From: Sudeep Holla <sudeep.holla@arm.com>
To: Yunfeng Ye <yeyunfeng@huawei.com>
Cc: catalin.marinas@arm.com, will@kernel.org,
kstewart@linuxfoundation.org, gregkh@linuxfoundation.org,
lorenzo.pieralisi@arm.com, tglx@linutronix.de,
David.Laight@ACULAB.COM, mark.rutland@arm.com,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, hushiyuan@huawei.com,
wuyun.wu@huawei.com, linfeilong@huawei.com
Subject: Re: [PATCH v4] arm64: psci: Reduce the waiting time for cpu_psci_cpu_kill()
Date: Fri, 18 Oct 2019 16:20:52 +0100 [thread overview]
Message-ID: <20191018152052.GA10312@bogus> (raw)
In-Reply-To: <04ab51e4-bc08-8250-4e70-4c87c58c8ad0@huawei.com>
On Fri, Oct 18, 2019 at 08:46:37PM +0800, Yunfeng Ye wrote:
> In case like suspend-to-disk and uspend-to-ram, a large number of CPU
s/case/cases/
s/uspend-to-ram/suspend-to-ram/
> cores need to be shut down. At present, the CPU hotplug operation is
> serialised, and the CPU cores can only be shut down one by one. In this
> process, if PSCI affinity_info() does not return LEVEL_OFF quickly,
> cpu_psci_cpu_kill() needs to wait for 10ms. If hundreds of CPU cores
> need to be shut down, it will take a long time.
>
> Normally, there is no need to wait 10ms in cpu_psci_cpu_kill(). So
> change the wait interval from 10 ms to max 1 ms and use usleep_range()
> instead of msleep() for more accurate timer.
>
> In addition, reducing the time interval will increase the messages
> output, so remove the "Retry ..." message, instead, put the number of
> waiting times to the sucessful message.
>
> Signed-off-by: Yunfeng Ye <yeyunfeng@huawei.com>
> ---
> v3 -> v4:
> - using time_before(jiffies, timeout) to check
> - update the comment as review suggest
>
> v2 -> v3:
> - update the comment
> - remove the busy-wait logic, modify the loop logic and output message
>
> v1 -> v2:
> - use usleep_range() instead of udelay() after waiting for a while
> arch/arm64/kernel/psci.c | 14 ++++++++------
> 1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/arch/arm64/kernel/psci.c b/arch/arm64/kernel/psci.c
> index c9f72b2665f1..77965c3ba477 100644
> --- a/arch/arm64/kernel/psci.c
> +++ b/arch/arm64/kernel/psci.c
> @@ -81,7 +81,8 @@ static void cpu_psci_cpu_die(unsigned int cpu)
>
> static int cpu_psci_cpu_kill(unsigned int cpu)
> {
> - int err, i;
> + int err, i = 0;
> + unsigned long timeout;
>
> if (!psci_ops.affinity_info)
> return 0;
> @@ -91,16 +92,17 @@ static int cpu_psci_cpu_kill(unsigned int cpu)
> * while it is dying. So, try again a few times.
> */
>
> - for (i = 0; i < 10; i++) {
> + timeout = jiffies + msecs_to_jiffies(100);
> + do {
> + i++;
> err = psci_ops.affinity_info(cpu_logical_map(cpu), 0);
> if (err == PSCI_0_2_AFFINITY_LEVEL_OFF) {
> - pr_info("CPU%d killed.\n", cpu);
> + pr_info("CPU%d killed (polled %d times)\n", cpu, i);
We can even drop loop counter completely, track time and log that
instead of loop counter that doesn't give any indication without looking
into the code.
start = jiffies, end = start + msecs_to_jiffies(100);
do {
....
pr_info("CPU%d killed (polled %u ms)\n", cpu,
jiffies_to_msecs(jiffies - start));
....
} while (time_before(jiffies, end));
Just my preference. Looks good otherwise.
--
Regards,
Sudeep
next prev parent reply other threads:[~2019-10-18 15:21 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-18 12:46 [PATCH v4] arm64: psci: Reduce the waiting time for cpu_psci_cpu_kill() Yunfeng Ye
2019-10-18 15:20 ` Sudeep Holla [this message]
2019-10-21 7:47 ` Yunfeng Ye
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=20191018152052.GA10312@bogus \
--to=sudeep.holla@arm.com \
--cc=David.Laight@ACULAB.COM \
--cc=catalin.marinas@arm.com \
--cc=gregkh@linuxfoundation.org \
--cc=hushiyuan@huawei.com \
--cc=kstewart@linuxfoundation.org \
--cc=linfeilong@huawei.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lorenzo.pieralisi@arm.com \
--cc=mark.rutland@arm.com \
--cc=tglx@linutronix.de \
--cc=will@kernel.org \
--cc=wuyun.wu@huawei.com \
--cc=yeyunfeng@huawei.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;
as well as URLs for NNTP newsgroup(s).