public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Vineet Gupta <vineetg@rivosinc.com>
To: Jeff Law <jeffreyalaw@gmail.com>, gcc-patches@gcc.gnu.org
Cc: kito.cheng@gmail.com, Palmer Dabbelt <palmer@rivosinc.com>,
	gnu-toolchain@rivosinc.com
Subject: Re: [PATCH] RISC-V: zicond: remove bogus opt2 pattern
Date: Thu, 31 Aug 2023 10:57:52 -0700	[thread overview]
Message-ID: <d24bb347-9677-46f6-984a-1af704ea9208@rivosinc.com> (raw)
In-Reply-To: <2b3c7a2a-6e7f-78f9-7453-c2f91ea40b4a@gmail.com>



On 8/31/23 06:51, Jeff Law wrote:
>
>
> On 8/30/23 15:57, Vineet Gupta wrote:
>> This was tripping up gcc.c-torture/execute/pr60003.c at -O1 since the
>> pattern semantics can't be expressed by zicond instructions.
>>
>> This involves test code snippet:
>>
>>        if (a == 0)
>>     return 0;
>>        else
>>     return x;
>>      }
>>
>> which is equivalent to:  "x = (a != 0) ? x : a"
> Isn't it
>
> x = (a == 0) ? 0 : x
>
> Which seems like it ought to fit zicond just fine.

Logically they are equivalent, but ....

>
> If we take yours;
>
> x = (a != 0) ? x : a
>
> And simplify with the known value of a on the false arm we get:
>
> x = (a != 0 ) ? x : 0;
>
> Which is equivalent to
>
> x = (a == 0) ? 0 : x;
>
> So ISTM this does fit zicond just fine.

I could very well be mistaken, but define_insn is a pattern match and 
opt2 has *ne* so the expression has to be in != form and thus needs to 
work with that condition. No ?

>> and matches define_insn "*czero.nez.<GPR:mode><X:mode>.opt2"
>>
>> | (insn 41 20 38 3 (set (reg/v:DI 136 [ x ])
>> |        (if_then_else:DI (ne (reg/v:DI 134 [ a ])
>> |                (const_int 0 [0]))
>> |            (reg/v:DI 136 [ x ])
>> |            (reg/v:DI 134 [ a ]))) {*czero.nez.didi.opt2}
>>
>> The corresponding asm pattern generates
>>      czero.nez x, x, a   ; %0, %2, %1
>> implying
>>      "x = (a != 0) ? 0 : a"
> I get this from the RTL pattern:
>
> x = (a != 0) ? x : a
> x = (a != 0) ? x : 0

This is the issue, for ne, czero.nez can only express
x = (a != 0) ? 0 : x

>
> I think you got the arms reversed.

What I meant was czero.nez as specified in RTL pattern would generate x 
= (a != 0) ? 0 : a, whereas pattern's desired semantics is (a != 0) ? x : 0
And that is a problem because after all equivalents/simplifications, a 
ternary operation's middle operand has to be zero to map to czero*, but 
it doesn't for the opt2 RTL semantics.

I've sat on this for 2 days, trying to convince myself I was wrong, but 
as it stands, it was generating wrong code in the test which is fixed 
after the patch.

Thx,
-Vineet

  reply	other threads:[~2023-08-31 17:57 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-30 21:57 Vineet Gupta
2023-08-31 13:51 ` Jeff Law
2023-08-31 17:57   ` Vineet Gupta [this message]
2023-09-01 13:13     ` Jeff Law
2023-09-01 18:55       ` Vineet Gupta
2023-09-01 17:40     ` Palmer Dabbelt
2023-09-01 19:17       ` Vineet Gupta

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=d24bb347-9677-46f6-984a-1af704ea9208@rivosinc.com \
    --to=vineetg@rivosinc.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=gnu-toolchain@rivosinc.com \
    --cc=jeffreyalaw@gmail.com \
    --cc=kito.cheng@gmail.com \
    --cc=palmer@rivosinc.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).