120 Commits
| Author | SHA1 | Message | Date |
|---|---|---|---|
Tom de Vries
|
5ed77fb3ed |
[libgomp, nvptx] Fix hang in gomp_team_barrier_wait_end
Consider the following omp fragment.
...
#pragma omp target
#pragma omp parallel num_threads (2)
#pragma omp task
;
...
This hangs at -O0 for nvptx.
Investigating the behaviour gives us the following trace of events:
- both threads execute GOMP_task, where they:
- deposit a task, and
- execute gomp_team_barrier_wake
- thread 1 executes gomp_team_barrier_wait_end and, not being the last thread,
proceeds to wait at the team barrier
- thread 0 executes gomp_team_barrier_wait_end and, being the last thread, it
calls gomp_barrier_handle_tasks, where it:
- executes both tasks and marks the team barrier done
- executes a gomp_team_barrier_wake which wakes up thread 1
- thread 1 exits the team barrier
- thread 0 returns from gomp_barrier_handle_tasks and goes to wait at
the team barrier.
- thread 0 hangs.
To understand why there is a hang here, it's good to understand how things
are setup for nvptx. The libgomp/config/nvptx/bar.c implementation is
a copy of the libgomp/config/linux/bar.c implementation, with uses of both
futex_wake and do_wait replaced with uses of ptx insn bar.sync:
...
if (bar->total > 1)
asm ("bar.sync 1, %0;" : : "r" (32 * bar->total));
...
The point where thread 0 goes to wait at the team barrier, corresponds in
the linux implementation with a do_wait. In the linux case, the call to
do_wait doesn't hang, because it's waiting for bar->generation to become
a certain value, and if bar->generation already has that value, it just
proceeds, without any need for coordination with other threads.
In the nvtpx case, the bar.sync waits until thread 1 joins it in the same
logical barrier, which never happens: thread 1 is lingering in the
thread pool at the thread pool barrier (using a different logical barrier),
waiting to join a new team.
The easiest way to fix this is to revert to the posix implementation for
bar.{c,h}. That however falls back on a busy-waiting approach, and
does not take advantage of the ptx bar.sync insn.
Instead, we revert to the linux implementation for bar.c,
and implement bar.c local functions futex_wait and futex_wake using the
bar.sync insn.
The bar.sync insn takes an argument specifying how many threads are
participating, and that doesn't play well with the futex syntax where it's
not clear in advance how many threads will be woken up.
This is solved by waking up all waiting threads each time a futex_wait or
futex_wake happens, and possibly going back to sleep with an updated thread
count.
Tested libgomp on x86_64 with nvptx accelerator.
libgomp/ChangeLog:
2021-04-20 Tom de Vries <tdevries@suse.de>
PR target/99555
* config/nvptx/bar.c (generation_to_barrier): New function, copied
from config/rtems/bar.c.
(futex_wait, futex_wake): New function.
(do_spin, do_wait): New function, copied from config/linux/wait.h.
(gomp_barrier_wait_end, gomp_barrier_wait_last)
(gomp_team_barrier_wake, gomp_team_barrier_wait_end):
(gomp_team_barrier_wait_cancel_end, gomp_team_barrier_cancel): Remove
and replace with include of config/linux/bar.c.
* config/nvptx/bar.h (gomp_barrier_t): Add fields waiters and lock.
(gomp_barrier_init): Init new fields.
* testsuite/libgomp.c-c++-common/task-detach-6.c: Remove nvptx-specific
workarounds.
* testsuite/libgomp.c/pr99555-1.c: Same.
* testsuite/libgomp.fortran/task-detach-6.f90: Same.
|
|
Marcel Vollweiler
|
0bd247bbbe |
libgomp, OpenMP: Fix issue for omp_get_device_num on gcn targets.
Currently omp_get_device_num does not work on gcn targets with more than one offload device. The reason is that GOMP_DEVICE_NUM_VAR is static in icv-device.c and thus "__gomp_device_num" is not visible in the offload image. This patch removes "static" such that "__gomp_device_num" is now part of the offload image and can now be found in GOMP_OFFLOAD_load_image in the plugin. This is not an issue for nvptx. There, "__gomp_device_num" is in the offload image even with "static". libgomp/ChangeLog: * config/gcn/icv-device.c: Make GOMP_DEVICE_NUM_VAR public (remove "static") to make the device num available in the offload image. |
|
Jakub Jelinek
|
7adcbafe45 | Update copyright years. | |
|
Jakub Jelinek
|
aea7238683 |
openmp: Add support for thread_limit clause on target
OpenMP 5.1 says that thread_limit clause can also appear on target, and similarly to teams should affect the thread-limit-var ICV. On combined target teams, the clause goes to both. We actually passed thread_limit internally on target already before, but only used it for gcn/ptx offloading to hint how many threads should be created and for ptx didn't set thread_limit_var in that case. Similarly for host fallback. Also, I found that we weren't copying the args array that contains encoded thread_limit and num_teams clause for target (etc.) for async target. 2021-11-15 Jakub Jelinek <jakub@redhat.com> gcc/ * gimplify.c (optimize_target_teams): Only add OMP_CLAUSE_THREAD_LIMIT to OMP_TARGET_CLAUSES if it isn't there already. gcc/c-family/ * c-omp.c (c_omp_split_clauses) <case OMP_CLAUSE_THREAD_LIMIT>: Duplicate to both OMP_TARGET and OMP_TEAMS. gcc/c/ * c-parser.c (OMP_TARGET_CLAUSE_MASK): Add PRAGMA_OMP_CLAUSE_THREAD_LIMIT. gcc/cp/ * parser.c (OMP_TARGET_CLAUSE_MASK): Add PRAGMA_OMP_CLAUSE_THREAD_LIMIT. libgomp/ * task.c (gomp_create_target_task): Copy args array as well. * target.c (gomp_target_fallback): Add args argument. Set gomp_icv (true)->thread_limit_var if thread_limit is present. (GOMP_target): Adjust gomp_target_fallback caller. (GOMP_target_ext): Likewise. (gomp_target_task_fn): Likewise. * config/nvptx/team.c (gomp_nvptx_main): Set gomp_global_icv.thread_limit_var. * testsuite/libgomp.c-c++-common/thread-limit-1.c: New test. |
|
|
Jakub Jelinek
|
9fa72756d9 |
libgomp, nvptx: Honor OpenMP 5.1 num_teams lower bound
Here is a PTX implementation of what I was talking about, that for num_teams_upper 0 or whenever num_teams_lower <= num_blocks, the current implementation is fine but if the user explicitly asks for more teams than we can provide in hardware, we need to stop assuming that omp_get_team_num () is equal to the hw team id, but instead need to use some team specific memory (it is .shared for PTX), or if none is provided, array indexed by the hw team id and run some teams serially within the same hw thread. 2021-11-15 Jakub Jelinek <jakub@redhat.com> * config/nvptx/team.c (__gomp_team_num): Define as __attribute__((shared)) var. (gomp_nvptx_main): Initialize __gomp_team_num to 0. * config/nvptx/target.c (__gomp_team_num): Declare as extern __attribute__((shared)) var. (GOMP_teams4): Use __gomp_team_num as the team number instead of %ctaid.x. If first, initialize it to %ctaid.x. If num_teams_lower is bigger than num_blocks, use num_teams_lower teams and arrange for bumping of __gomp_team_num if !first and returning false once we run out of teams. * config/nvptx/teams.c (__gomp_team_num): Declare as extern __attribute__((shared)) var. (omp_get_team_num): Return __gomp_team_num value instead of %ctaid.x. |
|
|
Jakub Jelinek
|
f49c7a4fb2 |
libgomp: Unbreak gcn offload build
My recent libgomp change apparently broke libgomp build for gcn offloading.
The problem is that gcn, unlike nvptx, doesn't override teams.c source file
and the patch I've committed assumed all the non-LIBGOMP_USE_PTHREADS targets
do not use it. My understanding is that gcn included omp_get_num_teams
and omp_get_team_num definitions in both icv-device.o and teams.o,
with the definitions only in the former working correctly.
This patch brings gcn into sync with how nvptx does it, that teams.c
is overridden, provides a dummy GOMP_teams_reg and omp_get_{num_teams,team_num}
definitions and icv-device.c doesn't provide those.
2021-11-12 Jakub Jelinek <jakub@redhat.com>
PR target/103201
* config/gcn/icv-device.c (omp_get_num_teams, omp_get_team_num): Move
to ...
* config/gcn/teams.c: ... here. New file.
|
|
|
Jakub Jelinek
|
7d6da11fce |
openmp: Honor OpenMP 5.1 num_teams lower bound
The following patch implements what I've been talking about earlier, honor that for explicit num_teams clause we create at least the lower-bound (if not specified, upper-bound) teams in the league. For host fallback, it still means we only have one thread doing all the teams, sequentially one after another. For PTX and GCN, I think the new teams-2.c test and maybe teams-4.c too will or might fail. For these offloads, I think it is ok to remove symbols no longer used from libgomp.a. If num_teams_lower is bigger than the provided num_blocks or num_workgroups, we should arrange for gomp_num_teams_var to be num_teams_lower - 1, stop using the %ctaid.x or __builtin_gcn_dim_pos (0) for omp_get_team_num () and instead use for it some .shared var that GOMP_teams4 initializes to %ctaid.x or __builtin_gcn_dim_pos (0) when first and for !first increment that by num_blocks or num_workgroups each time and only return false when we are above num_teams_lower. Any help with actually implementing this for the 2 architectures highly appreciated. 2021-11-12 Jakub Jelinek <jakub@redhat.com> gcc/ * omp-builtins.def (BUILT_IN_GOMP_TEAMS): Remove. (BUILT_IN_GOMP_TEAMS4): New. * builtin-types.def (BT_FN_VOID_UINT_UINT): Remove. (BT_FN_BOOL_UINT_UINT_UINT_BOOL): New. * omp-low.c (lower_omp_teams): Use GOMP_teams4 instead of GOMP_teams, pass to it also num_teams lower-bound expression or a dup of upper-bound if it is missing and a flag whether it is the first call or not. gcc/fortran/ * types.def (BT_FN_VOID_UINT_UINT): Remove. (BT_FN_BOOL_UINT_UINT_UINT_BOOL): New. libgomp/ * libgomp_g.h (GOMP_teams4): Declare. * libgomp.map (GOMP_5.1): Export GOMP_teams4. * target.c (GOMP_teams4): New function. * config/nvptx/target.c (GOMP_teams): Remove. (GOMP_teams4): New function. * config/gcn/target.c (GOMP_teams): Remove. (GOMP_teams4): New function. * testsuite/libgomp.c/teams-4.c (main): Expect exactly 2 teams instead of <= 2. * testsuite/libgomp.c-c++-common/teams-2.c: New test. |
|
|
Jakub Jelinek
|
3adcf7e104 |
openmp: Fix handling of numa_domains(1)
If numa-domains is used with num-places count, sometimes the function could create more places than requested and crash. This depended on the content of /sys/devices/system/node/online file, e.g. if the file contains 0-1,16-17 and all NUMA nodes contain at least one CPU in the cpuset of the program, then numa_domains(2) or numa_domains(4) (or 5+) work fine while numa_domains(1) or numa_domains(3) misbehave. I.e. the function was able to stop after reaching limit on the , separators (or trivially at the end), but not within in the ranges. 2021-10-18 Jakub Jelinek <jakub@redhat.com> * config/linux/affinity.c (gomp_affinity_init_numa_domains): Add && gomp_places_list_len < count after nfirst <= nlast loop condition. |
|
|
Jakub Jelinek
|
c057ed9c52 |
openmp: Fix up strtoul and strtoull uses in libgomp
Yesterday when working on numa_domains, I've noticed because of a bug
in my patch a hang on a large NUMA machine. I've fixed the bug, but
also discovered that the hang was a result of making wrong assumptions
about strtoul/strtoull. All the uses were for portability setting
errno = 0 before the calls and treating non-zero errno after the call
as invalid input, but for the case where there are no valid digits at
all strtoul may set errno to EINVAL, but doesn't have to and with
glibc doesn't do that. So, this patch goes through all the strtoul calls
and next to errno != 0 checks adds also endptr == startptr check.
Haven't done it in places where we immediately reject strtoul returning 0
the same as we reject errno != 0, because strtoul must return 0 in the
case where it sets endptr to the start pointer. In some spots the code
was using errno = 0; x = strtoul (p, &p, 10); if (errno) { /*invalid*/ }
and those spots had to be changed to
errno = 0; x = strtoul (p, &end, 10); if (errno || end == p) { /*invalid*/ }
p = end;
2021-10-15 Jakub Jelinek <jakub@redhat.com>
* env.c (parse_schedule): For strtoul or strtoull calls which don't
clearly reject return value 0 as invalid handle the case where end
pointer is the same as first argument as invalid.
(parse_unsigned_long_1): Likewise.
(parse_one_place): Likewise.
(parse_places_var): Likewise.
(parse_stacksize): Likewise.
(parse_spincount): Likewise.
(parse_affinity): Likewise.
(parse_gomp_openacc_dim): Likewise. Avoid strict aliasing violation.
Make code valid C89.
* config/linux/affinity.c (gomp_affinity_find_last_cache_level):
For strtoul calls which don't clearly reject return value 0 as
invalid handle the case where end pointer is the same as first
argument as invalid.
(gomp_affinity_init_level_1): Likewise.
(gomp_affinity_init_numa_domains): Likewise.
