From: Denis Efremov <efremov@ispras.ru>
To: Matthew Wilcox <willy@infradead.org>
Cc: akpm@linux-foundation.org, Akinobu Mita <akinobu.mita@gmail.com>,
Jan Kara <jack@suse.cz>,
linux-kernel@vger.kernel.org, Matthew Wilcox <matthew@wil.cx>,
dm-devel@redhat.com, linux-fsdevel@vger.kernel.org,
linux-media@vger.kernel.org, Erdem Tumurov <erdemus@gmail.com>,
Vladimir Shelekhov <vshel@iis.nsk.su>
Subject: Re: [PATCH v2] lib/memweight.c: open codes bitmap_weight()
Date: Sun, 25 Aug 2019 14:39:47 +0300 [thread overview]
Message-ID: <ba051566-0343-ea75-0484-8852f65a15da@ispras.ru> (raw)
In-Reply-To: <20190825061158.GC28002@bombadil.infradead.org>
On 25.08.2019 09:11, Matthew Wilcox wrote:
> On Sat, Aug 24, 2019 at 01:01:02PM +0300, Denis Efremov wrote:
>> This patch open codes the bitmap_weight() call. The direct
>> invocation of hweight_long() allows to remove the BUG_ON and
>> excessive "longs to bits, bits to longs" conversion.
>
> Honestly, that's not the problem with this function. Take a look
> at https://danluu.com/assembly-intrinsics/ for a _benchmarked_
> set of problems with popcnt.
>
>> BUG_ON was required to check that bitmap_weight() will return
>> a correct value, i.e. the computed weight will fit the int type
>> of the return value.
>
> What? No. Look at the _arguments_ of bitmap_weight():
>
> static __always_inline int bitmap_weight(const unsigned long *src, unsigned int nbits)
I'm not sure why it is INT_MAX then? I would expect in case we care only about arguments
something like:
BUG_ON(longs >= UINT_MAX / BITS_PER_LONG);
>
>> With this patch memweight() controls the
>> computation directly with size_t type everywhere. Thus, the BUG_ON
>> becomes unnecessary.
>
> Why are you bothering? How are you allocating half a gigabyte of memory?
> Why are you calling memweight() on half a gigabyte of memory?
>
No, we don't use such big arrays. However, it's possible to remove BUG_ON and make
the code more "straight". Why do we need to "artificially" limit this function
to arrays of a particular size if we can relatively simple omit this restriction?
>
> If you really must change anything, I'd rather see this turned into a
> loop:
>
> while (longs) {
> unsigned int nbits;
>
> if (longs >= INT_MAX / BITS_PER_LONG)
> nbits = INT_MAX + 1;
> else
> nbits = longs * BITS_PER_LONG;
>
> ret += bitmap_weight((unsigned long *)bitmap, sz);
> bytes -= nbits / 8;
> bitmap += nbits / 8;
> longs -= nbits / BITS_PER_LONG;
> }
>
> then we only have to use Dan Luu's optimisation in bitmap_weight()
> and not in memweight() as well.
I don't know how the implementation of this optimization will look like in it's
final shape, because of different hardware/compiler issues. It looks there are
a number of different ways to do it https://arxiv.org/pdf/1611.07612.pdf,
http://0x80.pl/articles/sse-popcount.html.
However, if it will be based on popcnt instruction I would expect that
hweight_long will also contain this intrinsics. Since version 4.9.2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62011#c13 GCC knows of the
false-dependency in popcnt and generates code to handle it
(e.g. xor https://godbolt.org/z/Q7AW_d) Thus, I would expect that it's
possible to use popcnt intrinsics in hweight_long that would be natively
optimized in all loops like "for (...) { res += hweight_long() }" without
requiring manual unrolling like in builtin_popcnt_unrolled_errata_manual
example of Dan Luu's optimization.
>
> Also, why does the trailer do this:
>
> for (; bytes > 0; bytes--, bitmap++)
> ret += hweight8(*bitmap);
>
> instead of calling hweight_long on *bitmap & mask?
>
Do you mean something like this?
longs = bytes;
bytes = do_div(longs, sizeof(long));
bitmap_long = (const unsigned long *)bitmap;
if (longs) {
for (; longs > 0; longs--, bitmap_long++)
ret += hweight_long(*bitmap_long);
}
if (bytes) {
ret += hweight_long(*bitmap_long &
((0x1 << bytes * BITS_PER_BYTE) - 1));
}
The *bitmap_long will lead to buffer overflow here.
Thanks,
Denis
next prev parent reply other threads:[~2019-08-25 11:39 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-21 7:42 [PATCH] lib/memweight.c: optimize by inlining bitmap_weight() Denis Efremov
2019-08-22 1:25 ` Andrew Morton
2019-08-22 7:30 ` Denis Efremov
2019-08-24 10:01 ` [PATCH v2] lib/memweight.c: open codes bitmap_weight() Denis Efremov
2019-08-25 6:11 ` Matthew Wilcox
2019-08-25 11:39 ` Denis Efremov [this message]
2019-08-26 18:39 ` Matthew Wilcox
2019-09-13 11:48 ` Denis Efremov
2019-09-13 13:41 ` efremov
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=ba051566-0343-ea75-0484-8852f65a15da@ispras.ru \
--to=efremov@ispras.ru \
--cc=akinobu.mita@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=dm-devel@redhat.com \
--cc=erdemus@gmail.com \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=matthew@wil.cx \
--cc=vshel@iis.nsk.su \
--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 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).