Message ID | 20210918184527.408540-28-richard.henderson@linaro.org |
---|---|
State | Superseded |
Headers | show |
Series | linux-user: Streamline handling of SIGSEGV | expand |
On 9/18/21 20:45, Richard Henderson wrote: > Record cr2, error_code, and exception_index. That last means > that we must exit to cpu_loop ourselves, instead of letting > exception_index being overwritten. > > Use the maperr parameter to properly set PG_ERROR_P_MASK. > > Signed-off-by: Richard Henderson <richard.henderson@linaro.org> > --- > target/i386/tcg/helper-tcg.h | 6 ++++++ > target/i386/tcg/tcg-cpu.c | 3 ++- > target/i386/tcg/user/excp_helper.c | 23 +++++++++++++++++------ > 3 files changed, 25 insertions(+), 7 deletions(-) > diff --git a/target/i386/tcg/user/excp_helper.c b/target/i386/tcg/user/excp_helper.c > index a89b5228fd..cd507e2a1b 100644 > --- a/target/i386/tcg/user/excp_helper.c > +++ b/target/i386/tcg/user/excp_helper.c > @@ -22,18 +22,29 @@ > #include "exec/exec-all.h" > #include "tcg/helper-tcg.h" > > -bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size, > - MMUAccessType access_type, int mmu_idx, > - bool probe, uintptr_t retaddr) > +void x86_cpu_record_sigsegv(CPUState *cs, vaddr addr, > + MMUAccessType access_type, > + bool maperr, uintptr_t ra) > { > X86CPU *cpu = X86_CPU(cs); > CPUX86State *env = &cpu->env; > > + /* > + * The error_code that hw reports as part of the exception frame > + * is copied to linux sigcontext.err. The exception_index is > + * copied to linux sigcontext.trapno. Short of inventing a new > + * place to store the trapno, we cannot let our caller raise the > + * signal and set exception_index to EXCP_INTERRUPT. > + */ > env->cr[2] = addr; > - env->error_code = (access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT; > - env->error_code |= PG_ERROR_U_MASK; > + env->error_code = ((access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT) > + | (maperr ? 0 : PG_ERROR_P_MASK) > + | PG_ERROR_U_MASK; > cs->exception_index = EXCP0E_PAGE; > + > + /* Disable do_interrupt_user. */ > env->exception_is_int = 0; > env->exception_next_eip = -1; > - cpu_loop_exit_restore(cs, retaddr); > + > + cpu_loop_exit_restore(cs, ra); > } > Better have an x86 expert also review this, but to the best of my knowledge: Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> And YAY! btw, thanks :>
> On Sep 18, 2021, at 12:45 PM, Richard Henderson <richard.henderson@linaro.org> wrote: > > Record cr2, error_code, and exception_index. That last means > that we must exit to cpu_loop ourselves, instead of letting > exception_index being overwritten. > > Use the maperr parameter to properly set PG_ERROR_P_MASK. > > Signed-off-by: Richard Henderson <richard.henderson@linaro.org> > --- > target/i386/tcg/helper-tcg.h | 6 ++++++ > target/i386/tcg/tcg-cpu.c | 3 ++- > target/i386/tcg/user/excp_helper.c | 23 +++++++++++++++++------ > 3 files changed, 25 insertions(+), 7 deletions(-) Reviewed by: Warner Losh <imp@bsdimp.com> I think this encoding is fine, but haven’t thought though how I’d implement this in bsd-user yet… I have the signal code queued up, but not ready to send off yet. > diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h > index 60ca09e95e..0a4401e917 100644 > --- a/target/i386/tcg/helper-tcg.h > +++ b/target/i386/tcg/helper-tcg.h > @@ -43,9 +43,15 @@ bool x86_cpu_exec_interrupt(CPUState *cpu, int int_req); > #endif > > /* helper.c */ > +#ifdef CONFIG_USER_ONLY > +void x86_cpu_record_sigsegv(CPUState *cs, vaddr addr, > + MMUAccessType access_type, > + bool maperr, uintptr_t ra); > +#else > bool x86_cpu_tlb_fill(CPUState *cs, vaddr address, int size, > MMUAccessType access_type, int mmu_idx, > bool probe, uintptr_t retaddr); > +#endif > > void breakpoint_handler(CPUState *cs); > > diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c > index aef050d089..3fab3676b1 100644 > --- a/target/i386/tcg/tcg-cpu.c > +++ b/target/i386/tcg/tcg-cpu.c > @@ -77,11 +77,12 @@ static const struct TCGCPUOps x86_tcg_ops = { > .synchronize_from_tb = x86_cpu_synchronize_from_tb, > .cpu_exec_enter = x86_cpu_exec_enter, > .cpu_exec_exit = x86_cpu_exec_exit, > - .tlb_fill = x86_cpu_tlb_fill, > #ifdef CONFIG_USER_ONLY > .fake_user_interrupt = x86_cpu_do_interrupt, > + .record_sigsegv = x86_cpu_record_sigsegv, > #else > .has_work = x86_cpu_has_work, > + .tlb_fill = x86_cpu_tlb_fill, > .do_interrupt = x86_cpu_do_interrupt, > .cpu_exec_interrupt = x86_cpu_exec_interrupt, > .debug_excp_handler = breakpoint_handler, > diff --git a/target/i386/tcg/user/excp_helper.c