From: "Kewen.Lin" <linkw@linux.ibm.com>
To: jeevitha <jeevitha@linux.vnet.ibm.com>
Cc: Peter Bergner <bergner@linux.ibm.com>,
GCC Patches <gcc-patches@gcc.gnu.org>,
Segher Boessenkool <segher@kernel.crashing.org>,
David Edelsohn <dje.gcc@gmail.com>
Subject: Re: [PATCH] rs6000: load high and low part of 128bit vector independently [PR110040]
Date: Wed, 22 May 2024 11:20:33 +0800 [thread overview]
Message-ID: <dca8b124-d86b-822d-b83d-cd130da94ae3@linux.ibm.com> (raw)
In-Reply-To: <1a62a215-98e2-4e3b-9059-681189157d0a@linux.vnet.ibm.com>
Hi,
on 2024/2/26 13:43, jeevitha wrote:
> Hi All,
>
> The following patch has been bootstrapped and regtested on powerpc64le-linux.
>
> PR110040 exposes an issue concerning moves from vector registers to GPRs.
> There are two moves, one for upper 64 bits and the other for the lower
> 64 bits. In the problematic test case, we are only interested in storing
> the lower 64 bits. However, the instruction for copying the upper 64 bits
> is still emitted and is dead code. This patch adds a splitter that splits
> apart the two move instructions so that DCE can remove the dead code after
> splitting.
>
> 2024-02-26 Jeevitha Palanisamy <jeevitha@linux.ibm.com>
>
> gcc/
> PR target/110040
> * config/rs6000/vsx.md (split pattern for V1TI to DI move): Defined.
>
> gcc/testsuite/
> PR target/110040
> * gcc.target/powerpc/pr110040-1.c: New testcase.
> * gcc.target/powerpc/pr110040-2.c: New testcase.
>
>
> diff --git a/gcc/config/rs6000/vsx.md b/gcc/config/rs6000/vsx.md
> index 6111cc90eb7..78457f8fb14 100644
> --- a/gcc/config/rs6000/vsx.md
> +++ b/gcc/config/rs6000/vsx.md
> @@ -6706,3 +6706,19 @@
> "vmsumcud %0,%1,%2,%3"
> [(set_attr "type" "veccomplex")]
> )
> +
> +(define_split
> + [(set (match_operand:V1TI 0 "int_reg_operand")
> + (match_operand:V1TI 1 "vsx_register_operand"))]
> + "reload_completed
> + && TARGET_DIRECT_MOVE_64BIT"
> + [(pc)]
> +{
> + rtx op0 = gen_rtx_REG (DImode, REGNO (operands[0]));
> + rtx op1 = gen_rtx_REG (V2DImode, REGNO (operands[1]));
> + rtx op2 = gen_rtx_REG (DImode, REGNO (operands[0]) + 1);
> + rtx op3 = gen_rtx_REG (V2DImode, REGNO (operands[1]));
Nit: op3 is the same as op1 (so useless and removable), just
call it as src_op? Maybe op0 and op2 as dest_op0, dest_op1
to match the extracted index?
> + emit_insn (gen_vsx_extract_v2di (op0, op1, GEN_INT (0)));
> + emit_insn (gen_vsx_extract_v2di (op2, op3, GEN_INT (1)));
Nit: GEN_INT ({0,1}) can be substituted with const{0,1}_rtx.
> + DONE;
> +})
> diff --git a/gcc/testsuite/gcc.target/powerpc/pr110040-1.c b/gcc/testsuite/gcc.target/powerpc/pr110040-1.c
> new file mode 100644
> index 00000000000..fb3bd254636
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/pr110040-1.c
> @@ -0,0 +1,14 @@
> +/* PR target/110040 */
> +/* { dg-do compile } */
> +/* { dg-require-effective-target powerpc_p9vector_ok } */
> +/* { dg-options "-O2 -mdejagnu-cpu=power9" } */
powerpc_p9vector_ok doesn't exist any more, you have to rebase.
This also requires int128 effective target.
Change it to:
/* { dg-options "-O2" } */
/* { dg-require-effective-target int128 } */
/* { dg-require-effective-target powerpc_vsx } */
Since we just check no mfvsrd, I think we can even drop the
"-mdejagnu-cpu=power9", could you test if it works on multiple
environments?
> +/* { dg-final { scan-assembler-not {\mmfvsrd\M} } } */
> +
> +#include <altivec.h>
> +
> +void
> +foo (signed long *dst, vector signed __int128 src)
> +{
> + *dst = (signed long) src[0];
> +}
> +
> diff --git a/gcc/testsuite/gcc.target/powerpc/pr110040-2.c b/gcc/testsuite/gcc.target/powerpc/pr110040-2.c
> new file mode 100644
> index 00000000000..f3aa22be4e8
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/pr110040-2.c
> @@ -0,0 +1,13 @@
> +/* PR target/110040 */
> +/* { dg-do compile } */
> +/* { dg-require-effective-target power10_ok } */
> +/* { dg-options "-O2 -mdejagnu-cpu=power10" } */
> +/* { dg-final { scan-assembler-not {\mmfvsrd\M} } } */
Similar to the above:
/* { dg-options "-O2 -mdejagnu-cpu=power10" } */
/* { dg-require-effective-target int128 } */
/* { dg-require-effective-target powerpc_vsx } */
Also adding one comment why it requires power10.
BR,
Kewen
> +
> +#include <altivec.h>
> +
> +void
> +foo (signed int *dst, vector signed __int128 src)
> +{
> + __builtin_vec_xst_trunc (src, 0, dst);
> +}
>
>
next prev parent reply other threads:[~2024-05-22 3:20 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <a18cccd5-9465-4cc3-85ed-156a72a7236d@linux.vnet.ibm.com>
2024-02-26 5:43 ` jeevitha
2024-03-07 15:24 ` [PING^1][PATCH] " jeevitha
2024-03-26 4:53 ` [PING^2][PATCH] " jeevitha
2024-04-17 9:16 ` [PING^3][PATCH] " jeevitha
2024-05-21 4:57 ` [PING^4][PATCH] " jeevitha
2024-05-22 3:20 ` Kewen.Lin [this message]
2024-06-19 12:39 ` [PATCH V2] " jeevitha
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=dca8b124-d86b-822d-b83d-cd130da94ae3@linux.ibm.com \
--to=linkw@linux.ibm.com \
--cc=bergner@linux.ibm.com \
--cc=dje.gcc@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=jeevitha@linux.vnet.ibm.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).