Tom de Vries 065e25f633 [libgomp, testsuite, nvptx] Fix dg-output test in vector-length-128-7.c ... When running test-case libgomp.oacc-c-c++-common/vector-length-128-7.c on an RTX A2000 (sm_86) with driver 510.60.02 I run into: ... FAIL: libgomp.oacc-c/../libgomp.oacc-c-c++-common/vector-length-128-7.c \ -DACC_DEVICE_TYPE_nvidia=1 -DACC_MEM_SHARED=0 -foffload=nvptx-none -O0 \ output pattern test ... The failing check verifies the launch dimensions: ... /* { dg-output "nvptx_exec: kernel main\\\$_omp_fn\\\$0: \ launch gangs=1, workers=8, vectors=128" } */ ... which fails because (as we can see with GOMP_DEBUG=1) the actual num_workers is 6: ... nvptx_exec: kernel main$_omp_fn$0: launch gangs=1, workers=6, vectors=128 ... This is due to the result of cuOccupancyMaxPotentialBlockSize (which suggests 'a launch configuration with reasonable occupancy') printed just before: ... cuOccupancyMaxPotentialBlockSize: grid = 52, block = 768 ... [ Note: 6 * 128 == 768. ] Fix this by updating the check to allow num_workers in the range 1 to 8. Tested on x86_64 with nvptx accelerator. libgomp/ChangeLog: 2022-04-01 Tom de Vries <tdevries@suse.de> * testsuite/libgomp.oacc-c-c++-common/vector-length-128-7.c: Fix num_workers check.		2022-04-01 13:22:07 +02:00
..
abort-1.c
abort-2.c
abort-3.c
abort-4.c
abort-5.c
acc_free-pr92503-1.c
acc_free-pr92503-2.c
acc_free-pr92503-3-2.c
acc_free-pr92503-3.c
acc_free-pr92503-4-2.c
acc_free-pr92503-4.c
acc_get_property-aux.c
acc_get_property-gcn.c
acc_get_property-host.c
acc_get_property-nvptx.c
acc_get_property.c
acc_map_data-device_already-1.c
acc_map_data-device_already-2.c
acc_map_data-device_already-3.c
acc_map_data-host_already-1.c
acc_map_data-host_already-2.c
acc_map_data-host_already-3.c
acc_on_device-1.c
acc_prof-dispatch-1.c
acc_prof-init-1.c	[OpenACC] Clarify sequencing of 'async' data copying vs. profiling events in 'libgomp.oacc-c-c++-common/acc_prof-{init,parallel}-1.c'	2021-07-27 11:16:25 +02:00
acc_prof-init-2.c
acc_prof-kernels-1.c	Enhance further testcases to verify Openacc 'kernels' decomposition	2022-03-17 08:51:32 +01:00
acc_prof-parallel-1.c	[OpenACC] Clarify sequencing of 'async' data copying vs. profiling events in 'libgomp.oacc-c-c++-common/acc_prof-{init,parallel}-1.c'	2021-07-27 11:16:25 +02:00
acc_prof-valid_bytes-1.c
acc_prof-version-1.c
acc_set_cuda_stream-1.c
acc_unmap_data-pr92840-1.c
acc_unmap_data-pr92840-2.c
acc_unmap_data-pr92840-3.c
acc-on-device-2.c
acc-on-device.c
async_queue-1.c
async-data-1-1.c	Fix OpenACC "ephemeral" asynchronous host-to-device copies	2021-07-27 11:16:27 +02:00
async-data-1-2.c	Don't use libgomp 'cbuf' buffering with OpenACC 'async'	2021-07-27 11:16:37 +02:00
asyncwait-1.c
asyncwait-nop-1.c
atomic_capture-1.c
atomic_capture-2.c
atomic_capture-3.c
atomic_rw-1.c
atomic_update-1.c
broadcast-1.c
broadcast-many.c	[libgomp, testsuite] Fix insufficient resources in test-cases	2022-02-01 08:15:00 +01:00
cache-1.c
clauses-1.c
clauses-2.c
collapse-1.c
collapse-2.c
collapse-3.c
collapse-4.c
combined-directives-1.c
combined-reduction.c
context-1.c
context-2.c
context-3.c
context-4.c
crash-1.c
data-1.c
data-2-lib.c
data-2.c
data-3.c
data-clauses-kernels-ipa-pta.c
data-clauses-kernels.c
data-clauses-parallel-ipa-pta.c
data-clauses-parallel.c
data-clauses.h
data-firstprivate-1.c
