diff options
Diffstat (limited to 'target/arm/sve_helper.c')
| -rw-r--r-- | target/arm/sve_helper.c | 232 |
1 files changed, 18 insertions, 214 deletions
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index 3bdcd4ce9d..1654c0bbf9 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -21,12 +21,12 @@ #include "cpu.h" #include "internals.h" #include "exec/exec-all.h" -#include "exec/cpu_ldst.h" #include "exec/helper-proto.h" #include "tcg/tcg-gvec-desc.h" #include "fpu/softfloat.h" #include "tcg/tcg.h" #include "vec_internal.h" +#include "sve_ldst_internal.h" /* Return a value for NZCV as per the ARM PredTest pseudofunction. @@ -103,44 +103,6 @@ uint32_t HELPER(sve_predtest)(void *vd, void *vg, uint32_t words) return flags; } -/* - * Expand active predicate bits to bytes, for byte elements. - * (The data table itself is in vec_helper.c as MVE also needs it.) - */ -static inline uint64_t expand_pred_b(uint8_t byte) -{ - return expand_pred_b_data[byte]; -} - -/* Similarly for half-word elements. - * for (i = 0; i < 256; ++i) { - * unsigned long m = 0; - * if (i & 0xaa) { - * continue; - * } - * for (j = 0; j < 8; j += 2) { - * if ((i >> j) & 1) { - * m |= 0xfffful << (j << 3); - * } - * } - * printf("[0x%x] = 0x%016lx,\n", i, m); - * } - */ -static inline uint64_t expand_pred_h(uint8_t byte) -{ - static const uint64_t word[] = { - [0x01] = 0x000000000000ffff, [0x04] = 0x00000000ffff0000, - [0x05] = 0x00000000ffffffff, [0x10] = 0x0000ffff00000000, - [0x11] = 0x0000ffff0000ffff, [0x14] = 0x0000ffffffff0000, - [0x15] = 0x0000ffffffffffff, [0x40] = 0xffff000000000000, - [0x41] = 0xffff00000000ffff, [0x44] = 0xffff0000ffff0000, - [0x45] = 0xffff0000ffffffff, [0x50] = 0xffffffff00000000, - [0x51] = 0xffffffff0000ffff, [0x54] = 0xffffffffffff0000, - [0x55] = 0xffffffffffffffff, - }; - return word[byte & 0x55]; -} - /* Similarly for single word elements. */ static inline uint64_t expand_pred_s(uint8_t byte) { @@ -5302,111 +5264,6 @@ void HELPER(sve_fcmla_zpzzz_d)(void *vd, void *vn, void *vm, void *va, */ /* - * Load one element into @vd + @reg_off from @host. - * The controlling predicate is known to be true. - */ -typedef void sve_ldst1_host_fn(void *vd, intptr_t reg_off, void *host); - -/* - * Load one element into @vd + @reg_off from (@env, @vaddr, @ra). - * The controlling predicate is known to be true. - */ -typedef void sve_ldst1_tlb_fn(CPUARMState *env, void *vd, intptr_t reg_off, - target_ulong vaddr, uintptr_t retaddr); - -/* - * Generate the above primitives. - */ - -#define DO_LD_HOST(NAME, H, TYPEE, TYPEM, HOST) \ -static void sve_##NAME##_host(void *vd, intptr_t reg_off, void *host) \ -{ \ - TYPEM val = HOST(host); \ - *(TYPEE *)(vd + H(reg_off)) = val; \ -} - -#define DO_ST_HOST(NAME, H, TYPEE, TYPEM, HOST) \ -static void sve_##NAME##_host(void *vd, intptr_t reg_off, void *host) \ -{ HOST(host, (TYPEM)*(TYPEE *)(vd + H(reg_off))); } - -#define DO_LD_TLB(NAME, H, TYPEE, TYPEM, TLB) \ -static void sve_##NAME##_tlb(CPUARMState *env, void *vd, intptr_t reg_off, \ - target_ulong addr, uintptr_t ra) \ -{ \ - *(TYPEE *)(vd + H(reg_off)) = \ - (TYPEM)TLB(env, useronly_clean_ptr(addr), ra); \ -} - -#define DO_ST_TLB(NAME, H, TYPEE, TYPEM, TLB) \ -static void sve_##NAME##_tlb(CPUARMState *env, void *vd, intptr_t reg_off, \ - target_ulong addr, uintptr_t ra) \ -{ \ - TLB(env, useronly_clean_ptr(addr), \ - (TYPEM)*(TYPEE *)(vd + H(reg_off)), ra); \ -} - -#define DO_LD_PRIM_1(NAME, H, TE, TM) \ - DO_LD_HOST(NAME, H, TE, TM, ldub_p) \ - DO_LD_TLB(NAME, H, TE, TM, cpu_ldub_data_ra) - -DO_LD_PRIM_1(ld1bb, H1, uint8_t, uint8_t) -DO_LD_PRIM_1(ld1bhu, H1_2, uint16_t, uint8_t) -DO_LD_PRIM_1(ld1bhs, H1_2, uint16_t, int8_t) -DO_LD_PRIM_1(ld1bsu, H1_4, uint32_t, uint8_t) -DO_LD_PRIM_1(ld1bss, H1_4, uint32_t, int8_t) -DO_LD_PRIM_1(ld1bdu, H1_8, uint64_t, uint8_t) -DO_LD_PRIM_1(ld1bds, H1_8, uint64_t, int8_t) - -#define DO_ST_PRIM_1(NAME, H, TE, TM) \ - DO_ST_HOST(st1##NAME, H, TE, TM, stb_p) \ - DO_ST_TLB(st1##NAME, H, TE, TM, cpu_stb_data_ra) - -DO_ST_PRIM_1(bb, H1, uint8_t, uint8_t) -DO_ST_PRIM_1(bh, H1_2, uint16_t, uint8_t) -DO_ST_PRIM_1(bs, H1_4, uint32_t, uint8_t) -DO_ST_PRIM_1(bd, H1_8, uint64_t, uint8_t) - -#define DO_LD_PRIM_2(NAME, H, TE, TM, LD) \ - DO_LD_HOST(ld1##NAME##_be, H, TE, TM, LD##_be_p) \ - DO_LD_HOST(ld1##NAME##_le, H, TE, TM, LD##_le_p) \ - DO_LD_TLB(ld1##NAME##_be, H, TE, TM, cpu_##LD##_be_data_ra) \ - DO_LD_TLB(ld1##NAME##_le, H, TE, TM, cpu_##LD##_le_data_ra) - -#define DO_ST_PRIM_2(NAME, H, TE, TM, ST) \ - DO_ST_HOST(st1##NAME##_be, H, TE, TM, ST##_be_p) \ - DO_ST_HOST(st1##NAME##_le, H, TE, TM, ST##_le_p) \ - DO_ST_TLB(st1##NAME##_be, H, TE, TM, cpu_##ST##_be_data_ra) \ - DO_ST_TLB(st1##NAME##_le, H, TE, TM, cpu_##ST##_le_data_ra) - -DO_LD_PRIM_2(hh, H1_2, uint16_t, uint16_t, lduw) -DO_LD_PRIM_2(hsu, H1_4, uint32_t, uint16_t, lduw) -DO_LD_PRIM_2(hss, H1_4, uint32_t, int16_t, lduw) -DO_LD_PRIM_2(hdu, H1_8, uint64_t, uint16_t, lduw) -DO_LD_PRIM_2(hds, H1_8, uint64_t, int16_t, lduw) - -DO_ST_PRIM_2(hh, H1_2, uint16_t, uint16_t, stw) -DO_ST_PRIM_2(hs, H1_4, uint32_t, uint16_t, stw) -DO_ST_PRIM_2(hd, H1_8, uint64_t, uint16_t, stw) - -DO_LD_PRIM_2(ss, H1_4, uint32_t, uint32_t, ldl) -DO_LD_PRIM_2(sdu, H1_8, uint64_t, uint32_t, ldl) -DO_LD_PRIM_2(sds, H1_8, uint64_t, int32_t, ldl) - -DO_ST_PRIM_2(ss, H1_4, uint32_t, uint32_t, stl) -DO_ST_PRIM_2(sd, H1_8, uint64_t, uint32_t, stl) - -DO_LD_PRIM_2(dd, H1_8, uint64_t, uint64_t, ldq) -DO_ST_PRIM_2(dd, H1_8, uint64_t, uint64_t, stq) - -#undef DO_LD_TLB -#undef DO_ST_TLB -#undef DO_LD_HOST -#undef DO_LD_PRIM_1 -#undef DO_ST_PRIM_1 -#undef DO_LD_PRIM_2 -#undef DO_ST_PRIM_2 - -/* * Skip through a sequence of inactive elements in the guarding predicate @vg, * beginning at @reg_off bounded by @reg_max. Return the offset of the active * element >= @reg_off, or @reg_max if there were no active elements at all. @@ -5446,16 +5303,9 @@ static intptr_t find_next_active(uint64_t *vg, intptr_t reg_off, * exit via page fault exception. */ -typedef struct { - void *host; - int flags; - MemTxAttrs attrs; -} SVEHostPage; - -static bool sve_probe_page(SVEHostPage *info, bool nofault, - CPUARMState *env, target_ulong addr, - int mem_off, MMUAccessType access_type, - int mmu_idx, uintptr_t retaddr) +bool sve_probe_page(SVEHostPage *info, bool nofault, CPUARMState *env, + target_ulong addr, int mem_off, MMUAccessType access_type, + int mmu_idx, uintptr_t retaddr) { int flags; @@ -5511,59 +5361,13 @@ static bool sve_probe_page(SVEHostPage *info, bool nofault, return true; } - -/* - * Analyse contiguous data, protected by a governing predicate. - */ - -typedef enum { - FAULT_NO, - FAULT_FIRST, - FAULT_ALL, -} SVEContFault; - -typedef struct { - /* - * First and last element wholly