aboutsummaryrefslogtreecommitdiff
path: root/libgomp
AgeCommit message (Collapse)AuthorFilesLines
2023-02-07Fix 'libgomp.fortran/reverse-offload-6.f90' nvptx offloading compilationThomas Schwinge1-0/+2
Fix-up for recent commit 0b1ce70a813b98ef2893779d14ad6c90c5d06a71 "libgomp: Fix reverse offload issues". libgomp/ * testsuite/libgomp.fortran/reverse-offload-6.f90: Fix nvptx offloading compilation.
2023-02-04Daily bump.GCC Administrator1-0/+13
2023-02-03libgomp: Fix reverse offload issuesTobias Burnus2-4/+36
If there is nothing to map, skip the mapping and avoid attempting to copy 0 bytes from addrs, sizes and kinds. Additionally, it could happen that a non-allocated address was deallocated, such as a pointer set, leading to a free for the actual data. libgomp/ * target.c (gomp_target_rev): Handle mapnum == 0 and avoid freeing not allocated memory. * testsuite/libgomp.fortran/reverse-offload-6.f90: New test.
2023-02-03libgomp: enable reverse offload for AMDGCNTobias Burnus2-6/+10
libgomp/ChangeLog: * libgomp.texi (5.0 Impl. Status, gcn specifics): Update for reverse offload. * plugin/plugin-gcn.c (GOMP_OFFLOAD_get_num_devices): Accept reverse-offload requirement.
2023-02-03Daily bump.GCC Administrator1-0/+39
2023-02-02amdgcn, libgomp: Manually allocated stacksAndrew Stubbs4-120/+179
Switch from using stacks in the "private segment" to using a memory block allocated on the host side. The primary reason is to permit the reverse offload implementation to access values located on the device stack, but there may also be performance benefits, especially with repeated kernel invocations. This implementation unifies the stacks with the "team arena" optimization feature, and now allows both to have run-time configurable sizes. A new ABI is needed, so all libraries must be rebuilt, and newlib must be version 4.3.0.20230120 or newer. gcc/ChangeLog: * config/gcn/gcn-run.cc: Include libgomp-gcn.h. (struct kernargs): Replace the common content with kernargs_abi. (struct heap): Delete. (main): Read GCN_STACK_SIZE envvar. Allocate space for the device stacks. Write the new kernargs fields. * config/gcn/gcn.cc (gcn_option_override): Remove stack_size_opt. (default_requested_args): Remove PRIVATE_SEGMENT_BUFFER_ARG and PRIVATE_SEGMENT_WAVE_OFFSET_ARG. (gcn_addr_space_convert): Mask the QUEUE_PTR_ARG content. (gcn_expand_prologue): Move the TARGET_PACKED_WORK_ITEMS to the top. Set up the stacks from the values in the kernargs, not private. (gcn_expand_builtin_1): Match the stack configuration in the prologue. (gcn_hsa_declare_function_name): Turn off the private segment. (gcn_conditional_register_usage): Ensure QUEUE_PTR is fixed. * config/gcn/gcn.h (FIXED_REGISTERS): Fix the QUEUE_PTR register. * config/gcn/gcn.opt (mstack-size): Change the description. include/ChangeLog: * gomp-constants.h (GOMP_VERSION_GCN): Bump. libgomp/ChangeLog: * config/gcn/libgomp-gcn.h (DEFAULT_GCN_STACK_SIZE): New define. (DEFAULT_TEAM_ARENA_SIZE): New define. (struct heap): Move to this file. (struct kernargs_abi): Likewise. * config/gcn/team.c (gomp_gcn_enter_kernel): Use team arena size from the kernargs. * libgomp.h: Include libgomp-gcn.h. (TEAM_ARENA_SIZE): Remove. (team_malloc): Update the error message. * plugin/plugin-gcn.c (struct kernargs): Move common content to struct kernargs_abi. (struct agent_info): Rename team arenas to ephemeral memories. (struct team_arena_list): Rename .... (struct ephemeral_memories_list): to this. (struct heap): Delete. (team_arena_size): New variable. (stack_size): New variable. (print_kernel_dispatch): Update debug messages. (init_environment_variables): Read GCN_TEAM_ARENA_SIZE. Read GCN_STACK_SIZE. (get_team_arena): Rename ... (configure_ephemeral_memories): ... to this, and set up stacks. (release_team_arena): Rename ... (release_ephemeral_memories): ... to this. (destroy_team_arenas): Rename ... (destroy_ephemeral_memories): ... to this. (create_kernel_dispatch): Add num_threads parameter. Adjust for kernargs_abi refactor and ephemeral memories. (release_kernel_dispatch): Adjust for ephemeral memories. (run_kernel): Pass thread-count to create_kernel_dispatch. (GOMP_OFFLOAD_init_device): Adjust for ephemeral memories. (GOMP_OFFLOAD_fini_device): Adjust for ephemeral memories. gcc/testsuite/ChangeLog: * gcc.c-torture/execute/pr47237.c: Xfail on amdgcn. * gcc.dg/builtin-apply3.c: Xfail for amdgcn. * gcc.dg/builtin-apply4.c: Xfail for amdgcn. * gcc.dg/torture/stackalign/builtin-apply-3.c: Xfail for amdgcn. * gcc.dg/torture/stackalign/builtin-apply-4.c: Xfail for amdgcn.
