TCG_REG_ZERO, a1); + break; + case INDEX_op_neg_i64: + tcg_out_opc_sub_d(s, a0, TCG_REG_ZERO, a1); + break; + case INDEX_op_mul_i32: tcg_out_opc_mul_w(s, a0, a1, a2); break; @@ -2049,7 +2059,7 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op) case INDEX_op_qemu_ld_a32_i128: case INDEX_op_qemu_ld_a64_i128: - return C_O2_I1(r, r, r); + return C_N2_I1(r, r, r); case INDEX_op_qemu_st_a32_i128: case INDEX_op_qemu_st_a64_i128: @@ -2073,6 +2083,8 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op) case INDEX_op_extrl_i64_i32: case INDEX_op_extrh_i64_i32: case INDEX_op_ext_i32_i64: + case INDEX_op_neg_i32: + case INDEX_op_neg_i64: case INDEX_op_not_i32: case INDEX_op_not_i64: case INDEX_op_extract_i32: @@ -2309,10 +2321,6 @@ static void tcg_target_init(TCGContext *s) exit(EXIT_FAILURE); } - if (hwcap & HWCAP_LOONGARCH_LSX) { - use_lsx_instructions = 1; - } - tcg_target_available_regs[TCG_TYPE_I32] = ALL_GENERAL_REGS; tcg_target_available_regs[TCG_TYPE_I64] = ALL_GENERAL_REGS; @@ -2328,7 +2336,7 @@ static void tcg_target_init(TCGContext *s) tcg_regset_reset_reg(tcg_target_call_clobber_regs, TCG_REG_S8); tcg_regset_reset_reg(tcg_target_call_clobber_regs, TCG_REG_S9); - if (use_lsx_instructions) { + if (cpuinfo & CPUINFO_LSX) { tcg_target_available_regs[TCG_TYPE_V128] = ALL_VECTOR_REGS; tcg_regset_reset_reg(tcg_target_call_clobber_regs, TCG_REG_V24); tcg_regset_reset_reg(tcg_target_call_clobber_regs, TCG_REG_V25); diff --git a/tcg/loongarch64/tcg-target.h b/tcg/loongarch64/tcg-target.h index 03017672f6..9c70ebfefc 100644 --- a/tcg/loongarch64/tcg-target.h +++ b/tcg/loongarch64/tcg-target.h @@ -29,6 +29,8 @@ #ifndef LOONGARCH_TCG_TARGET_H #define LOONGARCH_TCG_TARGET_H +#include "host/cpuinfo.h" + #define TCG_TARGET_INSN_UNIT_SIZE 4 #define TCG_TARGET_NB_REGS 64 @@ -85,8 +87,6 @@ typedef enum { TCG_VEC_TMP0 = TCG_REG_V23, } TCGReg; -extern bool use_lsx_instructions; - /* used for function call generation */ #define TCG_REG_CALL_STACK TCG_REG_SP #define TCG_TARGET_STACK_ALIGN 16 @@ -97,7 +97,6 @@ extern bool use_lsx_instructions; #define TCG_TARGET_CALL_RET_I128 TCG_CALL_RET_NORMAL /* optional instructions */ -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 0 #define TCG_TARGET_HAS_div_i32 1 #define TCG_TARGET_HAS_rem_i32 1 @@ -120,7 +119,6 @@ extern bool use_lsx_instructions; #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 #define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_neg_i32 0 #define TCG_TARGET_HAS_andc_i32 1 #define TCG_TARGET_HAS_orc_i32 1 #define TCG_TARGET_HAS_eqv_i32 0 @@ -134,7 +132,6 @@ extern bool use_lsx_instructions; #define TCG_TARGET_HAS_qemu_st8_i32 0 /* 64-bit operations */ -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 0 #define TCG_TARGET_HAS_div_i64 1 #define TCG_TARGET_HAS_rem_i64 1 @@ -155,7 +152,6 @@ extern bool use_lsx_instructions; #define TCG_TARGET_HAS_bswap32_i64 1 #define TCG_TARGET_HAS_bswap64_i64 1 #define TCG_TARGET_HAS_not_i64 1 -#define TCG_TARGET_HAS_neg_i64 0 #define TCG_TARGET_HAS_andc_i64 1 #define TCG_TARGET_HAS_orc_i64 1 #define TCG_TARGET_HAS_eqv_i64 0 @@ -171,10 +167,10 @@ extern bool use_lsx_instructions; #define TCG_TARGET_HAS_muluh_i64 1 #define TCG_TARGET_HAS_mulsh_i64 1 -#define TCG_TARGET_HAS_qemu_ldst_i128 use_lsx_instructions +#define TCG_TARGET_HAS_qemu_ldst_i128 (cpuinfo & CPUINFO_LSX) #define TCG_TARGET_HAS_v64 0 -#define TCG_TARGET_HAS_v128 use_lsx_instructions +#define TCG_TARGET_HAS_v128 (cpuinfo & CPUINFO_LSX) #define TCG_TARGET_HAS_v256 0 #define TCG_TARGET_HAS_not_vec 1 diff --git a/tcg/mips/tcg-target.c.inc b/tcg/mips/tcg-target.c.inc index 328984ccff..8328dbdecc 100644 --- a/tcg/mips/tcg-target.c.inc +++ b/tcg/mips/tcg-target.c.inc @@ -871,71 +871,83 @@ static void tcg_out_addsub2(TCGContext *s, TCGReg rl, TCGReg rh, TCGReg al, } } -/* Bit 0 set if inversion required; bit 1 set if swapping required. */ -#define MIPS_CMP_INV 1 -#define MIPS_CMP_SWAP 2 - -static const uint8_t mips_cmp_map[16] = { - [TCG_COND_LT] = 0, - [TCG_COND_LTU] = 0, - [TCG_COND_GE] = MIPS_CMP_INV, - [TCG_COND_GEU] = MIPS_CMP_INV, - [TCG_COND_LE] = MIPS_CMP_INV | MIPS_CMP_SWAP, - [TCG_COND_LEU] = MIPS_CMP_INV | MIPS_CMP_SWAP, - [TCG_COND_GT] = MIPS_CMP_SWAP, - [TCG_COND_GTU] = MIPS_CMP_SWAP, -}; +#define SETCOND_INV TCG_TARGET_NB_REGS +#define SETCOND_NEZ (SETCOND_INV << 1) +#define SETCOND_FLAGS (SETCOND_INV | SETCOND_NEZ) -static void tcg_out_setcond(TCGContext *s, TCGCond cond, TCGReg ret, - TCGReg arg1, TCGReg arg2) +static int tcg_out_setcond_int(TCGContext *s, TCGCond cond, TCGReg ret, + TCGReg arg1, TCGReg arg2) { - MIPSInsn s_opc = OPC_SLTU; - int cmp_map; + int flags = 0; switch (cond) { - case TCG_COND_EQ: - if (arg2 != 0) { - tcg_out_opc_reg(s, OPC_XOR, ret, arg1, arg2); - arg1 = ret; - } - tcg_out_opc_imm(s, OPC_SLTIU, ret, arg1, 1); + case TCG_COND_EQ: /* -> NE */ + case TCG_COND_GE: /* -> LT */ + case TCG_COND_GEU: /* -> LTU */ + case TCG_COND_LE: /* -> GT */ + case TCG_COND_LEU: /* -> GTU */ + cond = tcg_invert_cond(cond); + flags ^= SETCOND_INV; break; + default: + break; + } + switch (cond) { case TCG_COND_NE: - if (arg2 != 0) { - tcg_out_opc_reg(s, OPC_XOR, ret, arg1, arg2); - arg1 = ret; + flags |= SETCOND_NEZ; + if (arg2 == 0) { + return arg1 | flags; } - tcg_out_opc_reg(s, OPC_SLTU, ret, TCG_REG_ZERO, arg1); + tcg_out_opc_reg(s, OPC_XOR, ret, arg1, arg2); break; - case TCG_COND_LT: - case TCG_COND_GE: - case TCG_COND_LE: - case TCG_COND_GT: - s_opc = OPC_SLT; - /* FALLTHRU */ - + tcg_out_opc_reg(s, OPC_SLT, ret, arg1, arg2); + break; case TCG_COND_LTU: - case TCG_COND_GEU: - case TCG_COND_LEU: + tcg_out_opc_reg(s, OPC_SLTU, ret, arg1, arg2); + break; + case TCG_COND_GT: + tcg_out_opc_reg(s, OPC_SLT, ret, arg2, arg1); + break; case TCG_COND_GTU: - cmp_map = mips_cmp_map[cond]; - if (cmp_map & MIPS_CMP_SWAP) { - TCGReg t = arg1; - arg1 = arg2; - arg2 = t; - } - tcg_out_opc_reg(s, s_opc, ret, arg1, arg2); - if (cmp_map & MIPS_CMP_INV) { - tcg_out_opc_imm(s, OPC_XORI, ret, ret, 1); - } + tcg_out_opc_reg(s, OPC_SLTU, ret, arg2, arg1); break; + default: + g_assert_not_reached(); + } + return ret | flags; +} + +static void tcg_out_setcond_end(TCGContext *s, TCGReg ret, int tmpflags) +{ + if (tmpflags != ret) { + TCGReg tmp = tmpflags & ~SETCOND_FLAGS; - default: - g_assert_not_reached(); - break; - } + switch (tmpflags & SETCOND_FLAGS) { + case SETCOND_INV: + /* Intermediate result is boolean: simply invert. */ + tcg_out_opc_imm(s, OPC_XORI, ret, tmp, 1); + break; + case SETCOND_NEZ: + /* Intermediate result is zero/non-zero: test != 0. */ + tcg_out_opc_reg(s, OPC_SLTU, ret, TCG_REG_ZERO, tmp); + break; + case SETCOND_NEZ | SETCOND_INV: + /* Intermediate result is zero/non-zero: test == 0. */ + tcg_out_opc_imm(s, OPC_SLTIU, ret, tmp, 1); + break; + default: + g_assert_not_reached(); + } + } +} + +static void tcg_out_setcond(TCGContext *s, TCGCond cond, TCGReg ret, + TCGReg arg1, TCGReg arg2) +{ + int tmpflags = tcg_out_setcond_int(s, cond, ret, arg1, arg2); + tcg_out_setcond_end(s, ret, tmpflags); } static void tcg_out_brcond(TCGContext *s, TCGCond cond, TCGReg arg1, @@ -948,9 +960,7 @@ static void tcg_out_brcond(TCGContext *s, TCGCond cond, TCGReg arg1, [TCG_COND_GE] = OPC_BGEZ, }; - MIPSInsn s_opc = OPC_SLTU; - MIPSInsn b_opc; - int cmp_map; + MIPSInsn b_opc = 0; switch (cond) { case TCG_COND_EQ: @@ -959,7 +969,6 @@ static void tcg_out_brcond(TCGContext *s, TCGCond cond, TCGReg arg1, case TCG_COND_NE: b_opc = OPC_BNE; break; - case TCG_COND_LT: case TCG_COND_GT: case TCG_COND_LE: @@ -968,133 +977,76 @@ static void tcg_out_brcond(TCGContext *s, TCGCond cond, TCGReg arg1, b_opc = b_zero[cond]; arg2 = arg1; arg1 = 0; - break; } - s_opc = OPC_SLT; - /* FALLTHRU */ - - case TCG_COND_LTU: - case TCG_COND_GTU: - case TCG_COND_LEU: - case TCG_COND_GEU: - cmp_map = mips_cmp_map[cond]; - if (cmp_map & MIPS_CMP_SWAP) { - TCGReg t = arg1; - arg1 = arg2; - arg2 = t; - } - tcg_out_opc_reg(s, s_opc, TCG_TMP0, arg1, arg2); - b_opc = (cmp_map & MIPS_CMP_INV ? OPC_BEQ : OPC_BNE); - arg1 = TCG_TMP0; - arg2 = TCG_REG_ZERO; break; - default: - g_assert_not_reached(); break; } - tcg_out_opc_br(s, b_opc, arg1, arg2); - tcg_out_reloc(s, s->code_ptr - 1, R_MIPS_PC16, l, 0); - tcg_out_nop(s); -} + if (b_opc == 0) { + int tmpflags = tcg_out_setcond_int(s, cond, TCG_TMP0, arg1, arg2); -static TCGReg tcg_out_reduce_eq2(TCGContext *s, TCGReg tmp0, TCGReg tmp1, - TCGReg al, TCGReg ah, - TCGReg bl, TCGReg bh) -{ - /* Merge highpart comparison into AH. */ - if (bh != 0) { - if (ah != 0) { - tcg_out_opc_reg(s, OPC_XOR, tmp0, ah, bh); - ah = tmp0; - } else { - ah = bh; - } - } - /* Merge lowpart comparison into AL. */ - if (bl != 0) { - if (al != 0) { - tcg_out_opc_reg(s, OPC_XOR, tmp1, al, bl); - al = tmp1; - } else { - al = bl; - } - } - /* Merge high and low part comparisons into AL. */ - if (ah != 0) { - if (al != 0) { - tcg_out_opc_reg(s, OPC_OR, tmp0, ah, al); - al = tmp0; - } else { - al = ah; - } + arg2 = TCG_REG_ZERO; + arg1 = tmpflags & ~SETCOND_FLAGS; + b_opc = tmpflags & SETCOND_INV ? OPC_BEQ : OPC_BNE; } - return al; + + tcg_out_reloc(s, s->code_ptr, R_MIPS_PC16, l, 0); + tcg_out_opc_br(s, b_opc, arg1, arg2); + tcg_out_nop(s); } -static void tcg_out_setcond2(TCGContext *s, TCGCond cond, TCGReg ret, - TCGReg al, TCGReg ah, TCGReg bl, TCGReg bh) +static int tcg_out_setcond2_int(TCGContext *s, TCGCond cond, TCGReg ret, + TCGReg al, TCGReg ah, TCGReg bl, TCGReg bh) { - TCGReg tmp0 = TCG_TMP0; - TCGReg tmp1 = ret; - - tcg_debug_assert(ret != TCG_TMP0); - if (ret == ah || ret == bh) { - tcg_debug_assert(ret != TCG_TMP1); - tmp1 = TCG_TMP1; - } + int flags = 0; switch (cond) { case TCG_COND_EQ: + flags |= SETCOND_INV; + /* fall through */ case TCG_COND_NE: - tmp1 = tcg_out_reduce_eq2(s, tmp0, tmp1, al, ah, bl, bh); - tcg_out_setcond(s, cond, ret, tmp1, TCG_REG_ZERO); + flags |= SETCOND_NEZ; + tcg_out_opc_reg(s, OPC_XOR, TCG_TMP0, al, bl); + tcg_out_opc_reg(s, OPC_XOR, TCG_TMP1, ah, bh); + tcg_out_opc_reg(s, OPC_OR, ret, TCG_TMP0, TCG_TMP1); break; default: - tcg_out_setcond(s, TCG_COND_EQ, tmp0, ah, bh); - tcg_out_setcond(s, tcg_unsigned_cond(cond), tmp1, al, bl); - tcg_out_opc_reg(s, OPC_AND, tmp1, tmp1, tmp0); - tcg_out_setcond(s, tcg_high_cond(cond), tmp0, ah, bh); - tcg_out_opc_reg(s, OPC_OR, ret, tmp1, tmp0); + tcg_out_setcond(s, TCG_COND_EQ, TCG_TMP0, ah, bh); + tcg_out_setcond(s, tcg_unsigned_cond(cond), TCG_TMP1, al, bl); + tcg_out_opc_reg(s, OPC_AND, TCG_TMP1, TCG_TMP1, TCG_TMP0); + tcg_out_setcond(s, tcg_high_cond(cond), TCG_TMP0, ah, bh); + tcg_out_opc_reg(s, OPC_OR, ret, TCG_TMP0, TCG_TMP1); break; } + return ret | flags; +} + +static void tcg_out_setcond2(TCGContext *s, TCGCond cond, TCGReg ret, + TCGReg al, TCGReg ah, TCGReg bl, TCGReg bh) +{ + int tmpflags = tcg_out_setcond2_int(s, cond, ret, al, ah, bl, bh); + tcg_out_setcond_end(s, ret, tmpflags); } static void tcg_out_brcond2(TCGContext *s, TCGCond cond, TCGReg al, TCGReg ah, TCGReg bl, TCGReg bh, TCGLabel *l) { - TCGCond b_cond = TCG_COND_NE; - TCGReg tmp = TCG_TMP1; - - /* With branches, we emit between 4 and 9 insns with 2 or 3 branches. - With setcond, we emit between 3 and 10 insns and only 1 branch, - which ought to get better branch prediction. */ - switch (cond) { - case TCG_COND_EQ: - case TCG_COND_NE: - b_cond = cond; - tmp = tcg_out_reduce_eq2(s, TCG_TMP0, TCG_TMP1, al, ah, bl, bh); - break; - - default: - /* Minimize code size by preferring a compare not requiring INV. */ - if (mips_cmp_map[cond] & MIPS_CMP_INV) { - cond = tcg_invert_cond(cond); - b_cond = TCG_COND_EQ; - } - tcg_out_setcond2(s, cond, tmp, al, ah, bl, bh); - break; - } + int tmpflags = tcg_out_setcond2_int(s, cond, TCG_TMP0, al, ah, bl, bh); + TCGReg tmp = tmpflags & ~SETCOND_FLAGS; + MIPSInsn b_opc = tmpflags & SETCOND_INV ? OPC_BEQ : OPC_BNE; - tcg_out_brcond(s, b_cond, tmp, TCG_REG_ZERO, l); + tcg_out_reloc(s, s->code_ptr, R_MIPS_PC16, l, 0); + tcg_out_opc_br(s, b_opc, tmp, TCG_REG_ZERO); + tcg_out_nop(s); } static void tcg_out_movcond(TCGContext *s, TCGCond cond, TCGReg ret, TCGReg c1, TCGReg c2, TCGReg v1, TCGReg v2) { - bool eqz = false; + int tmpflags; + bool eqz; /* If one of the values is zero, put it last to match SEL*Z instructions */ if (use_mips32r6_instructions && v1 == 0) { @@ -1103,27 +1055,9 @@ static void tcg_out_movcond(TCGContext *s, TCGCond cond, TCGReg ret, cond = tcg_invert_cond(cond); } - switch (cond) { - case TCG_COND_EQ: - eqz = true; - /* FALLTHRU */ - case TCG_COND_NE: - if (c2 != 0) { - tcg_out_opc_reg(s, OPC_XOR, TCG_TMP0, c1, c2); - c1 = TCG_TMP0; - } - break; - - default: - /* Minimize code size by preferring a compare not requiring INV. */ - if (mips_cmp_map[cond] & MIPS_CMP_INV) { - cond = tcg_invert_cond(cond); - eqz = true; - } - tcg_out_setcond(s, cond, TCG_TMP0, c1, c2); - c1 = TCG_TMP0; - break; - } + tmpflags = tcg_out_setcond_int(s, cond, TCG_TMP0, c1, c2); + c1 = tmpflags & ~SETCOND_FLAGS; + eqz = tmpflags & SETCOND_INV; if (use_mips32r6_instructions) { MIPSInsn m_opc_t = eqz ? OPC_SELEQZ : OPC_SELNEZ; @@ -1136,13 +1070,22 @@ static void tcg_out_movcond(TCGContext *s, TCGCond cond, TCGReg ret, if (v2 != 0) { tcg_out_opc_reg(s, OPC_OR, ret, ret, TCG_TMP1); } - } else { - MIPSInsn m_opc = eqz ? OPC_MOVZ : OPC_MOVN; + return; + } - tcg_out_opc_reg(s, m_opc, ret, v1, c1); + /* This should be guaranteed via constraints */ + tcg_debug_assert(v2 == ret); - /* This should be guaranteed via constraints */ - tcg_debug_assert(v2 == ret); + if (use_movnz_instructions) { + MIPSInsn m_opc = eqz ? OPC_MOVZ : OPC_MOVN; + tcg_out_opc_reg(s, m_opc, ret, v1, c1); + } else { + /* Invert the condition in order to branch over the move. */ + MIPSInsn b_opc = eqz ? OPC_BNE : OPC_BEQ; + tcg_out_opc_imm(s, b_opc, c1, TCG_REG_ZERO, 2); + tcg_out_nop(s); + /* Open-code tcg_out_mov, without the nop-move check. */ + tcg_out_opc_reg(s, OPC_OR, ret, v1, TCG_REG_ZERO); } } @@ -1977,6 +1920,12 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_opc_reg(s, OPC_MFHI, a1, 0, 0); break; + case INDEX_op_neg_i32: + i1 = OPC_SUBU; + goto do_unary; + case INDEX_op_neg_i64: + i1 = OPC_DSUBU; + goto do_unary; case INDEX_op_not_i32: case INDEX_op_not_i64: i1 = OPC_NOR; @@ -2201,6 +2150,7 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op) case INDEX_op_ld16u_i32: case INDEX_op_ld16s_i32: case INDEX_op_ld_i32: + case INDEX_op_neg_i32: case INDEX_op_not_i32: case INDEX_op_bswap16_i32: case INDEX_op_bswap32_i32: @@ -2214,6 +2164,7 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op) case INDEX_op_ld32s_i64: case INDEX_op_ld32u_i64: case INDEX_op_ld_i64: + case INDEX_op_neg_i64: case INDEX_op_not_i64: case INDEX_op_bswap16_i64: case INDEX_op_bswap32_i64: diff --git a/tcg/mips/tcg-target.h b/tcg/mips/tcg-target.h index c0576f66d7..b98ffae1d0 100644 --- a/tcg/mips/tcg-target.h +++ b/tcg/mips/tcg-target.h @@ -154,7 +154,6 @@ extern bool use_mips32r2_instructions; #endif /* optional instructions detected at runtime */ -#define TCG_TARGET_HAS_movcond_i32 use_movnz_instructions #define TCG_TARGET_HAS_bswap16_i32 use_mips32r2_instructions #define TCG_TARGET_HAS_deposit_i32 use_mips32r2_instructions #define TCG_TARGET_HAS_extract_i32 use_mips32r2_instructions @@ -169,7 +168,6 @@ extern bool use_mips32r2_instructions; #define TCG_TARGET_HAS_qemu_st8_i32 0 #if TCG_TARGET_REG_BITS == 64 -#define TCG_TARGET_HAS_movcond_i64 use_movnz_instructions #define TCG_TARGET_HAS_bswap16_i64 use_mips32r2_instructions #define TCG_TARGET_HAS_bswap32_i64 use_mips32r2_instructions #define TCG_TARGET_HAS_bswap64_i64 use_mips32r2_instructions @@ -186,12 +184,10 @@ extern bool use_mips32r2_instructions; #endif /* optional instructions automatically implemented */ -#define TCG_TARGET_HAS_neg_i32 0 /* sub rd, zero, rt */ #define TCG_TARGET_HAS_ext8u_i32 0 /* andi rt, rs, 0xff */ #define TCG_TARGET_HAS_ext16u_i32 0 /* andi rt, rs, 0xffff */ #if TCG_TARGET_REG_BITS == 64 -#define TCG_TARGET_HAS_neg_i64 0 /* sub rd, zero, rt */ #define TCG_TARGET_HAS_ext8u_i64 0 /* andi rt, rs, 0xff */ #define TCG_TARGET_HAS_ext16u_i64 0 /* andi rt, rs, 0xffff */ #endif diff --git a/tcg/optimize.c b/tcg/optimize.c index 2db5177c32..f2d01654c5 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -25,6 +25,7 @@ #include "qemu/osdep.h" #include "qemu/int128.h" +#include "qemu/interval-tree.h" #include "tcg/tcg-op-common.h" #include "tcg-internal.h" @@ -37,10 +38,18 @@ glue(glue(case INDEX_op_, x), _i64): \ glue(glue(case INDEX_op_, x), _vec) +typedef struct MemCopyInfo { + IntervalTreeNode itree; + QSIMPLEQ_ENTRY (MemCopyInfo) next; + TCGTemp *ts; + TCGType type; +} MemCopyInfo; + typedef struct TempOptInfo { bool is_const; TCGTemp *prev_copy; TCGTemp *next_copy; + QSIMPLEQ_HEAD(, MemCopyInfo) mem_copy; uint64_t val; uint64_t z_mask; /* mask bit is 0 if and only if value bit is 0 */ uint64_t s_mask; /* a left-aligned mask of clrsb(value) bits. */ @@ -51,6 +60,9 @@ typedef struct OptContext { TCGOp *prev_mb; TCGTempSet temps_used; + IntervalTreeRoot mem_copy; + QSIMPLEQ_HEAD(, MemCopyInfo) mem_free; + /* In flight values from optimization. */ uint64_t a_mask; /* mask bit is 0 iff value identical to first input */ uint64_t z_mask; /* mask bit is 0 iff value bit is 0 */ @@ -122,25 +134,9 @@ static inline bool ts_is_copy(TCGTemp *ts) return ts_info(ts)->next_copy != ts; } -/* Reset TEMP's state, possibly removing the temp for the list of copies. */ -static void reset_ts(TCGTemp *ts) -{ - TempOptInfo *ti = ts_info(ts); - TempOptInfo *pi = ts_info(ti->prev_copy); - TempOptInfo *ni = ts_info(ti->next_copy); - - ni->prev_copy = ti->prev_copy; - pi->next_copy = ti->next_copy; - ti->next_copy = ts; - ti->prev_copy = ts; - ti->is_const = false; - ti->z_mask = -1; - ti->s_mask = 0; -} - -static void reset_temp(TCGArg arg) +static TCGTemp *cmp_better_copy(TCGTemp *a, TCGTemp *b) { - reset_ts(arg_temp(arg)); + return a->kind < b->kind ? b : a; } /* Initialize and activate a temporary. */ @@ -162,6 +158,7 @@ static void init_ts_info(OptContext *ctx, TCGTemp *ts) ti->next_copy = ts; ti->prev_copy = ts; + QSIMPLEQ_INIT(&ti->mem_copy); if (ts->kind == TEMP_CONST) { ti->is_const = true; ti->val = ts->val; @@ -174,30 +171,133 @@ static void init_ts_info(OptContext *ctx, TCGTemp *ts) } } -static TCGTemp *find_better_copy(TCGContext *s, TCGTemp *ts) +static MemCopyInfo *mem_copy_first(OptContext *ctx, intptr_t s, intptr_t l) +{ + IntervalTreeNode *r = interval_tree_iter_first(&ctx->mem_copy, s, l); + return r ? container_of(r, MemCopyInfo, itree) : NULL; +} + +static MemCopyInfo *mem_copy_next(MemCopyInfo *mem, intptr_t s, intptr_t l) +{ + IntervalTreeNode *r = interval_tree_iter_next(&mem->itree, s, l); + return r ? container_of(r, MemCopyInfo, itree) : NULL; +} + +static void remove_mem_copy(OptContext *ctx, MemCopyInfo *mc) +{ + TCGTemp *ts = mc->ts; + TempOptInfo *ti = ts_info(ts); + + interval_tree_remove(&mc->itree, &ctx->mem_copy); + QSIMPLEQ_REMOVE(&ti->mem_copy, mc, MemCopyInfo, next); + QSIMPLEQ_INSERT_TAIL(&ctx->mem_free, mc, next); +} + +static void remove_mem_copy_in(OptContext *ctx, intptr_t s, intptr_t l) +{ + while (true) { + MemCopyInfo *mc = mem_copy_first(ctx, s, l); + if (!mc) { + break; + } + remove_mem_copy(ctx, mc); + } +} + +static void remove_mem_copy_all(OptContext *ctx) +{ + remove_mem_copy_in(ctx, 0, -1); + tcg_debug_assert(interval_tree_is_empty(&ctx->mem_copy)); +} + +static TCGTemp *find_better_copy(TCGTemp *ts) { - TCGTemp *i, *g, *l; + TCGTemp *i, *ret; /* If this is already readonly, we can't do better. */ if (temp_readonly(ts)) { return ts; } - g = l = NULL; + ret = ts; for (i = ts_info(ts)->next_copy; i != ts; i = ts_info(i)->next_copy) { - if (temp_readonly(i)) { - return i; - } else if (i->kind > ts->kind) { - if (i->kind == TEMP_GLOBAL) { - g = i; - } else if (i->kind == TEMP_TB) { - l = i; + ret = cmp_better_copy(ret, i); + } + return ret; +} + +static void move_mem_copies(TCGTemp *dst_ts, TCGTemp *src_ts) +{ + TempOptInfo *si = ts_info(src_ts); + TempOptInfo *di = ts_info(dst_ts); + MemCopyInfo *mc; + + QSIMPLEQ_FOREACH(mc, &si->mem_copy, next) { + tcg_debug_assert(mc->ts == src_ts); + mc->ts = dst_ts; + } + QSIMPLEQ_CONCAT(&di->mem_copy, &si->mem_copy); +} + +/* Reset TEMP's state, possibly removing the temp for the list of copies. */ +static void reset_ts(OptContext *ctx, TCGTemp *ts) +{ + TempOptInfo *ti = ts_info(ts); + TCGTemp *pts = ti->prev_copy; + TCGTemp *nts = ti->next_copy; + TempOptInfo *pi = ts_info(pts); + TempOptInfo *ni = ts_info(nts); + + ni->prev_copy = ti->prev_copy; + pi->next_copy = ti->next_copy; + ti->next_copy = ts; + ti->prev_copy = ts; + ti->is_const = false; + ti->z_mask = -1; + ti->s_mask = 0; + + if (!QSIMPLEQ_EMPTY(&ti->mem_copy)) { + if (ts == nts) { + /* Last temp copy being removed, the mem copies die. */ + MemCopyInfo *mc; + QSIMPLEQ_FOREACH(mc, &ti->mem_copy, next) { + interval_tree_remove(&mc->itree, &ctx->mem_copy); } + QSIMPLEQ_CONCAT(&ctx->mem_free, &ti->mem_copy); + } else { + move_mem_copies(find_better_copy(nts), ts); } } +} - /* If we didn't find a better representation, return the same temp. */ - return g ? g : l ? l : ts; +static void reset_temp(OptContext *ctx, TCGArg arg) +{ + reset_ts(ctx, arg_temp(arg)); +} + +static void record_mem_copy(OptContext *ctx, TCGType type, + TCGTemp *ts, intptr_t start, intptr_t last) +{ + MemCopyInfo *mc; + TempOptInfo *ti; + + mc = QSIMPLEQ_FIRST(&ctx->mem_free); + if (mc) { + QSIMPLEQ_REMOVE_HEAD(&ctx->mem_free, next); + } else { + mc = tcg_malloc(sizeof(*mc)); + } + + memset(mc, 0, sizeof(*mc)); + mc->itree.start = start; + mc->itree.last = last; + mc->type = type; + interval_tree_insert(&mc->itree, &ctx->mem_copy); + + ts = find_better_copy(ts); + ti = ts_info(ts); + mc->ts = ts; + QSIMPLEQ_INSERT_TAIL(&ti->mem_copy, mc, next); } static bool ts_are_copies(TCGTemp *ts1, TCGTemp *ts2) @@ -226,6 +326,33 @@ static bool args_are_copies(TCGArg arg1, TCGArg arg2) return ts_are_copies(arg_temp(arg1), arg_temp(arg2)); } +static TCGTemp *find_mem_copy_for(OptContext *ctx, TCGType type, intptr_t s) +{ + MemCopyInfo *mc; + + for (mc = mem_copy_first(ctx, s, s); mc; mc = mem_copy_next(mc, s, s)) { + if (mc->itree.start == s && mc->type == type) { + return find_better_copy(mc->ts); + } + } + return NULL; +} + +static TCGArg arg_new_constant(OptContext *ctx, uint64_t val) +{ + TCGType type = ctx->type; + TCGTemp *ts; + + if (type == TCG_TYPE_I32) { + val = (int32_t)val; + } + + ts = tcg_constant_internal(type, val); + init_ts_info(ctx, ts); + + return temp_arg(ts); +} + static bool tcg_opt_gen_mov(OptContext *ctx, TCGOp *op, TCGArg dst, TCGArg src) { TCGTemp *dst_ts = arg_temp(dst); @@ -239,7 +366,7 @@ static bool tcg_opt_gen_mov(OptContext *ctx, TCGOp *op, TCGArg dst, TCGArg src) return true; } - reset_ts(dst_ts); + reset_ts(ctx, dst_ts); di = ts_info(dst_ts); si = ts_info(src_ts); @@ -275,6 +402,11 @@ static bool tcg_opt_gen_mov(OptContext *ctx, TCGOp *op, TCGArg dst, TCGArg src) si->next_copy = dst_ts; di->is_const = si->is_const; di->val = si->val; + + if (!QSIMPLEQ_EMPTY(&si->mem_copy) + && cmp_better_copy(src_ts, dst_ts) == dst_ts) { + move_mem_copies(dst_ts, src_ts); + } } return true; } @@ -282,16 +414,8 @@ static bool tcg_opt_gen_mov(OptContext *ctx, TCGOp *op, TCGArg dst, TCGArg src) static bool tcg_opt_gen_movi(OptContext *ctx, TCGOp *op, TCGArg dst, uint64_t val) { - TCGTemp *tv; - - if (ctx->type == TCG_TYPE_I32) { - val = (int32_t)val; - } - /* Convert movi to mov with constant temp. */ - tv = tcg_constant_internal(ctx->type, val); - init_ts_info(ctx, tv); - return tcg_opt_gen_mov(ctx, op, dst, temp_arg(tv)); + return tcg_opt_gen_mov(ctx, op, dst, arg_new_constant(ctx, val)); } static uint64_t do_constant_folding_2(TCGOpcode op, uint64_t x, uint64_t y) @@ -672,12 +796,10 @@ static void init_arguments(OptContext *ctx, TCGOp *op, int nb_args) static void copy_propagate(OptContext *ctx, TCGOp *op, int nb_oargs, int nb_iargs) { - TCGContext *s = ctx->tcg; - for (int i = nb_oargs; i < nb_oargs + nb_iargs; i++) { TCGTemp *ts = arg_temp(op->args[i]); if (ts_is_copy(ts)) { - op->args[i] = temp_arg(find_better_copy(s, ts)); + op->args[i] = temp_arg(find_better_copy(ts)); } } } @@ -695,6 +817,7 @@ static void finish_folding(OptContext *ctx, TCGOp *op) ctx->prev_mb = NULL; if (!