Re: [BUG 2.6.20-rc2-mm1] init segfaults whenCONFIG_PROFILE_LIKELY=y

[Randy Dunlap]

On Sun, 31 Dec 2006 11:45:09 -0800 Daniel Walker wrote:

> On Sun, 2006-12-31 at 23:04 +0800, Fengguang Wu wrote:
> > Hi,
> > 
> > The following messages keeps popping up when CONFIG_PROFILE_LIKELY=y:
> > 
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > init[1]: segfault at ffffffff8118c110 rip ffffffff8118c110 rsp 00007fff9a9d14d8 error 15
> > 
> 
> 
> Does this seem like an appropriate solution? This just reconstitutes
> Ingo's patch by removing the unlikely calls that got added recently. 

How does this fix the problem?  (if it does)
What is the real cause of the problem?

> Maybe a comment into vsyscall.c that says to stay away from all macro's
> and possible debug code that could be added might be helpful ?

Why?

> Signed-Off-By: Daniel Walker <dwalker@xxxxxxxxxx>
> 
> ---
>  arch/x86_64/kernel/vsyscall.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> Index: linux-2.6.19/arch/x86_64/kernel/vsyscall.c
> ===================================================================
> --- linux-2.6.19.orig/arch/x86_64/kernel/vsyscall.c
> +++ linux-2.6.19/arch/x86_64/kernel/vsyscall.c
> @@ -111,7 +111,7 @@ static __always_inline void do_vgettimeo
>  		seq = read_seqbegin(&__vsyscall_gtod_data.lock);
>  
>  		vread = __vsyscall_gtod_data.clock.vread;
> -		if (unlikely(!__vsyscall_gtod_data.sysctl_enabled || !vread)) {
> +		if (!__vsyscall_gtod_data.sysctl_enabled || !vread) {
>  			gettimeofday(tv,0);
>  			return;
>  		}
> @@ -151,7 +151,7 @@ int __vsyscall(0) vgettimeofday(struct t
>   * unlikely */
>  time_t __vsyscall(1) vtime(time_t *t)
>  {
> -	if (unlikely(!__vsyscall_gtod_data.sysctl_enabled))
> +	if (!__vsyscall_gtod_data.sysctl_enabled)
>  		return time_syscall(t);
>  	else if (t)
>  		*t = __vsyscall_gtod_data.wall_time_tv.tv_sec;


---
~Randy
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/