RE: Re: [V4 PATCH 3/4] kexec: Fix race between panic() and crash_kexec() called directly
From: æåèå / KAWAIïHIDEHIRO
Date: Wed Sep 30 2015 - 22:05:10 EST
> On Mon, Sep 28, 2015 at 07:08:19AM +0000, æåèå / KAWAIïHIDEHIRO wrote:
> > > >> kernel/kexec_core.c:899:3: note: in expansion of macro 'atomic_xchg'
> > > atomic_xchg(&panic_cpu, -1);
> > > ^
> > I changed to use atomic_xchg() instead of atomic_set() in V3
> > because atomic_set() doesn't mean memory barrier. However,
> > I thought again and there is no need of barrier; there is no
> > problem if a competitor sees old value of panic_cpu or new one.
> > So, atomic_set() is sufficient and using it will remove this warning.
> > I will resend the fixed version later.
> So if you rely on the memory barrier; you should have also put a comment
> on explaining the ordering requirements.
I don't intend to use an explicit memory barrier. There is no
memory ordering requirement here. Also, atomic_set() which will be
used instead of atomic_xchg() is used as a RELEASE operation, so
I believe there is no problem.
> The following operations are potential problems as they do _not_ imply memory
> barriers, but might be used for implementing such things as RELEASE-class