(def->flags & TCG_OPF_COND_BRANCH)) { memset(&ctx->temps_used, 0, sizeof(ctx->temps_used)); + remove_mem_copy_all(ctx); } return; } @@ -702,7 +825,7 @@ static void finish_folding(OptContext *ctx, TCGOp *op) nb_oargs = def->nb_oargs; for (i = 0; i < nb_oargs; i++) { TCGTemp *ts = arg_temp(op->args[i]); - reset_ts(ts); + reset_ts(ctx, ts); /* * Save the corresponding known-zero/sign bits mask for the * first output argument (only one supported so far). @@ -921,8 +1044,10 @@ static bool fold_add_vec(OptContext *ctx, TCGOp *op) static bool fold_addsub2(OptContext *ctx, TCGOp *op, bool add) { - if (arg_is_const(op->args[2]) && arg_is_const(op->args[3]) && - arg_is_const(op->args[4]) && arg_is_const(op->args[5])) { + bool a_const = arg_is_const(op->args[2]) && arg_is_const(op->args[3]); + bool b_const = arg_is_const(op->args[4]) && arg_is_const(op->args[5]); + + if (a_const && b_const) { uint64_t al = arg_info(op->args[2])->val; uint64_t ah = arg_info(op->args[3])->val; uint64_t bl = arg_info(op->args[4])->val; @@ -966,6 +1091,21 @@ static bool fold_addsub2(OptContext *ctx, TCGOp *op, bool add) tcg_opt_gen_movi(ctx, op2, rh, ah); return true; } + + /* Fold sub2 r,x,i to add2 r,x,-i */ + if (!add && b_const) { + uint64_t bl = arg_info(op->args[4])->val; + uint64_t bh = arg_info(op->args[5])->val; + + /* Negate the two parts without assembling and disassembling. */ + bl = -bl; + bh = ~bh + !bl; + + op->opc = (ctx->type == TCG_TYPE_I32 + ? INDEX_op_add2_i32 : INDEX_op_add2_i64); + op->args[4] = arg_new_constant(ctx, bl); + op->args[5] = arg_new_constant(ctx, bh); + } return false; } @@ -1215,14 +1355,19 @@ static bool fold_call(OptContext *ctx, TCGOp *op) for (i = 0; i < nb_globals; i++) { if (test_bit(i, ctx->temps_used.l)) { - reset_ts(&ctx->tcg->temps[i]); + reset_ts(ctx, &ctx->tcg->temps[i]); } } } + /* If the function has side effects, reset mem data. */ + if (!(flags & TCG_CALL_NO_SIDE_EFFECTS)) { + remove_mem_copy_all(ctx); + } + /* Reset temp data for outputs. */ for (i = 0; i < nb_oargs; i++) { - reset_temp(op->args[i]); + reset_temp(ctx, op->args[i]); } /* Stop optimizing MB across calls. */ @@ -1310,7 +1455,7 @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) op->opc = and_opc; op->args[1] = op->args[2]; - op->args[2] = temp_arg(tcg_constant_internal(ctx->type, mask)); + op->args[2] = arg_new_constant(ctx, mask); ctx->z_mask = mask & arg_info(op->args[1])->z_mask; return false; } @@ -1321,7 +1466,7 @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) uint64_t mask = deposit64(-1, op->args[3], op->args[4], 0); op->opc = and_opc; - op->args[2] = temp_arg(tcg_constant_internal(ctx->type, mask)); + op->args[2] = arg_new_constant(ctx, mask); ctx->z_mask = mask & arg_info(op->args[1])->z_mask; return false; } @@ -2001,11 +2146,11 @@ static bool fold_sub_to_neg(OptContext *ctx, TCGOp *op) switch (ctx->type) { case TCG_TYPE_I32: neg_op = INDEX_op_neg_i32; - have_neg = TCG_TARGET_HAS_neg_i32; + have_neg = true; break; case TCG_TYPE_I64: neg_op = INDEX_op_neg_i64; - have_neg = TCG_TARGET_HAS_neg_i64; + have_neg = true; break; case TCG_TYPE_V64: case TCG_TYPE_V128: @@ -2038,7 +2183,19 @@ static bool fold_sub_vec(OptContext *ctx, TCGOp *op) static bool fold_sub(OptContext *ctx, TCGOp *op) { - return fold_const2(ctx, op) || fold_sub_vec(ctx, op); + if (fold_const2(ctx, op) || fold_sub_vec(ctx, op)) { + return true; + } + + /* Fold sub r,x,i to add r,x,-i */ + if (arg_is_const(op->args[2])) { + uint64_t val = arg_info(op->args[2])->val; + + op->opc = (ctx->type == TCG_TYPE_I32 + ? INDEX_op_add_i32 : INDEX_op_add_i64); + op->args[2] = arg_new_constant(ctx, -val); + } + return false; } static bool fold_sub2(OptContext *ctx, TCGOp *op) @@ -2077,6 +2234,96 @@ static bool fold_tcg_ld(OptContext *ctx, TCGOp *op) return false; } +static bool fold_tcg_ld_memcopy(OptContext *ctx, TCGOp *op) +{ + TCGTemp *dst, *src; + intptr_t ofs; + TCGType type; + + if (op->args[1] != tcgv_ptr_arg(tcg_env)) { + return false; + } + + type = ctx->type; + ofs = op->args[2]; + dst = arg_temp(op->args[0]); + src = find_mem_copy_for(ctx, type, ofs); + if (src && src->base_type == type) { + return tcg_opt_gen_mov(ctx, op, temp_arg(dst), temp_arg(src)); + } + + reset_ts(ctx, dst); + record_mem_copy(ctx, type, dst, ofs, ofs + tcg_type_size(type) - 1); + return true; +} + +static bool fold_tcg_st(OptContext *ctx, TCGOp *op) +{ + intptr_t ofs = op->args[2]; + intptr_t lm1; + + if (op->args[1] != tcgv_ptr_arg(tcg_env)) { + remove_mem_copy_all(ctx); + return false; + } + + switch (op->opc) { + CASE_OP_32_64(st8): + lm1 = 0; + break; + CASE_OP_32_64(st16): + lm1 = 1; + break; + case INDEX_op_st32_i64: + case INDEX_op_st_i32: + lm1 = 3; + break; + case INDEX_op_st_i64: + lm1 = 7; + break; + case INDEX_op_st_vec: + lm1 = tcg_type_size(ctx->type) - 1; + break; + default: + g_assert_not_reached(); + } + remove_mem_copy_in(ctx, ofs, ofs + lm1); + return false; +} + +static bool fold_tcg_st_memcopy(OptContext *ctx, TCGOp *op) +{ + TCGTemp *src; + intptr_t ofs, last; + TCGType type; + + if (op->args[1] != tcgv_ptr_arg(tcg_env)) { + fold_tcg_st(ctx, op); + return false; + } + + src = arg_temp(op->args[0]); + ofs = op->args[2]; + type = ctx->type; + + /* + * Eliminate duplicate stores of a constant. + * This happens frequently when the target ISA zero-extends. + */ + if (ts_is_const(src)) { + TCGTemp *prev = find_mem_copy_for(ctx, type, ofs); + if (src == prev) { + tcg_op_remove(ctx->tcg, op); + return true; + } + } + + last = ofs + tcg_type_size(type) - 1; + remove_mem_copy_in(ctx, ofs, last); + record_mem_copy(ctx, type, src, ofs, last); + return false; +} + static bool fold_xor(OptContext *ctx, TCGOp *op) { if (fold_const2_commutative(ctx, op) || @@ -2100,6 +2347,8 @@ void tcg_optimize(TCGContext *s) TCGOp *op, *op_next; OptContext ctx = { .tcg = s }; + QSIMPLEQ_INIT(&ctx.mem_free); + /* Array VALS has an element for each temp. If this temp holds a constant then its value is kept in VALS' element. If this temp is a copy of other ones then the other copies are @@ -2221,6 +2470,21 @@ void tcg_optimize(TCGContext *s) case INDEX_op_ld32u_i64: done = fold_tcg_ld(&ctx, op); break; + case INDEX_op_ld_i32: + case INDEX_op_ld_i64: + case INDEX_op_ld_vec: + done = fold_tcg_ld_memcopy(&ctx, op); + break; + CASE_OP_32_64(st8): + CASE_OP_32_64(st16): + case INDEX_op_st32_i64: + done = fold_tcg_st(&ctx, op); + break; + case INDEX_op_st_i32: + case INDEX_op_st_i64: + case INDEX_op_st_vec: + done = fold_tcg_st_memcopy(&ctx, op); + break; case INDEX_op_mb: done = fold_mb(&ctx, op); break; diff --git a/tcg/ppc/tcg-target.h b/tcg/ppc/tcg-target.h index 8bfb14998e..5295e4f9ab 100644 --- a/tcg/ppc/tcg-target.h +++ b/tcg/ppc/tcg-target.h @@ -83,7 +83,6 @@ typedef enum { #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 #define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_andc_i32 1 #define TCG_TARGET_HAS_orc_i32 1 #define TCG_TARGET_HAS_eqv_i32 1 @@ -96,7 +95,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i32 1 #define TCG_TARGET_HAS_sextract_i32 0 #define TCG_TARGET_HAS_extract2_i32 0 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_mulu2_i32 0 #define TCG_TARGET_HAS_muls2_i32 0 @@ -121,7 +119,6 @@ typedef enum { #define TCG_TARGET_HAS_bswap32_i64 1 #define TCG_TARGET_HAS_bswap64_i64 1 #define TCG_TARGET_HAS_not_i64 1 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_andc_i64 1 #define TCG_TARGET_HAS_orc_i64 1 #define TCG_TARGET_HAS_eqv_i64 1 @@ -134,7 +131,6 @@ typedef enum { #define TCG_TARGET_HAS_extract_i64 1 #define TCG_TARGET_HAS_sextract_i64 0 #define TCG_TARGET_HAS_extract2_i64 0 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_add2_i64 1 #define TCG_TARGET_HAS_sub2_i64 1 diff --git a/tcg/riscv/tcg-target.h b/tcg/riscv/tcg-target.h index c1132d178f..a4edc3dc74 100644 --- a/tcg/riscv/tcg-target.h +++ b/tcg/riscv/tcg-target.h @@ -87,7 +87,6 @@ extern bool have_zbb; #endif /* optional instructions */ -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_div_i32 1 #define TCG_TARGET_HAS_rem_i32 1 @@ -110,7 +109,6 @@ extern bool have_zbb; #define TCG_TARGET_HAS_bswap16_i32 have_zbb #define TCG_TARGET_HAS_bswap32_i32 have_zbb #define TCG_TARGET_HAS_not_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_andc_i32 have_zbb #define TCG_TARGET_HAS_orc_i32 have_zbb #define TCG_TARGET_HAS_eqv_i32 have_zbb @@ -123,7 +121,6 @@ extern bool have_zbb; #define TCG_TARGET_HAS_setcond2 1 #define TCG_TARGET_HAS_qemu_st8_i32 0 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_div_i64 1 #define TCG_TARGET_HAS_rem_i64 1 @@ -144,7 +141,6 @@ extern bool have_zbb; #define TCG_TARGET_HAS_bswap32_i64 have_zbb #define TCG_TARGET_HAS_bswap64_i64 have_zbb #define TCG_TARGET_HAS_not_i64 1 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_andc_i64 have_zbb #define TCG_TARGET_HAS_orc_i64 have_zbb #define TCG_TARGET_HAS_eqv_i64 have_zbb diff --git a/tcg/s390x/tcg-target.h b/tcg/s390x/tcg-target.h index 50e12ef9d6..e69b0d2ddd 100644 --- a/tcg/s390x/tcg-target.h +++ b/tcg/s390x/tcg-target.h @@ -82,7 +82,6 @@ extern uint64_t s390_facilities[3]; #define TCG_TARGET_HAS_bswap16_i32 1 #define TCG_TARGET_HAS_bswap32_i32 1 #define TCG_TARGET_HAS_not_i32 HAVE_FACILITY(MISC_INSN_EXT3) -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_andc_i32 HAVE_FACILITY(MISC_INSN_EXT3) #define TCG_TARGET_HAS_orc_i32 HAVE_FACILITY(MISC_INSN_EXT3) #define TCG_TARGET_HAS_eqv_i32 HAVE_FACILITY(MISC_INSN_EXT3) @@ -95,7 +94,6 @@ extern uint64_t s390_facilities[3]; #define TCG_TARGET_HAS_extract_i32 1 #define TCG_TARGET_HAS_sextract_i32 0 #define TCG_TARGET_HAS_extract2_i32 0 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_add2_i32 1 #define TCG_TARGET_HAS_sub2_i32 1 @@ -118,7 +116,6 @@ extern uint64_t s390_facilities[3]; #define TCG_TARGET_HAS_bswap32_i64 1 #define TCG_TARGET_HAS_bswap64_i64 1 #define TCG_TARGET_HAS_not_i64 HAVE_FACILITY(MISC_INSN_EXT3) -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_andc_i64 HAVE_FACILITY(MISC_INSN_EXT3) #define TCG_TARGET_HAS_orc_i64 HAVE_FACILITY(MISC_INSN_EXT3) #define TCG_TARGET_HAS_eqv_i64 HAVE_FACILITY(MISC_INSN_EXT3) @@ -131,7 +128,6 @@ extern uint64_t s390_facilities[3]; #define TCG_TARGET_HAS_extract_i64 1 #define TCG_TARGET_HAS_sextract_i64 0 #define TCG_TARGET_HAS_extract2_i64 0 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_add2_i64 1 #define TCG_TARGET_HAS_sub2_i64 1 diff --git a/tcg/sparc64/tcg-target.c.inc b/tcg/sparc64/tcg-target.c.inc index 19d9df4a09..a91defd0ac 100644 --- a/tcg/sparc64/tcg-target.c.inc +++ b/tcg/sparc64/tcg-target.c.inc @@ -529,6 +529,11 @@ static void tcg_out_extu_i32_i64(TCGContext *s, TCGReg rd, TCGReg rs) tcg_out_ext32u(s, rd, rs); } +static void tcg_out_extrl_i64_i32(TCGContext *s, TCGReg rd, TCGReg rs) +{ + tcg_out_ext32u(s, rd, rs); +} + static bool tcg_out_xchg(TCGContext *s, TCGType type, TCGReg r1, TCGReg r2) { return false; diff --git a/tcg/sparc64/tcg-target.h b/tcg/sparc64/tcg-target.h index 5cfc4b4679..f8cf145266 100644 --- a/tcg/sparc64/tcg-target.h +++ b/tcg/sparc64/tcg-target.h @@ -91,7 +91,6 @@ extern bool use_vis3_instructions; #define TCG_TARGET_HAS_ext16u_i32 0 #define TCG_TARGET_HAS_bswap16_i32 0 #define TCG_TARGET_HAS_bswap32_i32 0 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_not_i32 1 #define TCG_TARGET_HAS_andc_i32 1 #define TCG_TARGET_HAS_orc_i32 1 @@ -105,7 +104,6 @@ extern bool use_vis3_instructions; #define TCG_TARGET_HAS_extract_i32 0 #define TCG_TARGET_HAS_sextract_i32 0 #define TCG_TARGET_HAS_extract2_i32 0 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 1 #define TCG_TARGET_HAS_add2_i32 1 #define TCG_TARGET_HAS_sub2_i32 1 @@ -128,7 +126,6 @@ extern bool use_vis3_instructions; #define TCG_TARGET_HAS_bswap16_i64 0 #define TCG_TARGET_HAS_bswap32_i64 0 #define TCG_TARGET_HAS_bswap64_i64 0 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_not_i64 1 #define TCG_TARGET_HAS_andc_i64 1 #define TCG_TARGET_HAS_orc_i64 1 @@ -142,7 +139,6 @@ extern bool use_vis3_instructions; #define TCG_TARGET_HAS_extract_i64 0 #define TCG_TARGET_HAS_sextract_i64 0 #define TCG_TARGET_HAS_extract2_i64 0 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 1 #define TCG_TARGET_HAS_add2_i64 1 #define TCG_TARGET_HAS_sub2_i64 1 diff --git a/tcg/tcg-internal.h b/tcg/tcg-internal.h index 40a69e6e6e..6c9d9e48db 100644 --- a/tcg/tcg-internal.h +++ b/tcg/tcg-internal.h @@ -83,4 +83,22 @@ static inline TCGv_i64 TCGV128_HIGH(TCGv_i128 t) bool tcg_target_has_memory_bswap(MemOp memop); +/* + * Locate or create a read-only temporary that is a constant. + * This