> -----Original Message-----
> From: Richard Biener <rguent...@suse.de>
> Sent: Tuesday, March 18, 2025 10:48 AM
> To: Tamar Christina <tamar.christ...@arm.com>
> Cc: gcc-patches@gcc.gnu.org; nd <n...@arm.com>
> Subject: Re: [PATCH] testsuite: update early-break tests for non-load-lanes
> targets
> [PR119286]
>
> On Mon, 17 Mar 2025, Tamar Christina wrote:
>
> > Hi All,
> >
> > Broadly speaking, these tests were failing because the BB limitation for
> > SLP'ing
> > loads in an || in an early break makes the loads end up in different BBs
> > and so
> > today we can't SLP them. This results in load_lanes being required to
> > vectorize
> > them because the alternative is loads with permutes which we don't allow.
> >
> > The original checks were only checking partial vectors, which ended up
> > working
> > because e.g. Adv. SIMD isn't a partial vector target, so it failed, and SVE
> > was
> > a partial vector target but also has load lanes so it passes.
> >
> > GCN however is a partial vector target without load lanes which makes the
> > tests
> > fail. As we require load_lanes for now, also check for them.
> >
> > Bootstrapped Regtested on aarch64-none-linux-gnu,
> > arm-none-linux-gnueabihf, x86_64-pc-linux-gnu
> > -m32, -m64 and no issues.
>
> Does that reduce the number of PASSes on x86_64?
>
Nope:
13:30:29 Manifest:
/mnt/cache/gcc-regtest-cache/x86_64-linux-gnu/9d68a2a67351fc5b56262c0028ef8fd1d1466627.xfail
13:30:29 Getting actual results from build directory
/opt/buildAgent/temp/buildTmp
13:30:29
/opt/buildAgent/temp/buildTmp/x86_64-pc-linux-gnu/libitm/testsuite/libitm.sum
13:30:29
/opt/buildAgent/temp/buildTmp/x86_64-pc-linux-gnu/libstdc++-v3/testsuite/libstdc++.sum
13:30:29
/opt/buildAgent/temp/buildTmp/x86_64-pc-linux-gnu/libgomp/testsuite/libgomp.sum
13:30:29
/opt/buildAgent/temp/buildTmp/x86_64-pc-linux-gnu/libatomic/testsuite/libatomic.sum
13:30:29 /opt/buildAgent/temp/buildTmp/gcc/testsuite/objc/objc.sum
13:30:29 /opt/buildAgent/temp/buildTmp/gcc/testsuite/gfortran/gfortran.sum
13:30:29 /opt/buildAgent/temp/buildTmp/gcc/testsuite/gcc/gcc.sum
13:30:29 /opt/buildAgent/temp/buildTmp/gcc/testsuite/g++/g++.sum
13:30:29
13:30:29 SUCCESS: No unexpected failures.
13:30:29 Process exited with code 0
And in total Tests passed: 1060737, ignored: 13398
And on trunk Tests failed: 510, passed: 1060227, ignored: 13398 = Tests failed:
510, passed: 1060737, ignored: 13398
So no change in the results on any of the platforms I regtested above.
The reason for this is that we already couldn't vectorize these for non
load_lanes targets, and x86_64 is not a vect_partial_vectors
target. So we already didn't expect them to vectorize on x86_64 today. At
least until we can SLP the loads to create 1 group instead
of two.
Thanks,
Tamar
> > Cross checked the failing cases on amdgcn-amdhsa
> > and all pass now.
> >
> > Ok for master?
> >
> > Thanks,
> > Tamar
> >
> > gcc/testsuite/ChangeLog:
> >
> > PR target/119286
> > * gcc.dg/vect/bb-slp-41.c: Add pragma novector.
> > * gcc.dg/vect/vect-early-break_133_pfa11.c: Should never vectorize today
> > as indexes can be out of range.
> > * gcc.dg/vect/vect-early-break_128.c: Require load_lanes as well.
> > * gcc.dg/vect/vect-early-break_133_pfa10.c: Likewise.
> > * gcc.dg/vect/vect-early-break_133_pfa8.c: Likewise.
> > * gcc.dg/vect/vect-early-break_133_pfa9.c: Likewise.
> > * gcc.dg/vect/vect-early-break_22.c: Likewise.
> > * gcc.dg/vect/vect-early-break_26.c: Likewise.
