diff options
Diffstat (limited to 'linux-user')
35 files changed, 152 insertions, 25 deletions
diff --git a/linux-user/aarch64/vdso-be.so b/linux-user/aarch64/vdso-be.so index d43c3b19cd..4089838b30 100755 --- a/linux-user/aarch64/vdso-be.so +++ b/linux-user/aarch64/vdso-be.so Binary files differdiff --git a/linux-user/aarch64/vdso-le.so b/linux-user/aarch64/vdso-le.so index aaedc9d85e..240802821c 100755 --- a/linux-user/aarch64/vdso-le.so +++ b/linux-user/aarch64/vdso-le.so Binary files differdiff --git a/linux-user/aarch64/vdso.S b/linux-user/aarch64/vdso.S index a0ac1487b0..59dd94dc8f 100644 --- a/linux-user/aarch64/vdso.S +++ b/linux-user/aarch64/vdso.S @@ -71,5 +71,7 @@ vdso_syscall __kernel_clock_getres, __NR_clock_getres __kernel_rt_sigreturn: /* No BTI C insn here -- we arrive via RET. */ mov x8, #__NR_rt_sigreturn +sigreturn_region_start: svc #0 +sigreturn_region_end: endf __kernel_rt_sigreturn diff --git a/linux-user/alpha/cpu_loop.c b/linux-user/alpha/cpu_loop.c index bb8346b509..f93597c400 100644 --- a/linux-user/alpha/cpu_loop.c +++ b/linux-user/alpha/cpu_loop.c @@ -94,11 +94,6 @@ void cpu_loop(CPUAlphaState *env) break; case 0x86: /* IMB */ - /* ??? We can probably elide the code using page_unprotect - that is checking for self-modifying code. Instead we - could simply call tb_flush here. Until we work out the - changes required to turn off the extra write protection, - this can be a no-op. */ break; case 0x9E: /* RDUNIQUE */ diff --git a/linux-user/arm/vdso-be32.so b/linux-user/arm/vdso-be32.so index b896d3d545..6d71cd9c36 100755 --- a/linux-user/arm/vdso-be32.so +++ b/linux-user/arm/vdso-be32.so Binary files differdiff --git a/linux-user/arm/vdso-be8.so b/linux-user/arm/vdso-be8.so index 784b7bdb2a..6446a96fcf 100755 --- a/linux-user/arm/vdso-be8.so +++ b/linux-user/arm/vdso-be8.so Binary files differdiff --git a/linux-user/arm/vdso-le.so b/linux-user/arm/vdso-le.so index 38d3d51047..d34e577b69 100755 --- a/linux-user/arm/vdso-le.so +++ b/linux-user/arm/vdso-le.so Binary files differdiff --git a/linux-user/arm/vdso.S b/linux-user/arm/vdso.S index b3bb6491dc..d84d964730 100644 --- a/linux-user/arm/vdso.S +++ b/linux-user/arm/vdso.S @@ -140,6 +140,7 @@ SYSCALL __vdso_gettimeofday, __NR_gettimeofday .balign 16 sigreturn_codes: +sigreturn_region_start: /* [EO]ABI sigreturn */ slot 0 raw_syscall __NR_sigreturn @@ -172,3 +173,4 @@ sigreturn_codes: .balign 16 endf sigreturn_codes +sigreturn_region_end: diff --git a/linux-user/elfload.c b/linux-user/elfload.c index 26c090c95d..1370ec59be 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -1659,6 +1659,11 @@ static void load_elf_vdso(struct image_info *info, const VdsoImageInfo *vdso) if (vdso->rt_sigreturn_ofs) { default_rt_sigreturn = load_addr + vdso->rt_sigreturn_ofs; } + if (vdso->sigreturn_region_start_ofs) { + vdso_sigreturn_region_start = + load_addr + vdso->sigreturn_region_start_ofs; + vdso_sigreturn_region_end = load_addr + vdso->sigreturn_region_end_ofs; + } /* Remove write from VDSO segment. */ target_mprotect(info->start_data, info->end_data - info->start_data, @@ -1969,6 +1974,8 @@ int load_elf_binary(struct linux_binprm *bprm, struct image_info *info) setup_sigtramp(tramp_page); target_mprotect(tramp_page, TARGET_PAGE_SIZE, PROT_READ | PROT_EXEC); + vdso_sigreturn_region_start = tramp_page; + vdso_sigreturn_region_end = tramp_page + TARGET_PAGE_SIZE; } bprm->p = create_elf_tables(bprm->p, bprm->argc, bprm->envc, &ehdr, info, diff --git a/linux-user/gen-vdso-elfn.c.inc b/linux-user/gen-vdso-elfn.c.inc index b47019e136..c2677a146c 100644 --- a/linux-user/gen-vdso-elfn.c.inc +++ b/linux-user/gen-vdso-elfn.c.inc @@ -84,9 +84,12 @@ static void elfN(search_symtab)(ElfN(Shdr) *shdr, unsigned sym_idx, if (sigreturn_sym && strcmp(sigreturn_sym, name) == 0) { sigreturn_addr = sym.st_value; - } - if (rt_sigreturn_sym && strcmp(rt_sigreturn_sym, name) == 0) { + } else if (rt_sigreturn_sym && strcmp(rt_sigreturn_sym, name) == 0) { rt_sigreturn_addr = sym.st_value; + } else if (strcmp("sigreturn_region_start", name) == 0) { + sigreturn_region_start_addr = sym.st_value; + } else if (strcmp("sigreturn_region_end", name) == 0) { + sigreturn_region_end_addr = sym.st_value; } } } diff --git a/linux-user/gen-vdso.c b/linux-user/gen-vdso.c index aeaa927db8..d6a2cdaa83 100644 --- a/linux-user/gen-vdso.c +++ b/linux-user/gen-vdso.c @@ -36,6 +36,8 @@ static const char *rt_sigreturn_sym; static unsigned sigreturn_addr; static unsigned rt_sigreturn_addr; +static unsigned sigreturn_region_start_addr; +static unsigned sigreturn_region_end_addr; #define N 32 #define elfN(x) elf32_##x @@ -215,6 +217,10 @@ int main(int argc, char **argv) fprintf(outf, " .reloc_count = ARRAY_SIZE(%s_relocs),\n", prefix); fprintf(outf, " .sigreturn_ofs = 0x%x,\n", sigreturn_addr); fprintf(outf, " .rt_sigreturn_ofs = 0x%x,\n", rt_sigreturn_addr); + fprintf(outf, " .sigreturn_region_start_ofs = 0x%x,\n", + sigreturn_region_start_addr); + fprintf(outf, " .sigreturn_region_end_ofs = 0x%x,\n", + sigreturn_region_end_addr); fprintf(outf, "};\n"); ret = EXIT_SUCCESS; diff --git a/linux-user/hppa/vdso.S b/linux-user/hppa/vdso.S index 5be14d2f70..a6f8da2981 100644 --- a/linux-user/hppa/vdso.S +++ b/linux-user/hppa/vdso.S @@ -156,8 +156,10 @@ __kernel_sigtramp_rt: ldi 0, %r25 ldi __NR_rt_sigreturn, %r20 +sigreturn_region_start: be,l 0x100(%sr2, %r0), %sr0, %r31 nop +sigreturn_region_end: .cfi_endproc .size __kernel_sigtramp_rt, . - __kernel_sigtramp_rt diff --git a/linux-user/hppa/vdso.so b/linux-user/hppa/vdso.so index e1ddd70c37..68baf80fd3 100755 --- a/linux-user/hppa/vdso.so +++ b/linux-user/hppa/vdso.so Binary files differdiff --git a/linux-user/i386/vdso.S b/linux-user/i386/vdso.S index e7a1f333a1..8df77b5a94 100644 --- a/linux-user/i386/vdso.S +++ b/linux-user/i386/vdso.S @@ -114,6 +114,7 @@ vdso_syscall3 __vdso_getcpu, __NR_gettimeofday */ nop +sigreturn_region_start: __kernel_sigreturn: popl %eax /* pop sig */ .cfi_adjust_cfa_offset -4 @@ -128,6 +129,7 @@ __kernel_rt_sigreturn: movl $__NR_rt_sigreturn, %eax int $0x80 endf __kernel_rt_sigreturn +sigreturn_region_end: .cfi_endproc