From: Jason Gunthorpe <jgg@ziepe.ca>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Steven Rostedt <rostedt@goodmis.org>,
David Hildenbrand <david@kernel.org>,
Leon Romanovsky <leon@kernel.org>,
Masami Hiramatsu <mhiramat@kernel.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Huiwen He <hehuiwen@kylinos.cn>,
Jerome Marchand <jmarchan@redhat.com>,
Qing Wang <wangqing7171@gmail.com>,
Shengming Hu <hu.shengming@zte.com.cn>,
Linux-MM <linux-mm@kvack.org>,
linux-rdma <linux-rdma@vger.kernel.org>
Subject: Re: [GIT PULL] tracing: Fixes for 7.0
Date: Thu, 5 Mar 2026 15:07:50 -0400 [thread overview]
Message-ID: <20260305190750.GA1687929@ziepe.ca> (raw)
In-Reply-To: <CAHk-=wggaToeRYv6B5L9ob=wBdVW9_gFudYxH_WJDTuhyX_Ueg@mail.gmail.com>
On Thu, Mar 05, 2026 at 08:44:27AM -0800, Linus Torvalds wrote:
> Now, Debian code search does show some users (libfabric, libibverbs),
> and maybe they actually want the forking behavior for other reasons
> too.
DOFORK in libibverbs is a consequence of its automatic DONTFORK, is is
explcitily there to undo a prior DONTFORK only.
This is because the library wrappers would automatically DONTFORK
regions of memory based on library usage, and then DOFORK them back to
normal after the library is done with them.
This whole path is disabled on modern kernels in favour of the fixed
fork support.
I think your point that DOFORK should only take action on previous
DONTFORK is correct.
Jason
prev parent reply other threads:[~2026-03-05 19:07 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20260305103941.11f1b27d@gandalf.local.home>
2026-03-05 16:44 ` Linus Torvalds
2026-03-05 16:52 ` Steven Rostedt
2026-03-05 17:00 ` David Hildenbrand (Arm)
2026-03-05 17:17 ` Linus Torvalds
2026-03-05 18:59 ` David Hildenbrand (Arm)
2026-03-05 19:07 ` Jason Gunthorpe [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260305190750.GA1687929@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=david@kernel.org \
--cc=hehuiwen@kylinos.cn \
--cc=hu.shengming@zte.com.cn \
--cc=jmarchan@redhat.com \
--cc=leon@kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mhiramat@kernel.org \
--cc=rostedt@goodmis.org \
--cc=torvalds@linux-foundation.org \
--cc=wangqing7171@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox