[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230704090132.GP4253@hirez.programming.kicks-ass.net>
Date: Tue, 4 Jul 2023 11:01:32 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Borislav Petkov <bp@...en8.de>
Cc: X86 ML <x86@...nel.org>,
Kishon VijayAbraham <Kishon.VijayAbraham@....com>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] x86/barrier: Do not serialize MSR accesses on AMD
On Tue, Jul 04, 2023 at 09:46:31AM +0200, Borislav Petkov wrote:
> On Mon, Jul 03, 2023 at 02:54:19PM +0200, Peter Zijlstra wrote:
> > So you're saying that AMD tsc_deadline and x2apic MSRs *do* imply
> > ordering constraints unlike the Intel ones?
>
> Yah, that's the default situation. Only those two - TSC_DEADLINE and
> x2APIC MSRs - and on *Intel* are special.
So they are normal MSRs like all other? AMD doesn't have any exceptions
for MSRs, they all the same?
> > Both instructions are 3 bytes, a 6 byte nop would be better, no?
>
> Why? You wanna save the branch insn when sending IPIs through the
> x2APIC? Does that really matter? I doubt it...
Dunno, code density, speculation, many raisons to avoid jumps :-)
> > asm volatile (ALTERNATIVE("mfence; lfence;", "", X86_FEATURE_AMD));
>
> There's no X86_FEATURE_AMD :)
I know, but that's easily fixed.
Powered by blists - more mailing lists