declare-1.c
declare-2.c
declare-3.c
declare-4.c
declare-5.c
declare-vla.c	OpenACC 'kernels' decomposition: resolve wrong-code cases unless manually making certain variables addressable [PR100280, PR104892]	2022-03-12 15:37:27 +01:00
deep-copy-1.c
deep-copy-2.c
deep-copy-3.c
deep-copy-4.c
deep-copy-5.c
deep-copy-6.c
deep-copy-7.c
deep-copy-8.c
deep-copy-9.c
deep-copy-10.c
deep-copy-11.c
deep-copy-14.c
default-1.c	OpenACC 'kernels' decomposition: resolve wrong-code cases unless manually making certain variables addressable [PR100280, PR104892]	2022-03-12 15:37:27 +01:00
deviceptr-1.c
enter_exit-lib.c
enter-data.c
f-asyncwait-1.c	OpenACC 'kernels' decomposition: Move 'TREE_ADDRESSABLE' setting into OMP lowering [PR100280]	2022-03-04 14:21:01 +01:00
f-asyncwait-2.c	Enhance OpenACC 'kernels' decomposition testing	2022-01-13 10:42:17 +01:00
f-asyncwait-3.c	Enhance OpenACC 'kernels' decomposition testing	2022-01-13 10:42:17 +01:00
firstprivate-1.c
firstprivate-mappings-1.c
function-not-offloaded.c
gang-reduction-var-assignment.c
gang-static-1.c
gang-static-2.c
gomp-debug-env.c
host_data-1.c
host_data-2.c
host_data-4.c
host_data-5.c
host_data-6.c
host_data-7.c	OpenACC privatization diagnostics vs. 'assert' [PR102841]	2022-03-16 10:12:09 +01:00
if-1.c
insufficient-resources.c
kernels-alias-ipa-pta-2.c
kernels-alias-ipa-pta-3.c
kernels-alias-ipa-pta.c
kernels-decompose-1.c	OpenACC 'kernels' decomposition: resolve wrong-code cases unless manually making certain variables addressable [PR100280, PR104892]	2022-03-12 15:37:27 +01:00
kernels-empty.c
kernels-loop-2.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
kernels-loop-3.c
kernels-loop-and-seq-2.c
kernels-loop-and-seq-3.c
kernels-loop-and-seq-4.c
kernels-loop-and-seq-5.c
kernels-loop-and-seq-6.c
kernels-loop-and-seq.c
kernels-loop-clauses.c
kernels-loop-collapse.c
kernels-loop-data-2.c
kernels-loop-data-enter-exit-2.c
kernels-loop-data-enter-exit.c
kernels-loop-data-update.c
kernels-loop-data.c
kernels-loop-g.c	Enhance further testcases to verify Openacc 'kernels' decomposition	2022-03-17 08:51:32 +01:00
kernels-loop-mod-not-zero.c
kernels-loop-n.c
kernels-loop-nest.c
kernels-loop.c
kernels-parallel-loop-data-enter-exit.c
kernels-private-vars-local-worker-1.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-local-worker-2.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-local-worker-3.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-local-worker-4.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-local-worker-5.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-gang-1.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-gang-2.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-gang-3.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-gang-4.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-gang-5.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-gang-6.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-vector-1.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-vector-2.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-1.c	Test 'libgomp.oacc-*/kernels-private-vars-*' with '--param=openacc-kernels=decompose' [PR104784]	2022-03-04 15:47:06 +01:00