* config/rtems/proc.c (parse_thread_pools): Likewise.
|
|
|
Jakub Jelinek
|
4764049dd6 |
openmp: Fix up handling of OMP_PLACES=threads(1)
When writing the places-*.c tests, I've noticed that we mishandle threads abstract name with specified num-places if num-places isn't a multiple of number of hw threads in a core. It then happily ignores the maximum count and overwrites for the remaining hw threads in a core further places that haven't been allocated. 2021-10-15 Jakub Jelinek <jakub@redhat.com> * config/linux/affinity.c (gomp_affinity_init_level_1): For level 1 after creating count places clean up and return immediately. * testsuite/libgomp.c/places-6.c: New test. * testsuite/libgomp.c/places-7.c: New test. * testsuite/libgomp.c/places-8.c: New test. * testsuite/libgomp.c/places-9.c: New test. * testsuite/libgomp.c/places-10.c: New test. |
|
|
Jakub Jelinek
|
e7ce32c783 |
openmp: Add support for OMP_PLACES=numa_domains
This adds support for numa_domains abstract name in OMP_PLACES, also new in OpenMP 5.1. Way to test this is OMP_PLACES=numa_domains OMP_DISPLAY_ENV=true LD_PRELOAD=.libs/libgomp.so.1 /bin/true and see what it prints on OMP_PLACES line. For non-NUMA machines it should print a single place that covers all CPUs, for NUMA machine one place for each NUMA node with corresponding CPUs. 2021-10-15 Jakub Jelinek <jakub@redhat.com> * env.c (parse_places_var): Handle numa_domains as level 5. * config/linux/affinity.c (gomp_affinity_init_numa_domains): New function. (gomp_affinity_init_level): Use it instead of gomp_affinity_init_level_1 for level == 5. * testsuite/libgomp.c/places-5.c: New test. |
|
|
Jakub Jelinek
|
5809be05a2 |
openmp: Add support for OMP_PLACES=ll_caches
This patch implements support for ll_caches abstract name in OMP_PLACES, which stands for places where logical cpus in each place share the last level cache. This seems to work fine for me on x86 and kernel sources show that it is in common code, but on some machines on CompileFarm the files I'm using, i.e. /sys/devices/system/cpu/cpuN/cache/indexN/level /sys/devices/system/cpu/cpuN/cache/indexN/shared_cpu_list don't exist, is that because they have too old kernel and newer kernels are fine or should I implement some fallback methods (which)? E.g. on gcc112.fsffrance.org I see just shared_cpu_map and not shared_cpu_list (with shared_cpu_map being harder to parse) and on another box I didn't even see the cache subdirectories. Way to test this is OMP_PLACES=ll_caches OMP_DISPLAY_ENV=true LD_PRELOAD=.libs/libgomp.so.1 /bin/true and see what it prints on OMP_PLACES line. 2021-10-15 Jakub Jelinek <jakub@redhat.com> * env.c (parse_places_var): Handle ll_caches as level 4. * config/linux/affinity.c (gomp_affinity_find_last_cache_level): New function. (gomp_affinity_init_level_1): Handle level 4 as logical cpus sharing last level cache. (gomp_affinity_init_level): Likewise. * testsuite/libgomp.c/places-1.c: New test. * testsuite/libgomp.c/places-2.c: New test. * testsuite/libgomp.c/places-3.c: New test. * testsuite/libgomp.c/places-4.c: New test. |
|
Thomas Schwinge
|
a5416bf369 |
Make the OpenMP 'error' directive work for nvptx offloading
... and add a minimum amount of offloading testing.
(Leaving aside that 'fwrite' to 'stderr' probably wouldn't work anyway) the
'fwrite' calls in 'libgomp/error.c:GOMP_warning', 'libgomp/error.c:GOMP_error'
drag in 'isatty', which isn't provided by my nvptx newlib build at present, so
we get, for example:
[...]
FAIL: libgomp.c/../libgomp.c-c++-common/declare_target-1.c (test for excess errors)
Excess errors:
unresolved symbol isatty
mkoffload: fatal error: [...]/build-gcc/./gcc/x86_64-pc-linux-gnu-accel-nvptx-none-gcc returned 1 exit status
[...]
..., and many more.
Fix up for recent commit
|
|
Chung-Lin Tang
|
0bac793ed6 |
openmp: Implement omp_get_device_num routine
This patch implements the omp_get_device_num library routine, specified in
OpenMP 5.0.
GOMP_DEVICE_NUM_VAR is a macro symbol which defines name of a "device number"
variable, is defined on the device-side libgomp, has it's address returned to
host-side libgomp during device initialization, and the host libgomp then
sets its value to the designated device number.
libgomp/ChangeLog:
* icv-device.c (omp_get_device_num): New API function, host side.
* fortran.c (omp_get_device_num_): New interface function.
* libgomp-plugin.h (GOMP_DEVICE_NUM_VAR): Define macro symbol.
* libgomp.map (OMP_5.0.2): New version space with omp_get_device_num,
omp_get_device_num_.
* libgomp.texi (omp_get_device_num): Add documentation for new API
function.
* omp.h.in (omp_get_device_num): Add declaration.
* omp_lib.f90.in (omp_get_device_num): Likewise.
* omp_lib.h.in (omp_get_device_num): Likewise.
* target.c (gomp_load_image_to_device): If additional entry for device
number exists at end of returned entries from 'load_image_func' hook,
copy the assigned device number over to the device variable.
* config/gcn/icv-device.c (GOMP_DEVICE_NUM_VAR): Define static global.
(omp_get_device_num): New API function, device side.
* plugin/plugin-gcn.c ("symcat.h"): Add include.
(GOMP_OFFLOAD_load_image): Add addresses of device GOMP_DEVICE_NUM_VAR
at end of returned 'target_table' entries.
* config/nvptx/icv-device.c (GOMP_DEVICE_NUM_VAR): Define static global.
(omp_get_device_num): New API function, device side.
* plugin/plugin-nvptx.c ("symcat.h"): Add include.
(GOMP_OFFLOAD_load_image): Add addresses of device GOMP_DEVICE_NUM_VAR
at end of returned 'target_table' entries.
* testsuite/lib/libgomp.exp
(check_effective_target_offload_target_intelmic): New function for
testing for intelmic offloading.
* testsuite/libgomp.c-c++-common/target-45.c: New test.
* testsuite/libgomp.fortran/target10.f90: New test.
|
|
|
Thomas Schwinge
|
8168338684 |
[gcn] Work-around libgomp 'error: array subscript 0 is outside array bounds of ‘__lds struct gomp_thread * __lds[0]’ [-Werror=array-bounds]' some more [PR101484]
With yesterday's commit |
|
|
Thomas Schwinge
|
9f2bc5077d |
[gcn] Work-around libgomp 'error: array subscript 0 is outside array bounds of ‘__lds struct gomp_thread * __lds[0]’ [-Werror=array-bounds]' [PR101484]
... seen as of commit
|
|
|
Jakub Jelinek
|
42f10ba5b5 |
libgomp: Don't include limits.h instead of hidden visibility block
sem.h is included in between # pragma GCC visibility push(hidden) and # pragma GCC visibility pop and includes limits.h there, which since the introduction of sysconf declaration in recent glibcs in there causes trouble. libgomp assumes it is compiled by gcc, so we don't really need to include limits.h there and can use -__INT_MAX__ - 1 instead (which clang and icc support too for years). 2021-07-13 Jakub Jelinek <jakub@redhat.com> Florian Weimer <fweimer@redhat.com> * config/linux/sem.h: Don't include limits.h. (SEM_WAIT): Define to -__INT_MAX__ - 1 instead of INT_MIN. * config/linux/affinity.c: Include limits.h. |
|
|
Jakub Jelinek
|
95d6776217 |
openmp: Fix up handling of target constructs in offloaded routines [PR100573]
OpenMP Nesting of Regions restrictions say:
- If a target update, target data, target enter data, or target exit data
construct is encountered during execution of a target region, the behavior is unspecified.
- If a target construct is encountered during execution of a target region and a device
clause in which the ancestor device-modifier appears is not present on the construct, the
behavior is unspecified.
That wording is about the dynamic (runtime) behavior, not about lexical nesting,
so while it is UB if omp target * is encountered in the target region, we need to make
it compile and link (for lexical nesting of target * inside of target we actually
emit a warning).
To make this work, I had to do multiple changes.
One was to mark .omp_data_{sizes,kinds}.* variables when static as "omp declare target".
Another one was to add stub GOMP_target* entrypoints to nvptx and gcn libgomp.a.
The entrypoint functions shouldn't be called or passed in the offload regions,
otherwise
libgomp: cuLaunchKernel error: too many resources requested for launch
was reported; fixed by changing those arguments of calls to GOMP_target_ext
to NULL.
And we didn't mark the entrypoints "omp target entrypoint" when the caller
has been "omp declare target".
2021-05-26 Jakub Jelinek <jakub@redhat.com>
PR libgomp/100573
gcc/
* omp-low.c: Include omp-offload.h.
(create_omp_child_function): If current_function_decl has
"omp declare target" attribute and is_gimple_omp_offloaded,
remove that attribute from the copy of attribute list and
add "omp target entrypoint" attribute instead.
(lower_omp_target): Mark .omp_data_sizes.* and .omp_data_kinds.*
variables for offloading if in omp_maybe_offloaded_ctx.
* omp-offload.c (pass_omp_target_link::execute): Nullify second
argument to GOMP_target_data_ext in offloaded code.
libgomp/
* config/nvptx/target.c (GOMP_target_ext, GOMP_target_data_ext,
GOMP_target_end_data, GOMP_target_update_ext,
GOMP_target_enter_exit_data): New dummy entrypoints.
* config/gcn/target.c (GOMP_target_ext, GOMP_target_data_ext,
GOMP_target_end_data, GOMP_target_update_ext,
GOMP_target_enter_exit_data): Likewise.
* testsuite/libgomp.c-c++-common/for-3.c (DO_PRAGMA, OMPTEAMS,
OMPFROM, OMPTO): Define.
(main): Remove #pragma omp target teams around all the tests.
* testsuite/libgomp.c-c++-common/target-41.c: New test.
* testsuite/libgomp.c-c++-common/target-42.c: New test.
|
|
Uros Bizjak
|
5e40542f87 |
libgomp/i386: Revert the type of syscall wrappers output back to long.
Linux man-pages 5.07 wrongly declares syscall output type as int. This error was fixed in release 5.10, so this patch reverts my recent change. 2021-02-11 Uroš Bizjak <ubizjak@gmail.com> libgomp/ * config/linux/x86/futex.h (__futex_wait): Revert output type back to long. (__futex_wake): Ditto. (futex_wait): Update for revert. (futex_wake): Ditto. |
|
|
Uros Bizjak
|
c36ad24e8a |
libgomp/i386: Move syscall asms to static inline wrapper.
Move syscall asms to static inline wrapper functions to improve #ifdeffery. Also correct output type to int and timeout type to void *. 2021-02-11 Uroš Bizjak <ubizjak@gmail.com> libgomp/ * config/linux/x86/futex.h (__futex_wait): New static inline wrapper function. Correct output type to int and timeout type to void *. (__futex_wake): New static inline wrapper function. Correct output type to int. (futex_wait): Use __futex_wait. (futex_wake): Use __futex_wake. |
|
Sebastian Huber
|
0f951b3dd3 |
RTEMS: Fix libgomp build
libgomp/ * config/rtems/sem.h (gomp_sem_getcount): New function. |
|
|
Jakub Jelinek
|
d3b41bde96 |
libgomp: Don't access gomp_sem_t as int using atomics unconditionally
This patch introduces gomp_sem_getcount wrapper, which uses sem_getvalue for POSIX and atomic loads for linux futex and accel. rtems for now remains broken. 2021-01-18 Jakub Jelinek <jakub@redhat.com> * config/linux/sem.h (gomp_sem_getcount): New function. * config/posix/sem.h (gomp_sem_getcount): New function. * config/posix/sem.c (gomp_sem_getcount): New function. * config/accel/sem.h (gomp_sem_getcount): New function. * task.c (task_fulfilled_p): Use gomp_sem_getcount. (omp_fulfill_event): Likewise. |
|
|
Jakub Jelinek
|
99dee82307 | Update copyright years. | |
|
Jakub Jelinek
|
74c9882b80 |
openmp: Change omp_get_initial_device () to match OpenMP 5.1 requirements
> Therefore, I think until omp_get_initial_device () value is changed, we The following so far untested patch implements that change. OpenMP 4.5 said for omp_get_initial_device: The value of the device number is implementation defined. If it is between 0 and one less than omp_get_num_devices() then it is valid for use with all device constructs and routines; if it is outside that range, then it is only valid for use with the device memory routines and not in the device clause. and OpenMP 5.0 similarly, but OpenMP 5.1 says: The value of the device number is the value returned by the omp_get_num_devices routine. As the new value is compatible with what has been required earlier, I think we can change it already now. 2020-10-22 Jakub Jelinek <jakub@redhat.com> * icv.c (omp_get_initial_device): Remove including corresponding ialias. * icv-device.c (omp_get_initial_device): New function. Return gomp_get_num_devices (). Add ialias. * target.c (resolve_device): Don't fail with OMP_TARGET_OFFLOAD=mandatory if device_id is equal to gomp_get_num_devices (). (omp_target_alloc, omp_target_free, omp_target_is_present, omp_target_memcpy, omp_target_memcpy_rect, omp_target_associate_ptr, omp_target_disassociate_ptr, omp_pause_resource): Use gomp_get_num_devices () instead of GOMP_DEVICE_HOST_FALLBACK on the first use in the functions, in uses dominated by the gomp_get_num_devices call use num_devices_openmp instead. * libgomp.texi (omp_get_initial_device): Document. * config/gcn/icv-device.c (omp_get_initial_device): New function. Add ialias. * config/nvptx/icv-device.c (omp_get_initial_device): Likewise. * testsuite/libgomp.c/target-40.c: New test. |
|
Clément Chigot
|
4eaf96c56c |
aix: remove libgomp and libatomic archives before creating FAT archives
AIX caches shared objects in archives with read-other permission. libgomp and libatomic might be in use during the build or testing, which may cause archiver operations on them to fail. This patch adjusts the Makefile fragments to delete the library archives before creating fresh archives containing both the 32 bit and 64 bit shared objects. libatomic/ChangeLog: 2020-10-11 Clement Chigot <clement.chigot@atos.net> * config/t-aix: Delete and recreate libatomic before creating FAT library. libgomp/ChangeLog: 2020-10-11 Clement Chigot <clement.chigot@atos.net> * config/t-aix: Delete and recreate libgomp before creating FAT library. |
|
Andrew Stubbs
|
6f51395197 |
libgomp: disable barriers in nested teams
Both GCN and NVPTX allow nested parallel regions, but the barrier implementation did not allow the nested teams to run independently of each other (due to hardware limitations). This patch fixes that, under the assumption that each thread will create a new subteam of one thread, by simply not using barriers when there's no other thread to synchronise. libgomp/ChangeLog: * config/gcn/bar.c (gomp_barrier_wait_end): Skip the barrier if the total number of threads is one. (gomp_team_barrier_wake): Likewise. (gomp_team_barrier_wait_end): Likewise. (gomp_team_barrier_wait_cancel_end): Likewise. * config/nvptx/bar.c (gomp_barrier_wait_end): Likewise. (gomp_team_barrier_wake): Likewise. (gomp_team_barrier_wait_end): Likewise. (gomp_team_barrier_wait_cancel_end): Likewise. * testsuite/libgomp.c-c++-common/nested-parallel-unbalanced.c: New test. |
|
|
Clément Chigot
|
3c11f25fb8 |
aix: Use $(AR) without -X32_64 to build FAT libraries.
