linux-kernel - Re: [PATCH 5.15 000/402] 5.15.161-rc1 review

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: <2024061530-dress-powdery-c464@gregkh>
Date: Sat, 15 Jun 2024 13:05:31 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: Harshit Mogalapalli <harshit.m.mogalapalli@...cle.com>
Cc: stable@...r.kernel.org, patches@...ts.linux.dev,
	linux-kernel@...r.kernel.org, torvalds@...ux-foundation.org,
	akpm@...ux-foundation.org, linux@...ck-us.net, shuah@...nel.org,
	patches@...nelci.org, lkft-triage@...ts.linaro.org,
	Kamalesh Babulal <kamalesh.babulal@...cle.com>, pavel@...x.de,
	jonathanh@...dia.com, f.fainelli@...il.com,
	sudipm.mukherjee@...il.com, srw@...dewatkins.net, rwarsow@....de,
	conor@...nel.org, allen.lkml@...il.com, broonie@...nel.org,
	acme@...hat.com, namhyung@...nel.org, gpavithrasha@...il.com,
	irogers@...gle.com, Vegard Nossum <vegard.nossum@...cle.com>,
	Darren Kenny <darren.kenny@...cle.com>
Subject: Re: [PATCH 5.15 000/402] 5.15.161-rc1 review

On Fri, Jun 14, 2024 at 02:10:26PM +0530, Harshit Mogalapalli wrote:
> Hello Greg and Sasha,
> 
> On 13/06/24 16:59, Greg Kroah-Hartman wrote:
> > This is the start of the stable review cycle for the 5.15.161 release.
> > There are 402 patches in this series, all will be posted as a response
> > to this one.  If anyone has any issues with these being applied, please
> > let me know.
> > 
> > Responses should be made by Sat, 15 Jun 2024 11:31:50 +0000.
> > Anything received after that time might be too late.
> > 
> 
> perf build breaks on 5.15.161:(and on 5.4.278, 5.10.219)
> 
> 
> In file included from util/cache.h:7,
>                  from builtin-annotate.c:13:
> util/../ui/ui.h:5:10: fatal error: ../util/mutex.h: No such file or
> directory
>     5 | #include "../util/mutex.h"
>       |          ^~~~~~~~~~~~~~~~~
> compilation terminated.
> make[3]: *** [/home/linux-stable-rc/tools/build/Makefile.build:97:
> builtin-annotate.o] Error 1
> make[2]: *** [Makefile.perf:658: perf-in.o] Error 2
> make[2]: *** Waiting for unfinished jobs....
> 
> 
> 
> From the git log:
> 
> commit 83185fafbd143274c0313897fd8fda41aecffc93
> Author: Ian Rogers <irogers@...gle.com>
> Date:   Fri Aug 26 09:42:33 2022 -0700
> 
>     perf ui: Update use of pthread mutex
> 
>     [ Upstream commit 82aff6cc070417f26f9b02b26e63c17ff43b4044 ]
> 
>     Switch to the use of mutex wrappers that provide better error checking.
> 
> 
> I think building perf while adding perf patches would help us prevent from
> running into this issue. cd tools/perf/ ; make -j$(nproc) all

Maybe, but I can't seem to build perf at all for quite a while, so I
doubt that anyone really cares here, right?

> We can choose one of the three ways to solve this :
> 
> 1. Drop this patch and resolve conflicts in the next patch by keeping
> pthread_mutex_*, but this might not help future backports.

Let me just drop all of the perf patches for now from 5.15 and then I'll
take some tested backports if really needed.

Otherwise, why not just use perf from the latest 6.9 tree?

> 2. Add another dependency patch which introduces header file in util folder,
> that is also not clean backport due to a missing commit, but I have tried
> preparing a backport. I am not sure if that is a preferred way but with the
> backport inserted before: commit 83185fafbd143274c0313897fd8fda41aecffc93
> (between PATCH 224 and 225 in this series). Attached the backport. [
> 0001-perf-mutex-Wrapped-usage-of-mutex-and-cond.patch ]
> 
> 3. Clean cherry-pick way: instead of resolving conflict add one more
> prerequisite patch:
> just before commit 83185fafbd14 in 5.15.y: Cherry-pick:
> 	a. git cherry-pick -s 92ec3cc94c2c  // list_sort.h addition
> 	b. git cherry-pick -s e57d897703c3  // mutex.h addition
> 
> tools/perf builds with option 2/3, tested it.
> 
> For 5.10.y: Option 2 and 3 works.

I'll just drop this from 5.10 and 5.15.

> For 5.4.y we need other way to fix this.

Again, do you really want to use the 5.4 version of perf?

thanks,

greg k-h