2023-02-02libgomp.texi (OpenMP TR11 impl. status): Fix 'strict' itemTobias Burnus1-2/+1
Fix the 'strict' modifier status: it is already listed (as 'Y') for OpenMP 5.1 for num_task and grainsize; only strict on num_threads is new with TR11. libgomp/ * libgomp.texi (OpenMP TR11): Fix item for 'strict' modifier.
2023-02-02Daily bump.GCC Administrator1-0/+12
2023-02-01Fortran: Extend align-clause checks of OpenMP's allocate directiveTobias Burnus2-2/+44
gcc/fortran/ChangeLog: * openmp.cc (resolve_omp_clauses): Check also for power of two. libgomp/ChangeLog: * testsuite/libgomp.fortran/allocate-3.f90: Fix ALIGN usage, remove unused -fdump-tree-original. * testsuite/libgomp.fortran/allocate-4.f90: New.
2023-02-01libgomp.texi: Reverse-offload updatesTobias Burnus1-7/+19
libgomp/ * libgomp.texi (5.0 Impl. Status): Update 'requires' and 'ancestor'. (GCN): Add item about 'omp requires'. (nvptx): Likewise; add item about reverse offload.
2023-01-28Daily bump.GCC Administrator1-0/+5
2023-01-27OpenMP/Fortran: Fix has_device_addr clause splitting [PR108558]Tobias Burnus1-0/+59
gcc/fortran/ChangeLog: PR fortran/108558 * trans-openmp.cc (gfc_split_omp_clauses): Handle has_device_addr. libgomp/ChangeLog: PR fortran/108558 * testsuite/libgomp.fortran/has_device_addr.f90: New test.
2023-01-24Daily bump.GCC Administrator1-0/+5
2023-01-23libgomp.texi: Impl. status - non-rect loop nest only partialTobias Burnus1-1/+1
libgomp/ * libgomp.texi (OpenMP 5.0): Set non-rectangular loop nest back to 'P' as Fortran support is incomplete.
2023-01-20Daily bump.GCC Administrator1-0/+5
2023-01-19openmp: Fix up OpenMP expansion of non-rectangular loops [PR108459]Jakub Jelinek1-0/+41
expand_omp_for_init_counts was using for the case where collapse(2) inner loop has init expression dependent on non-constant multiple of the outer iterator and the condition upper bound expression doesn't depend on the outer iterator fold_unary (NEGATE_EXPR, ...). This will just return NULL if it can't be folded, we need fold_build1 instead. 2023-01-19 Jakub Jelinek <jakub@redhat.com> PR middle-end/108459 * omp-expand.cc (expand_omp_for_init_counts): Use fold_build1 rather than fold_unary for NEGATE_EXPR. * testsuite/libgomp.c/pr108459.c: New test.
2023-01-18Daily bump.GCC Administrator1-0/+5
2023-01-17Regenerate Makefile.in files.Martin Liska2-2/+2
libbacktrace/ChangeLog: * Makefile.in: Regenerate. libgomp/ChangeLog: * Makefile.in: Regenerate. * configure: Regenerate. libphobos/ChangeLog: * Makefile.in: Regenerate. * libdruntime/Makefile.in: Regenerate. libstdc++-v3/ChangeLog: * src/libbacktrace/Makefile.in: Regenerate.
