From: George Spelvin <lkml@SDF.ORG>
To: Dan Williams <dan.j.williams@intel.com>
Cc: Kees Cook <keescook@chromium.org>, Linux MM <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
lkml@sdf.org
Subject: Re: [PATCH v2] mm/shuffle.c: Fix races in add_to_free_area_random()
Date: Wed, 18 Mar 2020 19:29:34 +0000 [thread overview]
Message-ID: <20200318192934.GD2281@SDF.ORG> (raw)
In-Reply-To: <CAPcyv4jxq3VQ+MZKM7p8RiiWUCfB2CaxOu+haAp+7UchgHv2+g@mail.gmail.com>
On Wed, Mar 18, 2020 at 10:36:10AM -0700, Dan Williams wrote:
> On Wed, Mar 18, 2020 at 1:20 AM George Spelvin <lkml@sdf.org> wrote:
>> On Tue, Mar 17, 2020 at 08:53:55PM -0700, Dan Williams wrote:
>>> I had the impression that unless unlikely is "mostly never" then it
>>> can do more harm than good. Is a branch guaranteed to be taken every
>>> BITS_PER_LONG'th occurrence really a candidate for unlikely()
>>> annotation?
>>
>> I had to look this up. GCC manual:
>>
>> For the purposes of branch prediction optimizations, the probability
>> that a '__builtin_expect' expression is 'true' is controlled by GCC's
>> 'builtin-expect-probability' parameter, which defaults to 90%. You can
>> also use '__builtin_expect_with_probability' to explicitly assign a
>> probability value to individual expressions.
>>
>> So I think that <= 10% is good enough, which is true in this case.
>>
>> I was tring to encourage the compiler to:
>> * Place this code path out of line, and
>> * Not do the stack manipulations (build a frame, spill registers)
>> needed for a non-leaf function if this path isn't taken.
>
> Understood, I think it's ok in this case because the shuffling only
> happens for order-10 page free events by default so it will be
> difficult to measure the perf impact either way. But in other kernel
> contexts I think unlikely() annotation should come with numbers, 90%
> not taken is not sufficient in and of itself.
I'm not sure I fully understand your point. I *think* you're
editorializing on unlikely() in general and not this specific code,
but it's a little hard to follow.
Your mention of "order-10 page free events" is confusing. Do you mean
"(order-10 page) free events", i.e. freeing of 1024 consecutive pages?
Or are you using "order" as a synonym for "approximately" and you mean
"approximately 10 (page free event)s"?
We both agree (I hope) that the number here is obvious on brief
inspection: 1/BITS_PER_LONG.
GCC's heuristics are tuned to value cycles on the fast path 9x as much as
cycles on the slow path, so it will spend up to 9 cycles on the slow path
to save a cycle on the fast path.
I've found one comment (https://pastebin.com/S8Y8tqZy) saying that
GCC < 9.x was a lot sloppier on the cost ratio and could pessimize
the code if the branch was more than ~ 1% taken. Perhaps that's what
you're remembering?
Fortunately, 1/64 = 1.56% is fairly close to 1%. so I'm not too
worried.
> You can add:
>
> Acked-by: Dan Williams <dan.j.williams@intel.com>
Thank you!
next prev parent reply other threads:[~2020-03-18 19:29 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-17 13:50 [PATCH] mm/shuffle.c: optimize add_to_free_area_random() George Spelvin
2020-03-17 21:44 ` Kees Cook
2020-03-17 23:06 ` George Spelvin
2020-03-17 23:38 ` Kees Cook
2020-03-18 1:44 ` [PATCH v2] mm/shuffle.c: Fix races in add_to_free_area_random() George Spelvin
2020-03-18 1:49 ` Randy Dunlap
2020-03-18 3:53 ` Dan Williams
2020-03-18 8:20 ` George Spelvin
2020-03-18 17:36 ` Dan Williams
2020-03-18 19:29 ` George Spelvin [this message]
2020-03-18 19:40 ` Dan Williams
2020-03-18 21:02 ` George Spelvin
2020-03-18 3:58 ` Kees Cook
2020-03-18 15:26 ` Alexander Duyck
2020-03-18 18:35 ` George Spelvin
2020-03-18 19:17 ` Alexander Duyck
2020-03-18 20:06 ` George Spelvin
2020-03-18 20:39 ` [PATCH v3] " George Spelvin
2020-03-18 21:34 ` Alexander Duyck
2020-03-18 22:49 ` George Spelvin
2020-03-18 22:57 ` Dan Williams
2020-03-18 23:18 ` George Spelvin
2020-03-19 12:05 ` [PATCH v4] " George Spelvin
2020-03-19 17:49 ` Alexander Duyck
2020-03-20 17:58 ` Kees Cook
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=20200318192934.GD2281@SDF.ORG \
--to=lkml@sdf.org \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=keescook@chromium.org \
--cc=linux-mm@kvack.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).