Re: [Qemu-devel] [PATCH v8 10/23] RISC-V Linux User Emulation

[Michael Clark]

On Thu, Apr 5, 2018 at 12:44 AM, Laurent Vivier <address@hidden> wrote:

> Le 02/03/2018 à 14:51, Michael Clark a écrit :
> > Implementation of linux user emulation for RISC-V.
> >
> > Reviewed-by: Richard Henderson <address@hidden>
> > Signed-off-by: Sagar Karandikar <address@hidden>
> > Signed-off-by: Michael Clark <address@hidden>
> > ---
> >  linux-user/elfload.c              |  22 +++
> >  linux-user/main.c                 |  99 +++++++++++++
> >  linux-user/riscv/syscall_nr.h     | 287 ++++++++++++++++++++++++++++++
> ++++++++
> >  linux-user/riscv/target_cpu.h     |  18 +++
> >  linux-user/riscv/target_elf.h     |  14 ++
> >  linux-user/riscv/target_signal.h  |  23 +++
> >  linux-user/riscv/target_structs.h |  46 ++++++
> >  linux-user/riscv/target_syscall.h |  56 ++++++++
> >  linux-user/riscv/termbits.h       | 222 +++++++++++++++++++++++++++++
> >  linux-user/signal.c               | 203 ++++++++++++++++++++++++++-
> >  linux-user/syscall.c              |   2 +
> >  linux-user/syscall_defs.h         |  13 +-
> >  target/riscv/cpu_user.h           |  13 ++
> >  13 files changed, 1012 insertions(+), 6 deletions(-)
> >  create mode 100644 linux-user/riscv/syscall_nr.h
> >  create mode 100644 linux-user/riscv/target_cpu.h
> >  create mode 100644 linux-user/riscv/target_elf.h
> >  create mode 100644 linux-user/riscv/target_signal.h
> >  create mode 100644 linux-user/riscv/target_structs.h
> >  create mode 100644 linux-user/riscv/target_syscall.h
> >  create mode 100644 linux-user/riscv/termbits.h
> >  create mode 100644 target/riscv/cpu_user.h
> >
> ...
> > diff --git a/linux-user/signal.c b/linux-user/signal.c
> > index 9a380b9..4d3f244 100644
> > --- a/linux-user/signal.c
> > +++ b/linux-user/signal.c
> ...
> > +static abi_ulong get_sigframe(struct target_sigaction *ka,
> > +                              CPURISCVState *regs, size_t framesize)
> > +{
> > +    abi_ulong sp = regs->gpr[xSP];
> > +    int onsigstack = on_sig_stack(sp);
> > +
> > +    /* redzone */
> > +    /* This is the X/Open sanctioned signal stack switching.  */
> > +    if ((ka->sa_flags & TARGET_SA_ONSTACK) != 0 && !onsigstack) {
> > +        sp = target_sigaltstack_used.ss_sp + target_sigaltstack_used.ss_
> size;
> > +    }
> > +
> > +    sp -= framesize;
> > +    sp &= ~3UL; /* align sp on 4-byte boundary */
>
> kernel aligns using 0xf. Why do you use a different alignment?


Thanks for reviewing this! This looks like a bug!

I'm raising this as an issue on the riscv-qemu tracker so that its easier
than searching through the mail archive:

- https://github.com/riscv/riscv-qemu/issues/129

> +
> > +    /* If we are on the alternate signal stack and would overflow it,
> don't.
> > +       Return an always-bogus address instead so we will die with
> SIGSEGV. */
> > +    if (onsigstack && !likely(on_sig_stack(sp))) {
> > +        return -1L;
> > +    }
> > +
> > +    return sp;
> > +}
> Other question why don't you use the same logic as in kernel?
>
> 1- check for signal stack overflow
> 2- check for X/Open sanctioned signal stack switching
>
> static inline void __user *get_sigframe(struct ksignal *ksig,
>         struct pt_regs *regs, size_t framesize)
> {
>         unsigned long sp;
>         /* Default to using normal stack */
>         sp = regs->sp;
>
>         /*
>          * If we are on the alternate signal stack and would overflow
> it, don't.
>          * Return an always-bogus address instead so we will die with
> SIGSEGV.
>          */
>         if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize)))
>                 return (void __user __force *)(-1UL);
>
>         /* This is the X/Open sanctioned signal stack switching. */
>         sp = sigsp(sp, ksig) - framesize;
>
>         /* Align the stack frame. */
>         sp &= ~0xfUL;
>
>         return (void __user *)sp;
> }
>
> Thanks,
> Laurent
>