2023-01-16Update copyright years.Jakub Jelinek151-151/+151
2023-01-08Daily bump.GCC Administrator1-0/+6
2023-01-07Always define `WIN32_LEAN_AND_MEAN` before <windows.h>LIU Hao1-0/+1
Recently, mingw-w64 has got updated <msxml.h> from Wine which is included indirectly by <windows.h> if `WIN32_LEAN_AND_MEAN` is not defined. The `IXMLDOMDocument` class has a member function named `abort()`, which gets affected by our `abort()` macro in "system.h". `WIN32_LEAN_AND_MEAN` should, nevertheless, always be defined. This can exclude 'APIs such as Cryptography, DDE, RPC, Shell, and Windows Sockets' [1], and speed up compilation of these files a bit. [1] https://learn.microsoft.com/en-us/windows/win32/winprog/using-the-windows-headers gcc/ PR middle-end/108300 * config/xtensa/xtensa-dynconfig.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. * diagnostic-color.cc: Likewise. * plugin.cc: Likewise. * prefix.cc: Likewise. gcc/ada/ PR middle-end/108300 * adaint.c: Define `WIN32_LEAN_AND_MEAN` before `#include <windows.h>`. * cio.c: Likewise. * ctrl_c.c: Likewise. * expect.c: Likewise. * gsocket.h: Likewise. * mingw32.h: Likewise. * mkdir.c: Likewise. * rtfinal.c: Likewise. * rtinit.c: Likewise. * seh_init.c: Likewise. * sysdep.c: Likewise. * terminals.c: Likewise. * tracebak.c: Likewise. gcc/jit/ PR middle-end/108300 * jit-w32.h: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libatomic/ PR middle-end/108300 * config/mingw/lock.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libffi/ PR middle-end/108300 * src/aarch64/ffi.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libgcc/ PR middle-end/108300 * config/i386/enable-execute-stack-mingw32.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. * libgcc2.c: Likewise. * unwind-generic.h: Likewise. libgfortran/ PR middle-end/108300 * intrinsics/sleep.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libgomp/ PR middle-end/108300 * config/mingw32/proc.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libiberty/ PR middle-end/108300 * make-temp-file.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. * pex-win32.c: Likewise. libssp/ PR middle-end/108300 * ssp.c: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. libstdc++-v3/ PR middle-end/108300 * src/c++11/system_error.cc: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. * src/c++11/thread.cc: Likewise. * src/c++17/fs_ops.cc: Likewise. * src/filesystem/ops.cc: Likewise. libvtv/ PR middle-end/108300 * vtv_malloc.cc: Define `WIN32_LEAN_AND_MEAN` before <windows.h>. * vtv_rts.cc: Likewise. * vtv_utils.cc: Likewise.
2023-01-06Daily bump.GCC Administrator1-0/+5
2023-01-05openmp: Fix up finish_omp_target_clauses [PR108286]Jakub Jelinek1-0/+29
The comment in the loop says that we shouldn't add a map clause if such a clause exists already, but the loop was actually using OMP_CLAUSE_DECL on any clause. Target construct can have various clauses which don't have OMP_CLAUSE_DECL at all (e.g. nowait, device or if) or clause where it means something different (e.g. privatization clauses, allocate, depend). So, only check OMP_CLAUSE_DECL on OMP_CLAUSE_MAP clauses. 2023-01-05 Jakub Jelinek <jakub@redhat.com> PR c++/108286 * semantics.cc (finish_omp_target_clauses): Ignore clauses other than OMP_CLAUSE_MAP. * testsuite/libgomp.c++/pr108286.C: New test.
2023-01-03Daily bump.GCC Administrator1-0/+4
2023-01-02Update copyright dates.Jakub Jelinek1-1/+1
Manual part of copyright year updates. 2023-01-02 Jakub Jelinek <jakub@redhat.com> gcc/ * gcc.cc (process_command): Update copyright notice dates. * gcov-dump.cc (print_version): Ditto. * gcov.cc (print_version): Ditto. * gcov-tool.cc (print_version): Ditto. * gengtype.cc (create_file): Ditto. * doc/cpp.texi: Bump @copying's copyright year. * doc/cppinternals.texi: Ditto. * doc/gcc.texi: Ditto. * doc/gccint.texi: Ditto. * doc/gcov.texi: Ditto. * doc/install.texi: Ditto. * doc/invoke.texi: Ditto. gcc/ada/ * gnat_ugn.texi: Bump @copying's copyright year. * gnat_rm.texi: Likewise. gcc/d/ * gdc.texi: Bump @copyrights-d year. gcc/fortran/ * gfortranspec.cc (lang_specific_driver): Update copyright notice dates. * gfc-internals.texi: Bump @copying's copyright year. * gfortran.texi: Ditto. * intrinsic.texi: Ditto. * invoke.texi: Ditto. gcc/go/ * gccgo.texi: Bump @copyrights-go year. libgomp/ * libgomp.texi: Bump @copying's copyright year. libitm/ * libitm.texi: Bump @copying's copyright year. libquadmath/ * libquadmath.texi: Bump @copying's copyright year.