kind of temporary need not be freed, but for convenience + * will be silently ignored by tcg_temp_free_*. + */ +TCGTemp *tcg_constant_internal(TCGType type, int64_t val); + +void tcg_gen_op1(TCGOpcode, TCGArg); +void tcg_gen_op2(TCGOpcode, TCGArg, TCGArg); +void tcg_gen_op3(TCGOpcode, TCGArg, TCGArg, TCGArg); +void tcg_gen_op4(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg); +void tcg_gen_op5(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg); +void tcg_gen_op6(TCGOpcode, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg, TCGArg); + +void vec_gen_2(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg); +void vec_gen_3(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg, TCGArg); +void vec_gen_4(TCGOpcode, TCGType, unsigned, TCGArg, TCGArg, TCGArg, TCGArg); + #endif /* TCG_INTERNAL_H */ diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c index feb2d3686b..bb88943f79 100644 --- a/tcg/tcg-op-gvec.c +++ b/tcg/tcg-op-gvec.c @@ -561,7 +561,6 @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, tcg_gen_dupi_vec(vece, t_vec, in_c); } do_dup_store(type, dofs, oprsz, maxsz, t_vec); - tcg_temp_free_vec(t_vec); return; } @@ -1024,11 +1023,10 @@ static void expand_2_vec(unsigned vece, uint32_t dofs, uint32_t aofs, bool load_dest, void (*fni)(unsigned, TCGv_vec, TCGv_vec)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); if (load_dest) { tcg_gen_ld_vec(t1, tcg_env, dofs + i); @@ -1036,8 +1034,6 @@ static void expand_2_vec(unsigned vece, uint32_t dofs, uint32_t aofs, fni(vece, t1, t0); tcg_gen_st_vec(t1, tcg_env, dofs + i); } - tcg_temp_free_vec(t0); - tcg_temp_free_vec(t1); } /* Expand OPSZ bytes worth of two-vector operands and an immediate operand @@ -1047,11 +1043,10 @@ static void expand_2i_vec(unsigned vece, uint32_t dofs, uint32_t aofs, int64_t c, bool load_dest, void (*fni)(unsigned, TCGv_vec, TCGv_vec, int64_t)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); if (load_dest) { tcg_gen_ld_vec(t1, tcg_env, dofs + i); @@ -1059,8 +1054,6 @@ static void expand_2i_vec(unsigned vece, uint32_t dofs, uint32_t aofs, fni(vece, t1, t0, c); tcg_gen_st_vec(t1, tcg_env, dofs + i); } - tcg_temp_free_vec(t0); - tcg_temp_free_vec(t1); } static void expand_2s_vec(unsigned vece, uint32_t dofs, uint32_t aofs, @@ -1068,11 +1061,10 @@ static void expand_2s_vec(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_vec c, bool scalar_first, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); if (scalar_first) { fni(vece, t1, c, t0); @@ -1081,8 +1073,6 @@ static void expand_2s_vec(unsigned vece, uint32_t dofs, uint32_t aofs, } tcg_gen_st_vec(t1, tcg_env, dofs + i); } - tcg_temp_free_vec(t0); - tcg_temp_free_vec(t1); } /* Expand OPSZ bytes worth of three-operand operations using host vectors. */ @@ -1091,12 +1081,11 @@ static void expand_3_vec(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t tysz, TCGType type, bool load_dest, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - TCGv_vec t2 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); + TCGv_vec t2 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); tcg_gen_ld_vec(t1, tcg_env, bofs + i); if (load_dest) { @@ -1105,9 +1094,6 @@ static void expand_3_vec(unsigned vece, uint32_t dofs, uint32_t aofs, fni(vece, t2, t0, t1); tcg_gen_st_vec(t2, tcg_env, dofs + i); } - tcg_temp_free_vec(t2); - tcg_temp_free_vec(t1); - tcg_temp_free_vec(t0); } /* @@ -1120,12 +1106,11 @@ static void expand_3i_vec(unsigned vece, uint32_t dofs, uint32_t aofs, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec, int64_t)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - TCGv_vec t2 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); + TCGv_vec t2 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); tcg_gen_ld_vec(t1, tcg_env, bofs + i); if (load_dest) { @@ -1134,9 +1119,6 @@ static void expand_3i_vec(unsigned vece, uint32_t dofs, uint32_t aofs, fni(vece, t2, t0, t1, c); tcg_gen_st_vec(t2, tcg_env, dofs + i); } - tcg_temp_free_vec(t0); - tcg_temp_free_vec(t1); - tcg_temp_free_vec(t2); } /* Expand OPSZ bytes worth of four-operand operations using host vectors. */ @@ -1146,13 +1128,12 @@ static void expand_4_vec(unsigned vece, uint32_t dofs, uint32_t aofs, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec, TCGv_vec)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - TCGv_vec t2 = tcg_temp_new_vec(type); - TCGv_vec t3 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); + TCGv_vec t2 = tcg_temp_new_vec(type); + TCGv_vec t3 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t1, tcg_env, aofs + i); tcg_gen_ld_vec(t2, tcg_env, bofs + i); tcg_gen_ld_vec(t3, tcg_env, cofs + i); @@ -1162,10 +1143,6 @@ static void expand_4_vec(unsigned vece, uint32_t dofs, uint32_t aofs, tcg_gen_st_vec(t1, tcg_env, aofs + i); } } - tcg_temp_free_vec(t3); - tcg_temp_free_vec(t2); - tcg_temp_free_vec(t1); - tcg_temp_free_vec(t0); } /* @@ -1178,23 +1155,18 @@ static void expand_4i_vec(unsigned vece, uint32_t dofs, uint32_t aofs, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_vec, TCGv_vec, int64_t)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - TCGv_vec t2 = tcg_temp_new_vec(type); - TCGv_vec t3 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); + TCGv_vec t2 = tcg_temp_new_vec(type); + TCGv_vec t3 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t1, tcg_env, aofs + i); tcg_gen_ld_vec(t2, tcg_env, bofs + i); tcg_gen_ld_vec(t3, tcg_env, cofs + i); fni(vece, t0, t1, t2, t3, c); tcg_gen_st_vec(t0, tcg_env, dofs + i); } - tcg_temp_free_vec(t3); - tcg_temp_free_vec(t2); - tcg_temp_free_vec(t1); - tcg_temp_free_vec(t0); } /* Expand a vector two-operand operation. */ @@ -1732,7 +1704,6 @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_vec t_vec = tcg_temp_new_vec(type); tcg_gen_dup_mem_vec(vece, t_vec, tcg_env, aofs); do_dup_store(type, dofs, oprsz, maxsz, t_vec); - tcg_temp_free_vec(t_vec); } else if (vece <= MO_32) { TCGv_i32 in = tcg_temp_ebb_new_i32(); switch (vece) { @@ -1766,7 +1737,6 @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, for (i = (aofs == dofs) * 16; i < oprsz; i += 16) { tcg_gen_st_vec(in, tcg_env, dofs + i); } - tcg_temp_free_vec(in); } else { TCGv_i64 in0 = tcg_temp_ebb_new_i64(); TCGv_i64 in1 = tcg_temp_ebb_new_i64(); @@ -1796,7 +1766,6 @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, for (i = (aofs == dofs) * 32; i < oprsz; i += 32) { tcg_gen_st_vec(in, tcg_env, dofs + i); } - tcg_temp_free_vec(in); } else if (TCG_TARGET_HAS_v128) { TCGv_vec in0 = tcg_temp_new_vec(TCG_TYPE_V128); TCGv_vec in1 = tcg_temp_new_vec(TCG_TYPE_V128); @@ -1807,8 +1776,6 @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, tcg_gen_st_vec(in0, tcg_env, dofs + i); tcg_gen_st_vec(in1, tcg_env, dofs + i + 16); } - tcg_temp_free_vec(in0); - tcg_temp_free_vec(in1); } else { TCGv_i64 in[4]; int j; @@ -3136,15 +3103,14 @@ static void expand_2sh_vec(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i32 shift, void (*fni)(unsigned, TCGv_vec, TCGv_vec, TCGv_i32)) { - TCGv_vec t0 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); - fni(vece, t0, t0, shift); - tcg_gen_st_vec(t0, tcg_env, dofs + i); + fni(vece, t1, t0, shift); + tcg_gen_st_vec(t1, tcg_env, dofs + i); } - tcg_temp_free_vec(t0); } static void @@ -3720,18 +3686,16 @@ static void expand_cmp_vec(unsigned vece, uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t oprsz, uint32_t tysz, TCGType type, TCGCond cond) { - TCGv_vec t0 = tcg_temp_new_vec(type); - TCGv_vec t1 = tcg_temp_new_vec(type); - uint32_t i; + for (uint32_t i = 0; i < oprsz; i += tysz) { + TCGv_vec t0 = tcg_temp_new_vec(type); + TCGv_vec t1 = tcg_temp_new_vec(type); + TCGv_vec t2 = tcg_temp_new_vec(type); - for (i = 0; i < oprsz; i += tysz) { tcg_gen_ld_vec(t0, tcg_env, aofs + i); tcg_gen_ld_vec(t1, tcg_env, bofs + i); - tcg_gen_cmp_vec(cond, vece, t0, t0, t1); - tcg_gen_st_vec(t0, tcg_env, dofs + i); + tcg_gen_cmp_vec(cond, vece, t2, t0, t1); + tcg_gen_st_vec(t2, tcg_env, dofs + i); } - tcg_temp_free_vec(t1); - tcg_temp_free_vec(t0); } void tcg_gen_gvec_cmp(TCGCond cond, unsigned vece, uint32_t dofs, diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c index 828eb9ee46..aa6bc6f57d 100644 --- a/tcg/tcg-op.c +++ b/tcg/tcg-op.c @@ -31,20 +31,26 @@ #include "tcg-internal.h" -void tcg_gen_op1(TCGOpcode opc, TCGArg a1) +/* + * Encourage the compiler to tail-call to a function, rather than inlining. + * Minimizes code size across 99 bottles of beer on the wall. + */ +#define NI __attribute__((noinline)) + +void NI tcg_gen_op1(TCGOpcode opc, TCGArg a1) { TCGOp *op = tcg_emit_op(opc, 1); op->args[0] = a1; } -void tcg_gen_op2(TCGOpcode opc, TCGArg a1, TCGArg a2) +void NI tcg_gen_op2(TCGOpcode opc, TCGArg a1, TCGArg a2) { TCGOp *op = tcg_emit_op(opc, 2); op->args[0] = a1; op->args[1] = a2; } -void tcg_gen_op3(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3) +void NI tcg_gen_op3(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3) { TCGOp *op = tcg_emit_op(opc, 3); op->args[0] = a1; @@ -52,7 +58,7 @@ void tcg_gen_op3(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3) op->args[2] = a3; } -void tcg_gen_op4(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, TCGArg a4) +void NI tcg_gen_op4(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, TCGArg a4) { TCGOp *op = tcg_emit_op(opc, 4); op->args[0] = a1; @@ -61,8 +67,8 @@ void tcg_gen_op4(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, TCGArg a4) op->args[3] = a4; } -void tcg_gen_op5(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, - TCGArg a4, TCGArg a5) +void NI tcg_gen_op5(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, + TCGArg a4, TCGArg a5) { TCGOp *op = tcg_emit_op(opc, 5); op->args[0] = a1; @@ -72,8 +78,8 @@ void tcg_gen_op5(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, op->args[4] = a5; } -void tcg_gen_op6(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, - TCGArg a4, TCGArg a5, TCGArg a6) +void NI tcg_gen_op6(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, + TCGArg a4, TCGArg a5, TCGArg a6) { TCGOp *op = tcg_emit_op(opc, 6); op->args[0] = a1; @@ -84,8 +90,195 @@ void tcg_gen_op6(TCGOpcode opc, TCGArg a1, TCGArg a2, TCGArg a3, op->args[5] = a6; } +/* + * With CONFIG_DEBUG_TCG, tcgv_*_tmp via tcgv_*_arg, is an out-of-line + * assertion check. Force tail calls to avoid too much code expansion. + */ +#ifdef CONFIG_DEBUG_TCG +# define DNI NI +#else +# define DNI +#endif + +static void DNI tcg_gen_op1_i32(TCGOpcode opc, TCGv_i32 a1) +{ + tcg_gen_op1(opc, tcgv_i32_arg(a1)); +} + +static void DNI tcg_gen_op1_i64(TCGOpcode opc, TCGv_i64 a1) +{ + tcg_gen_op1(opc, tcgv_i64_arg(a1)); +} + +static void DNI tcg_gen_op1i(TCGOpcode opc, TCGArg a1) +{ + tcg_gen_op1(opc, a1); +} + +static void DNI tcg_gen_op2_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2) +{ + tcg_gen_op2(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2)); +} + +static void DNI tcg_gen_op2_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2) +{ + tcg_gen_op2(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2)); +} + +static void DNI tcg_gen_op3_i32(TCGOpcode opc, TCGv_i32 a1, + TCGv_i32 a2, TCGv_i32 a3) +{ + tcg_gen_op3(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), tcgv_i32_arg(a3)); +} + +static void DNI tcg_gen_op3_i64(TCGOpcode opc, TCGv_i64 a1, + TCGv_i64 a2, TCGv_i64 a3) +{ + tcg_gen_op3(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), tcgv_i64_arg(a3)); +} + +static void DNI tcg_gen_op3i_i32(TCGOpcode opc, TCGv_i32 a1, + TCGv_i32 a2, TCGArg a3) +{ + tcg_gen_op3(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), a3); +} + +static void DNI tcg_gen_op3i_i64(TCGOpcode opc, TCGv_i64 a1, + TCGv_i64 a2, TCGArg a3) +{ + tcg_gen_op3(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), a3); +} + +static void DNI tcg_gen_ldst_op_i32(TCGOpcode opc, TCGv_i32 val, + TCGv_ptr base, TCGArg offset) +{ + tcg_gen_op3(opc, tcgv_i32_arg(val), tcgv_ptr_arg(base), offset); +} + +static void DNI tcg_gen_ldst_op_i64(TCGOpcode opc, TCGv_i64 val, + TCGv_ptr base, TCGArg offset) +{ + tcg_gen_op3(opc, tcgv_i64_arg(val), tcgv_ptr_arg(base), offset); +} + +static void DNI tcg_gen_op4_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGv_i32 a4) +{ + tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), tcgv_i32_arg(a4)); +} + +static