public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r14-2997] i386: Split SUBREGs of SSE vector registers into vec_select insns.
@ 2023-08-04 15:24 Roger Sayle
0 siblings, 0 replies; only message in thread
From: Roger Sayle @ 2023-08-04 15:24 UTC (permalink / raw)
To: gcc-cvs
https://gcc.gnu.org/g:faa2202ee7fcf039b2016ce5766a2927526c5f78
commit r14-2997-gfaa2202ee7fcf039b2016ce5766a2927526c5f78
Author: Roger Sayle <roger@nextmovesoftware.com>
Date: Fri Aug 4 16:23:38 2023 +0100
i386: Split SUBREGs of SSE vector registers into vec_select insns.
This patch is the final piece in the series to improve the ABI issues
affecting PR 88873. The previous patches tackled inserting DFmode
values into V2DFmode registers, by introducing insvti_{low,high}part
patterns. This patch improves the extraction of DFmode values from
V2DFmode registers via TImode intermediates.
I'd initially thought this would require new extvti_{low,high}part
patterns to be defined, but all that's required is to recognize that
the SUBREG idioms produced by combine are equivalent to (forms of)
vec_select patterns. The target-independent middle-end can't be sure
that the appropriate vec_select instruction exists on the target,
hence doesn't canonicalize a SUBREG of a vector mode as a vec_select,
but the backend can provide a define_split stating where and when
this is useful, for example, considering whether the operand is in
memory, or whether !TARGET_SSE_MATH and the destination is i387.
For pr88873.c, gcc -O2 -march=cascadelake currently generates:
foo: vpunpcklqdq %xmm3, %xmm2, %xmm7
vpunpcklqdq %xmm1, %xmm0, %xmm6
vpunpcklqdq %xmm5, %xmm4, %xmm2
vmovdqa %xmm7, -24(%rsp)
vmovdqa %xmm6, %xmm1
movq -16(%rsp), %rax
vpinsrq $1, %rax, %xmm7, %xmm4
vmovapd %xmm4, %xmm6
vfmadd132pd %xmm1, %xmm2, %xmm6
vmovapd %xmm6, -24(%rsp)
vmovsd -16(%rsp), %xmm1
vmovsd -24(%rsp), %xmm0
ret
with this patch, we now generate:
foo: vpunpcklqdq %xmm1, %xmm0, %xmm6
vpunpcklqdq %xmm3, %xmm2, %xmm7
vpunpcklqdq %xmm5, %xmm4, %xmm2
vmovdqa %xmm6, %xmm1
vfmadd132pd %xmm7, %xmm2, %xmm1
vmovsd %xmm1, %xmm1, %xmm0
vunpckhpd %xmm1, %xmm1, %xmm1
ret
The improvement is even more dramatic when compared to the original
29 instructions shown in comment #8. GCC 13, for example, required
12 transfers to/from memory.
2023-08-04 Roger Sayle <roger@nextmovesoftware.com>
gcc/ChangeLog
* config/i386/sse.md (define_split): Convert highpart:DF extract
from V2DFmode register into a sse2_storehpd instruction.
(define_split): Likewise, convert lowpart:DF extract from V2DF
register into a sse2_storelpd instruction.
gcc/testsuite/ChangeLog
* gcc.target/i386/pr88873.c: Tweak to check for improved code.
Diff:
---
gcc/config/i386/sse.md | 16 ++++++++++++++++
gcc/testsuite/gcc.target/i386/pr88873.c | 2 ++
2 files changed, 18 insertions(+)
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index ab455c3e297..f1712b001f9 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -13554,6 +13554,14 @@
[(set_attr "type" "ssemov")
(set_attr "mode" "V2SF,V4SF,V2SF")])
+;; Convert highpart SUBREG in sse2_storehpd or *vec_extractv2df_1_sse.
+(define_split
+ [(set (match_operand:DF 0 "register_operand")
+ (subreg:DF (match_operand:V2DF 1 "register_operand") 8))]
+ "TARGET_SSE"
+ [(set (match_dup 0)
+ (vec_select:DF (match_dup 1) (parallel [(const_int 1)])))])
+
;; Avoid combining registers from different units in a single alternative,
;; see comment above inline_secondary_memory_needed function in i386.cc
(define_insn "sse2_storelpd"
@@ -13599,6 +13607,14 @@
[(set_attr "type" "ssemov")
(set_attr "mode" "V2SF,V4SF,V2SF")])
+;; Convert lowpart SUBREG into sse2_storelpd or *vec_extractv2df_0_sse.
+(define_split
+ [(set (match_operand:DF 0 "register_operand")
+ (subreg:DF (match_operand:V2DF 1 "register_operand") 0))]
+ "TARGET_SSE"
+ [(set (match_dup 0)
+ (vec_select:DF (match_dup 1) (parallel [(const_int 0)])))])
+
(define_expand "sse2_loadhpd_exp"
[(set (match_operand:V2DF 0 "nonimmediate_operand")
(vec_concat:V2DF
diff --git a/gcc/testsuite/gcc.target/i386/pr88873.c b/gcc/testsuite/gcc.target/i386/pr88873.c
index d893aac714e..a3a7ef242e2 100644
--- a/gcc/testsuite/gcc.target/i386/pr88873.c
+++ b/gcc/testsuite/gcc.target/i386/pr88873.c
@@ -9,3 +9,5 @@ s_t foo (s_t a, s_t b, s_t c)
}
/* { dg-final { scan-assembler-times "vpunpcklqdq" 3 } } */
+/* { dg-final { scan-assembler "vunpckhpd" } } */
+/* { dg-final { scan-assembler-not "rsp" } } */
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2023-08-04 15:24 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-04 15:24 [gcc r14-2997] i386: Split SUBREGs of SSE vector registers into vec_select insns Roger Sayle
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).