2023-01-02Update Copyright year in ChangeLog filesJakub Jelinek1-1/+1
2022 -> 2023
2022-12-22Daily bump.GCC Administrator1-0/+26
2022-12-21nvptx: reimplement libgomp barriers [PR99555]Chung-Lin Tang2-103/+124
Instead of trying to have the GPU do CPU-with-OS-like things, this new barriers implementation for NVPTX uses simplistic bar.* synchronization instructions. Tasks are processed after threads have joined, and only if team->task_count != 0 It is noted that: there might be a little bit of performance forfeited for cases where earlier arriving threads could've been used to process tasks ahead of other threads, but that has the requirement of implementing complex futex-wait/wake like behavior, which is what we're try to avoid with this patch. It is deemed that task processing is not what GPU target offloading is usually used for. Implementation highlight notes: 1. gomp_team_barrier_wake() is now an empty function (threads never "wake" in the usual manner) 2. gomp_team_barrier_cancel() now uses the "exit" PTX instruction. 3. gomp_barrier_wait_last() now is implemented using "bar.arrive" 4. gomp_team_barrier_wait_end()/gomp_team_barrier_wait_cancel_end(): The main synchronization is done using a 'bar.red' instruction. This reduces across all threads the condition (team->task_count != 0), to enable the task processing down below if any thread created a task. (this bar.red usage means that this patch is dependent on the prior NVPTX bar.red GCC patch) PR target/99555 libgomp/ChangeLog: * config/nvptx/bar.c (generation_to_barrier): Remove. (futex_wait,futex_wake,do_spin,do_wait): Remove. (GOMP_WAIT_H): Remove. (#include "../linux/bar.c"): Remove. (gomp_barrier_wait_end): New function. (gomp_barrier_wait): Likewise. (gomp_barrier_wait_last): Likewise. (gomp_team_barrier_wait_end): Likewise. (gomp_team_barrier_wait): Likewise. (gomp_team_barrier_wait_final): Likewise. (gomp_team_barrier_wait_cancel_end): Likewise. (gomp_team_barrier_wait_cancel): Likewise. (gomp_team_barrier_cancel): Likewise. * config/nvptx/bar.h (gomp_barrier_t): Remove waiters, lock fields. (gomp_barrier_init): Remove init of waiters, lock fields. (gomp_team_barrier_wake): Remove prototype, add new static inline function.
2022-12-21openmp: Don't try to destruct DECL_OMP_PRIVATIZED_MEMBER vars [PR108180]Jakub Jelinek1-0/+55
DECL_OMP_PRIVATIZED_MEMBER vars are artificial vars with DECL_VALUE_EXPR of this->field used just during gimplification and omp lowering/expansion to privatize individual fields in methods when needed. As the following testcase shows, when not in templates, they were handled right, but in templates we actually called cp_finish_decl on them and that can result in their destruction, which is obviously undesirable, we should only destruct the privatized copies of them created in omp lowering. Fixed thusly. 2022-12-21 Jakub Jelinek <jakub@redhat.com> PR c++/108180 * pt.cc (tsubst_expr): Don't call cp_finish_decl on DECL_OMP_PRIVATIZED_MEMBER vars. * testsuite/libgomp.c++/pr108180.C: New test.
2022-12-17Daily bump.GCC Administrator1-0/+6
2022-12-16Remove libgomp/testsuite/libgomp.fortran/allocate-4.f90 [PR108056]Tobias Burnus1-42/+0
Commit r13-4716-ge205ec03f0794aeac3e8a89e947c12624d5a274e accidentally included a testcase of another patch that is pending review: https://gcc.gnu.org/pipermail/gcc-patches/2022-December/608401.html libgomp/ PR libfortran/108056 * testsuite/libgomp.fortran/allocate-4.f90: Remove accidentally added file.
