From: Frederic Weisbecker <fweisbec@gmail.com>
To: Joe Perches <joe@perches.com>
Cc: Andrew Morton <akpm@linux-foundation.org>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/3] softirq: Use ffs in __do_softirq
Date: Mon, 9 Dec 2013 19:44:55 +0100 [thread overview]
Message-ID: <20131209184454.GA1811@localhost.localdomain> (raw)
In-Reply-To: <8d10077333476c5c0e054cf060006999cf150f0c.1384656563.git.joe@perches.com>
On Sun, Nov 17, 2013 at 01:55:10AM -0800, Joe Perches wrote:
> Possible speed improvement of the __do_softirq function by using ffs
> instead of using a while loop with an & 1 test then single bit shift.
>
> Signed-off-by: Joe Perches <joe@perches.com>
> ---
> kernel/softirq.c | 43 ++++++++++++++++++++++---------------------
> 1 file changed, 22 insertions(+), 21 deletions(-)
>
> diff --git a/kernel/softirq.c b/kernel/softirq.c
> index 11025cc..7be95d7 100644
> --- a/kernel/softirq.c
> +++ b/kernel/softirq.c
> @@ -219,6 +219,7 @@ asmlinkage void __do_softirq(void)
> int cpu;
> unsigned long old_flags = current->flags;
> int max_restart = MAX_SOFTIRQ_RESTART;
> + int softirq_bit;
>
> /*
> * Mask out PF_MEMALLOC s current task context is borrowed for the
> @@ -242,30 +243,30 @@ restart:
>
> h = softirq_vec;
>
> - do {
> - if (pending & 1) {
> - unsigned int vec_nr = h - softirq_vec;
> - int prev_count = preempt_count();
> -
> - kstat_incr_softirqs_this_cpu(vec_nr);
> -
> - trace_softirq_entry(vec_nr);
> - h->action(h);
> - trace_softirq_exit(vec_nr);
> - if (unlikely(prev_count != preempt_count())) {
> - printk(KERN_ERR "huh, entered softirq %u %s %p"
> - "with preempt_count %08x,"
> - " exited with %08x?\n", vec_nr,
> - softirq_to_name[vec_nr], h->action,
> - prev_count, preempt_count());
> - preempt_count_set(prev_count);
> - }
> + while ((softirq_bit = ffs(pending))) {
> + unsigned int vec_nr;
> + int prev_count;
> +
> + h += softirq_bit - 1;
Perhaps using for_each_set_bit() would simplify that more?
> +
> + vec_nr = h - softirq_vec;
> + prev_count = preempt_count();
>
> - rcu_bh_qs(cpu);
> + kstat_incr_softirqs_this_cpu(vec_nr);
> +
> + trace_softirq_entry(vec_nr);
> + h->action(h);
> + trace_softirq_exit(vec_nr);
> + if (unlikely(prev_count != preempt_count())) {
> + printk(KERN_ERR "huh, entered softirq %u %s %p with preempt_count %08x, exited with %08x?\n",
> + vec_nr, softirq_to_name[vec_nr], h->action,
> + prev_count, preempt_count());
> + preempt_count_set(prev_count);
> }
> + rcu_bh_qs(cpu);
> h++;
> - pending >>= 1;
> - } while (pending);
> + pending >>= softirq_bit;
> + }
>
> local_irq_disable();
>
> --
> 1.8.1.2.459.gbcd45b4.dirty
>
next prev parent reply other threads:[~2013-12-09 18:45 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-17 9:55 [PATCH 0/3] softirq: possible speedup and neatenings Joe Perches
2013-11-17 9:55 ` [PATCH 1/3] softirq: Use ffs in __do_softirq Joe Perches
2013-12-09 18:44 ` Frederic Weisbecker [this message]
2013-12-09 18:56 ` Joe Perches
2013-12-09 19:13 ` Frederic Weisbecker
2013-11-17 9:55 ` [PATCH 2/3] softirq: Convert printks to pr_<level> Joe Perches
2013-11-17 9:55 ` [PATCH 3/3] softirq: Use const char * const for softirq_to_name, whitespace neatening Joe Perches
2013-12-24 15:19 ` Wang YanQing
2013-12-24 15:27 ` Joe Perches
2013-12-09 17:22 ` [PATCH 0/3] softirq: possible speedup and neatenings Joe Perches
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=20131209184454.GA1811@localhost.localdomain \
--to=fweisbec@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=joe@perches.com \
--cc=linux-kernel@vger.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