diff --git a/linux-user/i386/vdso.so b/linux-user/i386/vdso.so index bdece5dfcf..e01c3818d0 100755 --- a/linux-user/i386/vdso.so +++ b/linux-user/i386/vdso.so Binary files differdiff --git a/linux-user/loader.h b/linux-user/loader.h index e42b8fa1e3..da9ad28db5 100644 --- a/linux-user/loader.h +++ b/linux-user/loader.h @@ -117,6 +117,8 @@ typedef struct { unsigned reloc_count; unsigned sigreturn_ofs; unsigned rt_sigreturn_ofs; + unsigned sigreturn_region_start_ofs; + unsigned sigreturn_region_end_ofs; } VdsoImageInfo; /* Note that both Elf32_Word and Elf64_Word are uint32_t. */ diff --git a/linux-user/loongarch64/vdso.S b/linux-user/loongarch64/vdso.S index 780a5fda12..2409d95476 100644 --- a/linux-user/loongarch64/vdso.S +++ b/linux-user/loongarch64/vdso.S @@ -125,6 +125,8 @@ vdso_syscall __vdso_getcpu, __NR_getcpu __vdso_rt_sigreturn: li.w $a7, __NR_rt_sigreturn +sigreturn_region_start: syscall 0 +sigreturn_region_end: .cfi_endproc endf __vdso_rt_sigreturn diff --git a/linux-user/loongarch64/vdso.so b/linux-user/loongarch64/vdso.so index 7c2de6c50e..3704834f0d 100755 --- a/linux-user/loongarch64/vdso.so +++ b/linux-user/loongarch64/vdso.so Binary files differdiff --git a/linux-user/main.c b/linux-user/main.c index 4ddfc9a619..db751c0757 100644 --- a/linux-user/main.c +++ b/linux-user/main.c @@ -228,6 +228,8 @@ void init_task_state(TaskState *ts) ts->start_boottime += bt.tv_nsec * (uint64_t) ticks_per_sec / NANOSECONDS_PER_SECOND; } + + ts->sys_dispatch_len = -1; } CPUArchState *cpu_copy(CPUArchState *env) diff --git a/linux-user/mmap.c b/linux-user/mmap.c index 002e1e668e..847092a28a 100644 --- a/linux-user/mmap.c +++ b/linux-user/mmap.c @@ -22,8 +22,6 @@ #include "exec/log.h" #include "exec/page-protection.h" #include "exec/mmap-lock.h" -#include "exec/tb-flush.h" -#include "exec/translation-block.h" #include "qemu.h" #include "user/page-protection.h" #include "user-internals.h" @@ -1007,11 +1005,7 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, int target_prot, * be atomic with respect to an external process. */ if (ret != -1 && (flags & MAP_TYPE) != MAP_PRIVATE) { - CPUState *cpu = thread_cpu; - if (!tcg_cflags_has(cpu, CF_PARALLEL)) { - tcg_cflags_set(cpu, CF_PARALLEL); - tb_flush(cpu); - } + begin_parallel_context(thread_cpu); } return ret; @@ -1448,10 +1442,7 @@ abi_ulong target_shmat(CPUArchState *cpu_env, int shmid, * supported by the host -- anything that requires EXCP_ATOMIC will not * be atomic with respect to an external process. */ - if (!tcg_cflags_has(cpu, CF_PARALLEL)) { - tcg_cflags_set(cpu, CF_PARALLEL); - tb_flush(cpu); - } + begin_parallel_context(cpu); if (qemu_loglevel_mask(CPU_LOG_PAGE)) { FILE *f = qemu_log_trylock(); diff --git a/linux-user/ppc/vdso-32.so b/linux-user/ppc/vdso-32.so index 0dc55e0ddd..03476052fb 100755 --- a/linux-user/ppc/vdso-32.so +++ b/linux-user/ppc/vdso-32.so Binary files differdiff --git a/linux-user/ppc/vdso-64.so b/linux-user/ppc/vdso-64.so index ac1ab2582e..b89f2a0e01 100755 --- a/linux-user/ppc/vdso-64.so +++ b/linux-user/ppc/vdso-64.so Binary