> > * gcc.dg/vect/vect-early-break_43.c: Likewise.
> > * gcc.dg/vect/vect-early-break_44.c: Likewise.
> > * gcc.dg/vect/vect-early-break_6.c: Likewise.
> > * gcc.dg/vect/vect-early-break_56.c: Expect failures on group misalign.
> >
> > ---
> > diff --git a/gcc/testsuite/gcc.dg/vect/bb-slp-41.c
> b/gcc/testsuite/gcc.dg/vect/bb-slp-41.c
> > index
> 72245115f305de8ef26c9c481f160a05db8c3dcb..5a2bd4d2a33e41dbb33dc105
> b72f0ae6fd0e81ef 100644
> > --- a/gcc/testsuite/gcc.dg/vect/bb-slp-41.c
> > +++ b/gcc/testsuite/gcc.dg/vect/bb-slp-41.c
> > @@ -51,6 +51,7 @@ int main ()
> > foo (a1, b);
> > bar (a2, b);
> >
> > +#pragma GCC novector
> > for (i = 0; i < ARR_SIZE; i++)
> > if (a1[i] != a2[i])
> > return 1;
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_128.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_128.c
> > index
> ed6baf2d451f3887076a1e9143035363128efe70..3d51d52358fc2308ca3614bd
> 3c6fe079f9d839fa 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_128.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_128.c
> > @@ -3,8 +3,8 @@
> > /* { dg-require-effective-target vect_early_break } */
> > /* { dg-require-effective-target vect_int } */
> >
> > -/* { dg-final { scan-tree-dump "vectorizing stmts using SLP" "vect" {
> > target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "vectorizing stmts using SLP" "vect" {
> > target {
> ! vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "vectorizing stmts using SLP" "vect" {
> > target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "vectorizing stmts using SLP" "vect" {
> > target {
> { ! vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > /* { dg-final { scan-tree-dump "Loop contains only SLP stmts" "vect" } } */
> >
> > #ifndef N
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa10.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa10.c
> > index
> dd05046982524f15662be8df517716b581b8a2d9..2a58fb01401282fa857ffd56
> b79fc6e25936d03e 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa10.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa10.c
> > @@ -6,9 +6,9 @@
> > /* { dg-additional-options "-Ofast" } */
> >
> > /* Alignment requirement too big, load lanes targets can't safely
> > vectorize this.
> */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target {
> vect_partial_vectors || vect_load_lanes } } } } */
> > -/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { ! {
> vect_partial_vectors || vect_load_lanes } } } } } */
> > -/* { dg-final { scan-tree-dump-not "Alignment of access forced using
> > peeling"
> "vect" { target { ! { vect_partial_vectors || vect_load_lanes } } } } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target {
> vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { !
> vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "Alignment of access forced using
> > peeling"
> "vect" { target { ! vect_load_lanes } } } } */
> >
> > unsigned test4(char x, char *restrict vect_a, char *restrict vect_b, int n)
> > {
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa11.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa11.c
> > index
> 085dd9b81bb6943440f34d044cbd24ee2121657c..514bd37191744bf1df29c49
> 67d52a9acee605205 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa11.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa11.c
> > @@ -3,9 +3,8 @@
> > /* { dg-require-effective-target vect_early_break } */
> > /* { dg-require-effective-target vect_int } */
> >
> > -/* Gathers and scatters are not save to speculate across early breaks. */
> > -/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { !
> vect_partial_vectors } } } } */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target
> vect_partial_vectors } } } */
> > +/* Gathers and scatters are not safe to speculate across early breaks. */
> > +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" } } */
> >
> > #define N 1024
> > int vect_a[N];
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa8.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa8.c
> > index
> 25d3a62356baf127c89187b150810e4d31567c6f..90e59986dea8ff7a52f16d7b
> 89aa40e26e6b8b15 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa8.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa8.c
> > @@ -5,8 +5,8 @@
> >
> > /* { dg-additional-options "-Ofast" } */
> >
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { !
> vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { { !
> vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> >
> > char vect_a[1025];
> > char vect_b[1025];
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa9.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa9.c
> > index
> 10eb98b726acb32a0d1de4daf202724995bfa1a6..fa2b32e9b8599e4bdfbffe7eaa
> 1f9a80ede7a74c 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa9.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_133_pfa9.c
> > @@ -6,8 +6,8 @@
> > /* { dg-additional-options "-Ofast" } */
> >
> > /* Group size is uneven and second group is misaligned. Needs partial
> > vectors.
