[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <86c8a026-4df7-4ba8-b0e5-6289ae584289@intel.com>
Date: Thu, 30 May 2024 08:33:18 -0700
From: Dave Hansen <dave.hansen@...el.com>
To: Jörn Heusipp <osmanx@...sipp.de>,
linux-kernel@...r.kernel.org, Dave Hansen <dave.hansen@...ux.intel.com>
Cc: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
stable@...r.kernel.org, x86@...nel.org, Thomas Gleixner
<tglx@...utronix.de>, Ingo Molnar <mingo@...hat.com>,
Borislav Petkov <bp@...en8.de>, Andy Lutomirski <luto@...nel.org>,
Peter Zijlstra <peterz@...radead.org>, Adam Dunlap <acdunlap@...gle.com>
Subject: Re: [tip: x86/urgent] x86/cpu: Provide default cache line size if not
enumerated
On 5/30/24 08:05, Jörn Heusipp wrote:
>>
>> + /* Provide a sane default if not enumerated: */
>> + if (!c->x86_clflush_size)
>> + c->x86_clflush_size = 32;
>> } else {
>> cpuid(0x80000008, &eax, &ebx, &ecx, &edx);
>>
>
> That honestly looks like a mis-merge to me. This is not what I did test.
> x86_clflush_size needs to get set in the top-level else branch in which
> it is not already set. commit 95bfb35269b2e85cff0dd2c957b2d42ebf95ae5f
> had switched the if branches around.
You're totally right. Thank you for catching that.
I think a variable removal flipped that code around and it confused me
when I did the rebase. It should be fixed up now.
Again, thanks a bunch for the report, the testing, and catching my
screwup quickly!
Powered by blists - more mailing lists