lkml.org 
[lkml]   [2024]   [Jan]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: [PATCH] dmaengine: idxd: Change wmb() to smp_wmb() when copying completion record to user space
Date


> -----Original Message-----
> From: Yu, Fenghua <fenghua.yu@intel.com>
> Sent: Wednesday, January 31, 2024 2:12 AM
> To: Hansen, Dave <dave.hansen@intel.com>; Boqun Feng
> <boqun.feng@gmail.com>; Mark Rutland <mark.rutland@arm.com>
> Cc: Vinod Koul <vkoul@kernel.org>; Jiang, Dave <dave.jiang@intel.com>;
> dmaengine@vger.kernel.org; linux-kernel <linux-kernel@vger.kernel.org>; Rao,
> Nikhil <nikhil.rao@intel.com>; Zhu, Tony <tony.zhu@intel.com>; Mathieu
> Desnoyers <mathieu.desnoyers@efficios.com>; Thomas Gleixner
> <tglx@linutronix.de>; Ingo Molnar <mingo@redhat.com>; Borislav Petkov
> <bp@alien8.de>; Dave Hansen <dave.hansen@linux.intel.com>;
> x86@kernel.org
> Subject: Re: [PATCH] dmaengine: idxd: Change wmb() to smp_wmb() when
> copying completion record to user space
>
> Hi, Dave, Boqun, and Mark,
>
> On 1/30/24 12:30, Dave Hansen wrote:
> > On 1/30/24 11:53, Boqun Feng wrote:
> >>>> Fixes: b022f59725f0 ("dmaengine: idxd: add idxd_copy_cr() to copy
> >>>> user completion record during page fault handling")
> >>>> Suggested-by: Nikhil Rao <nikhil.rao@intel.com>
> >>>> Tested-by: Tony Zhu <tony.zhu@intel.com>
> >> Since it has a "Fixes" tag and a "Tested-by" tag, I'd assume there
> >> has been a test w/ and w/o this patch showing it can resolve a real
> >> issue *constantly*? If so, I think x86 might be broken somewhere.
> >>
> >> [Cc x86 maintainers]
> >
> > Fenghua, could you perhaps explain how this problem affects end users?
> > What symptom was observed that made it obvious something was broken
> > and what changes with this patch?
>
> There is no issue found by any test. This wmb() code was reviewed and was
> "thought" that it may have a potential issue.

I had made this suggestion since the code only needed a smp_wmb(), If the review refers to my suggestion, sorry if my message indicated a potential issue, that certainly wasn't my intention.
memory-barriers.txt does say that mandatory barriers (of which wmb() is one) should not be used to control SMP effects.

Nikhil
\
 
 \ /
  Last update: 2024-05-27 14:41    [W:0.051 / U:0.788 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site