From: Boaz Harrosh <openosd@gmail.com>
To: Matthew Wilcox <willy@infradead.org>,
Dan Williams <dan.j.williams@intel.com>
Cc: Jan Kara <jack@suse.cz>, linux-nvdimm <linux-nvdimm@lists.01.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
stable <stable@vger.kernel.org>,
Robert Barror <robert.barror@intel.com>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Seema Pandit <seema.pandit@intel.com>
Subject: Re: [PATCH] dax: Fix missed PMD wakeups
Date: Thu, 4 Jul 2019 16:00:00 +0300 [thread overview]
Message-ID: <f23a1c71-d1b1-b279-c922-ce0f48cb4448@gmail.com> (raw)
In-Reply-To: <20190704032728.GK1729@bombadil.infradead.org>
On 04/07/2019 06:27, Matthew Wilcox wrote:
> On Wed, Jul 03, 2019 at 02:28:41PM -0700, Dan Williams wrote:
<>
>>> +#ifdef CONFIG_XARRAY_MULTI
>>> + unsigned int sibs = xas->xa_sibs;
>>> +
>>> + while (sibs) {
>>> + order++;
>>> + sibs /= 2;
>>> + }
>>
>> Use ilog2() here?
>
> Thought about it. sibs is never going to be more than 31, so I don't
> know that it's worth eliminating 5 add/shift pairs in favour of whatever
> the ilog2 instruction is on a given CPU. In practice, on x86, sibs is
> going to be either 0 (PTEs) or 7 (PMDs). We could also avoid even having
> this function by passing PMD_ORDER or PTE_ORDER into get_unlocked_entry().
>
> It's probably never going to be noticable in this scenario because it's
> the very last thing checked before we put ourselves on a waitqueue and
> go to sleep.
>
Matthew you must be kidding an ilog2 in binary is zero clocks
(Return the highest bit or something like that)
In any way. It took me 5 minutes to understand what you are doing
here. And I only fully got it when Dan gave his comment. So please for
the sake of stupid guys like me could you please make it ilog2() so
to make it easier to understand?
(And please don't do the compiler's job. If in some arch the loop
is the fastest let the compiler decide?)
Thanks
Boaz
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
WARNING: multiple messages have this Message-ID (diff)
From: Boaz Harrosh <openosd@gmail.com>
To: Matthew Wilcox <willy@infradead.org>,
Dan Williams <dan.j.williams@intel.com>
Cc: linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Jan Kara <jack@suse.cz>, stable <stable@vger.kernel.org>,
Robert Barror <robert.barror@intel.com>,
Seema Pandit <seema.pandit@intel.com>,
linux-nvdimm <linux-nvdimm@lists.01.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] dax: Fix missed PMD wakeups
Date: Thu, 4 Jul 2019 16:00:00 +0300 [thread overview]
Message-ID: <f23a1c71-d1b1-b279-c922-ce0f48cb4448@gmail.com> (raw)
In-Reply-To: <20190704032728.GK1729@bombadil.infradead.org>
On 04/07/2019 06:27, Matthew Wilcox wrote:
> On Wed, Jul 03, 2019 at 02:28:41PM -0700, Dan Williams wrote:
<>
>>> +#ifdef CONFIG_XARRAY_MULTI
>>> + unsigned int sibs = xas->xa_sibs;
>>> +
>>> + while (sibs) {
>>> + order++;
>>> + sibs /= 2;
>>> + }
>>
>> Use ilog2() here?
>
> Thought about it. sibs is never going to be more than 31, so I don't
> know that it's worth eliminating 5 add/shift pairs in favour of whatever
> the ilog2 instruction is on a given CPU. In practice, on x86, sibs is
> going to be either 0 (PTEs) or 7 (PMDs). We could also avoid even having
> this function by passing PMD_ORDER or PTE_ORDER into get_unlocked_entry().
>
> It's probably never going to be noticable in this scenario because it's
> the very last thing checked before we put ourselves on a waitqueue and
> go to sleep.