void DNI tcg_gen_op4_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGv_i64 a4) +{ + tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), tcgv_i64_arg(a4)); +} + +static void DNI tcg_gen_op4i_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGArg a4) +{ + tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), a4); +} + +static void DNI tcg_gen_op4i_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGArg a4) +{ + tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), a4); +} + +static void DNI tcg_gen_op4ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGArg a3, TCGArg a4) +{ + tcg_gen_op4(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), a3, a4); +} + +static void DNI tcg_gen_op4ii_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGArg a3, TCGArg a4) +{ + tcg_gen_op4(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), a3, a4); +} + +static void DNI tcg_gen_op5_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGv_i32 a4, TCGv_i32 a5) +{ + tcg_gen_op5(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5)); +} + +static void DNI tcg_gen_op5_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGv_i64 a4, TCGv_i64 a5) +{ + tcg_gen_op5(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5)); +} + +static void DNI tcg_gen_op5ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGArg a4, TCGArg a5) +{ + tcg_gen_op5(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), a4, a5); +} + +static void DNI tcg_gen_op5ii_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGArg a4, TCGArg a5) +{ + tcg_gen_op5(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), a4, a5); +} + +static void DNI tcg_gen_op6_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGv_i32 a4, + TCGv_i32 a5, TCGv_i32 a6) +{ + tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5), + tcgv_i32_arg(a6)); +} + +static void DNI tcg_gen_op6_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGv_i64 a4, + TCGv_i64 a5, TCGv_i64 a6) +{ + tcg_gen_op6(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5), + tcgv_i64_arg(a6)); +} + +static void DNI tcg_gen_op6i_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGv_i32 a4, + TCGv_i32 a5, TCGArg a6) +{ + tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), tcgv_i32_arg(a4), tcgv_i32_arg(a5), a6); +} + +static void DNI tcg_gen_op6i_i64(TCGOpcode opc, TCGv_i64 a1, TCGv_i64 a2, + TCGv_i64 a3, TCGv_i64 a4, + TCGv_i64 a5, TCGArg a6) +{ + tcg_gen_op6(opc, tcgv_i64_arg(a1), tcgv_i64_arg(a2), + tcgv_i64_arg(a3), tcgv_i64_arg(a4), tcgv_i64_arg(a5), a6); +} + +static void DNI tcg_gen_op6ii_i32(TCGOpcode opc, TCGv_i32 a1, TCGv_i32 a2, + TCGv_i32 a3, TCGv_i32 a4, + TCGArg a5, TCGArg a6) +{ + tcg_gen_op6(opc, tcgv_i32_arg(a1), tcgv_i32_arg(a2), + tcgv_i32_arg(a3), tcgv_i32_arg(a4), a5, a6); +} + /* Generic ops. */ +void gen_set_label(TCGLabel *l) +{ + l->present = 1; + tcg_gen_op1(INDEX_op_set_label, label_arg(l)); +} + static void add_last_as_label_use(TCGLabel *l) { TCGLabelUse *u = tcg_malloc(sizeof(TCGLabelUse)); @@ -119,13 +312,40 @@ void tcg_gen_mb(TCGBar mb_type) } } +void tcg_gen_plugin_cb_start(unsigned from, unsigned type, unsigned wr) +{ + tcg_gen_op3(INDEX_op_plugin_cb_start, from, type, wr); +} + +void tcg_gen_plugin_cb_end(void) +{ + tcg_emit_op(INDEX_op_plugin_cb_end, 0); +} + /* 32 bit ops */ +void tcg_gen_discard_i32(TCGv_i32 arg) +{ + tcg_gen_op1_i32(INDEX_op_discard, arg); +} + +void tcg_gen_mov_i32(TCGv_i32 ret, TCGv_i32 arg) +{ + if (ret != arg) { + tcg_gen_op2_i32(INDEX_op_mov_i32, ret, arg); + } +} + void tcg_gen_movi_i32(TCGv_i32 ret, int32_t arg) { tcg_gen_mov_i32(ret, tcg_constant_i32(arg)); } +void tcg_gen_add_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_add_i32, ret, arg1, arg2); +} + void tcg_gen_addi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { /* some cases can be optimized here */ @@ -136,11 +356,15 @@ void tcg_gen_addi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) } } +void tcg_gen_sub_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_sub_i32, ret, arg1, arg2); +} + void tcg_gen_subfi_i32(TCGv_i32 ret, int32_t arg1, TCGv_i32 arg2) { - if (arg1 == 0 && TCG_TARGET_HAS_neg_i32) { - /* Don't recurse with tcg_gen_neg_i32. */ - tcg_gen_op2_i32(INDEX_op_neg_i32, ret, arg2); + if (arg1 == 0) { + tcg_gen_neg_i32(ret, arg2); } else { tcg_gen_sub_i32(ret, tcg_constant_i32(arg1), arg2); } @@ -148,12 +372,17 @@ void tcg_gen_subfi_i32(TCGv_i32 ret, int32_t arg1, TCGv_i32 arg2) void tcg_gen_subi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { - /* some cases can be optimized here */ - if (arg2 == 0) { - tcg_gen_mov_i32(ret, arg1); - } else { - tcg_gen_sub_i32(ret, arg1, tcg_constant_i32(arg2)); - } + tcg_gen_addi_i32(ret, arg1, -arg2); +} + +void tcg_gen_neg_i32(TCGv_i32 ret, TCGv_i32 arg) +{ + tcg_gen_op2_i32(INDEX_op_neg_i32, ret, arg); +} + +void tcg_gen_and_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_and_i32, ret, arg1, arg2); } void tcg_gen_andi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) @@ -184,6 +413,11 @@ void tcg_gen_andi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) tcg_gen_and_i32(ret, arg1, tcg_constant_i32(arg2)); } +void tcg_gen_or_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_or_i32, ret, arg1, arg2); +} + void tcg_gen_ori_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { /* Some cases can be optimized here. */ @@ -196,6 +430,11 @@ void tcg_gen_ori_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) } } +void tcg_gen_xor_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_xor_i32, ret, arg1, arg2); +} + void tcg_gen_xori_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { /* Some cases can be optimized here. */ @@ -209,6 +448,20 @@ void tcg_gen_xori_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) } } +void tcg_gen_not_i32(TCGv_i32 ret, TCGv_i32 arg) +{ + if (TCG_TARGET_HAS_not_i32) { + tcg_gen_op2_i32(INDEX_op_not_i32, ret, arg); + } else { + tcg_gen_xori_i32(ret, arg, -1); + } +} + +void tcg_gen_shl_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_shl_i32, ret, arg1, arg2); +} + void tcg_gen_shli_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { tcg_debug_assert(arg2 >= 0 && arg2 < 32); @@ -219,6 +472,11 @@ void tcg_gen_shli_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) } } +void tcg_gen_shr_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_shr_i32, ret, arg1, arg2); +} + void tcg_gen_shri_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { tcg_debug_assert(arg2 >= 0 && arg2 < 32); @@ -229,6 +487,11 @@ void tcg_gen_shri_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) } } +void tcg_gen_sar_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_sar_i32, ret, arg1, arg2); +} + void tcg_gen_sari_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { tcg_debug_assert(arg2 >= 0 && arg2 < 32); @@ -297,6 +560,11 @@ void tcg_gen_negsetcondi_i32(TCGCond cond, TCGv_i32 ret, tcg_gen_negsetcond_i32(cond, ret, arg1, tcg_constant_i32(arg2)); } +void tcg_gen_mul_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) +{ + tcg_gen_op3_i32(INDEX_op_mul_i32, ret, arg1, arg2); +} + void tcg_gen_muli_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) { if (arg2 == 0) { @@ -864,17 +1132,8 @@ void tcg_gen_movcond_i32(TCGCond cond, TCGv_i32 ret, TCGv_i32 c1, tcg_gen_mov_i32(ret, v1); } else if (cond == TCG_COND_NEVER) { tcg_gen_mov_i32(ret, v2); - } else if (TCG_TARGET_HAS_movcond_i32) { - tcg_gen_op6i_i32(INDEX_op_movcond_i32, ret, c1, c2, v1, v2, cond); } else { - TCGv_i32 t0 = tcg_temp_ebb_new_i32(); - TCGv_i32 t1 = tcg_temp_ebb_new_i32(); - tcg_gen_negsetcond_i32(cond, t0, c1, c2); - tcg_gen_and_i32(t1, v1, t0); - tcg_gen_andc_i32(ret, v2, t0); - tcg_gen_or_i32(ret, ret, t1); - tcg_temp_free_i32(t0); - tcg_temp_free_i32(t1); + tcg_gen_op6i_i32(INDEX_op_movcond_i32, ret, c1, c2, v1, v2, cond); } } @@ -1155,154 +1414,281 @@ void tcg_gen_abs_i32(TCGv_i32 ret, TCGv_i32 a) tcg_temp_free_i32(t); } -/* 64-bit ops */ +void tcg_gen_ld8u_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_ld8u_i32, ret, arg2, offset); +} + +void tcg_gen_ld8s_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_ld8s_i32, ret, arg2, offset); +} + +void tcg_gen_ld16u_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_ld16u_i32, ret, arg2, offset); +} + +void tcg_gen_ld16s_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_ld16s_i32, ret, arg2, offset); +} + +void tcg_gen_ld_i32(TCGv_i32 ret, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_ld_i32, ret, arg2, offset); +} + +void tcg_gen_st8_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_st8_i32, arg1, arg2, offset); +} + +void tcg_gen_st16_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_st16_i32, arg1, arg2, offset); +} -#if TCG_TARGET_REG_BITS == 32 -/* These are all inline for TCG_TARGET_REG_BITS == 64. */ +void tcg_gen_st_i32(TCGv_i32 arg1, TCGv_ptr arg2, tcg_target_long offset) +{ + tcg_gen_ldst_op_i32(INDEX_op_st_i32, arg1, arg2, offset); +} + + +/* 64-bit ops */ void tcg_gen_discard_i64(TCGv_i64 arg) { - tcg_gen_discard_i32(TCGV_LOW(arg)); - tcg_gen_discard_i32(TCGV_HIGH(arg)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op1_i64(INDEX_op_discard, arg); + } else { + tcg_gen_discard_i32(TCGV_LOW(arg)); + tcg_gen_discard_i32(TCGV_HIGH(arg)); + } } void tcg_gen_mov_i64(TCGv_i64 ret, TCGv_i64 arg) { - TCGTemp *ts = tcgv_i64_temp(arg); - - /* Canonicalize TCGv_i64 TEMP_CONST into TCGv_i32 TEMP_CONST. */ - if (ts->kind == TEMP_CONST) { - tcg_gen_movi_i64(ret, ts->val); + if (ret == arg) { + return; + } + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op2_i64(INDEX_op_mov_i64, ret, arg); } else { - tcg_gen_mov_i32(TCGV_LOW(ret), TCGV_LOW(arg)); - tcg_gen_mov_i32(TCGV_HIGH(ret), TCGV_HIGH(arg)); + TCGTemp *ts = tcgv_i64_temp(arg); + + /* Canonicalize TCGv_i64 TEMP_CONST into TCGv_i32 TEMP_CONST. */ + if (ts->kind == TEMP_CONST) { + tcg_gen_movi_i64(ret, ts->val); + } else { + tcg_gen_mov_i32(TCGV_LOW(ret), TCGV_LOW(arg)); + tcg_gen_mov_i32(TCGV_HIGH(ret), TCGV_HIGH(arg)); + } } } void tcg_gen_movi_i64(TCGv_i64 ret, int64_t arg) { - tcg_gen_movi_i32(TCGV_LOW(ret), arg); - tcg_gen_movi_i32(TCGV_HIGH(ret), arg >> 32); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_mov_i64(ret, tcg_constant_i64(arg)); + } else { + tcg_gen_movi_i32(TCGV_LOW(ret), arg); + tcg_gen_movi_i32(TCGV_HIGH(ret), arg >> 32); + } } void tcg_gen_ld8u_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld8u_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld8u_i64, ret, arg2, offset); + } else { + tcg_gen_ld8u_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + } } void tcg_gen_ld8s_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld8s_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld8s_i64, ret, arg2, offset); + } else { + tcg_gen_ld8s_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + } } void tcg_gen_ld16u_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld16u_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld16u_i64, ret, arg2, offset); + } else { + tcg_gen_ld16u_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + } } void tcg_gen_ld16s_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld16s_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld16s_i64, ret, arg2, offset); + } else { + tcg_gen_ld16s_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + } } void tcg_gen_ld32u_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld32u_i64, ret, arg2, offset); + } else { + tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_movi_i32(TCGV_HIGH(ret), 0); + } } void tcg_gen_ld32s_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld32s_i64, ret, arg2, offset); + } else { + tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31); + } } void tcg_gen_ld_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset) { - /* Since arg2 and ret have different types, - they cannot be the same temporary */ -#if HOST_BIG_ENDIAN - tcg_gen_ld_i32(TCGV_HIGH(ret), arg2, offset); - tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset + 4); -#else - tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); - tcg_gen_ld_i32(TCGV_HIGH(ret), arg2, offset + 4); -#endif + /* + * For 32-bit host, since arg2 and ret have different types, + * they cannot