AIX FAT libraries should be built with the version of AR chosen by configure. The GNU Make $(AR) variable includes the AIX -X32_64 option needed by the default Makefile rules to accept both 32 bit and 64 bit object files. The -X32_64 option conflicts with ar archiving objects of the same name used to build FAT libraries. This patch changes the Makefile fragments for AIX FAT libraries to use $(AR), but strips the -X32_64 option from the Make variable. libgcc/ChangeLog: 2020-09-27 Clement Chigot <clement.chigot@atos.net> * config/rs6000/t-slibgcc-aix: Use $(AR) without -X32_64. libatomic/ChangeLog: 2020-09-27 Clement Chigot <clement.chigot@atos.net> * config/t-aix: Use $(AR) without -X32_64. libgomp/ChangeLog: 2020-09-27 Clement Chigot <clement.chigot@atos.net> * config/t-aix: Use $(AR) without -X32_64. libstdc++-v3/ChangeLog: 2020-09-27 Clement Chigot <clement.chigot@atos.net> * config/os/aix/t-aix: Use $(AR) without -X32_64. libgfortran/ChangeLog: 2020-09-27 Clement Chigot <clement.chigot@atos.net> * config/t-aix: Use $(AR) without -X32_64. |
|
|
Tom de Vries
|
4ac7b66958 |
[libgomp, nvptx] Add __sync_compare_and_swap_16
As reported here ( https://gcc.gnu.org/pipermail/gcc-patches/2020-September/553070.html ), when running test-case libgomp.c-c++-common/reduction-16.c for powerpc host with nvptx accelerator, we run into: ... unresolved symbol __sync_val_compare_and_swap_16 ... I can reproduce the problem on x86_64 with a trigger patch that: - initializes ix86_isa_flags2 to TARGET_ISA2_CX16 - enables define_expand "atomic_load<mode>" in gcc/config/i386/sync.md for TImode The problem is that omp-expand.c generates atomic builtin calls based on checks whether those are supported on the host, which forces the target to support these, even though those checks fail for the accelerator target. Fix this by: - adding a __sync_val_compare_and_swap_16 in libgomp for nvptx, which falls back onto libatomic's __atomic_compare_and_swap_16 - adding -foffload=-latomic in the test-case Tested libgomp on x86_64-linux with nvptx accelerator. Tested libgomp with trigger patch on x86_64-linux with nvptx accelerator. libgomp/ChangeLog: * config/nvptx/atomic.c: New file. Add __sync_val_compare_and_swap_16. * testsuite/libgomp.c-c++-common/reduction-16.c: Add -latomic for target offload_target_nvptx. |
|
David Edelsohn
|
4f97bed9a7 |
aix: FAT libraries: test native compiler mode directly
The FAT libraries config fragments need to know which library is native and which is a multilib to choose the correct multilib from which to append the additional object file or shared object file. Testing the top-level archive is fragile because it will fail if rebuilding. This patch tests the compiler preprocessing macros for the 64 bit AIX specific __64BIT__ to determine the native mode of the compiler in MULTILIBTOP. 2020-07-14 David Edelsohn <dje.gcc@gmail.com> libatomic/ChangeLog * config/t-aix: Set BITS from compiler cpp macro. libgcc/ChangeLog * config/rs6000/t-slibgcc-aix: Set BITS from compiler cpp macro. libgfortran/ChangeLog * config/t-aix: Set BITS from compiler cpp macro. libgomp/ChangeLog * config/t-aix: Set BITS from compiler cpp macro. libstdc++-v3/ChangeLog * config/os/aix/t-aix: Set BITS from compiler cpp macro. |
|
|
David Edelsohn
|
47ddb895df |
aix: Add GCC64 configuration and FAT target libraries.
This patch adds the ability to configure GCC on AIX to build as a 64 bit application and to build target libraries "FAT" libraries in both 32 bit and 64 bit mode. The patch adds makefile fragment hooks to target libraries that allows them to include target-specific rules. The target specific rules for AIX place both 32 bit and 64 bit objects and shared objects in archives at the top-level, not multilib subdirectories. The multilibs are built in subdirectories, but must be combined during the last parts of the target library build process. Because of the way that GCC bootstrap works, the libraries must be combined during the multiple stages of GCC bootstrap, not solely when installed in the final destination, so the libraries are correct at the end of each target library build stage, not solely an install recipe. gcc/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * config.gcc: Use t-aix64, biarch64 and default64 for cpu_is_64bit. * config/rs6000/aix72.h (ASM_SPEC): Remove aix64 option. (ASM_SPEC32): New. (ASM_SPEC64): New. (ASM_CPU_SPEC): Remove vsx and altivec options. (CPP_SPEC_COMMON): Rename from CPP_SPEC. (CPP_SPEC32): New. (CPP_SPEC64): New. (CPLUSPLUS_CPP_SPEC): Rename to CPLUSPLUS_CPP_SPEC_COMMON.. (TARGET_DEFAULT): Only define if not BIARCH. (LIB_SPEC_COMMON): Rename from LIB_SPEC. (LIB_SPEC32): New. (LIB_SPEC64): New. (LINK_SPEC_COMMON): Rename from LINK_SPEC. (LINK_SPEC32): New. (LINK_SPEC64): New. (STARTFILE_SPEC): Add 64 bit version of crtcxa and crtdbase. (ASM_SPEC): Define 32 and 64 bit alternatives using DEFAULT_ARCH64_P. (CPP_SPEC): Same. (CPLUSPLUS_CPP_SPEC): Same. (LIB_SPEC): Same. (LINK_SPEC): Same. (SUBTARGET_EXTRA_SPECS): Add new 32/64 specs. * config/rs6000/defaultaix64.h: New file. * config/rs6000/t-aix64: New file. libgcc/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * config.host (extra_parts): Add crtcxa_64 and crtdbase_64. * config/rs6000/t-aix-cxa: Explicitly compile 32 bit with -maix32 and 64 bit with -maix64. * config/rs6000/t-slibgcc-aix: Remove extra @multilib_dir@ level. Build and install AIX-style FAT libraries. libgomp/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * Makefile.am (tmake_file): Build and install AIX-style FAT libraries. * Makefile.in: Regenerate * configure.ac (tmake_file): Substitute. * configure: Regenerate. * configure.tgt (powerpc-ibm-aix*): Define tmake_file. * config/t-aix: New file. libstdc++-v3/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * Makefile.am (tmake_file): Build and install AIX-style FAT libraries. * Makefile.in: Regenerate. * configure.ac (tmake_file): Substitute. * configure: Regenerate. * configure.host (aix*): Define tmake_file. * config/os/aix/t-aix: New file. libatomic/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * Makefile.am (tmake_file): Build and install AIX-style FAT libraries. * Makefile.in: Regenerate. * configure.ac (tmake_file): Substitute. * configure: Regenerate. * configure.tgt (powerpc-ibm-aix*): Define tmake_file. * config/t-aix: New file. libgfortran/ChangeLog 2020-06-21 David Edelsohn <dje.gcc@gmail.com> * Makefile.am (tmake_file): Build and install AIX-style FAT libraries. * Makefile.in: Regenerate. * configure.ac (tmake_file): Substitute. * configure: Regenerate. * configure.host: Add system configury stanza. Define tmake_file. * config/t-aix: New file. |
|
|
Thomas Schwinge
|
a5d0bc12e1 |
[OpenACC] Set 'acc_device_current = -1'
There's no point in using value '-3', and even though not directly related, value '-1' does match 'GOMP_DEVICE_ICV'. libgomp/ * config/accel/openacc.f90 (acc_device_current): Set to '-1'. * openacc.f90 (acc_device_current): Likewise. * openacc.h (acc_device_current): Likewise. * openacc_lib.h (acc_device_current): Likewise. |
|
Tobias Burnus
|
8d1a1cb1b8 |
libgomp: Fixes + cleanup for OpenACC's Fortran module + openacc_lib.h
2020-02-19 Tobias Burnus <tobias@codesourcery.com> * .gitattributes: New; whitespace handling for Fortran's openacc_lib.h. * config/accel/openacc.f90 (openacc_kinds): Add acc_device_current. (openacc_internal, acc_on_device_h): Fix argument name; minor cleanup. * libgomp.texi (Enabling OpenACC): No longer mark as experimental. (acc_set_device_num): Fix Fortran argument name, use same name for C. (acc_get_property): Update Fortran interface to post-OpenACC 3.0 corrections; add note about the previous interface and named constant. (OpenACC library and environment variables): Fix two typos. * openacc.f90: Use for all procedures the argument names from the spec as for …_h they are user visible. (openacc_kinds): Rename acc_device_property to acc_device_property_kinds and change value to int32 ; and update users. Re-add acc_device_property for for backward compatibility. (acc_get_property_string_h): Clean up as acc_device_property_kind changed. (acc_get_property_h): Likewise and return c_size_t instead of acc_device_property. (openacc): Also export acc_device_property_kinds. (acc_async_test_h, acc_async_test_all_h, acc_on_device_h, acc_is_present_32_h, acc_is_present_64_h): Simplify logical-return-value handling; check against /= 0 instead of == 1 to match C. * openacc_lib.h: Use for all procedures the argument names from the spec as for …_h they are user visible. Place !GCC$ into the first column to be active also for fixed-form souce form. (acc_device_current, acc_device_property_kind, acc_device_property, acc_property_memory, acc_property_free_memory, acc_property_name, acc_property_vendor, acc_property_driver): New named constants. (acc_get_property, acc_get_property_string): New generic interface. |
|
|
Andrew Stubbs
|
6687d13a87 |
Rename acc_device_gcn to acc_device_radeon
2020-01-17 Andrew Stubbs <ams@codesourcery.com> libgomp/ * config/accel/openacc.f90 (openacc_kinds): Rename acc_device_gcn to acc_device_radeon. (openacc): Likewise. * openacc.f90 (openacc_kinds): Likewise. (openacc): Likewise. * openacc.h (acc_device_t): Likewise. * openacc_lib.h: Likewise. * testsuite/lib/libgomp.exp (check_effective_target_openacc_amdgcn_accel_present): Likewise. * testsuite/libgomp.oacc-c-c++-common/acc_prof-init-1.c (cb_compute_construct_end): Likewise. * testsuite/libgomp.oacc-c-c++-common/acc_prof-kernels-1.c (cb_enqueue_launch_start): Likewise. * testsuite/libgomp.oacc-c-c++-common/acc_prof-parallel-1.c (cb_enter_data_end): Likewise. (cb_exit_data_start): Likewise. (cb_exit_data_end): Likewise. (cb_compute_construct_end): Likewise. (cb_enqueue_launch_start): Likewise. (cb_enqueue_launch_end): Likewise. * testsuite/libgomp.oacc-c-c++-common/asyncwait-nop-1.c (main): Likewise. |
|
Jakub Jelinek
|
8d9254fc8a |
Update copyright years.
From-SVN: r279813 |
|
|
Tobias Burnus
|
0b8499a480 |
libgomp/openacc.f90 – clean-up public/private attributes
* config/accel/openacc.f90 (module openacc_kinds): Use 'PUBLIC' to mark
all symbols as public except for the 'use …, only' imported symbol,
which is private.
(module openacc): Default to 'PRIVATE' to exclude openacc_internal; mark
all symbols from module openacc_kinds as PUBLIC
* openacc.f90: Add comment with crossref to that file and openmp_lib.h;
fix comment typo.
* openacc_lib.h (acc_device_gcn): Add this PARAMETER.
From-SVN: r279456
|
|
Tobias Burnus
|
93d9021987 |
libgomp – spelling fixes, incl. omp_lib.h.in
* omp_lib.h.in: Fix spelling of function declaration
omp_get_cancell(l)ation.
* libgomp.texi (acc_is_present, acc_async_test, acc_async_test_all):
Fix typos.
* env.c: Fix comment typos.
* oacc-host.c: Likewise.
* ordered.c: Likewise.
* task.c: Likewise.
* team.c: Likewise.
* config/gcn/task.c: Likewise.
* config/gcn/team.c: Likewise.
* config/nvptx/task.c: Likewise.
* config/nvptx/team.c: Likewise.
* plugin/plugin-gcn.c: Likewise.
* testsuite/libgomp.fortran/jacobi.f: Likewise.
* testsuite/libgomp.hsa.c/tiling-2.c: Likewise.
* testsuite/libgomp.oacc-c-c++-common/enter_exit-lib.c: Likewise.