> */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { !
> vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { { !
> vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > /* { dg-final { scan-tree-dump-not "Alignment of access forced using
> > peeling"
> "vect" } } */
> >
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_22.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_22.c
> > index
> f8f84fab97ab586847000af8b89448b0885ef5fc..8e91ac6b843bb8718d41f0c0c
> e5ec531039ec996 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_22.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_22.c
> > @@ -43,5 +43,5 @@ main ()
> > }
> >
> > /* This will fail because we cannot SLP the load groups yet. */
> > -/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 2
> > "vect" {
> target vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 1
> > "vect" {
> target { ! vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 2
> > "vect" {
> target { vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 1
> > "vect" {
> target { { ! vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_26.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_26.c
> > index
> 643016b2ccfea29ba36d65c8070f255cb8179481..f6688fc8ead168bae473485c8
> ac43dc35c1ee0c0 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_26.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_26.c
> > @@ -42,5 +42,5 @@ main ()
> > }
> >
> > /* This will fail because we cannot SLP the load groups yet. */
> > -/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 2
> > "vect" {
> target vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 1
> > "vect" {
> target { ! vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 2
> > "vect" {
> target { vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-times "vectorized 1 loops in function" 1
> > "vect" {
> target { { ! vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_43.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_43.c
> > index
> 0cfa2428cc61d5f4ea0784367acea6436736970f..e6866deaa08f423db3f00240e
> 7b4aaff69f38acb 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_43.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_43.c
> > @@ -28,5 +28,5 @@ unsigned test4(unsigned x)
> > }
> >
> > /* This will fail because we cannot SLP the load groups yet. */
> > -/* { dg-final { scan-tree-dump "vectorized 1 loops in function" "vect" {
> > target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "vectorized 1 loops in function" "vect"
> > {
> target { ! vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "vectorized 1 loops in function" "vect" {
> > target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "vectorized 1 loops in function" "vect"
> > {
> target { { ! vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_44.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_44.c
> > index
> 0cfa2428cc61d5f4ea0784367acea6436736970f..e6866deaa08f423db3f00240e
> 7b4aaff69f38acb 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_44.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_44.c
> > @@ -28,5 +28,5 @@ unsigned test4(unsigned x)
> > }
> >
> > /* This will fail because we cannot SLP the load groups yet. */
> > -/* { dg-final { scan-tree-dump "vectorized 1 loops in function" "vect" {
> > target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "vectorized 1 loops in function" "vect"
> > {
> target { ! vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "vectorized 1 loops in function" "vect" {
> > target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "vectorized 1 loops in function" "vect"
> > {
> target { { ! vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_56.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_56.c
> > index
> b35e737fa3b9137cd745c14f7ad915a3f81c38c4..76a1f9b3f55c23100064e151a
> 2bb831d04d5bdc1 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_56.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_56.c
> > @@ -99,5 +99,5 @@ int main (void)
> > return 0;
> > }
> >
> > -/* { dg-final { scan-tree-dump-times "vectorized 2 loops" 2 "vect" { xfail
> > {
> vect_early_break && { ! vect_hw_misalign } } } } } */
> > +/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 2 "vect" { xfail
> > {
> vect_early_break && { ! vect_hw_misalign } } } } } */
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_6.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_6.c
> > index
> c7cce81f52c80d83bd2c1face8cbd13f93834531..485cee63c63e772e5c134568e
> 7c288ea260b69f5 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_6.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_6.c
> > @@ -6,8 +6,8 @@
> > /* { dg-additional-options "-Ofast" } */
> >
> > /* This will fail because we cannot SLP the load groups yet. */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target
> vect_partial_vectors } } } */
> > -/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { !
> vect_partial_vectors } } } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target {
> vect_partial_vectors && vect_load_lanes } } } } */
> > +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" { target { { !
> vect_partial_vectors } || { ! vect_load_lanes } } } } } */
> >
> > #define N 1024
> > unsigned vect_a[N];
> >
> >
> >
>
> --
> Richard Biener <rguent...@suse.de>
> SUSE Software Solutions Germany GmbH,
> Frankenstrasse 146, 90461 Nuernberg, Germany;
> GF: Ivo Totev, Andrew McDonald, Werner Knoblich; (HRB 36809, AG Nuernberg)
