bug-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 07/10] hurd: Fix x86_64 _hurd_tls_fork


From: Samuel Thibault
Subject: Re: [PATCH 07/10] hurd: Fix x86_64 _hurd_tls_fork
Date: Wed, 17 May 2023 23:01:10 +0200
User-agent: NeoMutt/20170609 (1.8.3)

Applied, thanks!

Sergey Bugaev via Libc-alpha, le mer. 17 mai 2023 22:14:33 +0300, a ecrit:
> It is illegal to call thread_get_state () on mach_thread_self (), so
> this codepath cannot be used as-is to fork the calling thread's TLS.
> Fortunately we can use THREAD_SELF (aka %fs:0x0) to find out the value
> of our fs_base without calling into the kernel.
> 
> Fixes: f6cf701efc61c9ad910372bda14b9a235db310a8
> "hurd: Implement TLS for x86_64"
> 
> Checked on x86_64-gnu: fork () now works!
> 
> Signed-off-by: Sergey Bugaev <bugaevc@gmail.com>
> ---
> It would be nice if GNU Mach allowed
> 
> thread_get_state (mach_thread_self (), i386_FSGS_BASE_STATE)
> 
> like it already allows thread_set_state for this case. But even if
> gnumach adds suppport for this, doing it in userspace without an extra
> RPC is still better.
> 
>  sysdeps/mach/hurd/x86_64/tls.h | 25 +++++++++++++++++++------
>  1 file changed, 19 insertions(+), 6 deletions(-)
> 
> diff --git a/sysdeps/mach/hurd/x86_64/tls.h b/sysdeps/mach/hurd/x86_64/tls.h
> index 35dcef44..6487ed35 100644
> --- a/sysdeps/mach/hurd/x86_64/tls.h
> +++ b/sysdeps/mach/hurd/x86_64/tls.h
> @@ -140,12 +140,25 @@ _hurd_tls_fork (thread_t child, thread_t orig,
>    error_t err;
>    struct i386_fsgs_base_state state;
>    mach_msg_type_number_t state_count = i386_FSGS_BASE_STATE_COUNT;
> -  err = __thread_get_state (orig, i386_FSGS_BASE_STATE,
> -                            (thread_state_t) &state,
> -                            &state_count);
> -  if (err)
> -    return err;
> -  assert (state_count == i386_FSGS_BASE_STATE_COUNT);
> +
> +  extern thread_t hurd_thread_self (void);
> +  if (orig != hurd_thread_self ())
> +    {
> +      err = __thread_get_state (orig, i386_FSGS_BASE_STATE,
> +                                (thread_state_t) &state,
> +                                &state_count);
> +      if (err)
> +        return err;
> +      assert (state_count == i386_FSGS_BASE_STATE_COUNT);
> +    }
> +  else
> +    {
> +      /* It is illegal to call thread_get_state () on mach_thread_self ().
> +         But we're only interested in the value of fs_base, and since we're
> +         this thread, we know it points to our TCB.  */
> +      state.fs_base = (unsigned long) THREAD_SELF;
> +      state.gs_base = 0;
> +    }
>  
>    return __thread_set_state (child, i386_FSGS_BASE_STATE,
>                               (thread_state_t) &state,
> -- 
> 2.40.1
> 

-- 
Samuel
---
Pour une évaluation indépendante, transparente et rigoureuse !
Je soutiens la Commission d'Évaluation de l'Inria.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]