linux-kernel - Re: [PATCH v3] mm: add thp_utilization metrics to /proc/thp

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4875F747-54AC-4820-A3DA-51B5A96776FA@fb.com>
Date:   Fri, 5 Aug 2022 21:46:42 +0000
From:   "Alex Zhu (Kernel)" <alexlzhu@...com>
To:     William Kucharski <william.kucharski@...cle.com>
CC:     "linux-mm@...ck.org" <linux-mm@...ck.org>,
        Kernel Team <Kernel-team@...com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "akpm@...ux-foundation.org" <akpm@...ux-foundation.org>
Subject: Re: [PATCH v3] mm: add thp_utilization metrics to
 /proc/thp_utilization


> 
> Out of curiosity, do you have any statistics on how much of a performance impact scanning may have?
> 
> We've reached the point where we've finally been able to show previous admonitions to not enable THP due to a negative performance impact no longer hold true, and I'd hate to see systems on which enabling THP now would cause a performance hit due to this scanner.

There are no noticeable performance issues from running the scanner itself from our experiments. As written, we check 256 pfn s at a time every second, which does not appear to be an issue.