Commit Graph

594 Commits

Author SHA1 Message Date
Jonas Hahnfeld 5872f1e97f [test] Fix uninitialized memory in omp_taskloop_grainsize.c
result was never initialized to zero which sometimes failed the test.

llvm-svn: 314513
2017-09-29 13:53:03 +00:00
Jonathan Peyton bd3a7633f1 Remove unnecessary semicolons
Removes semicolons after if {} blocks, function definitions, etc.
I was able to apply the large OMPT patch cleanly on top of this one
with no conflicts.

llvm-svn: 314340
2017-09-27 20:36:27 +00:00
Jonathan Peyton 8f3d7448b9 Allow printing of KMP_TOPOLOGY_METHOD when KMP_SETTINGS=true
llvm-svn: 314243
2017-09-26 20:33:53 +00:00
Jonathan Peyton 6de85b1565 Remove unused t_single_lock
Add padding inside team structure to keep same structure size.

llvm-svn: 314242
2017-09-26 20:12:16 +00:00
Jonathan Peyton 52527cd2c1 Read blocktime value set by kmp_set_blocktime() before reading from KMP_BLOCKTIME
Patch by Hansang Bae

Differential Revision: https://reviews.llvm.org/D37403

llvm-svn: 312539
2017-09-05 15:45:48 +00:00
Jonathan Peyton 6a393f75f4 Minor code cleanup of Klocwork issues
Minor code cleanup of Klocwork issues. Fatal messages are given no return
attribute. Define and use KMP_NORETURN to work for multiple C++ versions.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D37275

llvm-svn: 312538
2017-09-05 15:43:58 +00:00
Jonathan Peyton 0447708f8d Use va_copy instead of __va_copy to fix building libomp against musl libc
Fixes https://bugs.llvm.org/show_bug.cgi?id=34040

Patch by Peter Levine

Differential Revision: https://reviews.llvm.org/D36343

llvm-svn: 311269
2017-08-19 23:53:36 +00:00
Jonathan Peyton d4daf4540a Remove BUILD_TV
Cleanup code to remove BUILD_TV and unused code bracketed by it.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D36011

llvm-svn: 311114
2017-08-17 19:09:28 +00:00
Sergey Dmitriev b305d26b57 [OpenMP] libomptarget: move debugging dumps under control of env var LIBOMPTARGET_DEBUG
Disable default debugging dumps for libomptarget and plugins and move dumps
under control of environment variable LIBOMPTARGET_DEBUG=<integer>. Dumps
are enabled when LIBOMPTARGET_DEBUG is set to a positive integer value.

Debugging dumps are available only in debug build; release build does not
support it.

Differential Revision: https://reviews.llvm.org/D33227

llvm-svn: 310841
2017-08-14 15:09:59 +00:00
Paul Osmialowski a016279422 OMP_PROC_BIND: better spread
This change improves the way threads are spread across cores
when OMP_PROC_BIND=spread is set and no unusual affinity masks are in use.

Differential Revision: https://reviews.llvm.org/D36510

llvm-svn: 310670
2017-08-10 23:04:11 +00:00
Jonathan Peyton 038855ade8 Exclude version symbols for static libomp
We use symbol versioning for GNU-compatibility but libgomp has versioned symbols
only in the shared library but not in the static.
Moreover, version symbols in the static library can cause an error at link time.

Patch by Olga Malysheva

Differential Revision: https://reviews.llvm.org/D36225

llvm-svn: 309877
2017-08-02 20:10:00 +00:00
Jonathan Peyton 1b536724d9 Move lock acquire/release functions in task deque cleanup code
The original locations can be reached without initializing the lock variable
(td_deque_lock), so it is potentially unsafe.  It is guaranteed that the lock
is initialized if the deque (td_deque) is not NULL, and lock functions can be
safely called.