2022-12-16Daily bump.GCC Administrator1-0/+5
2022-12-15libgfortran's ISO_Fortran_binding.c: Use GCC11 version for backward-only ↵Tobias Burnus1-0/+42
code [PR108056] Since GCC 12, the conversion between the array descriptors formats - the internal (GFC) and the C binding one (CFI) - moved to the compiler itself such that the cfi_desc_to_gfc_desc/gfc_desc_to_cfi_desc functions are only used with older code (GCC 9 to 11). The newly added checks caused asserts as older code did not pass the proper values (e.g. real(4) as effective argument arrived as BT_ASSUME type as the effective type got lost inbetween). As proposed in the PR, revert to the GCC 11 version - known bugs is better than some fixes and new issues. Still, GCC 12 is much better in terms of TS29113 support and should really be used. This patch uses the current libgomp version of the GCC 11 branch, except it fixes the GFC version number (which is 0), uses calloc instead of malloc, and sets the lower bound to 1 instead of keeping it as is for CFI_attribute_other. libgfortran/ChangeLog: PR libfortran/108056 * runtime/ISO_Fortran_binding.c (cfi_desc_to_gfc_desc, gfc_desc_to_cfi_desc): Mostly revert to GCC 11 version for those backward-compatiblity-only functions.
2022-12-15Daily bump.GCC Administrator1-0/+5
2022-12-14OpenMP/Fortran: Combined directives with map/firstprivate of same symbolJulian Brown1-0/+41
This patch fixes a case where a combined directive (e.g. "!$omp target parallel ...") contains both a map and a firstprivate clause for the same variable. When the combined directive is split into two nested directives, the outer "target" gets the "map" clause, and the inner "parallel" gets the "firstprivate" clause, like so: !$omp target parallel map(x) firstprivate(x) --> !$omp target map(x) !$omp parallel firstprivate(x) ... When there is no map of the same variable, the firstprivate is distributed to both directives, e.g. for 'y' in: !$omp target parallel map(x) firstprivate(y) --> !$omp target map(x) firstprivate(y) !$omp parallel firstprivate(y) ... This is not a recent regression, but appear to fix a long-standing ICE. (The included testcase is based on one by Tobias.) 2022-12-06 Julian Brown <julian@codesourcery.com> gcc/fortran/ * trans-openmp.cc (gfc_add_firstprivate_if_unmapped): New function. (gfc_split_omp_clauses): Call above. libgomp/ * testsuite/libgomp.fortran/combined-directive-splitting-1.f90: New test.
2022-12-11Daily bump.GCC Administrator1-0/+37
2022-12-10libgomp: Handle OpenMP's reverse offloadsTobias Burnus11-39/+1139
This commit enabled reverse offload for nvptx such that gomp_target_rev actually gets called. And it fills the latter function to do all of the following: finding the host function to the device func ptr and copying the arguments to the host, processing the mapping/firstprivate, calling the host function, copying back the data and freeing as needed. The data handling is made easier by assuming that all host variables either existed before (and are in the mapping) or that those are devices variables not yet available on the host. Thus, the reverse mapping can do without refcounts etc. Note that the spec disallows inside a target region device-affecting constructs other than target plus ancestor device-modifier and it also limits the clauses permitted on this construct. For the function addresses, an additional splay tree is used; for the lookup of mapped variables, the existing splay-tree is used. Unfortunately, its data structure requires a full walk of the tree; Additionally, the just mapped variables are recorded in a separate data structure an extra lookup. While the lookup is slow, assuming that only few variables get mapped in each reverse offload construct and that reverse offload is the exception and not performance critical, this seems to be acceptable. libgomp/ChangeLog: * libgomp.h (struct target_mem_desc): Predeclare; move below after 'reverse_splay_tree_node' and add rev_array member. (struct reverse_splay_tree_key_s, reverse_splay_compare): New. (reverse_splay_tree_node, reverse_splay_tree, reverse_splay_tree_key): New typedef. (struct gomp_device_descr): Add mem_map_rev member. * oacc-host.c (host_dispatch): NULL init .mem_map_rev. * plugin/plugin-nvptx.c (GOMP_OFFLOAD_get_num_devices): Claim support for GOMP_REQUIRES_REVERSE_OFFLOAD. * splay-tree.h (splay_tree_callback_stop): New typedef; like splay_tree_callback but returning int not void. (splay_tree_foreach_lazy): Define; like splay_tree_foreach but taking splay_tree_callback_stop as argument. * splay-tree.c (splay_tree_foreach_internal_lazy, splay_tree_foreach_lazy): New; but early exit if callback returns nonzero. * target.c: Instatiate splay_tree_c with splay_tree_prefix 'reverse'. (gomp_map_lookup_rev): New. (gomp_load_image_to_device): Handle reverse-offload function lookup table. (gomp_unload_image_from_device): Free devicep->mem_map_rev. (struct gomp_splay_tree_rev_lookup_data, gomp_splay_tree_rev_lookup, gomp_map_rev_lookup, struct cpy_data, gomp_map_cdata_lookup_int, gomp_map_cdata_lookup): New auxiliary structs and functions for gomp_target_rev. (gomp_target_rev): Implement reverse offloading and its mapping. (gomp_target_init): Init current_device.mem_map_rev.root. * testsuite/libgomp.fortran/reverse-offload-2.f90: New test. * testsuite/libgomp.fortran/reverse-offload-3.f90: New test. * testsuite/libgomp.fortran/reverse-offload-4.f90: New test. * testsuite/libgomp.fortran/reverse-offload-5.f90: New test. * testsuite/libgomp.fortran/reverse-offload-5a.f90: New test without mapping of on-device allocated variables.