From-SVN: r279218
|
|
Kwok Cheung Yeung
|
3ad4a03ac6 |
Fix Fortran linker errors on AMD GCN
2019-12-06 Kwok Cheung Yeung <kcy@codesourcery.com> libgomp/ * config/accel/proc.c (omp_get_num_procs): Apply ialias macro. From-SVN: r279060 |
|
|
Andrew Stubbs
|
cee1645106 |
Optimize GCN OpenMP malloc performance
2019-11-13 Andrew Stubbs <ams@codesourcery.com> libgomp/ * config/gcn/team.c (gomp_gcn_enter_kernel): Set up the team arena and use team_malloc variants. (gomp_gcn_exit_kernel): Use team_free. * libgomp.h (TEAM_ARENA_SIZE): Define. (TEAM_ARENA_START): Define. (TEAM_ARENA_FREE): Define. (TEAM_ARENA_END): Define. (team_malloc): New function. (team_malloc_cleared): New function. (team_free): New function. * team.c (gomp_new_team): Initialize and use team_malloc. (free_team): Use team_free. (gomp_free_thread): Use team_free. (gomp_pause_host): Use team_free. * work.c (gomp_init_work_share): Use team_malloc. (gomp_fini_work_share): Use team_free. From-SVN: r278136 |
|
|
Andrew Stubbs
|
fa4999953d |
GCN libgomp port
2019-11-13 Andrew Stubbs <ams@codesourcery.com> Kwok Cheung Yeung <kcy@codesourcery.com> Julian Brown <julian@codesourcery.com> Tom de Vries <tom@codesourcery.com> include/ * gomp-constants.h (GOMP_DEVICE_GCN): Define. (GOMP_VERSION_GCN): Define. libgomp/ * Makefile.am (libgomp_la_SOURCES): Add oacc-target.c. * Makefile.in: Regenerate. * config.h.in (PLUGIN_GCN): Add new undef. * config/accel/openacc.f90 (acc_device_gcn): New parameter. * config/gcn/affinity-fmt.c: New file. * config/gcn/bar.c: New file. * config/gcn/bar.h: New file. * config/gcn/doacross.h: New file. * config/gcn/icv-device.c: New file. * config/gcn/oacc-target.c: New file. * config/gcn/simple-bar.h: New file. * config/gcn/target.c: New file. * config/gcn/task.c: New file. * config/gcn/team.c: New file. * config/gcn/time.c: New file. * configure.ac: Add amdgcn*-*-*. * configure: Regenerate. * configure.tgt: Add amdgcn*-*-*. * libgomp-plugin.h (offload_target_type): Add OFFLOAD_TARGET_TYPE_GCN. * libgomp.h (gcn_thrs): Add amdgcn variant. (set_gcn_thrs): Likewise. (gomp_thread): Likewise. * oacc-int.h (goacc_thread): Likewise. * oacc-target.c: New file. * openacc.f90 (acc_device_gcn): New parameter. * openacc.h (acc_device_t): Add acc_device_gcn. * team.c (gomp_free_pool_helper): Add amdgcn support. Co-Authored-By: Julian Brown <julian@codesourcery.com> Co-Authored-By: Kwok Cheung Yeung <kcy@codesourcery.com> Co-Authored-By: Tom de Vries <tom@codesourcery.com> From-SVN: r278135 |
|
|
Andrew Stubbs
|
b3d14b3aa3 |
Move generic libgomp files from nvptx to accel
2019-11-13 Andrew Stubbs <ams@codesourcery.com> libgomp/ * configure.tgt (nvptx*-*-*): Add "accel" directory. * config/nvptx/libgomp-plugin.c: Move ... * config/accel/libgomp-plugin.c: ... to here. * config/nvptx/lock.c: Move ... * config/accel/lock.c: ... to here. * config/nvptx/mutex.c: Move ... * config/accel/mutex.c: ... to here. * config/nvptx/mutex.h: Move ... * config/accel/mutex.h: ... to here. * config/nvptx/oacc-async.c: Move ... * config/accel/oacc-async.c: ... to here. * config/nvptx/oacc-cuda.c: Move ... * config/accel/oacc-cuda.c: ... to here. * config/nvptx/oacc-host.c: Move ... * config/accel/oacc-host.c: ... to here. * config/nvptx/oacc-init.c: Move ... * config/accel/oacc-init.c: ... to here. * config/nvptx/oacc-mem.c: Move ... * config/accel/oacc-mem.c: ... to here. * config/nvptx/oacc-plugin.c: Move ... * config/accel/oacc-plugin.c: ... to here. * config/nvptx/omp-lock.h: Move ... * config/accel/omp-lock.h: ... to here. * config/nvptx/openacc.f90: Move ... * config/accel/openacc.f90: ... to here. * config/nvptx/pool.h: Move ... * config/accel/pool.h: ... to here. * config/nvptx/proc.c: Move ... * config/accel/proc.c: ... to here. * config/nvptx/ptrlock.c: Move ... * config/accel/ptrlock.c: ... to here. * config/nvptx/ptrlock.h: Move ... * config/accel/ptrlock.h: ... to here. * config/nvptx/sem.c: Move ... * config/accel/sem.c: ... to here. * config/nvptx/sem.h: Move ... * config/accel/sem.h: ... to here. * config/nvptx/thread-stacksize.h: Move ... * config/accel/thread-stacksize.h: ... to here. From-SVN: r278132 |
|
|
Sebastian Huber
|
cb87fec331 |
libgomp: Reduce copy and paste for RTEMS
libgomp/ * config/rtems/bar.c: Include "../linux/bar.c" and delete copy and paste code. From-SVN: r267752 |
|
|
Sebastian Huber
|
30b4d0d0b9 |
libgomp: Avoid complex dependencies for RTEMS
libgomp/ * config/rtems/affinity-fmt.c: New file. Include affinity-fmt.c, undefining HAVE_GETPID and HAVE_GETHOSTNAME, and mapping fwrite to write. From-SVN: r267751 |
|
|
Jakub Jelinek
|
a554497024 |
Update copyright years.
From-SVN: r267494 |
|
|
Tom de Vries
|
fe0827eed0 |
[libgomp, nvptx] Fix libgomp.c/target-5.c compilation
Libgomp test-case libgomp.c/target-5.c is failing to compile when building for x86_64 with nvptx accelerator due to missing: - getpid - gethostname - isatty (pulled in by fwrite) in the nvptx newlib. This patch fixes the build failure by: - adding a function gomp_print_string which limits the use of fwrite to a single location (in affinity-fmt.c), and - creating an nvptx version of affinity-fmt.c, which: - overrides the configure test results HAVE_GETPID and HAVE_GETHOSTNAME, and - implements fwrite using write. Build and reg-tested on x86_64 with nvptx accelerator. 2018-12-13 Tom de Vries <tdevries@suse.de> * affinity-fmt.c (gomp_print_string): New function, factored out of ... (omp_display_affinity, gomp_display_affinity_thread): ... here, and ... * fortran.c (omp_display_affinity_): ... here. * libgomp.h (gomp_print_string): Declare. * config/nvptx/affinity-fmt.c: New file. Include affinity-fmt.c, undefining HAVE_GETPID and HAVE_GETHOSTNAME, and mapping fwrite to write. From-SVN: r267100 |
|
Andreas Schwab
|
9c47a01994 |
Don't mark stack pointer as clobbered in asm
* config/linux/ia64/futex.h (sys_futex0): Don't mark r12 as clobbered. From-SVN: r267052 |
|
|
Jakub Jelinek
|
72d3bb76c9 |
re PR bootstrap/88106 (Bootstrap failure on mingw32)
PR bootstrap/88106 * config/mingw32/affinity-fmt.c: New file. From-SVN: r266309 |
|
|
Jakub Jelinek
|
28567c40e2 |
builtin-types.def (BT_FN_VOID_BOOL, [...]): New.
* builtin-types.def (BT_FN_VOID_BOOL, BT_FN_VOID_SIZE_SIZE_PTR,
BT_FN_UINT_UINT_PTR_PTR, BT_FN_UINT_OMPFN_PTR_UINT_UINT,
BT_FN_BOOL_UINT_LONGPTR_LONG_LONG_LONGPTR_LONGPTR_PTR_PTR,
BT_FN_BOOL_UINT_ULLPTR_LONG_ULL_ULLPTR_ULLPTR_PTR_PTR,
BT_FN_BOOL_LONG_LONG_LONG_LONG_LONG_LONGPTR_LONGPTR_PTR_PTR,
BT_FN_BOOL_BOOL_ULL_ULL_ULL_LONG_ULL_ULLPTR_ULLPTR_PTR_PTR): New.
* gengtype.c (open_base_files): Add omp-general.h.
* gimple.c (gimple_build_omp_critical):
(gimple_build_omp_taskgroup): Add CLAUSES argument. Call
gimple_omp_taskgroup_set_clauses.
(gimple_build_omp_atomic_load): Add mo argument, call
gimple_omp_atomic_set_memory_order.
(gimple_build_omp_atomic_store): Likewise.
(gimple_copy): Adjust handling of GIMPLE_OMP_TASKGROUP.
* gimple.def (GIMPLE_OMP_TASKGROUP): Use GSS_OMP_SINGLE_LAYOUT
instead of GSS_OMP.
(GIMPLE_OMP_TEAMS): Use GSS_OMP_PARALLEL_LAYOUT instead
of GSS_OMP_SINGLE_LAYOUT, adjust comments.
* gimple.h (enum gf_mask): Add GF_OMP_TEAMS_HOST, GF_OMP_TASK_TASKWAIT
and GF_OMP_ATOMIC_MEMORY_ORDER. Remove GF_OMP_ATOMIC_SEQ_CST, use
different value for GF_OMP_ATOMIC_NEED_VALUE.
(struct gimple_statement_omp_taskreg): Add GIMPLE_OMP_TEAMS to
comments.
(struct gimple_statement_omp_single_layout): And remove here.
(struct gomp_teams): Inherit from gimple_statement_omp_taskreg rather
than gimple_statement_omp_single_layout.
(is_a_helper <gimple_statement_omp_taskreg *>::test): Allow
GIMPLE_OMP_TEAMS.
(is_a_helper <const gimple_statement_omp_taskreg *>::test): Likewise.
(gimple_omp_subcode): Formatting fix.
(gimple_omp_teams_child_fn, gimple_omp_teams_child_fn_ptr,
gimple_omp_teams_set_child_fn, gimple_omp_teams_data_arg,
gimple_omp_teams_data_arg_ptr, gimple_omp_teams_set_data_arg,
gimple_omp_teams_host, gimple_omp_teams_set_host,
gimple_omp_task_taskwait_p, gimple_omp_task_set_taskwait_p,
gimple_omp_taskgroup_clauses, gimple_omp_taskgroup_clauses_ptr,
gimple_omp_taskgroup_set_clauses): New inline functions.
(gimple_build_omp_atomic_load): Add enum omp_memory_order argument.
(gimple_build_omp_atomic_store): Likewise.
(gimple_omp_atomic_seq_cst_p): Remove.
(gimple_omp_atomic_memory_order): New function.
(gimple_omp_atomic_set_seq_cst): Remove.
(gimple_omp_atomic_set_memory_order): New function.
(gimple_build_omp_taskgroup): Add clauses argument.
* gimple-pretty-print.c (dump_gimple_omp_taskgroup): New function.
(dump_gimple_omp_task): Print taskwait with depend clauses.
(dump_gimple_omp_atomic_load, dump_gimple_omp_atomic_store): Use
dump_omp_atomic_memory_order.
(pp_gimple_stmt_1): Handle GIMPLE_OMP_TASKGROUP.
* gimplify.c (enum gimplify_omp_var_data): Add GOVD_MAP_ALLOC_ONLY,
GOVD_MAP_FROM_ONLY and GOVD_NONTEMPORAL.
(enum omp_region_type): Reserve bits 1 and 2 for auxiliary flags,
renumber values of most of ORT_* enumerators, add ORT_HOST_TEAMS,
ORT_COMBINED_HOST_TEAMS, ORT_TASKGROUP, ORT_TASKLOOP and
ORT_UNTIED_TASKLOOP enumerators.
(enum gimplify_defaultmap_kind): New.
(struct gimplify_omp_ctx): Remove target_map_scalars_firstprivate and
target_map_pointers_as_0len_arrays members, add defaultmap.
(new_omp_context): Initialize defaultmap member.
(gimple_add_tmp_var): Handle ORT_TASKGROUP like ORT_WORKSHARE.
(maybe_fold_stmt): Don't fold even in host teams regions.
(omp_firstprivatize_variable): Handle ORT_TASKGROUP like
ORT_WORKSHARE. Test ctx->defaultmap[GDMK_SCALAR] instead of
ctx->omp_firstprivatize_variable.
(omp_add_variable): Don't add private/firstprivate for VLAs in
ORT_TASKGROUP.
(omp_default_clause): Print "taskloop" rather than "task" if
ORT_*TASKLOOP.
(omp_notice_variable): Handle ORT_TASKGROUP like ORT_WORKSHARE.
Handle new defaultmap clause kinds.
(omp_is_private): Handle ORT_TASKGROUP like ORT_WORKSHARE. Allow simd
iterator to be lastprivate or private. Fix up diagnostics if linear
is used on collapse>1 simd iterator.
(omp_check_private): Handle ORT_TASKGROUP like ORT_WORKSHARE.
(gimplify_omp_depend): New function.
(gimplify_scan_omp_clauses): Add shared clause on parallel for
combined parallel master taskloop{, simd} if taskloop has
firstprivate, lastprivate or reduction clause. Handle
OMP_CLAUSE_REDUCTION_TASK diagnostics. Adjust tests for
ORT_COMBINED_TEAMS. Gimplify depend clauses with iterators. Handle
cancel and simd OMP_CLAUSE_IF_MODIFIERs. Handle
OMP_CLAUSE_NONTEMPORAL. Handle new defaultmap clause kinds. Handle
OMP_CLAUSE_{TASK,IN}_REDUCTION. Diagnose invalid conditional
lastprivate.
(gimplify_adjust_omp_clauses_1): Ignore GOVD_NONTEMPORAL. Handle
GOVD_MAP_ALLOC_ONLY and GOVD_MAP_FROM_ONLY.
(gimplify_adjust_omp_clauses): Handle OMP_CLAUSE_NONTEMPORAL. Handle
OMP_CLAUSE_{TASK,IN}_REDUCTION.
(gimplify_omp_task): Handle taskwait with depend clauses.
(gimplify_omp_for): Add shared clause on parallel for combined
parallel master taskloop{, simd} if taskloop has firstprivate,
lastprivate or reduction clause. Use ORT_TASKLOOP or
ORT_UNTIED_TASKLOOP instead of ORT_TASK or ORT_UNTIED_TASK. Adjust
tests for ORT_COMBINED_TEAMS. Handle C++ range for loops with
NULL TREE_PURPOSE in OMP_FOR_ORIG_DECLS. Firstprivatize
__for_end and __for_range temporaries on OMP_PARALLEL for
distribute parallel for{, simd}. Move OMP_CLAUSE_REDUCTION
and OMP_CLAUSE_IN_REDUCTION from taskloop to the task construct
sandwiched in between two taskloops.
(computable_teams_clause): Test ctx->defaultmap[GDMK_SCALAR]
instead of ctx->omp_firstprivatize_variable.
(gimplify_omp_workshare): Set ort to ORT_HOST_TEAMS or
ORT_COMBINED_HOST_TEAMS if not inside of target construct. If
host teams, use gimplify_and_return_first etc. for body like
for target or target data constructs, and at the end call
gimple_omp_teams_set_host on the GIMPLE_OMP_TEAMS object.
(gimplify_omp_atomic): Use OMP_ATOMIC_MEMORY_ORDER instead
of OMP_ATOMIC_SEQ_CST, pass it as new argument to
gimple_build_omp_atomic_load and gimple_build_omp_atomic_store, remove
gimple_omp_atomic_set_seq_cst calls.
(gimplify_expr) <case OMP_TASKGROUP>: Move handling into a separate
case, handle taskgroup clauses.
* lto-streamer-out.c (hash_tree): Handle
OMP_CLAUSE_{TASK,IN}_REDUCTION.
* Makefile.in (GTFILES): Add omp-general.h.
* omp-builtins.def (BUILT_IN_GOMP_TASKWAIT_DEPEND,
BUILT_IN_GOMP_LOOP_NONMONOTONIC_RUNTIME_START,
BUILT_IN_GOMP_LOOP_MAYBE_NONMONOTONIC_RUNTIME_START,
BUILT_IN_GOMP_LOOP_START, BUILT_IN_GOMP_LOOP_ORDERED_START,
BUILT_IN_GOMP_LOOP_DOACROSS_START,
BUILT_IN_GOMP_LOOP_NONMONOTONIC_RUNTIME_NEXT,
BUILT_IN_GOMP_LOOP_MAYBE_NONMONOTONIC_RUNTIME_NEXT,
BUILT_IN_GOMP_LOOP_ULL_NONMONOTONIC_RUNTIME_START,
BUILT_IN_GOMP_LOOP_ULL_MAYBE_NONMONOTONIC_RUNTIME_START,
BUILT_IN_GOMP_LOOP_ULL_START, BUILT_IN_GOMP_LOOP_ULL_ORDERED_START,
BUILT_IN_GOMP_LOOP_ULL_DOACROSS_START,
BUILT_IN_GOMP_LOOP_ULL_NONMONOTONIC_RUNTIME_NEXT,
BUILT_IN_GOMP_LOOP_ULL_MAYBE_NONMONOTONIC_RUNTIME_NEXT,
BUILT_IN_GOMP_PARALLEL_LOOP_NONMONOTONIC_RUNTIME,
BUILT_IN_GOMP_PARALLEL_LOOP_MAYBE_NONMONOTONIC_RUNTIME,
BUILT_IN_GOMP_PARALLEL_REDUCTIONS, BUILT_IN_GOMP_SECTIONS2_START,
BUILT_IN_GOMP_TEAMS_REG, BUILT_IN_GOMP_TASKGROUP_REDUCTION_REGISTER,
BUILT_IN_GOMP_TASKGROUP_REDUCTION_UNREGISTER,
BUILT_IN_GOMP_TASK_REDUCTION_REMAP,
BUILT_IN_GOMP_WORKSHARE_TASK_REDUCTION_UNREGISTER): New builtins.