>
Matthew you must be kidding an ilog2 in binary is zero clocks
(Return the highest bit or something like that)
In any way. It took me 5 minutes to understand what you are doing
here. And I only fully got it when Dan gave his comment. So please for
the sake of stupid guys like me could you please make it ilog2() so
to make it easier to understand?
(And please don't do the compiler's job. If in some arch the loop
is the fastest let the compiler decide?)
Thanks
Boaz
next prev parent reply other threads:[~2019-07-04 13:00 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-03 7:24 [PATCH] dax: Fix missed PMD wakeups Dan Williams
2019-07-03 7:24 ` Dan Williams
2019-07-03 12:17 ` Matthew Wilcox
2019-07-03 12:17 ` Matthew Wilcox
2019-07-03 17:01 ` Dan Williams
2019-07-03 17:01 ` Dan Williams
2019-07-03 19:53 ` Matthew Wilcox
2019-07-03 19:53 ` Matthew Wilcox
2019-07-03 21:28 ` Dan Williams
2019-07-03 21:28 ` Dan Williams
2019-07-04 3:27 ` Matthew Wilcox
2019-07-04 3:27 ` Matthew Wilcox
2019-07-04 13:00 ` Boaz Harrosh [this message]
2019-07-04 13:00 ` Boaz Harrosh
2019-07-04 13:58 ` Matthew Wilcox
2019-07-04 13:58 ` Matthew Wilcox
2019-07-04 14:32 ` Boaz Harrosh
2019-07-04 14:32 ` Boaz Harrosh
2019-07-04 16:54 ` Jan Kara
2019-07-04 16:54 ` Jan Kara
2019-07-04 19:14 ` Matthew Wilcox
2019-07-04 19:14 ` Matthew Wilcox
2019-07-04 23:27 ` Dan Williams
2019-07-04 23:27 ` Dan Williams
2019-07-05 19:10 ` Matthew Wilcox
2019-07-05 19:10 ` Matthew Wilcox
2019-07-05 20:47 ` Dan Williams
2019-07-05 20:47 ` Dan Williams
2019-07-10 19:02 ` Jan Kara
2019-07-10 19:02 ` Jan Kara
2019-07-10 20:15 ` Matthew Wilcox
2019-07-10 20:15 ` Matthew Wilcox
2019-07-10 20:26 ` Jan Kara
2019-07-10 20:26 ` Jan Kara
2019-07-11 14:13 ` Matthew Wilcox
2019-07-11 14:13 ` Matthew Wilcox
2019-07-11 15:25 ` Matthew Wilcox
2019-07-11 15:25 ` Matthew Wilcox
2019-07-11 15:41 ` Jan Kara
2019-07-11 15:41 ` Jan Kara
2019-07-17 3:39 ` Dan Williams
2019-07-17 3:39 ` Dan Williams
2019-07-29 12:02 ` Jan Kara
2019-07-29 12:02 ` Jan Kara
2019-07-29 15:18 ` Dan Williams
2019-07-29 15:18 ` Dan Williams
2019-07-11 3:08 ` Matthew Wilcox
2019-07-11 3:08 ` Matthew Wilcox
2019-07-11 7:48 ` Jan Kara
2019-07-11 7:48 ` Jan Kara
2019-07-11 13:28 ` Matthew Wilcox
2019-07-11 13:28 ` Matthew Wilcox
2019-07-11 3:35 ` Matthew Wilcox
2019-07-11 3:35 ` Matthew Wilcox
2019-07-11 8:06 ` Jan Kara
2019-07-11 8:06 ` Jan Kara
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=f23a1c71-d1b1-b279-c922-ce0f48cb4448@gmail.com \
--to=openosd@gmail.com \
--cc=dan.j.williams@intel.com \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@lists.01.org \
--cc=robert.barror@intel.com \
--cc=seema.pandit@intel.com \
--cc=stable@vger.kernel.org \
--cc=willy@infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.