Patch by Hansang Bae

Differential Revision: https://reviews.llvm.org/D36017

llvm-svn: 309875
2017-08-02 20:06:32 +00:00
Jonathan Peyton 4f90c82aec Add new envirable KMP_TEAMS_THREAD_LIMIT
This change adds a new environment variable, KMP_TEAMS_THREAD_LIMIT, which is
used to set a new global variable, __kmp_teams_max_nth, which is checked when
determining the size and quantity of teams that will be created in the teams
construct. Specifically, it is a limit on the total number of threads in a given
teams construct. It differentiates the limits for the teams construct from the
limits for regular parallel regions (KMP_DEVICE_THREAD_LIMIT/__kmp_max_nth and
OMP_THREAD_LIMIT/__kmp_cg_max_nth). When each individual team is formed, it is
still subject to those limits. After the clauses to the teams construct are
parsed and calculated, we check to make sure we are within this limit, and if
not, reduce num_threads per team and/or number of teams, accordingly. The
default value is set to the number of available processors on the system.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D36009

llvm-svn: 309874
2017-08-02 20:04:45 +00:00
Jonathan Peyton 644f4e3d11 Fix comments and build messages concerning TSX
llvm-svn: 309418
2017-07-28 19:05:17 +00:00
Jonathan Peyton f439246328 Fix implementation of OMP_THREAD_LIMIT
This change fixes the implementation of OMP_THREAD_LIMIT. The implementation of
this previously was not restricted to a contention group (but it should be,
according to the spec), and this is fixed here. A field is added to root thread
to store a counter of the threads in the contention group. An extra check is
added when reserving threads for a parallel region that checks this variable and
compares to threadlimit-var, which is implemented as a new global variable,
kmp_cg_max_nth. Associated settings changes were also made, and clean up of
comments that referred to OMP_THREAD_LIMIT, but should refer to the new
KMP_DEVICE_THREAD_LIMIT (added in an earlier patch).

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D35912

llvm-svn: 309319
2017-07-27 20:58:41 +00:00
Jonathan Peyton 09244f39dd Introduce KMP_DEVICE_THREAD_LIMIT
This change drops in KMP_DEVICE_THREAD_LIMIT to replace KMP_MAX_THREADS. It's
possible there will eventually be a OMP_DEVICE_THREAD_LIMIT, and we need
something to distinguish from OMP_THREAD_LIMIT, which is currently implemented
incorrectly (the fix for that will be added soon in a separate patch).
KMP_ALL_THREADS is deprecated here, but we can keep the "all" option on
KMP_DEVICE_THREAD_LIMIT to support that functionality. KMP_DEVICE_THREAD_LIMIT
now has priority over its deprecated rival KMP_ALL_THREADS. I also cleaned up
some comments that incorrectly referred to non-existent kmp_max_threads variable
instead of kmp_max_nth.

I've left the name of where this setting eventually ends up as
__kmp_max_nth, for now.

This change does not change much in the way of functionality. It does NOT change
OMP_THREAD_LIMIT. It's just cleaning up and setting up for that.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D35860

llvm-svn: 309168
2017-07-26 20:07:58 +00:00
Jonas Hahnfeld 203c730719 [CMake] Disable building libomptarget and add CMake switch
Introduce OPENMP_ENABLE_LIBOMPTARGET which defaults to OFF at the moment.

libomptarget is not yet ready for prime time:
 - Offloading to NVIDIA GPUs is not completed yet (compiler, device RTL)
 - The generic ELF plugin for offloading to the host (meant for testing)
   uses a single instance of the OpenMP runtime (libomp). That is why
   omp_is_initial_device() returns 1 which makes the tests fail.
Because of these reasons, we want to disable building (and testing!)
for release 5.0.

See https://bugs.llvm.org/show_bug.cgi?id=33859

Differential Revision: https://reviews.llvm.org/D35719

llvm-svn: 309115
2017-07-26 13:55:00 +00:00
Jonathan Peyton d74d890247 Cleanup: __kmp_env_* variables
Removed unused __kmp_env_* variables. Also clangified other people's code.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D35808

llvm-svn: 309000
2017-07-25 18:20:16 +00:00
NAKAMURA Takumi 0c7d6ef459 Whitespace.
llvm-svn: 308693
2017-07-20 23:12:39 +00:00
Andrey Churbanov c7476ed0be OpenMP RTL cleanup: two PAUSEs per spin loop iteration replaced with single one
Differential Revision: https://reviews.llvm.org/D35490

llvm-svn: 308423
2017-07-19 09:26:13 +00:00
Dimitry Andric 0c7238b21c For KMP_PAGE_SIZE, use getpagesize() on Unix, GetSystemInfo() on Windows
Summary:
The kmp_os.h header is defining the `PAGE_SIZE` macro unconditionally,
even while it is only used directly after its definition, for the
Windows implementation of the `KMP_GET_PAGE_SIZE()` macro.