* omp-expand.c (workshare_safe_to_combine_p): Return false for
non-worksharing loops.
(omp_adjust_chunk_size): Don't adjust anything if chunk_size is zero.
(determine_parallel_type): Don't combine parallel with worksharing
which has _reductemp_ clause.
(expand_parallel_call): Emit the GOMP_*nonmonotonic_runtime* or
GOMP_*maybe_nonmonotonic_runtime* builtins instead of GOMP_*runtime*
if there is nonmonotonic modifier or if there is no modifier and no
ordered clause. For dynamic and guided schedule without monotonic
and nonmonotonic modifier, default to nonmonotonic.
(expand_omp_for): Likewise. Adjust expand_omp_for_generic caller, use
GOMP_loop{,_ull}{,_ordered,_doacross}_start builtins if there are
task reductions.
(expand_task_call): Add GOMP_TASK_FLAG_REDUCTION flag to flags if
there are any reduction clauses.
(expand_taskwait_call): New function.
(expand_teams_call): New function.
(expand_omp_taskreg): Allow GIMPLE_OMP_TEAMS and call
expand_teams_call for it. Formatting fix. Handle taskwait with
depend clauses.
(expand_omp_for_generic): Add SCHED_ARG argument. Handle expansion
of worksharing loops with task reductions.
(expand_omp_for_static_nochunk, expand_omp_for_static_chunk): Handle
expansion of worksharing loops with task reductions.
(expand_omp_sections): Handle expansion of sections with task
reductions.
(expand_omp_synch): For host teams call expand_omp_taskreg.
(omp_memory_order_to_memmodel): New function.
(expand_omp_atomic_load, expand_omp_atomic_store,
expand_omp_atomic_fetch_op): Use it and gimple_omp_atomic_memory_order
instead of gimple_omp_atomic_seq_cst_p.
(build_omp_regions_1, omp_make_gimple_edges): Treat taskwait with
depend clauses as a standalone directive.
* omp-general.c (enum omp_requires): New variable.
(omp_extract_for_data): Initialize have_reductemp member. Allow
NE_EXPR even in OpenMP loops, transform them into LT_EXPR or
GT_EXPR loops depending on incr sign. Formatting fixes.
* omp-general.h (struct omp_for_data): Add have_reductemp member.
(enum omp_requires): New enum.
(omp_requires_mask): Declare.
* omp-grid.c (grid_eliminate_combined_simd_part): Formatting fix.
Fix comment typos.
* omp-low.c (struct omp_context): Add task_reductions and
task_reduction_map fields.
(is_host_teams_ctx): New function.
(is_taskreg_ctx): Return true also if is_host_teams_ctx.
(use_pointer_for_field): Use is_global_var instead of
TREE_STATIC || DECL_EXTERNAL, and apply only if not privatized
in outer contexts.
(build_outer_var_ref): Ignore taskgroup outer contexts.
(delete_omp_context): Release task_reductions and task_reduction_map.
(scan_sharing_clauses): Don't add any fields for reduction clause on
taskloop. Handle OMP_CLAUSE__REDUCTEMP_. Handle
OMP_CLAUSE_{IN,TASK}_REDUCTION and OMP_CLAUSE_REDUCTION with task
modifier. Don't ignore shared clauses in is_host_teams_ctx contexts.
Handle OMP_CLAUSE_NONTEMPORAL.
(add_taskreg_looptemp_clauses): Add OMP_CLAUSE__REDUCTEMP_ clause if
needed.
(scan_omp_parallel): Add _reductemp_ clause if there are any reduction
clauses with task modifier.
(scan_omp_task): Handle taskwait with depend clauses.
(finish_taskreg_scan): Move field corresponding to _reductemp_ clause
first. Move also OMP_CLAUSE__REDUCTEMP_ clause in front if present.
Handle GIMPLE_OMP_TEAMS like GIMPLE_OMP_PARALLEL.
(scan_omp_for): Fix comment formatting.
(scan_omp_teams): Handle host teams constructs.
(check_omp_nesting_restrictions): Allow teams with no outer
OpenMP context. Adjust diagnostics for teams strictly nested into
some explicit OpenMP construct other than target. Allow OpenMP atomics
inside of simd regions.
(scan_omp_1_stmt): Call scan_sharing_clauses for taskgroups.
(scan_omp_1_stmt) <case GIMPLE_OMP_TEAMS>: Temporarily bump
taskreg_nesting_level while scanning host teams construct.
(task_reduction_read): New function.
(lower_rec_input_clauses): Handle OMP_CLAUSE_REDUCTION on taskloop
construct. Handle OMP_CLAUSE_IN_REDUCTION and OMP_CLAUSE__REDUCTEMP_
clauses. Handle OMP_CLAUSE_REDUCTION with task modifier. Remove
second argument create_tmp_var if it is NULL. Don't ignore shared
clauses in is_host_teams_ctx contexts. Handle
OMP_CLAUSE_FIRSTPRIVATE_NO_REFERENCE on OMP_CLAUSE_FIRSTPRIVATE
clauses.
(lower_reduction_clauses): Ignore reduction clauses with task
modifier. Remove second argument create_tmp_var if it is NULL.
Initialize OMP_ATOMIC_MEMORY_ORDER to relaxed.
(lower_send_clauses): Ignore reduction clauses with task modifier.
Handle OMP_CLAUSE__REDUCTEMP_. Don't send anything for
OMP_CLAUSE_REDUCTION on taskloop. Handle OMP_CLAUSE_IN_REDUCTION.
(maybe_add_implicit_barrier_cancel): Add OMP_RETURN argument, don't
rely that it is the last stmt in body so far. Ignore outer taskgroup
contexts.
(omp_task_reductions_find_first, omp_task_reduction_iterate,
lower_omp_task_reductions): New functions.
(lower_omp_sections): Handle reduction clauses with taskgroup
modifiers. Adjust maybe_add_implicit_barrier_cancel caller.
(lower_omp_single): Adjust maybe_add_implicit_barrier_cancel caller.
(lower_omp_for): Likewise. Handle reduction clauses with taskgroup
modifiers.
(lower_omp_taskgroup): Handle taskgroup reductions.
(create_task_copyfn): Copy over OMP_CLAUSE__REDUCTEMP_ pointer.
Handle OMP_CLAUSE_IN_REDUCTION and OMP_CLAUSE_REDUCTION clauses.
(lower_depend_clauses): If there are any
OMP_CLAUSE_DEPEND_DEPOBJ or OMP_CLAUSE_DEPEND_MUTEXINOUTSET
depend clauses, use a new array format. If OMP_CLAUSE_DEPEND_LAST is
seen, assume lowering is done already and return early. Set kind
on artificial depend clause to OMP_CLAUSE_DEPEND_LAST.
(lower_omp_taskreg): Handle reduction clauses with task modifier on
parallel construct. Handle reduction clause on taskloop construct.
Handle taskwait with depend clauses.
(lower_omp_1): Use lower_omp_taskreg instead of lower_omp_teams
for host teams constructs.
* tree.c (omp_clause_num_ops): Add in_reduction, task_reduction,
nontemporal and _reductemp_ clause entries.
(omp_clause_code_name): Likewise.
(walk_tree_1): Handle OMP_CLAUSE_{IN,TASK}_REDUCTION,
OMP_CLAUSE_NONTEMPORAL and OMP_CLAUSE__REDUCTEMP_.
* tree-core.h (enum omp_clause_code): Add
OMP_CLAUSE_{{IN,TASK}_REDUCTION,NONTEMPORAL,_REDUCTEMP_}.
(enum omp_clause_defaultmap_kind, enum omp_memory_order): New.
(struct tree_base): Add omp_atomic_memory_order field into union.
Remove OMP_ATOMIC_SEQ_CST comment.
(enum omp_clause_depend_kind): Add OMP_CLAUSE_DEPEND_MUTEXINOUTSET
and OMP_CLAUSE_DEPEND_DEPOBJ.
(struct tree_omp_clause): Add subcode.defaultmap_kind.
* tree.def (OMP_TASKGROUP): Add another operand, move next to other
OpenMP constructs with body and clauses operands.
* tree.h (OMP_BODY): Use OMP_MASTER instead of OMP_TASKGROUP.
(OMP_CLAUSES): Use OMP_TASKGROUP instead of OMP_SINGLE.
(OMP_TASKGROUP_CLAUSES): Define.
(OMP_CLAUSE_DECL): Use OMP_CLAUSE__REDUCTEMP_ instead of
OMP_CLAUSE__LOOPTEMP_.
(OMP_ATOMIC_SEQ_CST): Remove.
(OMP_ATOMIC_MEMORY_ORDER, OMP_CLAUSE_FIRSTPRIVATE_NO_REFERENCE,
OMP_CLAUSE_LASTPRIVATE_CONDITIONAL): Define.
(OMP_CLAUSE_REDUCTION_CODE, OMP_CLAUSE_REDUCTION_INIT,
OMP_CLAUSE_REDUCTION_MERGE, OMP_CLAUSE_REDUCTION_PLACEHOLDER,
OMP_CLAUSE_REDUCTION_DECL_PLACEHOLDER,
OMP_CLAUSE_REDUCTION_OMP_ORIG_REF): Handle
OMP_CLAUSE_{,IN_,TASK_}REDUCTION.
(OMP_CLAUSE_REDUCTION_TASK, OMP_CLAUSE_REDUCTION_INSCAN,
OMP_CLAUSE_DEFAULTMAP_KIND, OMP_CLAUSE_DEFAULTMAP_CATEGORY,
OMP_CLAUSE_DEFAULTMAP_BEHAVIOR, OMP_CLAUSE_DEFAULTMAP_SET_KIND):
Define.
* tree-inline.c (remap_gimple_stmt): Remap taskgroup clauses.
* tree-nested.c (convert_nonlocal_omp_clauses): Handle
OMP_CLAUSE__REDUCTEMP_, OMP_CLAUSE_NONTEMPORAL.
(convert_local_omp_clauses): Likewise. Remove useless test.
* tree-parloops.c (create_call_for_reduction_1): Pass
OMP_MEMORY_ORDER_RELAXED as new argument to
dump_gimple_omp_atomic_load and dump_gimple_omp_atomic_store.
* tree-pretty-print.c (dump_omp_iterators): New function.
(dump_omp_clause): Handle OMP_CLAUSE__REDUCTEMP_,
OMP_CLAUSE_NONTEMPORAL, OMP_CLAUSE_{TASK,IN}_REDUCTION. Print
reduction modifiers. Handle OMP_CLAUSE_DEPEND_DEPOBJ and
OMP_CLAUSE_DEPEND_MUTEXINOUTSET. Print iterators in depend clauses.
Print __internal__ for OMP_CLAUSE_DEPEND_LAST. Handle cancel and
simd OMP_CLAUSE_IF_MODIFIERs. Handle new kinds of
OMP_CLAUSE_DEFAULTMAP. Print conditional: for
OMP_CLAUSE_LASTPRIVATE_CONDITIONAL.
(dump_omp_atomic_memory_order): New function.
(dump_generic_node): Use it. Print taskgroup clauses. Print
taskwait with depend clauses.
* tree-pretty-print.h (dump_omp_atomic_memory_order): Declare.
* tree-streamer-in.c (unpack_ts_omp_clause_value_fields):
Handle OMP_CLAUSE_{TASK,IN}_REDUCTION.
* tree-streamer-out.c (pack_ts_omp_clause_value_fields,
write_ts_omp_clause_tree_pointers): Likewise.
gcc/c-family/
* c-common.h (c_finish_omp_taskgroup): Add CLAUSES argument.
(c_finish_omp_atomic): Replace bool SEQ_CST argument with
enum omp_memory_order MEMORY_ORDER.
(c_finish_omp_flush): Add MO argument.
(c_omp_depend_t_p, c_finish_omp_depobj): Declare.
(c_finish_omp_for): Add FINAL_P argument.
* c-omp.c: Include memmodel.h.
(c_finish_omp_taskgroup): Add CLAUSES argument. Set
OMP_TASKGROUP_CLAUSES to it.
(c_finish_omp_atomic): Replace bool SEQ_CST argument with
enum omp_memory_order MEMORY_ORDER. Set OMP_ATOMIC_MEMORY_ORDER
instead of OMP_ATOMIC_SEQ_CST.
(c_omp_depend_t_p, c_finish_omp_depobj): New functions.
(c_finish_omp_flush): Add MO argument, if not MEMMODEL_LAST, emit
__atomic_thread_fence call with the given value.
(check_omp_for_incr_expr): Formatting fixes.
(c_finish_omp_for): Add FINAL_P argument. Allow NE_EXPR
even in OpenMP loops, diagnose if NE_EXPR and incr expression
is not constant expression 1 or -1. Transform NE_EXPR loops
with iterators pointers to VLA into LT_EXPR or GT_EXPR loops.
(c_omp_check_loop_iv_r): Look for orig decl of C++ range for
loops too.
(c_omp_split_clauses): Add support for combined
#pragma omp parallel master and
#pragma omp {,parallel }master taskloop{, simd} constructs.
Handle OMP_CLAUSE_IN_REDUCTION. Handle OMP_CLAUSE_REDUCTION_TASK.
Handle OMP_CLAUSE_NONTEMPORAL. Handle splitting OMP_CLAUSE_IF
also to OMP_SIMD. Copy OMP_CLAUSE_LASTPRIVATE_CONDITIONAL.
(c_omp_predetermined_sharing): Don't return
OMP_CLAUSE_DEFAULT_SHARED for const qualified decls.
* c-pragma.c (omp_pragmas): Add PRAGMA_OMP_DEPOBJ and
PRAGMA_OMP_REQUIRES.
* c-pragma.h (enum pragma_kind): Likewise.
(enum pragma_omp_clause): Add PRAGMA_OMP_CLAUSE_NONTEMPORAL
and PRAGMA_OMP_CLAUSE_{IN,TASK}_REDUCTION.
gcc/c/
* c-parser.c: Include memmode.h.
(c_parser_omp_depobj, c_parser_omp_requires): New functions.
(c_parser_pragma): Handle PRAGMA_OMP_DEPOBJ and PRAGMA_OMP_REQUIRES.
