linux-kernel - Re: [PATCH] perf: Remove setting of page->index and ->mapping

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

Message-ID: <Zr-Rm14B5AHZ4_UB@casper.infradead.org>
Date: Fri, 16 Aug 2024 18:51:23 +0100
From: Matthew Wilcox <willy@...radead.org>
To: kernel test robot <oliver.sang@...el.com>
Cc: oe-lkp@...ts.linux.dev, lkp@...el.com, linux-perf-users@...r.kernel.org,
	linux-kernel@...r.kernel.org, ying.huang@...el.com,
	feng.tang@...el.com, fengwei.yin@...el.com,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Namhyung Kim <namhyung@...nel.org>
Subject: Re: [PATCH] perf: Remove setting of page->index and ->mapping

On Fri, Aug 16, 2024 at 02:35:39PM +0800, kernel test robot wrote:
> from commit message, it seems just some useless setting have been removed.
> but quite some performance regression bisect by our bot finally point to this
> change (already removed some not so significant changes to avoid this report
> mail is too long).
> 
> we don't have enough knowledge to explain these, so just report FYI.
> if you have any suggestion for our tests or need us run more tests, please
> just let us know. thanks a lot!

Can you try this replacement patch?

diff --git a/kernel/events/core.c b/kernel/events/core.c
index c973e3c11e03..bc12f4fb7fa9 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -6221,10 +6221,9 @@ static vm_fault_t perf_mmap_fault(struct vm_fault *vmf)
 		goto unlock;
 
 	get_page(vmf->page);
-	vmf->page->mapping = vmf->vma->vm_file->f_mapping;
-	vmf->page->index   = vmf->pgoff;
+	lock_page(vmf->page);
 
-	ret = 0;
+	ret = VM_FAULT_LOCKED;
 unlock:
 	rcu_read_unlock();