2022-12-10Daily bump.GCC Administrator1-0/+7
2022-12-09Fortran/OpenMP: align/allocator modifiers to the allocate clauseTobias Burnus3-2/+55
gcc/fortran/ChangeLog: * dump-parse-tree.cc (show_omp_namelist): Improve OMP_LIST_ALLOCATE output. * gfortran.h (struct gfc_omp_namelist): Add 'align' to 'u'. (gfc_free_omp_namelist): Add bool arg. * match.cc (gfc_free_omp_namelist): Likewise; free 'u.align'. * openmp.cc (gfc_free_omp_clauses, gfc_match_omp_clause_reduction, gfc_match_omp_flush): Update call. (gfc_match_omp_clauses): Match 'align/allocate modifers in 'allocate' clause. (resolve_omp_clauses): Resolve align. * st.cc (gfc_free_statement): Update call * trans-openmp.cc (gfc_trans_omp_clauses): Handle 'align'. libgomp/ChangeLog: * libgomp.texi (5.1 Impl. Status): Split allocate clause/directive item about 'align'; mark clause as 'Y' and directive as 'N'. * testsuite/libgomp.fortran/allocate-2.f90: New test. * testsuite/libgomp.fortran/allocate-3.f90: New test.
2022-12-07Daily bump.GCC Administrator1-0/+43
2022-12-06OpenMP: omp_get_max_teams, omp_set_num_teams, and ↵Marcel Vollweiler14-47/+1006
omp_{gs}et_teams_thread_limit on offload devices This patch adds support for omp_get_max_teams, omp_set_num_teams, and omp_{gs}et_teams_thread_limit on offload devices. That includes the usage of device-specific ICV values (specified as environment variables or changed on a device). In order to reuse device-specific ICV values, a copy back mechanism is implemented that copies ICV values back from device to the host. Additionally, a limitation of the number of teams on gcn offload devices is implemented. The number of teams is limited by twice the number of compute units (one team is executed on one compute unit). This avoids queueing unnessecary many teams and a corresponding allocation of large amounts of memory. Without that limitation the memory allocation for a large number of user-specified teams can result in an "memory access fault". A limitation of the number of teams is already also implemented for nvptx devices (see nvptx_adjust_launch_bounds in libgomp/plugin/plugin-nvptx.c). gcc/ChangeLog: * gimplify.cc (optimize_target_teams): Set initial num_teams_upper to "-2" instead of "1" for non-existing num_teams clause in order to disambiguate from the case of an existing num_teams clause with value 1. libgomp/ChangeLog: * config/gcn/icv-device.c (omp_get_teams_thread_limit): Added to allow processing of device-specific values. (omp_set_teams_thread_limit): Likewise. (ialias): Likewise. * config/nvptx/icv-device.c (omp_get_teams_thread_limit): Likewise. (omp_set_teams_thread_limit): Likewise. (ialias): Likewise. * icv-device.c (omp_get_teams_thread_limit): Likewise. (ialias): Likewise. (omp_set_teams_thread_limit): Likewise. * icv.c (omp_set_teams_thread_limit): Removed. (omp_get_teams_thread_limit): Likewise. (ialias): Likewise. * libgomp.texi: Updated documentation for nvptx and gcn corresponding to the limitation of the number of teams. * plugin/plugin-gcn.c (limit_teams): New helper function that limits the number of teams by twice the number of compute units. (parse_target_attributes): Limit the number of teams on gcn offload devices. * target.c (get_gomp_offload_icvs): Added teams_thread_limit_var handling. (gomp_load_image_to_device): Added a size check for the ICVs struct variable. (gomp_copy_back_icvs): New function that is used in GOMP_target_ext to copy back