(c_parser_omp_clause_name): Handle nontemporal, in_reduction and
task_reduction clauses.
(c_parser_omp_variable_list): Handle OMP_CLAUSE_{IN,TASK}_REDUCTION.
For OMP_CLAUSE_DEPEND, parse clause operands as either an array
section, or lvalue assignment expression.
(c_parser_omp_clause_if): Handle cancel and simd modifiers.
(c_parser_omp_clause_lastprivate): Parse optional
conditional: modifier.
(c_parser_omp_clause_hint): Require constant integer expression rather
than just integer expression.
(c_parser_omp_clause_defaultmap): Parse new kinds of defaultmap
clause.
(c_parser_omp_clause_reduction): Add IS_OMP and KIND arguments.
Parse reduction modifiers. Pass KIND to c_parser_omp_variable_list.
(c_parser_omp_clause_nontemporal, c_parser_omp_iterators): New
functions.
(c_parser_omp_clause_depend): Parse iterator modifier and handle
iterators. Parse mutexinoutset and depobj kinds.
(c_parser_oacc_all_clauses): Adjust c_parser_omp_clause_reduction
callers.
(c_parser_omp_all_clauses): Likewise. Handle
PRAGMA_OMP_CLAUSE_NONTEMPORAL and
PRAGMA_OMP_CLAUSE_{IN,TASK}_REDUCTION.
(c_parser_omp_atomic): Parse hint and memory order clauses. Handle
default memory order from requires directive if any. Adjust
c_finish_omp_atomic caller.
(c_parser_omp_critical): Allow comma in between (name) and hint clause.
(c_parser_omp_flush): Parse flush with memory-order-clause.
(c_parser_omp_for_loop): Allow NE_EXPR even in
OpenMP loops, adjust c_finish_omp_for caller.
(OMP_SIMD_CLAUSE_MASK): Add if and nontemporal clauses.
(c_parser_omp_master): Add p_name, mask and cclauses arguments.
Allow to be called while parsing combined parallel master.
Parse combined master taskloop{, simd}.
(c_parser_omp_parallel): Parse combined
parallel master{, taskloop{, simd}} constructs.
(OMP_TASK_CLAUSE_MASK): Add in_reduction clause.
(OMP_TASKGROUP_CLAUSE_MASK): Define.
(c_parser_omp_taskgroup): Add LOC argument. Parse taskgroup clauses.
(OMP_TASKWAIT_CLAUSE_MASK): Define.
(c_parser_omp_taskwait): Handle taskwait with depend clauses.
(c_parser_omp_teams): Force a BIND_EXPR with BLOCK
around teams body. Use SET_EXPR_LOCATION.
(c_parser_omp_target_data): Allow target data
with only use_device_ptr clauses.
(c_parser_omp_target): Use SET_EXPR_LOCATION. Set
OMP_REQUIRES_TARGET_USED bit in omp_requires_mask.
(c_parser_omp_requires): New function.
(c_finish_taskloop_clauses): New function.
(OMP_TASKLOOP_CLAUSE_MASK): Add reduction and in_reduction clauses.
(c_parser_omp_taskloop): Use c_finish_taskloop_clauses. Add forward
declaration. Disallow in_reduction clause when combined with parallel
master.
(c_parser_omp_construct): Adjust c_parser_omp_master and
c_parser_omp_taskgroup callers.
* c-typeck.c (c_finish_omp_cancel): Diagnose if clause with modifier
other than cancel.
(handle_omp_array_sections_1): Handle OMP_CLAUSE_{IN,TASK}_REDUCTION
like OMP_CLAUSE_REDUCTION.
(handle_omp_array_sections): Likewise. Call save_expr on array
reductions before calling build_index_type. Handle depend clauses
with iterators.
(struct c_find_omp_var_s): New type.
(c_find_omp_var_r, c_omp_finish_iterators): New functions.
(c_finish_omp_clauses): Don't diagnose nonmonotonic clause
with static, runtime or auto schedule kinds. Call save_expr for whole
array reduction sizes. Diagnose reductions with zero sized elements
or variable length structures. Diagnose nogroup clause used with
reduction clause(s). Handle depend clause with
OMP_CLAUSE_DEPEND_DEPOBJ. Diagnose bit-fields. Require
omp_depend_t type for OMP_CLAUSE_DEPEND_DEPOBJ kinds and
some different type for other kinds. Use build_unary_op with
ADDR_EXPR and build_indirect_ref instead of c_mark_addressable.
Handle depend clauses with iterators. Remove no longer needed special
case that predetermined const qualified vars may be specified in
firstprivate clause. Complain if const qualified vars are mentioned
in data-sharing clauses other than firstprivate or shared. Use
error_at with OMP_CLAUSE_LOCATION (c) as first argument instead of
error. Formatting fix. Handle OMP_CLAUSE_NONTEMPORAL and
OMP_CLAUSE_{IN,TASK}_REDUCTION. Allow any lvalue as
OMP_CLAUSE_DEPEND operand (besides array section), adjust diagnostics.
gcc/cp/
* constexpr.c (potential_constant_expression_1): Handle OMP_DEPOBJ.
* cp-gimplify.c (cp_genericize_r): Handle
OMP_CLAUSE_{IN,TASK}_REDUCTION.
(cxx_omp_predetermined_sharing_1): Don't return
OMP_CLAUSE_DEFAULT_SHARED for const qualified decls with no mutable
member. Return OMP_CLAUSE_DEFAULT_FIRSTPRIVATE for this pointer.
* cp-objcp-common.c (cp_common_init_ts): Handle OMP_DEPOBJ.
* cp-tree.def (OMP_DEPOBJ): New tree code.
* cp-tree.h (OMP_ATOMIC_DEPENDENT_P): Return true also for first
argument being OMP_CLAUSE.
(OMP_DEPOBJ_DEPOBJ, OMP_DEPOBJ_CLAUSES): Define.
(cp_convert_omp_range_for, cp_finish_omp_range_for): Declare.
(finish_omp_atomic): Add LOC, CLAUSES and MO arguments. Remove
SEQ_CST argument.
(finish_omp_for_block): Declare.
(finish_omp_flush): Add MO argument.
(finish_omp_depobj): Declare.
* cxx-pretty-print.c (cxx_pretty_printer::statement): Handle
OMP_DEPOBJ.
* dump.c (cp_dump_tree): Likewise.
* lex.c (cxx_init): Likewise.
* parser.c: Include memmodel.h.
(cp_parser_for): Pass false as new is_omp argument to
cp_parser_range_for.
(cp_parser_range_for): Add IS_OMP argument, return before finalizing
if it is true.
(cp_parser_omp_clause_name): Handle nontemporal, in_reduction and
task_reduction clauses.
(cp_parser_omp_var_list_no_open): Handle
OMP_CLAUSE_{IN,TASK}_REDUCTION. For OMP_CLAUSE_DEPEND, parse clause
operands as either an array section, or lvalue assignment expression.
(cp_parser_omp_clause_if): Handle cancel and simd modifiers.
(cp_parser_omp_clause_defaultmap): Parse new kinds of defaultmap
clause.
(cp_parser_omp_clause_reduction): Add IS_OMP and KIND arguments.
Parse reduction modifiers. Pass KIND to c_parser_omp_variable_list.
(cp_parser_omp_clause_lastprivate, cp_parser_omp_iterators): New
functions.
(cp_parser_omp_clause_depend): Parse iterator modifier and handle
iterators. Parse mutexinoutset and depobj kinds.
(cp_parser_oacc_all_clauses): Adjust cp_parser_omp_clause_reduction
callers.
(cp_parser_omp_all_clauses): Likewise. Handle
PRAGMA_OMP_CLAUSE_NONTEMPORAL and
PRAGMA_OMP_CLAUSE_{IN,TASK}_REDUCTION. Call
cp_parser_omp_clause_lastprivate for OpenMP lastprivate clause.
(cp_parser_omp_atomic): Pass pragma_tok->location as
LOC to finish_omp_atomic. Parse hint and memory order clauses.
Handle default memory order from requires directive if any. Adjust
finish_omp_atomic caller.
(cp_parser_omp_critical): Allow comma in between (name) and hint
clause.
(cp_parser_omp_depobj): New function.
(cp_parser_omp_flush): Parse flush with memory-order-clause.
(cp_parser_omp_for_cond): Allow NE_EXPR even in OpenMP loops.
(cp_convert_omp_range_for, cp_finish_omp_range_for): New functions.
(cp_parser_omp_for_loop): Parse C++11 range for loops among omp
loops. Handle OMP_CLAUSE_IN_REDUCTION like OMP_CLAUSE_REDUCTION.
(OMP_SIMD_CLAUSE_MASK): Add if and nontemporal clauses.
(cp_parser_omp_simd, cp_parser_omp_for): Call keep_next_level before
begin_omp_structured_block and call finish_omp_for_block on
finish_omp_structured_block result.
(cp_parser_omp_master): Add p_name, mask and cclauses arguments.
Allow to be called while parsing combined parallel master.
Parse combined master taskloop{, simd}.
(cp_parser_omp_parallel): Parse combined
parallel master{, taskloop{, simd}} constructs.
(cp_parser_omp_single): Use SET_EXPR_LOCATION.
(OMP_TASK_CLAUSE_MASK): Add in_reduction clause.
(OMP_TASKWAIT_CLAUSE_MASK): Define.
(cp_parser_omp_taskwait): Handle taskwait with depend clauses.
(OMP_TASKGROUP_CLAUSE_MASK): Define.
(cp_parser_omp_taskgroup): Parse taskgroup clauses, adjust
c_finish_omp_taskgroup caller.
(cp_parser_omp_distribute): Call keep_next_level before
begin_omp_structured_block and call finish_omp_for_block on
finish_omp_structured_block result.
(cp_parser_omp_teams): Force a BIND_EXPR with BLOCK around teams
body.
(cp_parser_omp_target_data): Allow target data with only
use_device_ptr clauses.
(cp_parser_omp_target): Set OMP_REQUIRES_TARGET_USED bit in
omp_requires_mask.
(cp_parser_omp_requires): New function.
(OMP_TASKLOOP_CLAUSE_MASK): Add reduction and in_reduction clauses.
(cp_parser_omp_taskloop): Add forward declaration. Disallow
in_reduction clause when combined with parallel master. Call
keep_next_level before begin_omp_structured_block and call
finish_omp_for_block on finish_omp_structured_block result.
(cp_parser_omp_construct): Adjust cp_parser_omp_master caller.
(cp_parser_pragma): Handle PRAGMA_OMP_DEPOBJ and PRAGMA_OMP_REQUIRES.
* pt.c (tsubst_omp_clause_decl): Add iterators_cache argument.
Adjust recursive calls. Handle iterators.
(tsubst_omp_clauses): Handle OMP_CLAUSE_{IN,TASK}_REDUCTION and
OMP_CLAUSE_NONTEMPORAL. Adjust tsubst_omp_clause_decl callers.
(tsubst_decomp_names):
(tsubst_omp_for_iterator): Change orig_declv into a reference.
Handle range for loops. Move orig_declv handling after declv/initv
handling.
(tsubst_expr): Force a BIND_EXPR with BLOCK around teams body.
Adjust finish_omp_atomic caller. Call keep_next_level before
begin_omp_structured_block. Call cp_finish_omp_range_for for range
for loops and use {begin,finish}_omp_structured_block instead of
{push,pop}_stmt_list if there are any range for loops. Call
finish_omp_for_block on finish_omp_structured_block result.
Handle OMP_DEPOBJ. Handle taskwait with depend clauses. For
OMP_ATOMIC call tsubst_omp_clauses on clauses if any, adjust
finish_omp_atomic caller. Use OMP_ATOMIC_MEMORY_ORDER rather
than OMP_ATOMIC_SEQ_CST. Handle clauses on OMP_TASKGROUP.
(dependent_omp_for_p): Always return true for range for loops if
processing_template_decl. Return true if class type iterator
does not have INTEGER_CST increment.
* semantics.c: Include memmodel.h.
(handle_omp_array_sections_1): Handle OMP_CLAUSE_{IN,TASK}_REDUCTION
like OMP_CLAUSE_REDUCTION.
(handle_omp_array_sections): Likewise. Call save_expr on array
reductions before calling build_index_type. Handle depend clauses
with iterators.
(finish_omp_reduction_clause): Call save_expr for whole array
reduction sizes. Don't mark OMP_CLAUSE_DECL addressable if it has
reference type. Do mark decl_placeholder addressable if needed.
Use error_at with OMP_CLAUSE_LOCATION (c) as first argument instead
of error.
(cp_omp_finish_iterators): New function.
(finish_omp_clauses): Don't diagnose nonmonotonic clause with static,
runtime or auto schedule kinds. Diagnose nogroup clause used with
reduction clause(s). Handle depend clause with
OMP_CLAUSE_DEPEND_DEPOBJ. Diagnose bit-fields. Require
omp_depend_t type for OMP_CLAUSE_DEPEND_DEPOBJ kinds and
some different type for other kinds. Use cp_build_addr_expr
and cp_build_indirect_ref instead of cxx_mark_addressable.
Handle depend clauses with iterators. Only handle static data members
in the special case that const qualified vars may be specified in
firstprivate clause. Complain if const qualified vars without mutable
members are mentioned in data-sharing clauses other than firstprivate
or shared. Use error_at with OMP_CLAUSE_LOCATION (c) as first
argument instead of error. Diagnose more than one nontemporal clause
refering to the same variable. Use error_at rather than error for
priority and hint clause diagnostics. Fix pasto for hint clause.
Diagnose hint expression that doesn't fold into INTEGER_CST.
Diagnose if clause with modifier other than cancel. Handle
OMP_CLAUSE_{IN,TASK}_REDUCTION like OMP_CLAUSE_REDUCTION. Allow any
lvalue as OMP_CLAUSE_DEPEND operand (besides array section), adjust
diagnostics.
(handle_omp_for_class_iterator): Don't create a new TREE_LIST if one
has been created already for range for, just fill TREE_PURPOSE and
TREE_VALUE. Call cp_fully_fold on incr.
(finish_omp_for): Don't check cond/incr if cond is global_namespace.
Pass to c_omp_check_loop_iv_exprs orig_declv if non-NULL. Don't
use IS_EMPTY_STMT on NULL pre_body. Adjust c_finish_omp_for caller.
(finish_omp_for_block): New function.
(finish_omp_atomic): Add LOC argument, pass it through
to c_finish_omp_atomic and set it as location of OMP_ATOMIC* trees.
Remove SEQ_CST argument. Add CLAUSES and MO arguments. Adjust
c_finish_omp_atomic caller. Stick clauses if any into first argument
of wrapping OMP_ATOMIC.
(finish_omp_depobj): New function.
(finish_omp_flush): Add MO argument, if not
MEMMODEL_LAST, emit __atomic_thread_fence call with the given value.