files differdiff --git a/linux-user/ppc/vdso-64le.so b/linux-user/ppc/vdso-64le.so index 424abb4290..22499d2701 100755 --- a/linux-user/ppc/vdso-64le.so +++ b/linux-user/ppc/vdso-64le.so Binary files differdiff --git a/linux-user/ppc/vdso.S b/linux-user/ppc/vdso.S index 2e79ea9808..e9256a2dea 100644 --- a/linux-user/ppc/vdso.S +++ b/linux-user/ppc/vdso.S @@ -220,6 +220,7 @@ endf __kernel_sync_dicache nop +sigreturn_region_start: __kernel_sigtramp_rt: raw_syscall __NR_rt_sigreturn endf __kernel_sigtramp_rt @@ -235,5 +236,6 @@ __kernel_sigtramp32: raw_syscall __NR_sigreturn endf __kernel_sigtramp32 #endif +sigreturn_region_end: .cfi_endproc diff --git a/linux-user/qemu.h b/linux-user/qemu.h index e4dca0c20f..cabb7bd6a8 100644 --- a/linux-user/qemu.h +++ b/linux-user/qemu.h @@ -155,6 +155,11 @@ struct TaskState { /* This thread's sigaltstack, if it has one */ struct target_sigaltstack sigaltstack_used; + /* This thread's SYSCALL_USER_DISPATCH state, len=~0 means disabled */ + vaddr sys_dispatch; + vaddr sys_dispatch_selector; + abi_ulong sys_dispatch_len; + /* Start time of task after system boot in clock ticks */ uint64_t start_boottime; }; diff --git a/linux-user/riscv/vdso-32.so b/linux-user/riscv/vdso-32.so index c2ce2a4757..4818a994f0 100755 --- a/linux-user/riscv/vdso-32.so +++ b/linux-user/riscv/vdso-32.so Binary files differdiff --git a/linux-user/riscv/vdso-64.so b/linux-user/riscv/vdso-64.so index ae49f5b043..cc6f7e974b 100755 --- a/linux-user/riscv/vdso-64.so +++ b/linux-user/riscv/vdso-64.so Binary files differdiff --git a/linux-user/riscv/vdso.S b/linux-user/riscv/vdso.S index c37275233a..1d780db771 100644 --- a/linux-user/riscv/vdso.S +++ b/linux-user/riscv/vdso.S @@ -181,7 +181,9 @@ endf __vdso_flush_icache nop __vdso_rt_sigreturn: +sigreturn_region_start: raw_syscall __NR_rt_sigreturn +sigreturn_region_end: endf __vdso_rt_sigreturn .cfi_endproc diff --git a/linux-user/s390x/vdso.S b/linux-user/s390x/vdso.S index 3332492477..c60e9ed086 100644 --- a/linux-user/s390x/vdso.S +++ b/linux-user/s390x/vdso.S @@ -52,6 +52,7 @@ vdso_syscall __kernel_getcpu, __NR_getcpu * by all users. Without it we get the fallback signal frame handling. */ +sigreturn_region_start: __kernel_sigreturn: raw_syscall __NR_sigreturn endf __kernel_sigreturn @@ -59,3 +60,4 @@ endf __kernel_sigreturn __kernel_rt_sigreturn: raw_syscall __NR_rt_sigreturn endf __kernel_rt_sigreturn +sigreturn_region_end: diff --git a/linux-user/s390x/vdso.so b/linux-user/s390x/vdso.so index 64130f6f33..a669a6b7dd 100755 --- a/linux-user/s390x/vdso.so +++ b/linux-user/s390x/vdso.so Binary files differdiff --git a/linux-user/signal-common.h b/linux-user/signal-common.h index 196d2406f8..8a44714251 100644 --- a/linux-user/signal-common.h +++ b/linux-user/signal-common.h @@ -25,6 +25,13 @@ /* Fallback addresses into sigtramp page. */ extern abi_ulong default_sigreturn; extern abi_ulong default_rt_sigreturn; +extern abi_ulong vdso_sigreturn_region_start; +extern abi_ulong vdso_sigreturn_region_end; + +static inline bool is_vdso_sigreturn(abi_ulong