be the same temporary -- no chance of overlap. + */ + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_ld_i64, ret, arg2, offset); + } else if (HOST_BIG_ENDIAN) { + tcg_gen_ld_i32(TCGV_HIGH(ret), arg2, offset); + tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset + 4); + } else { + tcg_gen_ld_i32(TCGV_LOW(ret), arg2, offset); + tcg_gen_ld_i32(TCGV_HIGH(ret), arg2, offset + 4); + } } void tcg_gen_st8_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_st8_i32(TCGV_LOW(arg1), arg2, offset); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_st8_i64, arg1, arg2, offset); + } else { + tcg_gen_st8_i32(TCGV_LOW(arg1), arg2, offset); + } } void tcg_gen_st16_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_st16_i32(TCGV_LOW(arg1), arg2, offset); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_st16_i64, arg1, arg2, offset); + } else { + tcg_gen_st16_i32(TCGV_LOW(arg1), arg2, offset); + } } void tcg_gen_st32_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset) { - tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_st32_i64, arg1, arg2, offset); + } else { + tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset); + } } void tcg_gen_st_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset) { -#if HOST_BIG_ENDIAN - tcg_gen_st_i32(TCGV_HIGH(arg1), arg2, offset); - tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset + 4); -#else - tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset); - tcg_gen_st_i32(TCGV_HIGH(arg1), arg2, offset + 4); -#endif + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_ldst_op_i64(INDEX_op_st_i64, arg1, arg2, offset); + } else if (HOST_BIG_ENDIAN) { + tcg_gen_st_i32(TCGV_HIGH(arg1), arg2, offset); + tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset + 4); + } else { + tcg_gen_st_i32(TCGV_LOW(arg1), arg2, offset); + tcg_gen_st_i32(TCGV_HIGH(arg1), arg2, offset + 4); + } } void tcg_gen_add_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - tcg_gen_add2_i32(TCGV_LOW(ret), TCGV_HIGH(ret), TCGV_LOW(arg1), - TCGV_HIGH(arg1), TCGV_LOW(arg2), TCGV_HIGH(arg2)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_add_i64, ret, arg1, arg2); + } else { + tcg_gen_add2_i32(TCGV_LOW(ret), TCGV_HIGH(ret), TCGV_LOW(arg1), + TCGV_HIGH(arg1), TCGV_LOW(arg2), TCGV_HIGH(arg2)); + } } void tcg_gen_sub_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - tcg_gen_sub2_i32(TCGV_LOW(ret), TCGV_HIGH(ret), TCGV_LOW(arg1), - TCGV_HIGH(arg1), TCGV_LOW(arg2), TCGV_HIGH(arg2)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_sub_i64, ret, arg1, arg2); + } else { + tcg_gen_sub2_i32(TCGV_LOW(ret), TCGV_HIGH(ret), TCGV_LOW(arg1), + TCGV_HIGH(arg1), TCGV_LOW(arg2), TCGV_HIGH(arg2)); + } } void tcg_gen_and_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - tcg_gen_and_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); - tcg_gen_and_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_and_i64, ret, arg1, arg2); + } else { + tcg_gen_and_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); + tcg_gen_and_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + } } void tcg_gen_or_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - tcg_gen_or_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); - tcg_gen_or_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_or_i64, ret, arg1, arg2); + } else { + tcg_gen_or_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); + tcg_gen_or_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + } } void tcg_gen_xor_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - tcg_gen_xor_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); - tcg_gen_xor_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_xor_i64, ret, arg1, arg2); + } else { + tcg_gen_xor_i32(TCGV_LOW(ret), TCGV_LOW(arg1), TCGV_LOW(arg2)); + tcg_gen_xor_i32(TCGV_HIGH(ret), TCGV_HIGH(arg1), TCGV_HIGH(arg2)); + } } void tcg_gen_shl_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - gen_helper_shl_i64(ret, arg1, arg2); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_shl_i64, ret, arg1, arg2); + } else { + gen_helper_shl_i64(ret, arg1, arg2); + } } void tcg_gen_shr_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - gen_helper_shr_i64(ret, arg1, arg2); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_shr_i64, ret, arg1, arg2); + } else { + gen_helper_shr_i64(ret, arg1, arg2); + } } void tcg_gen_sar_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) { - gen_helper_sar_i64(ret, arg1, arg2); + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_sar_i64, ret, arg1, arg2); + } else { + gen_helper_sar_i64(ret, arg1, arg2); + } } void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) @@ -1310,6 +1696,12 @@ void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) TCGv_i64 t0; TCGv_i32 t1; + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op3_i64(INDEX_op_mul_i64, ret, arg1, arg2); + return; + } + + t0 = tcg_temp_ebb_new_i64(); t1 = tcg_temp_ebb_new_i32(); @@ -1326,15 +1718,6 @@ void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) tcg_temp_free_i32(t1); } -#else - -void tcg_gen_movi_i64(TCGv_i64 ret, int64_t arg) -{ - tcg_gen_mov_i64(ret, tcg_constant_i64(arg)); -} - -#endif /* TCG_TARGET_REG_SIZE == 32 */ - void tcg_gen_addi_i64(TCGv_i64 ret, TCGv_i64 arg1, int64_t arg2) { /* some cases can be optimized here */ @@ -1351,9 +1734,8 @@ void tcg_gen_addi_i64(TCGv_i64 ret, TCGv_i64 arg1, int64_t arg2) void tcg_gen_subfi_i64(TCGv_i64 ret, int64_t arg1, TCGv_i64 arg2) { - if (arg1 == 0 && TCG_TARGET_HAS_neg_i64) { - /* Don't recurse with tcg_gen_neg_i64. */ - tcg_gen_op2_i64(INDEX_op_neg_i64, ret, arg2); + if (arg1 == 0) { + tcg_gen_neg_i64(ret, arg2); } else if (TCG_TARGET_REG_BITS == 64) { tcg_gen_sub_i64(ret, tcg_constant_i64(arg1), arg2); } else { @@ -1365,15 +1747,17 @@ void tcg_gen_subfi_i64(TCGv_i64 ret, int64_t arg1, TCGv_i64 arg2) void tcg_gen_subi_i64(TCGv_i64 ret, TCGv_i64 arg1, int64_t arg2) { - /* some cases can be optimized here */ - if (arg2 == 0) { - tcg_gen_mov_i64(ret, arg1); - } else if (TCG_TARGET_REG_BITS == 64) { - tcg_gen_sub_i64(ret, arg1, tcg_constant_i64(arg2)); + tcg_gen_addi_i64(ret, arg1, -arg2); +} + +void tcg_gen_neg_i64(TCGv_i64 ret, TCGv_i64 arg) +{ + if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op2_i64(INDEX_op_neg_i64, ret, arg); } else { + TCGv_i32 zero = tcg_constant_i32(0); tcg_gen_sub2_i32(TCGV_LOW(ret), TCGV_HIGH(ret), - TCGV_LOW(arg1), TCGV_HIGH(arg1), - tcg_constant_i32(arg2), tcg_constant_i32(arg2 >> 32)); + zero, zero, TCGV_LOW(arg), TCGV_HIGH(arg)); } } @@ -2600,43 +2984,22 @@ void tcg_gen_movcond_i64(TCGCond cond, TCGv_i64 ret, TCGv_i64 c1, tcg_gen_mov_i64(ret, v1); } else if (cond == TCG_COND_NEVER) { tcg_gen_mov_i64(ret, v2); - } else if (TCG_TARGET_REG_BITS == 32) { + } else if (TCG_TARGET_REG_BITS == 64) { + tcg_gen_op6i_i64(INDEX_op_movcond_i64, ret, c1, c2, v1, v2, cond); + } else { TCGv_i32 t0 = tcg_temp_ebb_new_i32(); - TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 zero = tcg_constant_i32(0); + tcg_gen_op6i_i32(INDEX_op_setcond2_i32, t0, TCGV_LOW(c1), TCGV_HIGH(c1), TCGV_LOW(c2), TCGV_HIGH(c2), cond); - if (TCG_TARGET_HAS_movcond_i32) { - tcg_gen_movi_i32(t1, 0); - tcg_gen_movcond_i32(TCG_COND_NE, TCGV_LOW(ret), t0, t1, - TCGV_LOW(v1), TCGV_LOW(v2)); - tcg_gen_movcond_i32(TCG_COND_NE, TCGV_HIGH(ret), t0, t1, - TCGV_HIGH(v1), TCGV_HIGH(v2)); - } else { - tcg_gen_neg_i32(t0, t0); - - tcg_gen_and_i32(t1, TCGV_LOW(v1), t0); - tcg_gen_andc_i32(TCGV_LOW(ret), TCGV_LOW(v2), t0); - tcg_gen_or_i32(TCGV_LOW(ret), TCGV_LOW(ret), t1); + tcg_gen_movcond_i32(TCG_COND_NE, TCGV_LOW(ret), t0, zero, + TCGV_LOW(v1), TCGV_LOW(v2)); + tcg_gen_movcond_i32(TCG_COND_NE, TCGV_HIGH(ret), t0, zero, + TCGV_HIGH(v1), TCGV_HIGH(v2)); - tcg_gen_and_i32(t1, TCGV_HIGH(v1), t0); - tcg_gen_andc_i32(TCGV_HIGH(ret), TCGV_HIGH(v2), t0); - tcg_gen_or_i32(TCGV_HIGH(ret), TCGV_HIGH(ret), t1); - } tcg_temp_free_i32(t0); - tcg_temp_free_i32(t1); - } else if (TCG_TARGET_HAS_movcond_i64) { - tcg_gen_op6i_i64(INDEX_op_movcond_i64, ret, c1, c2, v1, v2, cond); - } else { - TCGv_i64 t0 = tcg_temp_ebb_new_i64(); - TCGv_i64 t1 = tcg_temp_ebb_new_i64(); - tcg_gen_negsetcond_i64(cond, t0, c1, c2); - tcg_gen_and_i64(t1, v1, t0); - tcg_gen_andc_i64(ret, v2, t0); - tcg_gen_or_i64(ret, ret, t1); - tcg_temp_free_i64(t0); - tcg_temp_free_i64(t1); } } @@ -2872,6 +3235,11 @@ void tcg_gen_extr32_i64(TCGv_i64 lo, TCGv_i64 hi, TCGv_i64 arg) tcg_gen_shri_i64(hi, arg, 32); } +void tcg_gen_concat32_i64(TCGv_i64 ret, TCGv_i64 lo, TCGv_i64 hi) +{ + tcg_gen_deposit_i64(ret, lo, hi, 32, 32); +} + void tcg_gen_extr_i128_i64(TCGv_i64 lo, TCGv_i64 hi, TCGv_i128 arg) { tcg_gen_mov_i64(lo, TCGV128_LOW(arg)); diff --git a/tcg/tcg.c b/tcg/tcg.c index 35158a0846..d2ea22b397 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -653,6 +653,7 @@ static void tcg_out_movext3(TCGContext *s, const TCGMovExtend *i1, #define C_O1_I4(O1, I1, I2, I3, I4) C_PFX5(c_o1_i4_, O1, I1, I2, I3, I4), #define C_N1_I2(O1, I1, I2) C_PFX3(c_n1_i2_, O1, I1, I2), +#define C_N2_I1(O1, O2, I1) C_PFX3(c_n2_i1_, O1, O2, I1), #define C_O2_I1(O1, O2, I1) C_PFX3(c_o2_i1_, O1, O2, I1), #define C_O2_I2(O1, O2, I1, I2) C_PFX4(c_o2_i2_, O1, O2, I1, I2), @@ -675,6 +676,7 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode); #undef C_O1_I3 #undef C_O1_I4 #undef C_N1_I2 +#undef C_N2_I1 #undef C_O2_I1 #undef C_O2_I2 #undef C_O2_I3 @@ -694,6 +696,7 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode); #define C_O1_I4(O1, I1, I2, I3, I4) { .args_ct_str = { #O1, #I1, #I2, #I3, #I4 } }, #define C_N1_I2(O1, I1, I2) { .args_ct_str = { "&" #O1, #I1, #I2 } }, +#define C_N2_I1(O1, O2, I1) { .args_ct_str = { "&" #O1, "&" #O2, #I1 } }, #define C_O2_I1(O1, O2, I1) { .args_ct_str = { #O1, #O2, #I1 } }, #define C_O2_I2(O1, O2, I1, I2) { .args_ct_str = { #O1, #O2, #I1, #I2 } }, @@ -715,6 +718,7 @@ static const TCGTargetOpDef constraint_sets[] = { #undef C_O1_I3 #undef C_O1_I4 #undef C_N1_I2 +#undef C_N2_I1 #undef C_O2_I1 #undef C_O2_I2 #undef C_O2_I3 @@ -734,6 +738,7 @@ static const TCGTargetOpDef constraint_sets[] = { #define C_O1_I4(O1, I1, I2, I3, I4) C_PFX5(c_o1_i4_, O1, I1, I2, I3, I4) #define C_N1_I2(O1, I1, I2) C_PFX3(c_n1_i2_, O1, I1, I2) +#define C_N2_I1(O1, O2, I1) C_PFX3(c_n2_i1_, O1, O2, I1) #define C_O2_I1(O1, O2, I1) C_PFX3(c_o2_i1_, O1, O2, I1) #define C_O2_I2(O1, O2, I1, I2) C_PFX4(c_o2_i2_, O1, O2, I1, I2) @@ -1567,8 +1572,8 @@ void tcg_set_frame(TCGContext *s, TCGReg reg, intptr_t start, intptr_t size) = tcg_global_reg_new_internal(s, TCG_TYPE_PTR, reg, "_frame"); } -TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, - intptr_t offset, const char *name) +static TCGTemp *tcg_global_mem_new_internal(TCGv_ptr base, intptr_t offset, + const char *name, TCGType type) { TCGContext *s = tcg_ctx; TCGTemp *base_ts = tcgv_ptr_temp(base); @@ -1627,7 +1632,25 @@ TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, return ts; } -TCGTemp *tcg_temp_new_internal(TCGType type, TCGTempKind kind) +TCGv_i32 tcg_global_mem_new_i32(TCGv_ptr reg, intptr_t off, const char *name) +{ + TCGTemp *ts = tcg_global_mem_new_internal(reg, off, name, TCG_TYPE_I32); + return temp_tcgv_i32(ts); +} + +TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t off, const char *name) +{ + TCGTemp *ts = tcg_global_mem_new_internal(reg, off, name, TCG_TYPE_I64); + return temp_tcgv_i64(ts); +} + +TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t off, const char *name) +{ + TCGTemp *ts = tcg_global_mem_new_internal(reg, off, name, TCG_TYPE_PTR); + return temp_tcgv_ptr(ts); +} + +static TCGTemp *tcg_temp_new_internal(TCGType type, TCGTempKind kind) { TCGContext *s = tcg_ctx; TCGTemp *ts; @@ -1691,6 +1714,46 @@ TCGTemp *tcg_temp_new_internal(TCGType type, TCGTempKind kind) return ts; } +TCGv_i32 tcg_temp_new_i32(void) +{ + return temp_tcgv_i32(tcg_temp_new_internal(TCG_TYPE_I32, TEMP_TB)); +} + +TCGv_i32 tcg_temp_ebb_new_i32(void) +{ + return temp_tcgv_i32(tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB)); +} + +TCGv_i64 tcg_temp_new_i64(void) +{ + return temp_tcgv_i64(tcg_temp_new_internal(TCG_TYPE_I64, TEMP_TB)); +} + +TCGv_i64 tcg_temp_ebb_new_i64(void) +{ + return temp_tcgv_i64(tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB)); +} + +TCGv_ptr tcg_temp_new_ptr(void) +{ + return temp_tcgv_ptr(tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_TB)); +} + +TCGv_ptr tcg_temp_ebb_new_ptr(void) +{ + return temp_tcgv_ptr(tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB)); +} + +TCGv_i128 tcg_temp_new_i128(void) +{ + return temp_tcgv_i128(tcg_temp_new_internal(TCG_TYPE_I128, TEMP_TB)); +} + +TCGv_i128 tcg_temp_ebb_new_i128(void) +{ + return temp_tcgv_i128(tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB)); +} + TCGv_vec tcg_temp_new_vec(TCGType type) { TCGTemp *t; @@ -1746,6 +1809,31 @@ void tcg_temp_free_internal(TCGTemp *ts) } } +void tcg_temp_free_i32(TCGv_i32 arg) +{ + tcg_temp_free_internal(tcgv_i32_temp(arg)); +} + +void tcg_temp_free_i64(TCGv_i64 arg) +{ + tcg_temp_free_internal(tcgv_i64_temp(arg)); +} + +void tcg_temp_free_i128(TCGv_i128 arg) +{ + tcg_temp_free_internal(tcgv_i128_temp(arg)); +} + +void tcg_temp_free_ptr(TCGv_ptr arg) +{ + tcg_temp_free_internal(tcgv_ptr_temp(arg)); +} + +void tcg_temp_free_vec(TCGv_vec arg) +{ + tcg_temp_free_internal(tcgv_vec_temp(arg)); +} + TCGTemp *tcg_constant_internal(TCGType type, int64_t val) { TCGContext *s = tcg_ctx; @@ -1801,6 +1889,21 @@ TCGTemp *tcg_constant_internal(TCGType type, int64_t val) return ts; } +TCGv_i32 tcg_constant_i32(int32_t val) +{ + return temp_tcgv_i32(tcg_constant_internal(TCG_TYPE_I32, val)); +} + +TCGv_i64 tcg_constant_i64(int64_t val) +{ + return temp_tcgv_i64(tcg_constant_internal(TCG_TYPE_I64, val)); +} + +TCGv_ptr tcg_constant_ptr_int(intptr_t val) +{ + return temp_tcgv_ptr(tcg_constant_internal(TCG_TYPE_PTR, val)); +} + TCGv_vec tcg_constant_vec(TCGType type, unsigned vece, int64_t val) { val = dup_const(vece, val); @@ -1874,6 +1977,7 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_mov_i32: case INDEX_op_setcond_i32: case INDEX_op_brcond_i32: + case INDEX_op_movcond_i32: case INDEX_op_ld8u_i32: case INDEX_op_ld8s_i32: case INDEX_op_ld16u_i32: @@ -1884,6 +1988,7 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_st_i32: case INDEX_op_add_i32: case INDEX_op_sub_i32: + case INDEX_op_neg_i32: case INDEX_op_mul_i32: case INDEX_op_and_i32: case INDEX_op_or_i32: @@ -1895,8 +2000,6 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_negsetcond_i32: return TCG_TARGET_HAS_negsetcond_i32; - case INDEX_op_movcond_i32: - return TCG_TARGET_HAS_movcond_i32; case INDEX_op_div_i32: case INDEX_op_divu_i32: return TCG_TARGET_HAS_div_i32; @@ -1943,8 +2046,6 @@ bool tcg_op_supported(TCGOpcode op) return TCG_TARGET_HAS_bswap32_i32; case INDEX_op_not_i32: return TCG_TARGET_HAS_not_i32; - case INDEX_op_neg_i32: - return TCG_TARGET_HAS_neg_i32; case INDEX_op_andc_i32: return TCG_TARGET_HAS_andc_i32; case INDEX_op_orc_i32: @@ -1969,6 +2070,7 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_mov_i64: case INDEX_op_setcond_i64: case INDEX_op_brcond_i64: + case INDEX_op_movcond_i64: case INDEX_op_ld8u_i64: case INDEX_op_ld8s_i64: case INDEX_op_ld16u_i64: @@ -1982,6 +2084,7 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_st_i64: case INDEX_op_add_i64: case INDEX_op_sub_i64: + case INDEX_op_neg_i64: case INDEX_op_mul_i64: case INDEX_op_and_i64: case INDEX_op_or_i64: @@ -1995,8 +2098,6 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_negsetcond_i64: return TCG_TARGET_HAS_negsetcond_i64; - case INDEX_op_movcond_i64: - return TCG_TARGET_HAS_movcond_i64; case INDEX_op_div_i64: case INDEX_op_divu_i64: return TCG_TARGET_HAS_div_i64; @@ -2040,8 +2141,6 @@ bool tcg_op_supported(TCGOpcode op) return TCG_TARGET_HAS_bswap64_i64; case INDEX_op_not_i64: return TCG_TARGET_HAS_not_i64; - case INDEX_op_neg_i64: - return TCG_TARGET_HAS_neg_i64; case INDEX_op_andc_i64: return TCG_TARGET_HAS_andc_i64; case INDEX_op_orc_i64: @@ -5927,11 +6026,6 @@ static void tcg_out_st_helper_args(TCGContext *s, const TCGLabelQemuLdst *ldst, tcg_out_helper_load_common_args(s, ldst, parm, info, next_arg); } -void tcg_dump_op_count(GString *buf) -{ - g_string_append_printf(buf, "[TCG profiler not compiled]\n"); -} - int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start) { int i, start_words, num_insns; @@ -6128,11 +6222,6 @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start) return tcg_current_code_size(s); } -void tcg_dump_info(GString *buf) -{ - g_string_append_printf(buf, "[TCG profiler not compiled]\n"); -} - #ifdef ELF_HOST_MACHINE /* In order to use this feature, the backend needs to do three things: diff --git a/tcg/tci.c b/tcg/tci.c index 4640902c88..3cc851b7bd 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -733,12 +733,10 @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_exec(CPUArchState *env, regs[r0] = ~regs[r1]; break; #endif -#if TCG_TARGET_HAS_neg_i32 || TCG_TARGET_HAS_neg_i64 CASE_32_64(neg) tci_args_rr(insn, &r0, &r1); regs[r0] = -regs[r1]; break; -#endif #if TCG_TARGET_REG_BITS == 64 /* Load/store operations (64 bit). */ diff --git a/tcg/tci/tcg-target.h b/tcg/tci/tcg-target.h index 91ca33b616..2a13816c8e 100644 --- a/tcg/tci/tcg-target.h +++ b/tcg/tci/tcg-target.h @@ -65,11 +65,9 @@ #define TCG_TARGET_HAS_clz_i32 1 #define TCG_TARGET_HAS_ctz_i32 1 #define TCG_TARGET_HAS_ctpop_i32 1 -#define TCG_TARGET_HAS_neg_i32 1 #define TCG_TARGET_HAS_not_i32 1 #define TCG_TARGET_HAS_orc_i32 1 #define TCG_TARGET_HAS_rot_i32 1 -#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_TARGET_HAS_negsetcond_i32 0 #define TCG_TARGET_HAS_muls2_i32 1 #define TCG_TARGET_HAS_muluh_i32 0 @@ -100,11 +98,9 @@ #define TCG_TARGET_HAS_clz_i64 1 #define TCG_TARGET_HAS_ctz_i64 1 #define TCG_TARGET_HAS_ctpop_i64 1 -#define TCG_TARGET_HAS_neg_i64 1 #define TCG_TARGET_HAS_not_i64 1 #define TCG_TARGET_HAS_orc_i64 1 #define TCG_TARGET_HAS_rot_i64 1 -#define TCG_TARGET_HAS_movcond_i64 1 #define TCG_TARGET_HAS_negsetcond_i64 0 #define TCG_TARGET_HAS_muls2_i64 1 #define TCG_TARGET_HAS_add2_i32 1 diff --git a/tests/data/acpi/virt/DBG2 b/tests/data/acpi/virt/DBG2 index 86e6314f7b..0a05e1a47f 100644 --- a/tests/data/acpi/virt/DBG2 +++ b/tests/data/acpi/virt/DBG2 Binary files differdiff --git a/tests/data/acpi/virt/SPCR b/tests/data/acpi/virt/SPCR index 24e0a579e7..cf0f2b7522 100644 --- a/tests/data/acpi/virt/SPCR +++ b/tests/data/acpi/virt/SPCR Binary files differdiff --git a/tests/qtest/qmp-cmd-test.c b/tests/qtest/qmp-cmd-test.c index 73a670e8fa..2c15f60958 100644 --- a/tests/qtest/qmp-cmd-test.c +++ b/tests/qtest/qmp-cmd-test.c @@ -45,6 +45,7 @@ static int query_error_class(const char *cmd) { "query-acpi-ospm-status", ERROR_CLASS_GENERIC_ERROR }, { "query-balloon", ERROR_CLASS_DEVICE_NOT_ACTIVE }, { "query-hotpluggable-cpus", ERROR_CLASS_GENERIC_ERROR }, + { "query-hv-balloon-status-report", ERROR_CLASS_GENERIC_ERROR }, { "query-vm-generation-id", ERROR_CLASS_GENERIC_ERROR }, /* Only valid with a USB bus added */ { "x-query-usb", ERROR_CLASS_GENERIC_ERROR }, diff --git a/tests/unit/meson.build b/tests/unit/meson.build index f33ae64b8d..e6c51e7a86 100644 --- a/tests/unit/meson.build +++ b/tests/unit/meson.build @@ -21,6 +21,7 @@ tests = { 'test-opts-visitor': [testqapi], 'test-visitor-serialization': [testqapi], 'test-bitmap': [], + 'test-resv-mem': [], # all code tested by test-x86-cpuid is inside topology.h 'test-x86-cpuid': [], 'test-cutils': [], diff --git a/tests/unit/test-resv-mem.c b/tests/unit/test-resv-mem.c new file mode 100644 index 0000000000..5963274e2c --- /dev/null +++ b/tests/unit/test-resv-mem.c @@ -0,0 +1,316 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * + * reserved-region/range.c unit-tests. + * + * Copyright (C) 2023, Red Hat, Inc. + * + * Author: Eric Auger <eric.auger@redhat.com> + */ + +#include "qemu/osdep.h" +#include "qemu/range.h" +#include "exec/memory.h" +#include "qemu/reserved-region.h" + +#define DEBUG 0 + +#if DEBUG +static void print_ranges(const char *prefix, GList *ranges) +{ + GList *l; + int i = 0; + + if (!g_list_length(ranges)) { + printf("%s is void\n", prefix); + return; + } + for (l = ranges; l; l = l->next) { + Range *r = (Range *)l->data; + + printf("%s rev[%i] = [0x%"PRIx64",0x%"PRIx64"]\n", + prefix, i, range_lob(r), range_upb(r)); + i++; + } +} +#endif + +static void compare_ranges(const char *prefix, GList *ranges, + GList *expected) +{ + GList *l, *e; + +#if DEBUG + print_ranges("out", ranges); + print_ranges("expected", expected); +#endif + g_assert_cmpint(g_list_length(ranges), ==, g_list_length(expected)); + for (l = ranges, e = expected; l ; l = l->next, e = e->next) { + Range *r = (Range *)l->data; + Range *er = (Range *)e->data; + + g_assert_true(range_lob(r) == range_lob(er) && + range_upb(r) == range_upb(er)); + } +} + +static GList *insert_sorted_range(GList *list, uint64_t lob, uint64_t upb) +{ + Range *new = g_new0(Range, 1); + + range_set_bounds(new, lob, upb); + return range_list_insert(list, new); +} + +static void reset(GList **in, GList **out, GList **expected) +{ + g_list_free_full(*in, g_free); + g_list_free_full(*out, g_free); + g_list_free_full(*expected, g_free); + *in = NULL; + *out = NULL; + *expected = NULL; +} + +static void +run_range_inverse_array(const char *prefix, GList **in, GList **expected, + uint64_t low, uint64_t high) +{ + GList *out = NULL; + range_inverse_array(*in, &out, low, high); + compare_ranges(prefix, out, *expected); + reset(in, &out, expected); +} + +static void check_range_reverse_array(void) +{ + GList *in = NULL, *expected = NULL; + + /* test 1 */ + + in = insert_sorted_range(in, 0x10000, UINT64_MAX); + expected = insert_sorted_range(expected, 0x0, 0xFFFF); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); + + /* test 2 */ + + in = insert_sorted_range(in, 0x10000, 0xFFFFFFFFFFFF); + expected = insert_sorted_range(expected, 0x0, 0xFFFF); + expected = insert_sorted_range(expected, 0x1000000000000, UINT64_MAX); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); + + /* test 3 */ + + in = insert_sorted_range(in, 0x0, 0xFFFF); + in = insert_sorted_range(in, 0x10000, 0x2FFFF); + expected = insert_sorted_range(expected, 0x30000, UINT64_MAX); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); + + /* test 4 */ + + in = insert_sorted_range(in, 0x50000, 0x5FFFF); + in = insert_sorted_range(in, 0x60000, 0xFFFFFFFFFFFF); + expected = insert_sorted_range(expected, 0x0, 0x4FFFF); + expected = insert_sorted_range(expected, 0x1000000000000, UINT64_MAX); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); + + /* test 5 */ + + in = insert_sorted_range(in, 0x0, UINT64_MAX); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); + + /* test 6 */ + in = insert_sorted_range(in, 0x10000, 0x1FFFF); + in = insert_sorted_range(in, 0x30000, 0x6FFFF); + in = insert_sorted_range(in, 0x90000, UINT64_MAX); + expected = insert_sorted_range(expected, 0x0, 0xFFFF); + expected = insert_sorted_range(expected, 0x20000, 0x2FFFF); + expected = insert_sorted_range(expected, 0x70000, 0x8FFFF); + run_range_inverse_array("test1", &in, &expected, 0x0, UINT64_MAX); +} + +static void check_range_reverse_array_low_end(void) +{ + GList *in = NULL, *expected = NULL; + + /* test 1 */ + in = insert_sorted_range(in, 0x0, UINT64_MAX); + run_range_inverse_array("test1", &in, &expected, 0x10000, 0xFFFFFF); + + /* test 2 */ + + in = insert_sorted_range(in, 0x0, 0xFFFF); + in = insert_sorted_range(in, 0x20000, 0x2FFFF); + expected = insert_sorted_range(expected, 0x40000, 0xFFFFFFFFFFFF); + run_range_inverse_array("test2", &in, &expected, 0x40000, 0xFFFFFFFFFFFF); + + /* test 3 */ + in = insert_sorted_range(in, 0x0, 0xFFFF); + in = insert_sorted_range(in, 0x20000, 0x2FFFF); + in = insert_sorted_range(in, 0x1000000000000, UINT64_MAX); + expected = insert_sorted_range(expected, 0x40000, 0xFFFFFFFFFFFF); + run_range_inverse_array("test3", &in, &expected, 0x40000, 0xFFFFFFFFFFFF); + + /* test 4 */ + + in = insert_sorted_range(in, 0x0, 0xFFFF); + in = insert_sorted_range(in, 0x20000, 0x2FFFF); + in = insert_sorted_range(in, 0x1000000000000, UINT64_MAX); + expected = insert_sorted_range(expected, 0x30000, 0xFFFFFFFFFFFF); + run_range_inverse_array("test4", &in, &expected, 0x20000, 0xFFFFFFFFFFFF); + + /* test 5 */ + + in = insert_sorted_range(in, 0x2000, 0xFFFF); + in = insert_sorted_range(in, 0x20000, 0x2FFFF); + in = insert_sorted_range(in, 0x100000000, 0x1FFFFFFFF); + expected = insert_sorted_range(expected, 0x1000, 0x1FFF); + expected = insert_sorted_range(expected, 0x10000, 0x1FFFF); + expected = insert_sorted_range(expected, 0x30000, 0xFFFFFFFF); + expected = insert_sorted_range(expected, 0x200000000, 0xFFFFFFFFFFFF); + run_range_inverse_array("test5", &in, &expected, 0x1000, 0xFFFFFFFFFFFF); + + /* test 6 */ + + in = insert_sorted_range(in, 0x10000000 , 0x1FFFFFFF); + in = insert_sorted_range(in, 0x100000000, 0x1FFFFFFFF); + expected = insert_sorted_range(expected, 0x0, 0xFFFF); + run_range_inverse_array("test6", &in, &expected, 0x0, 0xFFFF); +} + +static ReservedRegion *alloc_resv_mem(unsigned type, uint64_t lob, uint64_t upb) +{ + ReservedRegion *r; + + r = g_new0(ReservedRegion, 1); + r->type = type; + range_set_bounds(&r->range, lob, upb); + return r; +} + +static void print_resv_region_list(const char *prefix, GList *list, + uint32_t expected_length) +{ + int i = g_list_length(list); + + g_assert_cmpint(i, ==, expected_length); +#if DEBUG + i = 0; + for (GList *l = list; l; l = l->next) { + ReservedRegion *r = (ReservedRegion *)l->data; + Range *range = &r->range; + + printf("%s item[%d]=[0x%x, 0x%"PRIx64", 0x%"PRIx64"]\n", + prefix, i++, r->type, range_lob(range), range_upb(range)); + } +#endif +} + +static void free_resv_region(gpointer data) +{ + ReservedRegion *reg = (ReservedRegion *)data; + + g_free(reg); +} + +static void check_resv_region_list_insert(void) +{ + ReservedRegion *r[10]; + GList *l = NULL; + + r[0] = alloc_resv_mem(0xA, 0, 0xFFFF); + r[1] = alloc_resv_mem(0xA, 0x20000, 0x2FFFF); + l = resv_region_list_insert(l, r[0]); + l = resv_region_list_insert(l, r[1]); + print_resv_region_list("test1", l, 2); + + /* adjacent on left */ + r[2] = alloc_resv_mem(0xB, 0x0, 0xFFF); + l = resv_region_list_insert(l, r[2]); + /* adjacent on right */ + r[3] = alloc_resv_mem(0xC, 0x21000, 0x2FFFF); + l = resv_region_list_insert(l, r[3]); + print_resv_region_list("test2", l, 4); + + /* exact overlap of D into C*/ + r[4] = alloc_resv_mem(0xD, 0x21000, 0x2FFFF); + l = resv_region_list_insert(l, r[4]); + print_resv_region_list("test3", l, 4); + + /* in the middle */ + r[5] = alloc_resv_mem(0xE, 0x22000, 0x23FFF); + l = resv_region_list_insert(l, r[5]); + print_resv_region_list("test4", l, 6); + + /* overwrites several existing ones */ + r[6] = alloc_resv_mem(0xF, 0x10000, 0x2FFFF); + l = resv_region_list_insert(l, r[6]); + print_resv_region_list("test5", l, 3); + + /* contiguous at the end */ + r[7] = alloc_resv_mem(0x0, 0x30000, 0x40000); + l = resv_region_list_insert(l, r[7]); + print_resv_region_list("test6", l, 4); + + g_list_free_full(l, free_resv_region); + l = NULL; + + r[0] = alloc_resv_mem(0x0, 0x10000, 0x1FFFF); + l = resv_region_list_insert(l, r[0]); + /* insertion before the 1st item */ + r[1] = alloc_resv_mem(0x1, 0x0, 0xFF); + l = resv_region_list_insert(l, r[1]); + print_resv_region_list("test8", l, 2); + + /* collision on the left side */ + r[2] = alloc_resv_mem(0xA, 0x1200, 0x11FFF); + l = resv_region_list_insert(l, r[2]); + print_resv_region_list("test9", l, 3); + + /* collision on the right side */ + r[3] = alloc_resv_mem(0xA, 0x1F000, 0x2FFFF); + l = resv_region_list_insert(l, r[3]); + print_resv_region_list("test10", l, 4); + + /* override everything */ + r[4] = alloc_resv_mem(0xF, 0x0, UINT64_MAX); + l = resv_region_list_insert(l, r[4]); + print_resv_region_list("test11", l, 1); + + g_list_free_full(l, free_resv_region); + l = NULL; + + r[0] = alloc_resv_mem(0xF, 0x1000000000000, UINT64_MAX); + l = resv_region_list_insert(l, r[0]); + print_resv_region_list("test12", l, 1); + + r[1] = alloc_resv_mem(0xA, 0x0, 0xFFFFFFF); + l = resv_region_list_insert(l, r[1]); + print_resv_region_list("test12", l, 2); + + r[2] = alloc_resv_mem(0xB, 0x100000000, 0x1FFFFFFFF); + l = resv_region_list_insert(l, r[2]); + print_resv_region_list("test12", l, 3); + + r[3] = alloc_resv_mem(0x0, 0x010000000, 0x2FFFFFFFF); + l = resv_region_list_insert(l, r[3]); + print_resv_region_list("test12", l, 3); + + g_list_free_full(l, free_resv_region); +} + +int main(int argc, char **argv) +{ + g_test_init(&argc, &argv, NULL); + + g_test_add_func("/resv-mem/range_reverse_array", + check_range_reverse_array); + g_test_add_func("/resv-mem/range_reverse_array_low_end", + check_range_reverse_array_low_end); + g_test_add_func("/resv-mem/resv_region_list_insert", + check_resv_region_list_insert); + + g_test_run(); + + return 0; +} diff --git a/tests/unit/test-uuid.c b/tests/unit/test-uuid.c index aedc125ae9..739b91583c 100644 --- a/tests/unit/test-uuid.c +++ b/tests/unit/test-uuid.c @@ -145,7 +145,7 @@ static void test_uuid_unparse(void) int i; for (i = 0; i < ARRAY_SIZE(uuid_test_data); i++) { - char out[37]; + char out[UUID_STR_LEN]; if (!uuid_test_data[i].check_unparse) { continue; diff --git a/util/cpuinfo-loongarch.c b/util/cpuinfo-loongarch.c new file mode 100644 index 0000000000..08b6d7460c --- /dev/null +++ b/util/cpuinfo-loongarch.c @@ -0,0 +1,35 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * Host specific cpu identification for LoongArch. + */ + +#include "qemu/osdep.h" +#include "host/cpuinfo.h" + +#ifdef CONFIG_GETAUXVAL +# include <sys/auxv.h> +#else +# include "elf.h" +#endif +#include <asm/hwcap.h> + +unsigned cpuinfo; + +/* Called both as constructor and (possibly) via other constructors. */ +unsigned __attribute__((constructor)) cpuinfo_init(void) +{ + unsigned info = cpuinfo; + unsigned long hwcap; + + if (info) { + return info; + } + + hwcap = qemu_getauxval(AT_HWCAP); + + info = CPUINFO_ALWAYS; + info |= (hwcap & HWCAP_LOONGARCH_LSX ? CPUINFO_LSX : 0); + + cpuinfo = info; + return info; +} diff --git a/util/filemonitor-inotify.c b/util/filemonitor-inotify.c index 2c45f7f176..2121111f38 100644 --- a/util/filemonitor-inotify.c +++ b/util/filemonitor-inotify.c @@ -81,16 +81,25 @@ static void qemu_file_monitor_watch(void *arg) /* Loop over all events in the buffer */ while (used < len) { - struct inotify_event *ev = - (struct inotify_event *)(buf + used); - const char *name = ev->len ? ev->name : ""; - QFileMonitorDir *dir = g_hash_table_lookup(mon->idmap, - GINT_TO_POINTER(ev->wd)); - uint32_t iev = ev->mask & - (IN_CREATE | IN_MODIFY | IN_DELETE | IN_IGNORED | - IN_MOVED_TO | IN_MOVED_FROM | IN_ATTRIB); + const char *name; + QFileMonitorDir *dir; + uint32_t iev; int qev; gsize i; + struct inotify_event *ev = (struct inotify_event *)(buf + used); + + /* + * We trust the kenel to provide valid buffer with complete event + * records. + */ + assert(len - used >= sizeof(struct inotify_event)); + assert(len - used - sizeof(struct inotify_event) >= ev->len); + + name = ev->len ? ev->name : ""; + dir = g_hash_table_lookup(mon->idmap, GINT_TO_POINTER(ev->wd)); + iev = ev->mask & + (IN_CREATE | IN_MODIFY | IN_DELETE | IN_IGNORED | + IN_MOVED_TO | IN_MOVED_FROM | IN_ATTRIB); used += sizeof(struct inotify_event) + ev->len; diff --git a/util/meson.build b/util/meson.build index 769b24f2e0..c2322ef6e7 100644 --- a/util/meson.build +++ b/util/meson.build @@ -52,6 +52,7 @@ util_ss.add(files('qdist.c')) util_ss.add(files('qht.c')) util_ss.add(files('qsp.c')) util_ss.add(files('range.c')) +util_ss.add(files('reserved-region.c')) util_ss.add(files('stats64.c')) util_ss.add(files('systemd.c')) util_ss.add(files('transactions.c')) @@ -113,6 +114,8 @@ if cpu == 'aarch64' util_ss.add(files('cpuinfo-aarch64.c')) elif cpu in ['x86', 'x86_64'] util_ss.add(files('cpuinfo-i386.c')) +elif cpu == 'loongarch64' + util_ss.add(files('cpuinfo-loongarch.c')) elif cpu in ['ppc', 'ppc64'] util_ss.add(files('cpuinfo-ppc.c')) endif diff --git a/util/range.c b/util/range.c index 098d9d2dc0..9605ccfcbe 100644 --- a/util/range.c +++ b/util/range.c @@ -20,11 +20,7 @@ #include "qemu/osdep.h" #include "qemu/range.h" -/* - * Return -1 if @a < @b, 1 @a > @b, and 0 if they touch or overlap. - * Both @a and @b must not be empty. - */ -static inline int range_compare(Range *a, Range *b) +int range_compare(Range *a, Range *b) { assert(!range_is_empty(a) && !range_is_empty(b)); @@ -70,3 +66,58 @@ GList *range_list_insert(GList *list, Range *data) return list; } + +static inline +GList *append_new_range(GList *list, uint64_t lob, uint64_t upb) +{ + Range *new = g_new0(Range, 1); + + range_set_bounds(new, lob, upb); + return g_list_append(list, new); +} + + +void range_inverse_array(GList *in, GList **rev, + uint64_t low, uint64_t high) +{ + Range *r, *rn; + GList *l = in, *out = *rev; + + for (l = in; l && range_upb(l->data) < low; l = l->next) { + continue; + } + + if (!l) { + out = append_new_range(out, low, high); + goto exit; + } + r = (Range *)l->data; + + /* first range lob is greater than min, insert a first range */ + if (range_lob(r) > low) { + out = append_new_range(out, low, MIN(range_lob(r) - 1, high)); + } + + /* insert a range inbetween each original range until we reach high */ + for (; l->next; l = l->next) { + r = (Range *)l->data; + rn = (Range *)l->next->data; + if (range_lob(r) >= high) { + goto exit; + } + if (range_compare(r, rn)) { + out = append_new_range(out, range_upb(r) + 1, + MIN(range_lob(rn) - 1, high)); + } + } + + /* last range */ + r = (Range *)l->data; + + /* last range upb is less than max, insert a last range */ + if (range_upb(r) < high) { + out = append_new_range(out, range_upb(r) + 1, high); + } +exit: + *rev = out; +} diff --git a/util/reserved-region.c b/util/reserved-region.c new file mode 100644 index 0000000000..18f83eb4c6 --- /dev/null +++ b/util/reserved-region.c @@ -0,0 +1,91 @@ +/* + * QEMU ReservedRegion helpers + * + * Copyright (c) 2023 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/range.h" +#include "qemu/reserved-region.h" + +GList *resv_region_list_insert(GList *list, ReservedRegion *reg) +{ + ReservedRegion *resv_iter, *new_reg; + Range *r = ®->range; + Range *range_iter; + GList *l; + + for (l = list; l ; ) { + resv_iter = (ReservedRegion *)l->data; + range_iter = &resv_iter->range; + + /* Skip all list elements strictly less than range to add */ + if (range_compare(range_iter, r) < 0) { + l = l->next; + } else if (range_compare(range_iter, r) > 0) { + return g_list_insert_before(list, l, reg); + } else { /* there is an overlap */ + if (range_contains_range(r, range_iter)) { + /* new range contains current item, simply remove this latter */ + GList *prev = l->prev; + g_free(l->data); + list = g_list_delete_link(list, l); + if (prev) { + l = prev->next; + } else { + l = list; + } + } else if (range_contains_range(range_iter, r)) { + /* new region is included in the current region */ + if (range_lob(range_iter) == range_lob(r)) { + /* adjacent on the left side, derives into 2 regions */ + range_set_bounds(range_iter, range_upb(r) + 1, + range_upb(range_iter)); + return g_list_insert_before(list, l, reg); + } else if (range_upb(range_iter) == range_upb(r)) { + /* adjacent on the right side, derives into 2 regions */ + range_set_bounds(range_iter, range_lob(range_iter), + range_lob(r) - 1); + l = l->next; + } else { + uint64_t lob = range_lob(range_iter); + /* + * the new range is in the middle of an existing one, + * split this latter into 3 regs instead + */ + range_set_bounds(range_iter, range_upb(r) + 1, + range_upb(range_iter)); + new_reg = g_new0(ReservedRegion, 1); + new_reg->type = resv_iter->type; + range_set_bounds(&new_reg->range, + lob, range_lob(r) - 1); + list = g_list_insert_before(list, l, new_reg); + return g_list_insert_before(list, l, reg); + } + } else if (range_lob(r) < range_lob(range_iter)) { + range_set_bounds(range_iter, range_upb(r) + 1, + range_upb(range_iter)); + return g_list_insert_before(list, l, reg); + } else { /* intersection on the upper range */ + range_set_bounds(range_iter, range_lob(range_iter), + range_lob(r) - 1); + l = l->next; + } + } /* overlap */ + } + return g_list_append(list, reg); +} + diff --git a/util/uuid.c b/util/uuid.c index d71aa79e5e..234619dd5e 100644 --- a/util/uuid.c +++ b/util/uuid.c @@ -51,7 +51,7 @@ int qemu_uuid_is_equal(const QemuUUID *lhv, const QemuUUID *rhv) void qemu_uuid_unparse(const QemuUUID *uuid, char *out) { const unsigned char *uu = &uuid->data[0]; - snprintf(out, UUID_FMT_LEN + 1, UUID_FMT, + snprintf(out, UUID_STR_LEN, UUID_FMT, uu[0], uu[1], uu[2], uu[3], uu[4], uu[5], uu[6], uu[7], uu[8], uu[9], uu[10], uu[11], uu[12], uu[13], uu[14], uu[15]); } |