(finish_omp_cancel): Diagnose if clause with modifier other than
cancel.
gcc/fortran/
* trans-openmp.c (gfc_trans_omp_clauses): Use
OMP_CLAUSE_DEFAULTMAP_SET_KIND.
(gfc_trans_omp_atomic): Set OMP_ATOMIC_MEMORY_ORDER
rather than OMP_ATOMIC_SEQ_CST.
(gfc_trans_omp_taskgroup): Build OMP_TASKGROUP using
make_node instead of build1_loc.
* types.def (BT_FN_VOID_BOOL, BT_FN_VOID_SIZE_SIZE_PTR,
BT_FN_UINT_UINT_PTR_PTR, BT_FN_UINT_OMPFN_PTR_UINT_UINT,
BT_FN_BOOL_UINT_LONGPTR_LONG_LONG_LONGPTR_LONGPTR_PTR_PTR,
BT_FN_BOOL_UINT_ULLPTR_LONG_ULL_ULLPTR_ULLPTR_PTR_PTR,
BT_FN_BOOL_LONG_LONG_LONG_LONG_LONG_LONGPTR_LONGPTR_PTR_PTR,
BT_FN_BOOL_BOOL_ULL_ULL_ULL_LONG_ULL_ULLPTR_ULLPTR_PTR_PTR): New.
(BT_FN_VOID_INT_OMPFN_SIZE_PTR_PTR_PTR_UINT_PTR_PTR): Formatting fix.
gcc/testsuite/
* c-c++-common/gomp/atomic-17.c: New test.
* c-c++-common/gomp/atomic-18.c: New test.
* c-c++-common/gomp/atomic-19.c: New test.
* c-c++-common/gomp/atomic-20.c: New test.
* c-c++-common/gomp/atomic-21.c: New test.
* c-c++-common/gomp/atomic-22.c: New test.
* c-c++-common/gomp/clauses-1.c (r2): New variable.
(foo): Add ntm argument and test if and nontemporal clauses on
constructs with simd.
(bar): Put taskloop simd inside of taskgroup with task_reduction,
use in_reduction clause instead of reduction. Add another
taskloop simd without nogroup clause, but with reduction clause and
a new in_reduction. Add ntm and i3 arguments. Test if and
nontemporal clauses on constructs with simd. Change if clauses on
some constructs from specific to the particular constituents to one
without a modifier. Add new tests for combined host teams and for
new parallel master and {,parallel }master taskloop{, simd} combined
constructs.
(baz): New function with host teams tests.
* gcc.dg/gomp/combined-1.c: Moved to ...
* c-c++-common/gomp/combined-1.c: ... here. Adjust expected library
call.
* c-c++-common/gomp/combined-2.c: New test.
* c-c++-common/gomp/combined-3.c: New test.
* c-c++-common/gomp/critical-1.c: New test.
* c-c++-common/gomp/critical-2.c: New test.
* c-c++-common/gomp/default-1.c: New test.
* c-c++-common/gomp/defaultmap-1.c: New test.
* c-c++-common/gomp/defaultmap-2.c: New test.
* c-c++-common/gomp/defaultmap-3.c: New test.
* c-c++-common/gomp/depend-5.c: New test.
* c-c++-common/gomp/depend-6.c: New test.
* c-c++-common/gomp/depend-iterator-1.c: New test.
* c-c++-common/gomp/depend-iterator-2.c: New test.
* c-c++-common/gomp/depobj-1.c: New test.
* c-c++-common/gomp/flush-1.c: New test.
* c-c++-common/gomp/flush-2.c: New test.
* c-c++-common/gomp/for-1.c: New test.
* c-c++-common/gomp/for-2.c: New test.
* c-c++-common/gomp/for-3.c: New test.
* c-c++-common/gomp/for-4.c: New test.
* c-c++-common/gomp/for-5.c: New test.
* c-c++-common/gomp/for-6.c: New test.
* c-c++-common/gomp/for-7.c: New test.
* c-c++-common/gomp/if-1.c (foo): Add some further tests.
* c-c++-common/gomp/if-2.c (foo): Likewise. Expect slightly different
diagnostics wording in one case.
* c-c++-common/gomp/if-3.c: New test.
* c-c++-common/gomp/master-combined-1.c: New test.
* c-c++-common/gomp/master-combined-2.c: New test.
* c-c++-common/gomp/nontemporal-1.c: New test.
* c-c++-common/gomp/nontemporal-2.c: New test.
* c-c++-common/gomp/reduction-task-1.c: New test.
* c-c++-common/gomp/reduction-task-2.c: New test.
* c-c++-common/gomp/requires-1.c: New test.
* c-c++-common/gomp/requires-2.c: New test.
* c-c++-common/gomp/requires-3.c: New test.
* c-c++-common/gomp/requires-4.c: New test.
* c-c++-common/gomp/schedule-modifiers-1.c (bar): Don't expect
diagnostics for nonmonotonic modifier with static, runtime or auto
schedule kinds.
* c-c++-common/gomp/simd7.c: New test.
* c-c++-common/gomp/target-data-1.c: New test.
* c-c++-common/gomp/taskloop-reduction-1.c: New test.
* c-c++-common/gomp/taskwait-depend-1.c: New test.
* c-c++-common/gomp/teams-1.c: New test.
* c-c++-common/gomp/teams-2.c: New test.
* gcc.dg/gomp/appendix-a/a.24.1.c: Update from OpenMP examples. Add
shared(c) clause.
* gcc.dg/gomp/atomic-5.c (f1): Add another expected error.
* gcc.dg/gomp/clause-1.c: Adjust expected diagnostics for const
qualified vars without mutable member no longer being predeterined
shared.
* gcc.dg/gomp/sharing-1.c: Likewise.
* g++.dg/gomp/clause-3.C: Likewise.
* g++.dg/gomp/member-2.C: Likewise.
* g++.dg/gomp/predetermined-1.C: Likewise.
* g++.dg/gomp/private-1.C: Likewise.
* g++.dg/gomp/sharing-1.C: Likewise.
* g++.dg/gomp/sharing-2.C: Likewise. Add a few tests with aggregate
const static data member without mutable elements.
* gcc.dg/gomp/for-4.c: Expected nonmonotonic functions in the dumps.
* gcc.dg/gomp/for-5.c: Likewise.
* gcc.dg/gomp/for-6.c: Change expected library call.
* gcc.dg/gomp/pr39495-2.c (foo): Don't expect errors on !=.
* gcc.dg/gomp/reduction-2.c: New test.
* gcc.dg/gomp/simd-1.c: New test.
* gcc.dg/gomp/teams-1.c: Adjust expected diagnostic lines.
* g++.dg/gomp/atomic-18.C: New test.
* g++.dg/gomp/atomic-19.C: New test.
* g++.dg/gomp/atomic-5.C (f1): Adjust expected lines of read-only
variable messages. Add another expected error.
* g++.dg/gomp/critical-3.C: New test.
* g++.dg/gomp/depend-iterator-1.C: New test.
* g++.dg/gomp/depend-iterator-2.C: New test.
* g++.dg/gomp/depobj-1.C: New test.
* g++.dg/gomp/doacross-1.C: New test.
* g++.dg/gomp/for-21.C: New test.
* g++.dg/gomp/for-4.C: Expected nonmonotonic functions in the dumps.
* g++.dg/gomp/for-5.C: Likewise.
* g++.dg/gomp/for-6.C: Change expected library call.
* g++.dg/gomp/loop-4.C: New test.
* g++.dg/gomp/pr33372-1.C: Adjust location of the expected
diagnostics.
* g++.dg/gomp/pr33372-3.C: Likewise.
* g++.dg/gomp/pr39495-2.C (foo): Don't expect errors on !=.
* g++.dg/gomp/simd-2.C: New test.
* g++.dg/gomp/tpl-atomic-2.C: Adjust expected diagnostic lines.
include/
* gomp-constants.h (GOMP_TASK_FLAG_REDUCTION,
GOMP_DEPEND_IN, GOMP_DEPEND_OUT, GOMP_DEPEND_INOUT,
GOMP_DEPEND_MUTEXINOUTSET): Define.
libgomp/
* affinity.c (gomp_display_affinity_place): New function.
* affinity-fmt.c: New file.
* alloc.c (gomp_aligned_alloc, gomp_aligned_free): New functions.
* config/linux/affinity.c (gomp_display_affinity_place): New function.
* config/nvptx/icv-device.c (omp_get_num_teams, omp_get_team_num):
Move these functions to ...
* config/nvptx/teams.c: ... here. New file.
* config/nvptx/target.c (omp_pause_resource, omp_pause_resource_all):
New functions.
* config/nvptx/team.c (gomp_team_start, gomp_pause_host): New
functions.
* configure.ac: Check for aligned_alloc, posix_memalign, memalign
and _aligned_malloc.
(HAVE_UNAME, HAVE_GETHOSTNAME, HAVE_GETPID): Add new tests.
* configure.tgt: Add -DUSING_INITIAL_EXEC_TLS to XCFLAGS for Linux.
* env.c (gomp_display_affinity_var, gomp_affinity_format_var,
gomp_affinity_format_len): New variables.
(parse_schedule): Parse monotonic and nonmonotonic modifiers in
OMP_SCHEDULE variable. Set GFS_MONOTONIC for monotonic schedules.
(handle_omp_display_env): Display monotonic/nonmonotonic schedule
modifiers. Display (non-default) chunk sizes. Print
OMP_DISPLAY_AFFINITY and OMP_AFFINITY_FORMAT.
(initialize_env): Don't call pthread_attr_setdetachstate. Handle
OMP_DISPLAY_AFFINITY and OMP_AFFINITY_FORMAT env vars.
* fortran.c: Include stdio.h and string.h.
(omp_pause_resource, omp_pause_resource_all): Add ialias_redirect.
(omp_get_schedule_, omp_get_schedule_8_): Mask off GFS_MONOTONIC bit.
(omp_set_affinity_format_, omp_get_affinity_format_,
omp_display_affinity_, omp_capture_affinity_, omp_pause_resource_,
omp_pause_resource_all_): New functions.
* icv.c (omp_set_schedule): Mask off omp_sched_monotonic bit in
switch.
* icv-device.c (omp_get_num_teams, omp_get_team_num): Move these
functions to ...
* teams.c: ... here. New file.
* libgomp_g.h: Include gstdint.h.
(GOMP_loop_nonmonotonic_runtime_start,
GOMP_loop_maybe_nonmonotonic_runtime_start, GOMP_loop_start,
GOMP_loop_ordered_start, GOMP_loop_nonmonotonic_runtime_next,
GOMP_loop_maybe_nonmonotonic_runtime_next, GOMP_loop_doacross_start,
GOMP_parallel_loop_nonmonotonic_runtime,
GOMP_parallel_loop_maybe_nonmonotonic_runtime,
GOMP_loop_ull_nonmonotonic_runtime_start,
GOMP_loop_ull_maybe_nonmonotonic_runtime_start, GOMP_loop_ull_start,
GOMP_loop_ull_ordered_start, GOMP_loop_ull_nonmonotonic_runtime_next,
GOMP_loop_ull_maybe_nonmonotonic_runtime_next,
GOMP_loop_ull_doacross_start, GOMP_parallel_reductions,
GOMP_taskwait_depend, GOMP_taskgroup_reduction_register,
GOMP_taskgroup_reduction_unregister, GOMP_task_reduction_remap,
GOMP_workshare_task_reduction_unregister, GOMP_sections2_start,
GOMP_teams_reg): Declare.
* libgomp.h (GOMP_HAVE_EFFICIENT_ALIGNED_ALLOC): Define unless
gomp_aligned_alloc uses fallback implementation.
(gomp_aligned_alloc, gomp_aligned_free): Declare.
(enum gomp_schedule_type): Add GFS_MONOTONIC.
(struct gomp_doacross_work_share): Add extra field.
(struct gomp_work_share): Add task_reductions field.
(struct gomp_taskgroup): Add workshare and reductions fields.
(GOMP_NEEDS_THREAD_HANDLE): Define if needed.
(gomp_thread_handle): New typedef.
(gomp_display_affinity_place, gomp_set_affinity_format,
gomp_display_string, gomp_display_affinity,
gomp_display_affinity_thread): Declare.
(gomp_doacross_init, gomp_doacross_ull_init): Add size_t argument.
(gomp_parallel_reduction_register, gomp_workshare_taskgroup_start,
gomp_workshare_task_reduction_register): Declare.
(gomp_team_start): Add taskgroup argument.
(gomp_pause_host): Declare.
(gomp_init_work_share, gomp_work_share_start): Change bool argument
to size_t.
(gomp_thread_self, gomp_thread_to_pthread_t): New inline functions.
* libgomp.map (GOMP_5.0): Export GOMP_loop_start,
GOMP_loop_ordered_start, GOMP_loop_doacross_start,
GOMP_loop_ull_start, GOMP_loop_ull_ordered_start,
GOMP_loop_ull_doacross_start,
GOMP_workshare_task_reduction_unregister, GOMP_sections2_start,
GOMP_loop_maybe_nonmonotonic_runtime_next,
GOMP_loop_maybe_nonmonotonic_runtime_start,
GOMP_loop_nonmonotonic_runtime_next,
GOMP_loop_nonmonotonic_runtime_start,
GOMP_loop_ull_maybe_nonmonotonic_runtime_next,
GOMP_loop_ull_maybe_nonmonotonic_runtime_start,
GOMP_loop_ull_nonmonotonic_runtime_next,
GOMP_loop_ull_nonmonotonic_runtime_start,
GOMP_parallel_loop_maybe_nonmonotonic_runtime,
GOMP_parallel_loop_nonmonotonic_runtime, GOMP_parallel_reductions,
GOMP_taskgroup_reduction_register,
GOMP_taskgroup_reduction_unregister, GOMP_task_reduction_remap,
GOMP_teams_reg and GOMP_taskwait_depend.
(OMP_5.0): Export omp_pause_resource{,_all}{,_},
omp_{capture,display}_affinity{,_}, and
omp_[gs]et_affinity_format{,_}.
* loop.c: Include string.h.
(GOMP_loop_runtime_next): Add ialias.
(GOMP_taskgroup_reduction_register): Add ialias_redirect.
(gomp_loop_static_start, gomp_loop_dynamic_start,
gomp_loop_guided_start, gomp_loop_ordered_static_start,
gomp_loop_ordered_dynamic_start, gomp_loop_ordered_guided_start,
gomp_loop_doacross_static_start, gomp_loop_doacross_dynamic_start,
gomp_loop_doacross_guided_start): Adjust gomp_work_share_start
or gomp_doacross_init callers.
(gomp_adjust_sched, GOMP_loop_start, GOMP_loop_ordered_start,
GOMP_loop_doacross_start): New functions.