pc) +{ + return pc >= vdso_sigreturn_region_start && pc < vdso_sigreturn_region_end; +} void setup_sigtramp(abi_ulong tramp_page); diff --git a/linux-user/signal.c b/linux-user/signal.c index cd0e7398aa..804096bd44 100644 --- a/linux-user/signal.c +++ b/linux-user/signal.c @@ -50,6 +50,8 @@ static void host_signal_handler(int host_signum, siginfo_t *info, /* Fallback addresses into sigtramp page. */ abi_ulong default_sigreturn; abi_ulong default_rt_sigreturn; +abi_ulong vdso_sigreturn_region_start; +abi_ulong vdso_sigreturn_region_end; /* * System includes define _NSIG as SIGRTMAX + 1, but qemu (like the kernel) diff --git a/linux-user/syscall.c b/linux-user/syscall.c index 91360a072c..1a5f2a03f9 100644 --- a/linux-user/syscall.c +++ b/linux-user/syscall.c @@ -27,8 +27,6 @@ #include "target_mman.h" #include "exec/page-protection.h" #include "exec/mmap-lock.h" -#include "exec/tb-flush.h" -#include "exec/translation-block.h" #include <elf.h> #include <endian.h> #include <grp.h> @@ -6344,6 +6342,10 @@ abi_long do_arch_prctl(CPUX86State *env, int code, abi_ulong addr) #endif #ifndef PR_SET_SYSCALL_USER_DISPATCH # define PR_SET_SYSCALL_USER_DISPATCH 59 +# define PR_SYS_DISPATCH_OFF 0 +# define PR_SYS_DISPATCH_ON 1 +# define SYSCALL_DISPATCH_FILTER_ALLOW 0 +# define SYSCALL_DISPATCH_FILTER_BLOCK 1 #endif #ifndef PR_SME_SET_VL # define PR_SME_SET_VL 63 @@ -6398,6 +6400,36 @@ static abi_long do_prctl_inval1(CPUArchState *env, abi_long arg2) #define do_prctl_sme_set_vl do_prctl_inval1 #endif +static abi_long do_prctl_syscall_user_dispatch(CPUArchState *env, + abi_ulong arg2, abi_ulong arg3, + abi_ulong arg4, abi_ulong arg5) +{ + CPUState *cpu = env_cpu(env); + TaskState *ts = get_task_state(cpu); + + switch (arg2) { + case PR_SYS_DISPATCH_OFF: + if (arg3 || arg4 || arg5) { + return -TARGET_EINVAL; + } + ts->sys_dispatch_len = -1; + return 0; + case PR_SYS_DISPATCH_ON: + if (arg3 && arg3 + arg4 <= arg3) { + return -TARGET_EINVAL; + } + if (arg5 && !access_ok(cpu, VERIFY_READ, arg5, 1)) { + return -TARGET_EFAULT; + } + ts->sys_dispatch = arg3; + ts->sys_dispatch_len = arg4; + ts->sys_dispatch_selector = arg5; + return 0; + default: + return -TARGET_EINVAL; + } +} + static abi_long do_prctl(CPUArchState *env, abi_long option, abi_long arg2, abi_long arg3, abi_long arg4, abi_long arg5) { @@ -6473,6 +6505,9 @@ static abi_long do_prctl(CPUArchState *env, abi_long option, abi_long arg2, case PR_SET_UNALIGN: return do_prctl_set_unalign(env, arg2); + case PR_SET_SYSCALL_USER_DISPATCH: + return do_prctl_syscall_user_dispatch(env, arg2, arg3, arg4, arg5); + case PR_CAP_AMBIENT: case PR_CAPBSET_READ: case PR_CAPBSET_DROP: @@ -6527,7 +6562,6 @@ static abi_long do_prctl(CPUArchState *env, abi_long option, abi_long arg2, case PR_SET_MM: case PR_GET_SECCOMP: case PR_SET_SECCOMP: - case PR_SET_SYSCALL_USER_DISPATCH: case PR_GET_THP_DISABLE: case PR_SET_THP_DISABLE: case PR_GET_TSC: @@ -6631,10 +6665,7 @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp, * generate code for parallel execution and flush old translations. * Do this