From: Xie Yuanbin <qq570070308@gmail.com>
To: tglx@linutronix.de
Cc: acme@kernel.org, adrian.hunter@intel.com, agordeev@linux.ibm.com,
akpm@linux-foundation.org, alex@ghiti.fr,
alexander.shishkin@linux.intel.com, andreas@gaisler.com,
anna-maria@linutronix.de, aou@eecs.berkeley.edu,
borntraeger@linux.ibm.com, bp@alien8.de, bsegall@google.com,
dave.hansen@linux.intel.com, davem@davemloft.net,
david@redhat.com, dietmar.eggemann@arm.com, frederic@kernel.org,
gor@linux.ibm.com, hca@linux.ibm.com, hpa@zytor.com,
irogers@google.com, jolsa@kernel.org, juri.lelli@redhat.com,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
linux-riscv@lists.infradead.org, linux-s390@vger.kernel.org,
linux@armlinux.org.uk, lorenzo.stoakes@oracle.com,
luto@kernel.org, mark.rutland@arm.com,
mathieu.desnoyers@efficios.com, max.kellermann@ionos.com,
mgorman@suse.de, mingo@redhat.com, namhyung@kernel.org,
nysal@linux.ibm.com, palmer@dabbelt.com, paulmck@kernel.org,
peterz@infradead.org, pjw@kernel.org, qq570070308@gmail.com,
riel@surriel.com, rostedt@goodmis.org, ryan.roberts@arm.com,
segher@kernel.crashing.org, sparclinux@vger.kernel.org,
svens@linux.ibm.com, thuth@redhat.com, urezki@gmail.com,
vincent.guittot@linaro.org, vschneid@redhat.com, will@kernel.org,
x86@kernel.org
Subject: Re: [PATCH 3/3] Set the subfunctions called by finish_task_switch to be inline
Date: Sun, 26 Oct 2025 02:51:20 +0800 [thread overview]
Message-ID: <20251025185120.6760-1-qq570070308@gmail.com> (raw)
In-Reply-To: <87placw0dx.ffs@tglx>
On Fri, 24 Oct 2025 21:44:10 +0200, Thomas Gleixner wrote:
> What is exactly the point of this indirection. Why can't you just mark
> finish_arch_post_lock_switch() __always_inline and be done with it?
In this patch, I've added an always inline version of the function,
finish_arch_post_lock_switch_ainline. The original function,
finish_arch_post_lock_switch, retains its original inline attribute.
The reason for this is that this function is called not only during
context switches but also from other code, and I don't want to affect
those parts. In fact, with Os/Oz-level optimizations, if this function
is called multiple times within one .c file, it will most likely not be
inlined, even if it's marked as inline.
Context switching is a hot code, I hope it will be always inlined here to
improve performance. In other places, if it is not a performance-critical
function, then it can be not inlined to gain codesize benefits.
Look at your opinions. I have no objection to setting
finish_arch_post_lock_switch directly to __always_inline.
Xie Yuanbin
next prev parent reply other threads:[~2025-10-25 18:51 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-24 18:26 [PATCH 0/3] Optimize code generation during context switching Xie Yuanbin
2025-10-24 18:26 ` [PATCH 1/3] Change enter_lazy_tlb to inline on x86 Xie Yuanbin
2025-10-24 20:14 ` Rik van Riel
2025-10-24 18:35 ` [PATCH 2/3] Provide and use an always inline version of finish_task_switch Xie Yuanbin
2025-10-24 18:35 ` [PATCH 3/3] Set the subfunctions called by finish_task_switch to be inline Xie Yuanbin
2025-10-24 19:44 ` Thomas Gleixner
2025-10-25 18:51 ` Xie Yuanbin [this message]
2025-10-24 21:36 ` [PATCH 2/3] Provide and use an always inline version of finish_task_switch Rik van Riel
2025-10-25 14:36 ` Segher Boessenkool
2025-10-25 17:37 ` [PATCH 0/3] Optimize code generation during context Xie Yuanbin
2025-10-29 10:26 ` David Hildenbrand
2025-10-30 15:04 ` Xie Yuanbin
2025-10-25 19:18 ` [PATCH 2/3] Provide and use an always inline version of finish_task_switch Xie Yuanbin
2025-10-25 12:26 ` [PATCH 0/3] Optimize code generation during context switching Peter Zijlstra
2025-10-25 18:20 ` [PATCH 0/3] Optimize code generation during context Xie Yuanbin
2025-10-27 15:21 ` Xie Yuanbin
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=20251025185120.6760-1-qq570070308@gmail.com \
--to=qq570070308@gmail.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=agordeev@linux.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=alex@ghiti.fr \
--cc=alexander.shishkin@linux.intel.com \
--cc=andreas@gaisler.com \
--cc=anna-maria@linutronix.de \
--cc=aou@eecs.berkeley.edu \
--cc=borntraeger@linux.ibm.com \
--cc=bp@alien8.de \
--cc=bsegall@google.com \
--cc=dave.hansen@linux.intel.com \
--cc=davem@davemloft.net \
--cc=david@redhat.com \
--cc=dietmar.eggemann@arm.com \
--cc=frederic@kernel.org \
--cc=gor@linux.ibm.com \
--cc=hca@linux.ibm.com \
--cc=hpa@zytor.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=juri.lelli@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=linux-s390@vger.kernel.org \
--cc=linux@armlinux.org.uk \
--cc=lorenzo.stoakes@oracle.com \
--cc=luto@kernel.org \
--cc=mark.rutland@arm.com \
--cc=mathieu.desnoyers@efficios.com \
--cc=max.kellermann@ionos.com \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=nysal@linux.ibm.com \
--cc=palmer@dabbelt.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=pjw@kernel.org \
--cc=riel@surriel.com \
--cc=rostedt@goodmis.org \
--cc=ryan.roberts@arm.com \
--cc=segher@kernel.crashing.org \
--cc=sparclinux@vger.kernel.org \
--cc=svens@linux.ibm.com \
--cc=tglx@linutronix.de \
--cc=thuth@redhat.com \
--cc=urezki@gmail.com \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.com \
--cc=will@kernel.org \
--cc=x86@kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).