From: Jeff Law <jeffreyalaw@gmail.com>
To: Segher Boessenkool <segher@kernel.crashing.org>
Cc: gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] Always enable LRA
Date: Fri, 14 Oct 2022 11:07:43 -0600 [thread overview]
Message-ID: <c112f8aa-abd2-479b-ceb3-ad7979269529@gmail.com> (raw)
In-Reply-To: <20221014163700.GE25951@gate.crashing.org>
On 10/14/22 10:37, Segher Boessenkool wrote:
> Hi!
>
> On Thu, Oct 13, 2022 at 10:47:20PM -0600, Jeff Law wrote:
>> On 10/13/22 17:56, Segher Boessenkool wrote:
>>> h8300 fails during GCC build:
>>> /home/segher/src/gcc/libgcc/unwind.inc: In function
>>> '_Unwind_SjLj_RaiseException':
>>> /home/segher/src/gcc/libgcc/unwind.inc:141:1: error: could not split insn
>>> 141 | }
>>> | ^
>>> (insn 69 256 327 (set (mem/f:SI (pre_dec:SI (reg/f:SI 7 sp)) [12 S4 A32])
>>> (reg/f:SI 7 sp)) "/home/segher/src/gcc/libgcc/unwind.inc":118:12
>>> 19 {*movsi}
>>> (expr_list:REG_ARGS_SIZE (const_int 4 [0x4])
>>> (nil)))
>>> during RTL pass: final
>>> which looks like a backend bug, I don't see a pattern that could split
>>> this (without needing an extra clobber)?
>> Really smells like an LRA bug to me.
>>
>>
>> We have this as we leave IRA:
>>
>>
>> (insn 10 9 11 2 (set (reg/f:SI 30)
>> (plus:SI (reg/f:SI 11 fp)
>> (const_int -4 [0xfffffffffffffffc]))) "j.c":31:7 264
>> {*addsi}
>> (expr_list:REG_EQUIV (plus:SI (reg/f:SI 11 fp)
>> (const_int -4 [0xfffffffffffffffc]))
>> (nil)))
>> (insn 11 10 12 2 (set (mem/f:SI (pre_dec:SI (reg/f:SI 7 sp)) [3 S4 A32])
>> (reg/f:SI 30)) "j.c":31:7 19 {*movsi}
>> (expr_list:REG_DEAD (reg/f:SI 30)
>> (expr_list:REG_ARGS_SIZE (const_int 4 [0x4])
>> (nil))))
>>
>>
>> And as we leave LRA:
>>
>> (note 10 9 11 2 NOTE_INSN_DELETED)
>> (insn 11 10 13 2 (set (mem/f:SI (pre_dec:SI (reg/f:SI 7 sp)) [3 S4 A32])
>> (reg/f:SI 7 sp)) "j.c":31:7 19 {*movsi}
>> (expr_list:REG_ARGS_SIZE (const_int 4 [0x4])
>> (nil)))
> LRA only ever generates insns that pass recog. The backend allows this
> define_insn, requiring it to be split (it returns template "#"), but
> then somehow it doesn't match in any split pass?
Nope. The elimination code will just change one register without
re-recognizing. That's precisely what happens here.
>
>> Register elimination ultimately discovered that (reg 30) was the same as
>> the stack pointer and did the natural substitution. The natural
>> substitution results in invalid RTL and there's really no way to back
>> out and do something different AFAICT in lra-eliminations.cc.
>>
>> The only reason we fault is because the H8 backend knows this is invalid
>> RTL and refuses to split it. If we were to try and re-recognize the
>> insn in question it would fail to recognize after the substitution
>> because the auto-inc'd operand overlaps with the other operand.
> But it *did* recog? Or does LRA somehow not always recog() everything?
> I always thought that was one of the huge improvements over old reload
> (it does everything very locally instead of very globally)!
No, LRA does not force re-recognition in some cases, particularly for
register eliminations.
jeff
next prev parent reply other threads:[~2022-10-14 17:07 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-13 23:56 Segher Boessenkool
2022-10-14 0:36 ` Koning, Paul
2022-10-14 16:18 ` Segher Boessenkool
2022-10-14 16:48 ` Koning, Paul
2022-10-14 1:07 ` Jeff Law
2022-10-14 12:37 ` Koning, Paul
2022-10-14 14:38 ` Jeff Law
2022-10-14 16:37 ` Koning, Paul
2022-10-14 17:10 ` Jeff Law
2022-10-14 17:36 ` Koning, Paul
2022-10-14 21:15 ` Jeff Law
2022-10-14 21:21 ` Koning, Paul
2022-10-14 21:30 ` Jeff Law
2022-10-15 0:19 ` Jeff Law
2022-10-14 16:39 ` Richard Biener
2022-10-14 17:11 ` Jeff Law
2022-10-14 4:47 ` Jeff Law
2022-10-14 16:37 ` Segher Boessenkool
2022-10-14 17:07 ` Jeff Law [this message]
2022-10-14 17:35 ` Segher Boessenkool
2022-10-14 18:03 ` Jeff Law
2022-10-14 19:58 ` Koning, Paul
2022-10-14 20:12 ` Segher Boessenkool
2022-10-14 20:40 ` Koning, Paul
2022-10-14 6:20 ` Takayuki 'January June' Suwa
2022-10-14 16:25 ` Segher Boessenkool
2022-10-15 3:18 ` Takayuki 'January June' Suwa
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=c112f8aa-abd2-479b-ceb3-ad7979269529@gmail.com \
--to=jeffreyalaw@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--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).