kernels-private-vars-loop-worker-2.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-3.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-4.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-5.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-6.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-private-vars-loop-worker-7.c	[OpenACC privatization] Analyze 'lookup_decl'-translated DECL [PR90115, PR102330, PR104774]	2022-03-10 12:06:28 +01:00
kernels-reduction-1.c	OpenACC 'kernels' decomposition: resolve wrong-code cases unless manually making certain variables addressable [PR100280, PR104892]	2022-03-12 15:37:27 +01:00
kernels-reduction.c
lib-1.c
lib-2.c
lib-3.c
lib-4.c
lib-5.c
lib-6.c
lib-7.c
lib-8.c
lib-9.c
lib-10.c
lib-11.c
lib-12.c
lib-13.c
lib-14.c
lib-15.c
lib-16.c
lib-19.c
lib-20.c
lib-23.c
lib-24.c
lib-25.c
lib-26.c
lib-27.c
lib-31.c
lib-32.c
lib-33.c
lib-34.c
lib-35.c
lib-36.c
lib-37.c
lib-39.c
lib-40.c
lib-41.c
lib-42.c
lib-44.c
lib-45.c
lib-46.c
lib-48.c
lib-49.c
lib-51.c
lib-52.c
lib-53.c
lib-54.c
lib-55.c
lib-56.c
lib-57.c
lib-58.c
lib-59.c
lib-60.c
lib-61.c
lib-62.c
lib-63.c
lib-64.c
lib-65.c
lib-66.c
lib-67.c
lib-68.c
lib-69.c
lib-70.c
lib-72.c
lib-73.c
lib-74.c
lib-75.c
lib-76.c
lib-78.c
lib-79.c
lib-81.c
lib-82.c
lib-83.c
lib-84.c
lib-85.c
lib-86.c
lib-87.c
lib-88.c
lib-89.c
lib-90.c
lib-91.c
lib-92.c
lib-94.c
lib-95.c
loop-auto-1.c
loop-default-runtime.c
loop-default.h
loop-dim-default.c	amdgcn: Enable OpenACC worker partitioning for AMD GCN	2021-08-09 15:08:44 +02:00
loop-g-1.c
loop-g-2.c
loop-gwv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-gwv-2.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-g-1.c
loop-red-gwv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-v-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-v-2.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-w-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-w-2.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-red-wv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-v-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-w-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
loop-wv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
map-data-1.c
mapping-1.c
mdc-refcount-1.c
mdc-refcount-2.c
mdc-refcount-3.c
mode-transitions.c	Address '?:' issues in 'libgomp.oacc-c-c++-common/mode-transitions.c'	2021-08-16 12:12:09 +02:00
nested-1.c
nested-2.c
no_create-1.c
no_create-2.c
no_create-3.c
no_create-4.c
no_create-5.c
nvptx-merged-loop.c
nvptx-sese-1.c
offset-1.c
par-loop-comb-reduction-1.c
par-loop-comb-reduction-2.c
par-loop-comb-reduction-3.c
par-loop-comb-reduction-4.c	[libgomp, testsuite] Fix insufficient resources in test-cases	2022-02-01 08:15:00 +01:00
par-reduction-1.c
par-reduction-2.c
parallel-dims.c	[libgomp, testsuite] Scale down some OpenACC test-cases	2022-03-25 13:51:48 +01:00
parallel-empty.c
parallel-loop-1.c
parallel-loop-1.h
parallel-loop-2.h
parallel-reduction.c
pointer-align-1.c
pr70289.c
pr70373.c
pr70688.c
pr83046.c
pr83589.c
pr83920.c
pr84217.c
pr84955-1.c
pr84955.c
pr85381-2.c
pr85381-3.c
pr85381-4.c
pr85381-5.c
pr85381.c
pr85422.c
pr85486-2.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
pr85486-3.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
pr85486.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
pr85782.c
pr87835.c
pr88288.c
pr88941.c
pr88946.c