contained within the two pages. - * mem_off_first[0] and reg_off_first[0] are always set >= 0. - * reg_off_last[0] may be < 0 if the first element crosses pages. - * All of mem_off_first[1], reg_off_first[1] and reg_off_last[1] - * are set >= 0 only if there are complete elements on a second page. - * - * The reg_off_* offsets are relative to the internal vector register. - * The mem_off_first offset is relative to the memory address; the - * two offsets are different when a load operation extends, a store - * operation truncates, or for multi-register operations. - */ - int16_t mem_off_first[2]; - int16_t reg_off_first[2]; - int16_t reg_off_last[2]; - - /* - * One element that is misaligned and spans both pages, - * or -1 if there is no such active element. - */ - int16_t mem_off_split; - int16_t reg_off_split; - - /* - * The byte offset at which the entire operation crosses a page boundary. - * Set >= 0 if and only if the entire operation spans two pages. - */ - int16_t page_split; - - /* TLB data for the two pages. */ - SVEHostPage page[2]; -} SVEContLdSt; - /* * Find first active element on each page, and a loose bound for the * final element on each page. Identify any single element that spans * the page boundary. Return true if there are any active elements. */ -static bool sve_cont_ldst_elements(SVEContLdSt *info, target_ulong addr, - uint64_t *vg, intptr_t reg_max, - int esz, int msize) +bool sve_cont_ldst_elements(SVEContLdSt *info, target_ulong addr, uint64_t *vg, + intptr_t reg_max, int esz, int msize) { const int esize = 1 << esz; const uint64_t pg_mask = pred_esz_masks[esz]; @@ -5653,9 +5457,9 @@ static bool sve_cont_ldst_elements(SVEContLdSt *info, target_ulong addr, * Control the generation of page faults with @fault. Return false if * there is no work to do, which can only happen with @fault == FAULT_NO. */ -static bool sve_cont_ldst_pages(SVEContLdSt *info, SVEContFault fault, - CPUARMState *env, target_ulong addr, - MMUAccessType access_type, uintptr_t retaddr) +bool sve_cont_ldst_pages(SVEContLdSt *info, SVEContFault fault, + CPUARMState *env, target_ulong addr, + MMUAccessType access_type, uintptr_t retaddr) { int mmu_idx = cpu_mmu_index(env, false); int mem_off = info->mem_off_first[0]; @@ -5711,12 +5515,12 @@ static bool sve_cont_ldst_pages(SVEContLdSt *info, SVEContFault fault, return have_work; } -static void sve_cont_ldst_watchpoints(SVEContLdSt *info, CPUARMState *env, - uint64_t *vg, target_ulong addr, - int esize, int msize, int wp_access, - uintptr_t retaddr) -{ #ifndef CONFIG_USER_ONLY +void sve_cont_ldst_watchpoints(SVEContLdSt *info, CPUARMState *env, + uint64_t *vg, target_ulong addr, + int esize, int msize, int wp_access, + uintptr_t retaddr) +{ intptr_t mem_off, reg_off, reg_last; int flags0 = info->page[0].flags; int flags1 = info->page[1].flags; @@ -5772,12 +5576,12 @@ static void sve_cont_ldst_watchpoints(SVEContLdSt *info, CPUARMState *env, } while (reg_off & 63); } while (reg_off <= reg_last); } -#endif } +#endif -static void sve_cont_ldst_mte_check(SVEContLdSt *info, CPUARMState *env, - uint64_t *vg, target_ulong addr, int esize, - int msize, uint32_t mtedesc, uintptr_t ra) +void sve_cont_ldst_mte_check(SVEContLdSt *info, CPUARMState *env, + uint64_t *vg, target_ulong addr, int esize, + int msize, uint32_t mtedesc, uintptr_t ra) { intptr_t mem_off, reg_off, reg_last; |