(GOMP_loop_runtime_start, GOMP_loop_ordered_runtime_start,
GOMP_loop_doacross_runtime_start, GOMP_parallel_loop_runtime_start):
Mask off GFS_MONOTONIC bit.
(GOMP_loop_maybe_nonmonotonic_runtime_next,
GOMP_loop_maybe_nonmonotonic_runtime_start,
GOMP_loop_nonmonotonic_runtime_next,
GOMP_loop_nonmonotonic_runtime_start,
GOMP_parallel_loop_maybe_nonmonotonic_runtime,
GOMP_parallel_loop_nonmonotonic_runtime): New aliases or wrapper
functions.
(gomp_parallel_loop_start): Pass NULL as taskgroup to
gomp_team_start.
* loop_ull.c: Include string.h.
(GOMP_loop_ull_runtime_next): Add ialias.
(GOMP_taskgroup_reduction_register): Add ialias_redirect.
(gomp_loop_ull_static_start, gomp_loop_ull_dynamic_start,
gomp_loop_ull_guided_start, gomp_loop_ull_ordered_static_start,
gomp_loop_ull_ordered_dynamic_start,
gomp_loop_ull_ordered_guided_start,
gomp_loop_ull_doacross_static_start,
gomp_loop_ull_doacross_dynamic_start,
gomp_loop_ull_doacross_guided_start): Adjust gomp_work_share_start
and gomp_doacross_ull_init callers.
(gomp_adjust_sched, GOMP_loop_ull_start, GOMP_loop_ull_ordered_start,
GOMP_loop_ull_doacross_start): New functions.
(GOMP_loop_ull_runtime_start,
GOMP_loop_ull_ordered_runtime_start,
GOMP_loop_ull_doacross_runtime_start): Mask off GFS_MONOTONIC bit.
(GOMP_loop_ull_maybe_nonmonotonic_runtime_next,
GOMP_loop_ull_maybe_nonmonotonic_runtime_start,
GOMP_loop_ull_nonmonotonic_runtime_next,
GOMP_loop_ull_nonmonotonic_runtime_start): Likewise.
* Makefile.am (libgomp_la_SOURCES): Add teams.c and affinity-fmt.c.
* omp.h.in (enum omp_sched_t): Add omp_sched_monotonic.
(omp_pause_resource_t, omp_depend_t): New typedefs.
(enum omp_lock_hint_t): Renamed to ...
(enum omp_sync_hint_t): ... this. Define omp_sync_hint_*
enumerators using numbers and omp_lock_hint_* as their aliases.
(omp_lock_hint_t): New typedef. Rename to ...
(omp_sync_hint_t): ... this.
(omp_init_lock_with_hint, omp_init_nest_lock_with_hint): Use
omp_sync_hint_t instead of omp_lock_hint_t.
(omp_pause_resource, omp_pause_resource_all, omp_set_affinity_format,
omp_get_affinity_format, omp_display_affinity, omp_capture_affinity):
Declare.
(omp_target_is_present, omp_target_disassociate_ptr):
Change first argument from void * to const void *.
(omp_target_memcpy, omp_target_memcpy_rect): Change second argument
from void * to const void *.
(omp_target_associate_ptr): Change first and second arguments from
void * to const void *.
* omp_lib.f90.in (omp_pause_resource_kind, omp_pause_soft,
omp_pause_hard): New parameters.
(omp_pause_resource, omp_pause_resource_all, omp_set_affinity_format,
omp_get_affinity_format, omp_display_affinity, omp_capture_affinity):
New interfaces.
* omp_lib.h.in (omp_pause_resource_kind, omp_pause_soft,
omp_pause_hard): New parameters.
(omp_pause_resource, omp_pause_resource_all, omp_set_affinity_format,
omp_get_affinity_format, omp_display_affinity, omp_capture_affinity):
New externals.
* ordered.c (gomp_doacross_init, gomp_doacross_ull_init): Add
EXTRA argument. If not needed to prepare array, if extra is 0,
clear ws->doacross, otherwise allocate just doacross structure and
extra payload. If array is needed, allocate also extra payload.
(GOMP_doacross_post, GOMP_doacross_wait, GOMP_doacross_ull_post,
GOMP_doacross_ull_wait): Handle doacross->array == NULL like
doacross == NULL.
* parallel.c (GOMP_parallel_start): Pass NULL as taskgroup to
gomp_team_start.
(GOMP_parallel): Likewise. Formatting fix.
(GOMP_parallel_reductions): New function.
(GOMP_cancellation_point): If taskgroup has workshare
flag set, check cancelled of prev taskgroup if any.
(GOMP_cancel): If taskgroup has workshare flag set, set cancelled
on prev taskgroup if any.
* sections.c: Include string.h.
(GOMP_taskgroup_reduction_register): Add ialias_redirect.
(GOMP_sections_start): Adjust gomp_work_share_start caller.
(GOMP_sections2_start): New function.
(GOMP_parallel_sections_start, GOMP_parallel_sections):
Pass NULL as taskgroup to gomp_team_start.
* single.c (GOMP_single_start, GOMP_single_copy_start): Adjust
gomp_work_share_start callers.
* target.c (GOMP_target_update_ext, GOMP_target_enter_exit_data):
If taskgroup has workshare flag set, check cancelled on prev
taskgroup if any. Guard all cancellation tests with
gomp_cancel_var test.
(omp_target_is_present, omp_target_disassociate_ptr):
Change ptr argument from void * to const void *.
(omp_target_memcpy): Change src argument from void * to const void *.
(omp_target_memcpy_rect): Likewise.
(omp_target_memcpy_rect_worker): Likewise. Use const char * casts
instead of char * where needed.
(omp_target_associate_ptr): Change host_ptr and device_ptr arguments
from void * to const void *.
(omp_pause_resource, omp_pause_resource_all): New functions.
* task.c (gomp_task_handle_depend): Handle new depend array format
in addition to the old. Handle mutexinoutset kinds the same as
inout for now, handle unspecified kinds.
(gomp_create_target_task): If taskgroup has workshare flag set, check
cancelled on prev taskgroup if any. Guard all cancellation tests with
gomp_cancel_var test. Handle new depend array format count in
addition to the old.
(GOMP_task): Likewise. Adjust function comment.
(gomp_task_run_pre): If taskgroup has workshare flag set, check
cancelled on prev taskgroup if any. Guard all cancellation tests with
gomp_cancel_var test.
(GOMP_taskwait_depend): New function.
(gomp_task_maybe_wait_for_dependencies): Handle new depend array
format in addition to the old. Handle mutexinoutset kinds the same as
inout for now, handle unspecified kinds. Fix a function comment typo.
(gomp_taskgroup_init): New function.
(GOMP_taskgroup_start): Use it.
(gomp_reduction_register, gomp_create_artificial_team,
GOMP_taskgroup_reduction_register,
GOMP_taskgroup_reduction_unregister, GOMP_task_reduction_remap,
gomp_parallel_reduction_register,
gomp_workshare_task_reduction_register,
gomp_workshare_taskgroup_start,
GOMP_workshare_task_reduction_unregister): New functions.
* taskloop.c (GOMP_taskloop): If taskgroup has workshare flag set,
check cancelled on prev taskgroup if any. Guard all cancellation
tests with gomp_cancel_var test. Handle GOMP_TASK_FLAG_REDUCTION flag
by calling GOMP_taskgroup_reduction_register.
* team.c (gomp_thread_attr): Remove comment.
(struct gomp_thread_start_data): Add handle field.
(gomp_thread_start): Call pthread_detach.
(gomp_new_team): Adjust gomp_init_work_share caller.
(gomp_free_pool_helper): Call pthread_detach.
(gomp_team_start): Add taskgroup argument, initialize implicit
tasks' taskgroup field to that. Don't call
pthread_attr_setdetachstate. Handle OMP_DISPLAY_AFFINITY env var.
(gomp_team_end): Determine nesting by thr->ts.level != 0
rather than thr->ts.team != NULL.
(gomp_pause_pool_helper, gomp_pause_host): New functions.
* work.c (alloc_work_share): Use gomp_aligned_alloc instead of
gomp_malloc if GOMP_HAVE_EFFICIENT_ALIGNED_ALLOC is defined.
(gomp_init_work_share): Change ORDERED argument from bool to size_t,
if more than 1 allocate also extra payload at the end of array. Never
keep ordered_team_ids NULL, set it to inline_ordered_team_ids instead.
(gomp_work_share_start): Change ORDERED argument from bool to size_t,
return true instead of ws.
* Makefile.in: Regenerated.
* configure: Regenerated.
* config.h.in: Regenerated.
* testsuite/libgomp.c/cancel-for-2.c (foo): Use cancel modifier
in some cases.
* testsuite/libgomp.c-c++-common/cancel-parallel-1.c: New test.
* testsuite/libgomp.c-c++-common/cancel-taskgroup-3.c: New test.
* testsuite/libgomp.c-c++-common/depend-iterator-1.c: New test.
* testsuite/libgomp.c-c++-common/depend-iterator-2.c: New test.
* testsuite/libgomp.c-c++-common/depend-mutexinout-1.c: New test.
* testsuite/libgomp.c-c++-common/depend-mutexinout-2.c: New test.
* testsuite/libgomp.c-c++-common/depobj-1.c: New test.
* testsuite/libgomp.c-c++-common/display-affinity-1.c: New test.
* testsuite/libgomp.c-c++-common/for-10.c: New test.
* testsuite/libgomp.c-c++-common/for-11.c: New test.
* testsuite/libgomp.c-c++-common/for-12.c: New test.
* testsuite/libgomp.c-c++-common/for-13.c: New test.
* testsuite/libgomp.c-c++-common/for-14.c: New test.
* testsuite/libgomp.c-c++-common/for-15.c: New test.
* testsuite/libgomp.c-c++-common/for-2.h: If CONDNE macro is defined,
define a different N(test), don't define N(f0) to N(f14), but instead
define N(f20) to N(f34) using != comparisons.
* testsuite/libgomp.c-c++-common/for-7.c: New test.
* testsuite/libgomp.c-c++-common/for-8.c: New test.
* testsuite/libgomp.c-c++-common/for-9.c: New test.
* testsuite/libgomp.c-c++-common/master-combined-1.c: New test.
* testsuite/libgomp.c-c++-common/pause-1.c: New test.
* testsuite/libgomp.c-c++-common/pause-2.c: New test.
* testsuite/libgomp.c-c++-common/pr66199-10.c: New test.
* testsuite/libgomp.c-c++-common/pr66199-11.c: New test.
* testsuite/libgomp.c-c++-common/pr66199-12.c: New test.
* testsuite/libgomp.c-c++-common/pr66199-13.c: New test.
* testsuite/libgomp.c-c++-common/pr66199-14.c: New test.
* testsuite/libgomp.c-c++-common/simd-1.c: New test.
* testsuite/libgomp.c-c++-common/taskloop-reduction-1.c: New test.
* testsuite/libgomp.c-c++-common/taskloop-reduction-2.c: New test.
* testsuite/libgomp.c-c++-common/taskloop-reduction-3.c: New test.
* testsuite/libgomp.c-c++-common/taskloop-reduction-4.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-11.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-12.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-1.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-2.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-3.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-4.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-5.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-6.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-7.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-8.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-9.c: New test.
* testsuite/libgomp.c-c++-common/taskwait-depend-1.c: New test.
* testsuite/libgomp.c++/depend-1.C: New test.
* testsuite/libgomp.c++/depend-iterator-1.C: New test.
* testsuite/libgomp.c++/depobj-1.C: New test.
* testsuite/libgomp.c++/for-16.C: New test.
* testsuite/libgomp.c++/for-21.C: New test.
* testsuite/libgomp.c++/for-22.C: New test.
* testsuite/libgomp.c++/for-23.C: New test.
* testsuite/libgomp.c++/for-24.C: New test.
* testsuite/libgomp.c++/for-25.C: New test.
* testsuite/libgomp.c++/for-26.C: New test.
* testsuite/libgomp.c++/taskloop-reduction-1.C: New test.
* testsuite/libgomp.c++/taskloop-reduction-2.C: New test.
* testsuite/libgomp.c++/taskloop-reduction-3.C: New test.
* testsuite/libgomp.c++/taskloop-reduction-4.C: New test.
* testsuite/libgomp.c++/task-reduction-10.C: New test.
* testsuite/libgomp.c++/task-reduction-11.C: New test.
* testsuite/libgomp.c++/task-reduction-12.C: New test.
* testsuite/libgomp.c++/task-reduction-13.C: New test.
* testsuite/libgomp.c++/task-reduction-14.C: New test.
* testsuite/libgomp.c++/task-reduction-15.C: New test.
* testsuite/libgomp.c++/task-reduction-16.C: New test.
* testsuite/libgomp.c++/task-reduction-17.C: New test.
* testsuite/libgomp.c++/task-reduction-18.C: New test.
* testsuite/libgomp.c++/task-reduction-19.C: New test.
* testsuite/libgomp.c/task-reduction-1.c: New test.
* testsuite/libgomp.c++/task-reduction-1.C: New test.
* testsuite/libgomp.c/task-reduction-2.c: New test.
* testsuite/libgomp.c++/task-reduction-2.C: New test.
* testsuite/libgomp.c++/task-reduction-3.C: New test.
* testsuite/libgomp.c++/task-reduction-4.C: New test.
* testsuite/libgomp.c++/task-reduction-5.C: New test.
* testsuite/libgomp.c++/task-reduction-6.C: New test.
* testsuite/libgomp.c++/task-reduction-7.C: New test.
* testsuite/libgomp.c++/task-reduction-8.C: New test.
* testsuite/libgomp.c++/task-reduction-9.C: New test.
* testsuite/libgomp.c/teams-1.c: New test.
* testsuite/libgomp.c/teams-2.c: New test.
* testsuite/libgomp.c/thread-limit-4.c: New test.
* testsuite/libgomp.c/thread-limit-5.c: New test.
* testsuite/libgomp.fortran/display-affinity-1.f90: New test.
From-SVN: r265930
|
|
Cesar Philippidis
|
701d080a4b |
[libgomp] Truncate config/nvptx/oacc-parallel.c
libgomp/ * config/nvptx/oacc-parallel.c: Truncate. Co-Authored-By: James Norris <jnorris@codesourcery.com> Co-Authored-By: Thomas Schwinge <thomas@codesourcery.com> From-SVN: r263236 |
|
|
Jakub Jelinek
|
85ec4feb11 |
Update copyright years.
From-SVN: r256169 |
|
|
Jakub Jelinek
|
01275e1e5a |
re PR libgomp/80822 (libgomp incorrect affinity when OMP_PLACES=threads)
PR libgomp/80822 * config/linux/affinity.c (gomp_affinity_init_level_1): New function. (gomp_affinity_init_level): Use it. Always analyze the core and thread sibling lists, depending on level just pick up what CPUs to put together into a place vs. whether add multiple ordered places. From-SVN: r248683 |