From: Jan Hubicka <hubicka@kam.mff.cuni.cz>
To: Feng Xue OS <fxue@os.amperecomputing.com>
Cc: Richard Biener <rguenther@suse.de>,
Xionghu Luo <luoxhu@linux.ibm.com>,
"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>,
"segher@kernel.crashing.org" <segher@kernel.crashing.org>,
"wschmidt@linux.ibm.com" <wschmidt@linux.ibm.com>,
"guojiufu@linux.ibm.com" <guojiufu@linux.ibm.com>,
"linkw@gcc.gnu.org" <linkw@gcc.gnu.org>
Subject: Re: [PATCH] Fix loop split incorrect count and probability
Date: Tue, 26 Oct 2021 15:05:23 +0200 [thread overview]
Message-ID: <20211026130523.GA49464@kam.mff.cuni.cz> (raw)
In-Reply-To: <SN6PR01MB4958F0E90EF1ECED4E66F4E4F7F89@SN6PR01MB4958.prod.exchangelabs.com>
> >
> That said, likely the profile update cannot be done uniformly
> for all blocks of a loop?
For the loop:
for (i = 0; i < n; i = inc (i))
{
if (ga)
ga = do_something ();
}
to:
for (i = 0; i < x; i = inc (i))
{
if (true)
ga = do_something ();
if (!ga)
break;
}
for (; i < n; i = inc (i))
{
if (false)
ga = do_something ();
}
If probability of if (ga) being true is p, then you indeed can scale the
first loop by p and second loop by 1-p.
Imagine that loop has n iterations and it takes m iterations for ga to
become false, then probability of if(ga) is m/n and you get frequencies
with m=n*(m/n) for first loop and n-m=n*(1-n/m) for second loop.
Because the conditional becomes constant true, one needs to scale up the
basic block guarded by the if (true) up by n/m to compensate for the
change. With that the udpate should be right.
Ideally one can bypass scaling of basic block(s) containing
ga = do_something () since the scaling first scales down to m/n
and then scale sup to m/n. Which may not combine to noop.
Perhaps one wants to have a parameter specifying basic blocks on which
the scaling is performed while duplicating for this?
Honza
>
> Richard.
next prev parent reply other threads:[~2021-10-26 13:05 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-03 8:58 Xionghu Luo
2021-08-04 2:42 ` Xionghu Luo
2021-08-06 11:46 ` Richard Biener
2021-08-09 2:37 ` Xionghu Luo
2021-08-10 14:47 ` Richard Biener
2021-08-11 3:03 ` Feng Xue OS
2021-10-26 13:05 ` Jan Hubicka [this message]
2021-10-27 1:42 ` Xionghu Luo
2021-08-11 8:32 ` Xionghu Luo
2021-08-11 9:16 ` Richard Biener
2021-08-12 3:24 ` Xionghu Luo
2021-09-22 8:40 ` Xionghu Luo
2021-09-23 12:17 ` Richard Biener
2021-10-15 5:51 ` Xionghu Luo
2021-10-21 8:43 ` Xionghu Luo
2021-10-21 10:55 ` Richard Biener
2021-10-26 5:40 ` Xionghu Luo
2021-10-26 11:59 ` Richard Biener
2021-10-26 12:19 ` Jan Hubicka
2021-08-09 4:33 ` Feng Xue OS
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=20211026130523.GA49464@kam.mff.cuni.cz \
--to=hubicka@kam.mff.cuni.cz \
--cc=fxue@os.amperecomputing.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=guojiufu@linux.ibm.com \
--cc=linkw@gcc.gnu.org \
--cc=luoxhu@linux.ibm.com \
--cc=rguenther@suse.de \
--cc=segher@kernel.crashing.org \
--cc=wschmidt@linux.ibm.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 read-only IMAP folder(s) and NNTP newsgroup(s).