On at least FreeBSD, but likely all other BSDs too, this macro conflicts
with the one defined in system headers, so remove it, since nothing else
uses it.  Make all Unixes use `getpagesize()` instead, and use
`GetSystemInfo()` for the Windows case.

Reviewers: jlpeyton, jcownie, emaste, AndreyChurbanov

Reviewed By: AndreyChurbanov

Subscribers: AndreyChurbanov, hfinkel, zturner

Differential Revision: https://reviews.llvm.org/D35072

llvm-svn: 308355
2017-07-18 20:31:19 +00:00
Jonathan Peyton 1c50ee64a2 Fix failing taskloop tests by omitting gcc
We do not have GOMP interface support for taskloop yet.

llvm-svn: 308351
2017-07-18 20:16:25 +00:00
Jonathan Peyton 93e17cfe6c Add recursive task scheduling strategy to taskloop implementation
Summary:
Taskloop implementation is extended by using recursive task scheduling.
Envirable KMP_TASKLOOP_MIN_TASKS added as a manual threshold for the user
to switch from recursive to linear tasks scheduling.

Details:
* The calculations for the loop parameters are moved from __kmp_taskloop_linear
  upper level
* Initial calculation is done in the __kmpc_taskloop, further range splitting
  is done in the __kmp_taskloop_recur.
* Added threshold to switch from recursive to linear tasks scheduling;
* One half of split range is scheduled as an internal task which just moves
  sub-range parameters to the stealing thread that continues recursive
  scheduling (if number of tasks still enough), the other half is processed
  recursively;
* Internal task duplication routine fixed to assign parent task, that was not
  needed when all tasks were scheduled by same thread, but is needed now.

Patch by Andrey Churbanov

Differential Revision: https://reviews.llvm.org/D35273

llvm-svn: 308338
2017-07-18 18:50:13 +00:00
Andrey Churbanov 71483f2dda Fix sporadic segfaults in tasking tests.
Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D35535

llvm-svn: 308298
2017-07-18 11:56:16 +00:00
Andrey Churbanov ddc38722a4 OpenMP RTL cleanup: nullify pointer after memory freeing
Differential Revision: https://reviews.llvm.org/D35497

llvm-svn: 308274
2017-07-18 08:30:03 +00:00
Jonathan Peyton f6f2c6e47f Removed "duplicates" from verbose affinity output
The internal details of this setting are not meant to be user visible and only create confusion.

Differential Revision: https://reviews.llvm.org/D35269

llvm-svn: 308189
2017-07-17 17:06:43 +00:00
Andrey Churbanov 5ba90c7979 OpenMP RTL cleanup: eliminated warnings with -Wcast-qual, patch 2.
Changes are: got all atomics to accept volatile pointers that allowed
to simplify many type conversions. Windows specific code fixed correspondingly.

Differential Revision: https://reviews.llvm.org/D35417

llvm-svn: 308164
2017-07-17 09:03:14 +00:00
Jonas Hahnfeld 266ddafc68 [GOMP] Fix (un)tied tasks with the GCC
The first bit is actually the "untied" flag. That is why the condition was
wrong and has to be inverted to set the flag correctly.

Found and initial patch by Simon Convent!

llvm-svn: 307899
2017-07-13 10:38:11 +00:00
Dimitry Andric b9fb12291a Rename z_Linux_asm.s to z_Linux_asm.S
Summary:
On Unix, a .S file is normally an assembly source which must be
preprocessed with a C preprocessor, while a .s file is "plain" assembly.
The former is handled by the compiler driver (cc), the latter is
directly passed to the assembler binary (as).

Because z_Linux_asm.s is supposed to be preprocessed, rename it to .S,
so it can be automatically picked up correctly by build systems.

