[Top] [All Lists]

Re: [PATCH] MIPS: fix forced successful syscalls

To: Tanguy Bouzeloc <>
Subject: Re: [PATCH] MIPS: fix forced successful syscalls
From: Ralf Baechle <>
Date: Wed, 2 Oct 2013 11:19:09 +0200
In-reply-to: <>
List-archive: <>
List-help: <>
List-id: linux-mips <>
List-owner: <>
List-post: <>
List-software: Ecartis version 1.0.0
List-subscribe: <>
List-unsubscribe: <>
Original-recipient: rfc822;
References: <>
User-agent: Mutt/1.5.21 (2010-09-15)
On Mon, Sep 30, 2013 at 04:22:49PM +0200, Tanguy Bouzeloc wrote:
> Date:   Mon, 30 Sep 2013 16:22:49 +0200
> From: Tanguy Bouzeloc <>
> To:
> Cc:, Tanguy Bouzeloc <>
> Subject: [PATCH] MIPS: fix forced successful syscalls
> On mips any syscalls who return a value between -MAXERRNO (1133) and
> -1, is considered as an error (the error flag is set and return value
> is the positive value of the error number).
> But some syscalls can return values between -MAXERRNO and -1 like
> sys_time and sys_times. In this case the userspace return value is
> -return value of the syscall and the error flag set.
> This patch add a TIF_NOERROR thread flag which indicates that the
> return value of a syscall is always correct.

To my personal embarassment I have to admit that I knew about this since the
day the syscall wrapper was written - but was considering it an acceptable
bug ...

Where it really bits is sigreturn and similar which use the following

         * Don't let your children do this ...
        __asm__ __volatile__(
                "move\t$29, %0\n\t"
                :/* no outputs */
                :"r" (&regs));
        /* Unreached */

to keep the syscall return path from tampering with the return value.

The scall*.S part of your patch is clearing TIF_NOERROR using a non-atomic
LW/SW sequence.  This needs to be done atomically or the thread's flags
variable might get corrupted.  This is complicated by MIPS I, R5900 and
afair some older oddball not-quite MIPS II CPUs lacking LL/SC rsp. LLD/SCD.


<Prev in Thread] Current Thread [Next in Thread>