Re: [PATCH v4 00/18] kasan: x86: arm64: KASAN tag-based mode for x86

[Maciej Wieczor-Retman]

On 2025-08-13 at 12:05:47 +0100, Kiryl Shutsemau wrote:
>On Wed, Aug 13, 2025 at 12:39:35PM +0200, Maciej Wieczor-Retman wrote:
>> On 2025-08-13 at 09:16:29 +0100, Kiryl Shutsemau wrote:
>> >On Tue, Aug 12, 2025 at 03:23:36PM +0200, Maciej Wieczor-Retman wrote:
>> >> Compilation time comparison (10 cores):
>> >> * 7:27 for clean kernel
>> >> * 8:21/7:44 for generic KASAN (inline/outline)
>> >> * 8:20/7:41 for tag-based KASAN (inline/outline)
>> >
>> >It is not clear if it is compilation time of a kernel with different
>> >config options or compilation time of the same kernel running on machine
>> >with different kernels (KASAN-off/KASAN-generic/KASAN-tagged).
>> 
>> It's the first one, I'll reword this accordingly.
>> 
>> When you said a while ago this would be a good thing to measure, did you mean
>> the first or the second thing? I thought you meant the first one but now I have
>> doubts.
>
>I meant the second. We want to know how slow is it to run a workload
>under kernel with KASAN enabled.

Okay, thanks for confirming, I'll run these compilations on the system with the
tested kernels and attach results to v5 of the series.

>
>-- 
>  Kiryl Shutsemau / Kirill A. Shutemov

-- 
Kind regards
Maciej Wieczór-Retman