[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <921e1b98-af36-1f51-5abe-dea36425b706@redhat.com>
Date: Thu, 22 Jun 2023 08:27:13 -0400
From: Waiman Long <longman@...hat.com>
To: Josh Poimboeuf <jpoimboe@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
Dave Hansen <dave.hansen@...ux.intel.com>,
"H. Peter Anvin" <hpa@...or.com>,
Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>,
Jacob Pan <jacob.jun.pan@...ux.intel.com>,
Len Brown <lenb@...nel.org>, linux-kernel@...r.kernel.org,
x86@...nel.org, linux-pm@...r.kernel.org,
Robin Jarry <rjarry@...hat.com>, Joe Mario <jmario@...hat.com>
Subject: Re: [PATCH v3 1/3] x86/idle: Disable IBRS when cpu is offline
On 6/22/23 01:40, Josh Poimboeuf wrote:
> On Wed, Jun 21, 2023 at 08:36:01PM -0400, Waiman Long wrote:
>> +/*
>> + * naitve_play_dead() is essentially a __noreturn function, but it can't
>> + * be marked as such as the compiler may complain about it.
>> + */
> FWIW, we could in theory do so by marking the smp_ops.play_dead function
> pointer as __noreturn, but it would be tricky to teach objtool how to
> understand that.
I added the comment here because I had taken out the MSR restoration
code. We can always replace that later on if there is a better way to do
that.
>
>> void native_play_dead(void)
>> {
>> + if (cpu_feature_enabled(X86_FEATURE_KERNEL_IBRS)) {
>> + this_cpu_write(x86_spec_ctrl_current, 0);
>> + native_wrmsrl(MSR_IA32_SPEC_CTRL, 0);
>> + }
> Can update_spec_ctrl() be used instead?
Yes, the code is similar to what has been done in update_spec_ctrl().
Using it, however, will require exporting the function either by putting
it into a public header or making it a global function.
Cheers,
Longman
>
Powered by blists - more mailing lists