Reviewers: AndreyChurbanov, emaste, jlpeyton

Reviewed By: AndreyChurbanov

Subscribers: mgorny, openmp-commits

Differential Revision: https://reviews.llvm.org/D35171

llvm-svn: 307680
2017-07-11 18:04:56 +00:00
Dimitry Andric 79bf29ccb7 Add a .arcconfig file for openmp.
llvm-svn: 307474
2017-07-08 16:09:47 +00:00
Ed Maste 414544c9aa remove deprecated register storage class specifier
While importing libomp into the FreeBSD base system we encountered
Clang warnings that "'register' storage class specifier is deprecated
and incompatible with C++1z [-Wdeprecated-register]".

Differential Revision:	https://reviews.llvm.org/D35124

llvm-svn: 307441
2017-07-07 21:06:05 +00:00
Ed Maste 78b0f075f7 remove duplicate symbol version script entries
GNU ld ignores duplicates, but lld produces a warning.

Differential Revision:	https://reviews.llvm.org/D35121

llvm-svn: 307399
2017-07-07 13:45:41 +00:00
Jonathan Peyton d0494046c7 Fix wrong website in messages
Address user message bug where the messages were sending users to Intel's
website instead of the LLVM OpenMP runtime websites.

Bugzilla: https://bugs.llvm.org/show_bug.cgi?id=32892

Differential Revision: https://reviews.llvm.org/D35018

llvm-svn: 307206
2017-07-05 22:01:05 +00:00
Andrey Churbanov c47afcd9bb OpenMP RTL cleanup: eliminated warnings with -Wcast-qual.
Changes are: replaced C-style casts with cons_cast and reinterpret_cast;
type of several counters changed to signed; type of parameters of 32-bit and
64-bit AND and OR intrinsics changes to unsigned; changed files formatted
using clang-format version 3.8.1.

Differential Revision: https://reviews.llvm.org/D34759

llvm-svn: 307020
2017-07-03 11:24:08 +00:00
Hal Finkel 2bc3449d22 Make test/parallel/omp_nested.c not use so many threads
I've found it very difficult to get test/parallel/omp_nested.c to pass
consistently across my build environments. The problem is that it creates N^2
threads (it is testing nested parallel regions), and that often exceeds the
thread limits on systems with many cores. We do raise the process limits in
lit, and that often helps, but if running lit with a smaller number of threads
or on a system where we're otherwise resource constrained, this particular test
tends to fail (because the runtime cannot create a sufficient number of
threads).

This seems to work: if the maximum number of threads is more than some small
number, then cap the number of threads used for the parallel region. The choice
of 4 here is somewhat arbitrary.

Differential Revision: https://reviews.llvm.org/D32033

llvm-svn: 306357
2017-06-27 03:04:25 +00:00
Dimitry Andric 695c69316b Only use libdl when it is available
Summary: On BSDs, there is no `libdl.so`, and functions like `dlopen`
are implemented in the main C library instead.  Use the `CMAKE_DL_LIBS`
variable instead of hardcoding a dependency on the `dl` library.

Reviewers: grokos, joerg, emaste

Reviewed By: emaste

Subscribers: jlpeyton, mgorny, openmp-commits

Differential Revision: https://reviews.llvm.org/D34632

llvm-svn: 306319
2017-06-26 19:16:49 +00:00
Jonathan Peyton 072ccb7239 Set affinity to none/false in child processes
Reset affinity to none (false for proc-bind-var) so that threads in the child
processes are not bound tightly, unless the user explicitly sets this in
KMP_AFFINITY/OMP_PROC_BIND, in child processes. This can improve
performance for scripting languages which fork for parallelism like Python's
multiprocessing module.

Differential Revision: https://reviews.llvm.org/D34154

llvm-svn: 305513
2017-06-15 21:51:07 +00:00
Jonathan Peyton 492e0a33cb Replace platform macro with KMP_MIC_SUPPORTED
Differential Revision: https://reviews.llvm.org/D34119

llvm-svn: 305307
2017-06-13 17:17:26 +00:00
Jonathan Peyton d330e630db Reset initial affinity in children processes
If OpenMP is initialized before fork()-ing occurs and affinity is set to
something like compact, then the master thread will be pinned to a single HW
thread/core after initialization. If the master (or any other thread) then
forks N processes, all N processes will then be pinned to that same single HW
thread/core. To reset the affinity for the new child process, the atfork
handler for the child process can call kmp_set_thread_affinity_mask_initial()
to reset its affinity to the initial affinity of the application before it
re-initializes libomp. The parent process will not be affected and still
keeps its affinity setting.

