From: Andrea Corallo <andrea.corallo@arm.com>
To: Segher Boessenkool <segher@kernel.crashing.org>
Cc: Andrew Pinski <pinskia@gmail.com>,
Richard Earnshaw <richard.earnshaw@arm.com>, nd <nd@arm.com>,
GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH 2/2] Aarch64: Add branch diluter pass
Date: Fri, 24 Jul 2020 09:01:33 +0200 [thread overview]
Message-ID: <gkrwo2twg4y.fsf@arm.com> (raw)
In-Reply-To: <20200723224715.GK32057@gate.crashing.org> (Segher Boessenkool's message of "Thu, 23 Jul 2020 17:47:15 -0500")
Segher Boessenkool <segher@kernel.crashing.org> writes:
> On Wed, Jul 22, 2020 at 09:45:08PM +0200, Andrea Corallo wrote:
>> > Should that actually be a sliding window, or should there actually just
>> > not be more than N branches per aligned block of machine code? Like,
>> > per fetch group.
>> >
>> > Can you not use ASM_OUTPUT_ALIGN_WITH_NOP (or ASM_OUTPUT_MAX_SKIP_ALIGN
>> > even) then? GCC has infrastructure for that, already.
>>
>> Correct, it's a sliding window only because the real load address is not
>> known to the compiler and the algorithm is conservative. I believe we
>> could use ASM_OUTPUT_ALIGN_WITH_NOP if we align each function to (al
>> least) the granule size, then we should be able to insert 'nop aligned
>> labels' precisely.
>
> Yeah, we have similar issues on Power... Our "granule" (fetch group
> size, in our terminology) is 32 typically, but we align functions to
> just 16. This is causing some problems, but aligning to bigger
> boundaries isn't a very happy alternative either. WIP...
Interesting, I was expecting other CPUs to have a similar mechanism.
> (We don't have this exact same problem, because our non-ancient cores
> can just predict *all* branches in the same cycle).
>
>> My main fear is that given new cores tend to have big granules code size
>> would blow. One advantage of the implemented algorithm is that even if
>> slightly conservative it's impacting code size only where an high branch
>> density shows up.
>
> What is "big granules" for you?
N1 is 8 instructions so 32 bytes as well, I guess this may grow further
(my speculation).
Andrea
next prev parent reply other threads:[~2020-07-24 7:01 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-22 10:02 [PATCH 1/2] Add new RTX instruction class FILLER_INSN Andrea Corallo
2020-07-22 10:09 ` [PATCH 2/2] Aarch64: Add branch diluter pass Andrea Corallo
2020-07-22 10:39 ` Andrew Pinski
2020-07-22 13:53 ` Andrea Corallo
2020-07-22 16:43 ` Segher Boessenkool
2020-07-22 19:45 ` Andrea Corallo
2020-07-23 22:47 ` Segher Boessenkool
2020-07-24 7:01 ` Andrea Corallo [this message]
2020-07-24 11:53 ` Segher Boessenkool
2020-07-24 13:21 ` Andrea Corallo
2020-07-24 22:09 ` Segher Boessenkool
2020-07-28 18:55 ` Andrea Corallo
2020-07-28 22:07 ` Segher Boessenkool
2020-07-22 12:24 ` [PATCH 1/2] Add new RTX instruction class FILLER_INSN Richard Biener
2020-07-22 13:16 ` Richard Earnshaw (lists)
2020-07-22 14:51 ` Andrea Corallo
2020-07-22 18:41 ` Joseph Myers
2020-07-24 21:18 ` Segher Boessenkool
2020-07-26 18:19 ` Eric Botcazou
2020-07-28 19:29 ` Andrea Corallo
2020-08-19 9:13 ` Andrea Corallo
2020-08-19 10:52 ` Richard Sandiford
2020-08-19 17:28 ` Segher Boessenkool
2020-08-19 16:51 ` Segher Boessenkool
2020-08-19 17:47 ` Andrea Corallo
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=gkrwo2twg4y.fsf@arm.com \
--to=andrea.corallo@arm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=nd@arm.com \
--cc=pinskia@gmail.com \
--cc=richard.earnshaw@arm.com \
--cc=segher@kernel.crashing.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 read-only IMAP folder(s) and NNTP newsgroup(s).