b/target/i386/tcg/user/excp_helper.c > index a89b5228fd..cd507e2a1b 100644 > --- a/target/i386/tcg/user/excp_helper.c > +++ b/target/i386/tcg/user/excp_helper.c > @@ -22,18 +22,29 @@ > #include "exec/exec-all.h" > #include "tcg/helper-tcg.h" > > -bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size, > - MMUAccessType access_type, int mmu_idx, > - bool probe, uintptr_t retaddr) > +void x86_cpu_record_sigsegv(CPUState *cs, vaddr addr, > + MMUAccessType access_type, > + bool maperr, uintptr_t ra) > { > X86CPU *cpu = X86_CPU(cs); > CPUX86State *env = &cpu->env; > > + /* > + * The error_code that hw reports as part of the exception frame > + * is copied to linux sigcontext.err. The exception_index is > + * copied to linux sigcontext.trapno. Short of inventing a new > + * place to store the trapno, we cannot let our caller raise the > + * signal and set exception_index to EXCP_INTERRUPT. > + */ > env->cr[2] = addr; > - env->error_code = (access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT; > - env->error_code |= PG_ERROR_U_MASK; > + env->error_code = ((access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT) > + | (maperr ? 0 : PG_ERROR_P_MASK) > + | PG_ERROR_U_MASK; > cs->exception_index = EXCP0E_PAGE; > + > + /* Disable do_interrupt_user. */ > env->exception_is_int = 0; > env->exception_next_eip = -1; > - cpu_loop_exit_restore(cs, retaddr); > + > + cpu_loop_exit_restore(cs, ra); > } > -- > 2.25.1 > >
diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h index 60ca09e95e..0a4401e917 100644 --- a/target/i386/tcg/helper-tcg.h +++ b/target/i386/tcg/helper-tcg.h @@ -43,9 +43,15 @@ bool x86_cpu_exec_interrupt(CPUState *cpu, int int_req); #endif /* helper.c */ +#ifdef CONFIG_USER_ONLY +void x86_cpu_record_sigsegv(CPUState *cs, vaddr addr, + MMUAccessType access_type, + bool maperr, uintptr_t ra); +#else bool x86_cpu_tlb_fill(CPUState *cs, vaddr address, int size, MMUAccessType access_type, int mmu_idx, bool probe, uintptr_t retaddr); +#endif void breakpoint_handler(CPUState *cs); diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c index aef050d089..3fab3676b1 100644 --- a/target/i386/tcg/tcg-cpu.c +++ b/target/i386/tcg/tcg-cpu.c @@ -77,11 +77,12 @@ static const struct TCGCPUOps x86_tcg_ops = { .synchronize_from_tb = x86_cpu_synchronize_from_tb, .cpu_exec_enter = x86_cpu_exec_enter, .cpu_exec_exit = x86_cpu_exec_exit, - .tlb_fill = x86_cpu_tlb_fill, #ifdef CONFIG_USER_ONLY .fake_user_interrupt = x86_cpu_do_interrupt, + .record_sigsegv = x86_cpu_record_sigsegv, #else .has_work = x86_cpu_has_work, + .tlb_fill = x86_cpu_tlb_fill, .do_interrupt = x86_cpu_do_interrupt, .cpu_exec_interrupt = x86_cpu_exec_interrupt, .debug_excp_handler = breakpoint_handler, diff --git a/target/i386/tcg/user/excp_helper.c b/target/i386/tcg/user/excp_helper.c index a89b5228fd..cd507e2a1b 100644 --- a/target/i386/tcg/user/excp_helper.c +++ b/target/i386/tcg/user/excp_helper.c @@ -22,18 +22,29 @@ #include "exec/exec-all.h" #include "tcg/helper-tcg.h" -bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size, - MMUAccessType access_type, int mmu_idx, - bool probe, uintptr_t retaddr) +void x86_cpu_record_sigsegv(CPUState *cs, vaddr addr, + MMUAccessType access_type, + bool maperr, uintptr_t ra) { X86CPU *cpu = X86_CPU(cs); CPUX86State *env = &cpu->env; + /* + * The error_code that hw reports as part of the exception frame + * is copied to linux sigcontext.err. The exception_index is + * copied to linux sigcontext.trapno. Short of inventing a new + * place to store the trapno, we cannot let our caller raise the + * signal and set exception_index to EXCP_INTERRUPT. + */ env->cr[2] = addr; - env->error_code = (access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT; - env->error_code |= PG_ERROR_U_MASK; + env->error_code = ((access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT) + | (maperr ? 0 : PG_ERROR_P_MASK) + | PG_ERROR_U_MASK; cs->exception_index = EXCP0E_PAGE; + + /* Disable do_interrupt_user. */ env->exception_is_int = 0; env->exception_next_eip = -1; - cpu_loop_exit_restore(cs, retaddr); + + cpu_loop_exit_restore(cs, ra); }
Record cr2, error_code, and exception_index. That last means that we must exit to cpu_loop ourselves, instead of letting exception_index being overwritten. Use the maperr parameter to properly set PG_ERROR_P_MASK. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/tcg/helper-tcg.h | 6 ++++++ target/i386/tcg/tcg-cpu.c | 3 ++- target/i386/tcg/user/excp_helper.c | 23 +++++++++++++++++------ 3 files changed, 25 insertions(+), 7 deletions(-) -- 2.25.1