pr89376.c
pr90009.c
pr92726-1.c
pr92843-1.c
pr92848-1-d-a.c
pr92848-1-d-p.c
pr92848-1-r-a.c
pr92848-1-r-p.c
pr92854-1.c
pr92877-1.c
pr92970-1.c
pr92984-1.c
pr95270-1.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
pr95270-2.c
present-1.c
present-2.c
private-atomic-1-gang.c
private-atomic-1.c	[nvptx] Add some support for .local atomics	2022-02-01 19:28:24 +01:00
private-variables.c
reduction-1.c
reduction-2.c
reduction-3.c
reduction-4.c
reduction-5.c
reduction-6.c
reduction-7.c	[libgomp, testsuite] Fix insufficient resources in test-cases	2022-02-01 08:15:00 +01:00
reduction-8.c
reduction-cplx-dbl.c
reduction-cplx-flt.c
reduction-dbl.c
reduction-flt.c
reduction-initial-1.c
reduction.h
refcounting-1.c
refcounting-2.c
routine-1.c
routine-4.c
routine-g-1.c
routine-gwv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
routine-nohost-1.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
routine-nohost-2_2.c
routine-nohost-2.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
routine-v-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
routine-w-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
routine-wv-1.c	Strengthen a few OpenACC test cases	2022-01-21 18:45:30 +01:00
routine-wv-2.c	amdgcn: Enable OpenACC worker partitioning for AMD GCN	2021-08-09 15:08:44 +02:00
static-variable-1.c	Adjust 'libgomp.oacc-c-c++-common/static-variable-1.c'	2021-08-13 22:53:58 +02:00
struct-1.c
struct-3-1-1.c
struct-copyout-1.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
struct-copyout-2.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
structured-detach-underflow.c
structured-dynamic-lifetimes-1-lib.c
structured-dynamic-lifetimes-1.c
structured-dynamic-lifetimes-2-lib.c
structured-dynamic-lifetimes-2.c
structured-dynamic-lifetimes-3-lib.c
structured-dynamic-lifetimes-3.c
structured-dynamic-lifetimes-4-lib.c
structured-dynamic-lifetimes-4.c
structured-dynamic-lifetimes-5-lib.c
structured-dynamic-lifetimes-5.c
structured-dynamic-lifetimes-6-lib.c
structured-dynamic-lifetimes-6.c
structured-dynamic-lifetimes-7-lib.c
structured-dynamic-lifetimes-7.c
structured-dynamic-lifetimes-8-lib.c
structured-dynamic-lifetimes-8.c
subr.h
subr.ptx
subset-subarray-mappings-1-d-a.c
subset-subarray-mappings-1-d-p.c
subset-subarray-mappings-1-r-a.c
subset-subarray-mappings-1-r-p.c
subset-subarray-mappings-2.c
switch-conversion-2.c
switch-conversion.c
tile-1.c
timer.h
unmap-infinity-1.c
update-1.c
variable-not-offloaded.c
vector-length-64-1.c	nvptx: update fix for -Wformat-diag	2022-01-19 08:27:00 +01:00
vector-length-64-2.c
vector-length-64-3.c
vector-length-128-1.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-2.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-3.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-4.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-5.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-6.c	[OpenACC] Extract 'pass_oacc_loop_designation' out of 'pass_oacc_device_lower'	2021-07-29 09:19:44 +02:00
vector-length-128-7.c	[libgomp, testsuite, nvptx] Fix dg-output test in vector-length-128-7.c	2022-04-01 13:22:07 +02:00
vector-length-128-10.c
vector-loop.c
vector-type-1.c
vprop-2.c
vprop.c
vred2d-128.c	[libgomp, testsuite] Scale down some OpenACC test-cases	2022-03-25 13:51:48 +01:00
zero_length_subarrays.c