the ICV values from device to host. (GOMP_target_ext): Update the number of teams and threads in the kernel args also considering device-specific values. * testsuite/libgomp.c-c++-common/icv-4.c: Fixed an error in the reading of OMP_TEAMS_THREAD_LIMIT from the environment. * testsuite/libgomp.c-c++-common/icv-5.c: Extended. * testsuite/libgomp.c-c++-common/icv-6.c: Extended. * testsuite/libgomp.c-c++-common/icv-7.c: Extended. * testsuite/libgomp.c-c++-common/icv-9.c: New test. * testsuite/libgomp.fortran/icv-5.f90: New test. * testsuite/libgomp.fortran/icv-6.f90: New test. gcc/testsuite/ChangeLog: * c-c++-common/gomp/target-teams-1.c: Adapt expected values for num_teams from "1" to "-2" in cases without num_teams clause. * g++.dg/gomp/target-teams-1.C: Likewise. * gfortran.dg/gomp/defaultmap-4.f90: Likewise. * gfortran.dg/gomp/defaultmap-5.f90: Likewise. * gfortran.dg/gomp/defaultmap-6.f90: Likewise.
2022-12-06libgomp.texi: Fix a OpenMP 5.2 and a TR11 impl-status itemTobias Burnus1-1/+2
libgomp/ * libgomp.texi (OpenMP 5.2): Add missing 'the'. (TR11): Add missing '@tab N @tab'.
2022-12-01Daily bump.GCC Administrator1-0/+14
2022-11-30libgomp.texi: List GCN's 'gfx803' under OpenMP Context SelectorsTobias Burnus1-1/+2
libgomp/ChangeLog: * libgomp.texi (OpenMP Context Selectors): Add 'gfx803' to gcn's isa.
2022-11-30amdgcn: Support AMD-specific 'isa' traits in OpenMP context selectorsPaul-Antoine Arras7-0/+106
Add support for gfx803 as an alias for fiji. Add test cases for all supported 'isa' values. gcc/ChangeLog: * config/gcn/gcn.cc (gcn_omp_device_kind_arch_isa): Add gfx803. * config/gcn/t-omp-device: Add gfx803. libgomp/ChangeLog: * testsuite/libgomp.c/declare-variant-4-fiji.c: New test. * testsuite/libgomp.c/declare-variant-4-gfx803.c: New test. * testsuite/libgomp.c/declare-variant-4-gfx900.c: New test. * testsuite/libgomp.c/declare-variant-4-gfx906.c: New test. * testsuite/libgomp.c/declare-variant-4-gfx908.c: New test. * testsuite/libgomp.c/declare-variant-4-gfx90a.c: New test. * testsuite/libgomp.c/declare-variant-4.h: New header file.
2022-11-29Daily bump.GCC Administrator1-0/+4
2022-11-28OpenMP/Fortran: Permit end-clause on directiveTobias Burnus1-1/+1
gcc/fortran/ChangeLog: * openmp.cc (OMP_DO_CLAUSES, OMP_SCOPE_CLAUSES, OMP_SECTIONS_CLAUSES): Add 'nowait'. (OMP_SINGLE_CLAUSES): Add 'nowait' and 'copyprivate'. (gfc_match_omp_distribute_parallel_do, gfc_match_omp_distribute_parallel_do_simd, gfc_match_omp_parallel_do, gfc_match_omp_parallel_do_simd, gfc_match_omp_parallel_sections, gfc_match_omp_teams_distribute_parallel_do, gfc_match_omp_teams_distribute_parallel_do_simd): Disallow 'nowait'. (gfc_match_omp_workshare): Match 'nowait' clause. (gfc_match_omp_end_single): Use clause matcher for 'nowait'. (resolve_omp_clauses): Reject 'nowait' + 'copyprivate'. * parse.cc (decode_omp_directive): Break too long line. (parse_omp_do, parse_omp_structured_block): Diagnose duplicated 'nowait' clause. libgomp/ChangeLog: * libgomp.texi (OpenMP 5.2): Mark end-directive as Y. gcc/testsuite/ChangeLog: * gfortran.dg/gomp/copyprivate-1.f90: New test. * gfortran.dg/gomp/copyprivate-2.f90: New test. * gfortran.dg/gomp/nowait-2.f90: Move dg-error tests ... * gfortran.dg/gomp/nowait-4.f90: ... to this new file. * gfortran.dg/gomp/nowait-5.f90: New test. * gfortran.dg/gomp/nowait-6.f90: New test. * gfortran.dg/gomp/nowait-7.f90: New test. * gfortran.dg/gomp/nowait-8.f90: New test.