Differential Revision: https://reviews.llvm.org/D34118

llvm-svn: 305306
2017-06-13 17:16:12 +00:00
Samuel Antao 8933ffbb12 [OpenMP] Prevent unused-variable warning in libomptarget when compiling in Release mode.
llvm-svn: 305090
2017-06-09 16:46:07 +00:00
Jonathan Peyton ccfed2edb6 Fix static initializers for locks.
Fix static initializers to use the proper unlocked value for the poll
field of the tas and futex locks.

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D33794

llvm-svn: 304828
2017-06-06 20:24:41 +00:00
Andrey Churbanov d454c73cc3 OpenMP 4.5: implemented support of schedule(simd:guided) and
schedule(simd:runtime) - library part. Compiler generation should use newly
introduced scheduling kinds kmp_sch_guided_simd = 46, kmp_sch_runtime_simd = 47,
as parameters to __kmpc_dispatch_init_* entries.

Differential Revision: https://reviews.llvm.org/D31602

llvm-svn: 304724
2017-06-05 17:17:33 +00:00
George Rokos 0e86bfb5bb [OpenMP] libomptarget: eliminate compiler warnings at build
Thanks to Sergey Dmitriev for submitting the patch.

Differential Revision: https://reviews.llvm.org/D33851

llvm-svn: 304601
2017-06-02 22:41:35 +00:00
Andrey Churbanov b3b10c2fa5 Re-enable assertion after the problem that caused it to be hit had been fixed
Differential Revision: https://reviews.llvm.org/D31421

llvm-svn: 304443
2017-06-01 18:10:45 +00:00
Jonathan Peyton 642688b632 Fix minor formatting issues
Some code was restructured to move it under KMP_DEBUG.  The rest is
formatting changes to fix some things broken by clang-format

Patch by Terry Wilmarth

Differential Revision: https://reviews.llvm.org/D33744

llvm-svn: 304438
2017-06-01 16:46:36 +00:00
Jonathan Peyton e3e2aaf68d Fix for KMP_AFFINITY=disabled and KMP_TOPOLOGY_METHOD=hwloc
With these settings, the create_hwloc_map() method was being called causing an
assert(). After some consideration, it was determined that disabling affinity
explicitly should just disable hwloc as well. i.e., KMP_AFFINITY overrides
KMP_TOPOLOGY_METHOD. This lets the user know that the Hwloc mechanism is being
ignored when KMP_AFFINITY=disabled.

Differential Revision: https://reviews.llvm.org/D33208

llvm-svn: 304344
2017-05-31 20:35:22 +00:00
Jonathan Peyton 9f5df8b02e Address default pinning OpenMP process with multiple processor groups
This change checks if the initial affinity mask is equal to exactly one
Windows processor group's affinity mask. If it is, then the code does not
respect the initial affinity mask and uses the entire machine instead.
The reasoning behind this is that, by default, Windows assigns exactly one
processor group as the initial affinity mask even when there are multiple
Windows processor groups available. User's typically want to use the whole
machine, so we ignore this special case and use the entire machine.

If the initial affinity mask is a proper subset of one group, or spans multiple
groups, then the initial affinity mask is respected since we can assume that the
operating system did not assign this initial affinity mask. This change only
affects machines with multiple processor groups

Differential Revision: https://reviews.llvm.org/D33210

llvm-svn: 304343
2017-05-31 20:33:56 +00:00
Jonathan Peyton 586849918b Fix for KMP_AFFINITY=respect with multiple processor groups
An assert() was being tripped when KMP_AFFINITY=respect + Multiple Processor
Groups. Let __kmp_affinity_create_proc_group_map() function be able to create
address2os object which contains a single group by deleting restriction that
process affinity mask must span multiple groups.

llvm-svn: 303101
2017-05-15 19:05:59 +00:00
Jonathan Peyton 6da813336c Remove some outdated comments
llvm-svn: 303086
2017-05-15 17:39:16 +00:00
Jonathan Peyton 9e704efaa6 Add the .clang-format file which the formatting was based on
llvm-svn: 303079
2017-05-15 16:39:42 +00:00