now so that the copy gets CF_PARALLEL too. */ - if (!tcg_cflags_has(cpu, CF_PARALLEL)) { - tcg_cflags_set(cpu, CF_PARALLEL); - tb_flush(cpu); - } + begin_parallel_context(cpu); /* we create a new CPU instance. */ new_env = cpu_copy(env); @@ -13897,12 +13928,46 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1, return ret; } +static bool sys_dispatch(CPUState *cpu, TaskState *ts) +{ + abi_ptr pc; + + if (likely(ts->sys_dispatch_len == -1)) { + return false; + } + + pc = cpu->cc->get_pc(cpu); + if (likely(pc - ts->sys_dispatch < ts->sys_dispatch_len)) { + return false; + } + if (unlikely(is_vdso_sigreturn(pc))) { + return false; + } + if (likely(ts->sys_dispatch_selector)) { + uint8_t sb; + if (get_user_u8(sb, ts->sys_dispatch_selector)) { + force_sig(TARGET_SIGSEGV); + return true; + } + if (likely(sb == SYSCALL_DISPATCH_FILTER_ALLOW)) { + return false; + } + if (unlikely(sb != SYSCALL_DISPATCH_FILTER_BLOCK)) { + force_sig(TARGET_SIGSYS); + return true; + } + } + force_sig_fault(TARGET_SIGSYS, TARGET_SYS_USER_DISPATCH, pc); + return true; +} + abi_long do_syscall(CPUArchState *cpu_env, int num, abi_long arg1, abi_long arg2, abi_long arg3, abi_long arg4, abi_long arg5, abi_long arg6, abi_long arg7, abi_long arg8) { CPUState *cpu = env_cpu(cpu_env); + TaskState *ts = get_task_state(cpu); abi_long ret; #ifdef DEBUG_ERESTARTSYS @@ -13919,6 +13984,10 @@ abi_long do_syscall(CPUArchState *cpu_env, int num, abi_long arg1, } #endif + if (sys_dispatch(cpu, ts)) { + return -QEMU_ESIGRETURN; + } + record_syscall_start(cpu, num, arg1, arg2, arg3, arg4, arg5, arg6, arg7, arg8); diff --git a/linux-user/syscall_defs.h b/linux-user/syscall_defs.h index df26a2d28f..cd9ff709b8 100644 --- a/linux-user/syscall_defs.h +++ b/linux-user/syscall_defs.h @@ -690,6 +690,12 @@ typedef struct target_siginfo { #define TARGET_TRAP_UNK (5) /* undiagnosed trap */ /* + * SIGSYS si_codes + */ +#define TARGET_SYS_SECCOMP (1) /* seccomp triggered */ +#define TARGET_SYS_USER_DISPATCH (2) /* syscall user dispatch triggered */ + +/* * SIGEMT si_codes */ #define TARGET_EMT_TAGOVF 1 /* tag overflow */ diff --git a/linux-user/user-internals.h b/linux-user/user-internals.h index 691b9a1775..7099349ec8 100644 --- a/linux-user/user-internals.h +++ b/linux-user/user-internals.h @@ -20,6 +20,8 @@ #include "user/thunk.h" #include "qemu/log.h" +#include "exec/tb-flush.h" +#include "exec/translation-block.h" extern char *exec_path; void init_task_state(TaskState *ts); @@ -172,6 +174,20 @@ static inline int regpairs_aligned(CPUArchState *cpu_env, int num) { return 0; } */ void preexit_cleanup(CPUArchState *env, int code); +/** + * begin_parallel_context + * @cs: the CPU context + * + * Called when starting the second vcpu, or joining shared memory. + */ +static inline void begin_parallel_context(CPUState *cs) +{ + if (!tcg_cflags_has(cs, CF_PARALLEL)) { + tb_flush__exclusive_or_serial(); + tcg_cflags_set(cs, CF_PARALLEL); + } +} + /* * Include target-specific struct and function definitions; * they may need access to the target-independent structures |