2022-11-26Daily bump.GCC Administrator1-0/+19
2022-11-25OpenMP: Generate SIMD clones for functions with "declare target"Sandra Loosemore4-0/+123
This patch causes the IPA simdclone pass to generate clones for functions with the "omp declare target" attribute as if they had "omp declare simd", provided the function appears to be suitable for SIMD execution. The filter is conservative, rejecting functions that write memory or that call other functions not known to be safe. A new option -fopenmp-target-simd-clone is added to control this transformation; it's enabled for offload processing at -O2 and higher. gcc/ChangeLog: * common.opt (fopenmp-target-simd-clone): New option. (target_simd_clone_device): New enum to go with it. * doc/invoke.texi (-fopenmp-target-simd-clone): Document. * flag-types.h (enum omp_target_simd_clone_device_kind): New. * omp-simd-clone.cc (auto_simd_fail): New function. (auto_simd_check_stmt): New function. (plausible_type_for_simd_clone): New function. (ok_for_auto_simd_clone): New function. (simd_clone_create): Add force_local argument, make the symbol have internal linkage if it is true. (expand_simd_clones): Also check for cloneable functions with "omp declare target". Pass explicit_p argument to simd_clone.compute_vecsize_and_simdlen target hook. * opts.cc (default_options_table): Add -fopenmp-target-simd-clone. * target.def (TARGET_SIMD_CLONE_COMPUTE_VECSIZE_AND_SIMDLEN): Add bool explicit_p argument. * doc/tm.texi: Regenerated. * config/aarch64/aarch64.cc (aarch64_simd_clone_compute_vecsize_and_simdlen): Update. * config/gcn/gcn.cc (gcn_simd_clone_compute_vecsize_and_simdlen): Update. * config/i386/i386.cc (ix86_simd_clone_compute_vecsize_and_simdlen): Update. gcc/testsuite/ChangeLog: * g++.dg/gomp/target-simd-clone-1.C: New. * g++.dg/gomp/target-simd-clone-2.C: New. * gcc.dg/gomp/target-simd-clone-1.c: New. * gcc.dg/gomp/target-simd-clone-2.c: New. * gcc.dg/gomp/target-simd-clone-3.c: New. * gcc.dg/gomp/target-simd-clone-4.c: New. * gcc.dg/gomp/target-simd-clone-5.c: New. * gcc.dg/gomp/target-simd-clone-6.c: New. * gcc.dg/gomp/target-simd-clone-7.c: New. * gcc.dg/gomp/target-simd-clone-8.c: New. * lib/scanoffloadipa.exp: New. libgomp/ChangeLog: * testsuite/lib/libgomp.exp: Load scanoffloadipa.exp library. * testsuite/libgomp.c/target-simd-clone-1.c: New. * testsuite/libgomp.c/target-simd-clone-2.c: New. * testsuite/libgomp.c/target-simd-clone-3.c: New.
2022-11-25libgomp: Add no-target-region rev offload test + fix plugin-nvptxTobias Burnus2-12/+73
OpenMP permits that a 'target device(ancestor:1)' is called without being enclosed in a target region - using the current device (i.e. the host) in that case. This commit adds a testcase for this. In case of nvptx, the missing on-device 'GOMP_target_ext' call causes that it and also the associated on-device GOMP_REV_OFFLOAD_VAR variable are not linked in from nvptx's libgomp.a. Thus, handle the failing cuModuleGetGlobal gracefully by disabling reverse offload and assuming that the failure is fine. libgomp/ChangeLog: * plugin/plugin-nvptx.c (GOMP_OFFLOAD_load_image): Use unsigned int for 'i' to match 'fn_entries'; regard absent GOMP_REV_OFFLOAD_VAR as valid and the code having no reverse-offload code. * testsuite/libgomp.c-c++-common/reverse-offload-2.c: New test.
generated by cgit v1.1 at 2024-07-17 22:14:38 +0000