Hahnfeld (Jonas Hahnfeld)
User

Projects

User does not belong to any projects.

User Details

User Since
Apr 2 2015, 4:52 AM (189 w, 6 d)
RWTH Aachen University

Recent Activity

Oct 14 2018

Hahnfeld added a comment to D53249: Force Hexagon to use default (hexagon-link) linker.

As said on D53250 I think this is the right way to fix these tests: We already do the same for -stdlib=platform and -rtlib=platform.

Oct 14 2018, 3:32 AM
Hahnfeld added a comment to D53250: [ToolChain] Use default linker if the toolchain uses a custom one.

If I read that patch correctly, this will render -fuse-ld with non-absolute paths useless if a toolchain has DefaultLinker != "ld". I don't think that's what we want to do if the user explicitly sets a different linker.

Oct 14 2018, 3:31 AM

Oct 11 2018

Hahnfeld removed a reviewer for D53141: [OpenMP][libomptarget] Add runtime function for pushing coalesced global records: Hahnfeld.

I guess this will break the case when the DataSize passed to __kmpc_data_sharing_push_stack() needs additional alignment: With this change it is handled in data_sharing_push_stack_common() but __kmpc_data_sharing_push_stack() will determine PushSize without the adjustment and do the final pointer arithmetic.

Why DataSize might require an additional alignment? The DataSize must be already aligned.

Oct 11 2018, 12:02 PM
Hahnfeld added a comment to D53141: [OpenMP][libomptarget] Add runtime function for pushing coalesced global records.

I guess this will break the case when the DataSize passed to __kmpc_data_sharing_push_stack() needs additional alignment: With this change it is handled in data_sharing_push_stack_common() but __kmpc_data_sharing_push_stack() will determine PushSize without the adjustment and do the final pointer arithmetic.

Oct 11 2018, 8:47 AM

Oct 9 2018

Hahnfeld added a comment to D52700: [libomptarget-nvptx] Remove dead functions.

Ping

Oct 9 2018, 12:24 PM

Oct 5 2018

Hahnfeld accepted D52938: [CUDA] Use all 64 bits of GUID in __nv_module_id.

Out of interest: Is this fixing a particular issue?

Oct 5 2018, 11:07 AM

Oct 2 2018

Hahnfeld added a comment to D52733: [OpenMP][NVPTX] Avoid data sharing if in parallel region.

It might lead to increased register pressure, isn't it? Currently, I'm trying to emit the code that can be optimized out and, thus, may decrease the register pressure. That's why I tried to reduce the number of the runtime checks.

You are right, it's increasing register usage but I think it shouldn't: The generated code is always checking __kmpc_is_spmd_exec_mode first. So if LLVM would be able to optimize this out in SPMD mode, __kmpc_parallel_level should never be called.

I guess this doesn't work because it's illegal to hoist the load of execution_param across a barrier?

Even if we will be able to reduce register usage for SPMD, it still going to be high for non-SPMD constructs. The optimizer is not able to understand that it is in parallel region or not during the compilation phase

Oct 2 2018, 12:55 PM
Hahnfeld added a comment to D52733: [OpenMP][NVPTX] Avoid data sharing if in parallel region.

It might lead to increased register pressure, isn't it? Currently, I'm trying to emit the code that can be optimized out and, thus, may decrease the register pressure. That's why I tried to reduce the number of the runtime checks.

Oct 2 2018, 12:43 PM
Hahnfeld committed rL343618: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI..
[OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI.
Oct 2 2018, 12:15 PM
Hahnfeld committed rL343617: [OpenMP] Simplify code for reductions on distribute directives, NFC..
[OpenMP] Simplify code for reductions on distribute directives, NFC.
Oct 2 2018, 12:14 PM
Hahnfeld committed rC343618: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI..
[OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI.
Oct 2 2018, 12:14 PM
Hahnfeld closed D52732: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI..
Oct 2 2018, 12:14 PM
Hahnfeld committed rC343617: [OpenMP] Simplify code for reductions on distribute directives, NFC..
[OpenMP] Simplify code for reductions on distribute directives, NFC.
Oct 2 2018, 12:14 PM
Hahnfeld closed D52731: [OpenMP] Simplify code for reductions on distribute directives, NFC..
Oct 2 2018, 12:14 PM

Oct 1 2018

Hahnfeld added a dependent revision for D52732: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI.: D52733: [OpenMP][NVPTX] Avoid data sharing if in parallel region.
Oct 1 2018, 10:10 AM
Hahnfeld added a dependency for D52733: [OpenMP][NVPTX] Avoid data sharing if in parallel region: D52732: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI..
Oct 1 2018, 10:10 AM
Hahnfeld created D52733: [OpenMP][NVPTX] Avoid data sharing if in parallel region.
Oct 1 2018, 10:10 AM
Hahnfeld created D52732: [OpenMP][NVPTX] Simplify codegen for orphaned parallel, NFCI..
Oct 1 2018, 10:09 AM
Hahnfeld created D52731: [OpenMP] Simplify code for reductions on distribute directives, NFC..
Oct 1 2018, 10:09 AM
Hahnfeld added a comment to D52701: [libomptarget-nvptx] Enable asserts in bclib.

Looks good. I have no idea why this option was left there...

Oct 1 2018, 7:20 AM
Hahnfeld committed rL343477: [libomptarget-nvptx] Enable asserts in bclib.
[libomptarget-nvptx] Enable asserts in bclib
Oct 1 2018, 7:20 AM
Hahnfeld committed rOMP343477: [libomptarget-nvptx] Enable asserts in bclib.
[libomptarget-nvptx] Enable asserts in bclib
Oct 1 2018, 7:20 AM
Hahnfeld closed D52701: [libomptarget-nvptx] Enable asserts in bclib.
Oct 1 2018, 7:20 AM
Hahnfeld committed rL343476: [libomptarget-nvptx] reduction: Determine if runtime uninitialized.
[libomptarget-nvptx] reduction: Determine if runtime uninitialized
Oct 1 2018, 7:16 AM
Hahnfeld committed rOMP343476: [libomptarget-nvptx] reduction: Determine if runtime uninitialized.
[libomptarget-nvptx] reduction: Determine if runtime uninitialized
Oct 1 2018, 7:16 AM
Hahnfeld closed D52725: [libomptarget-nvptx] reduction: Determine if runtime uninitialized.
Oct 1 2018, 7:16 AM
Hahnfeld created D52725: [libomptarget-nvptx] reduction: Determine if runtime uninitialized.
Oct 1 2018, 7:11 AM

Sep 30 2018

Hahnfeld created D52701: [libomptarget-nvptx] Enable asserts in bclib.
Sep 30 2018, 2:49 AM
Hahnfeld created D52700: [libomptarget-nvptx] Remove dead functions.
Sep 30 2018, 2:48 AM
Hahnfeld added a comment to D29660: [OpenMP] Add flag for overwriting default PTX version for OpenMP targets.

Going through my list of reviews, this patch was reverted because of memory leaks in other changes. However, I don't think we need this anymore because Clang is raising the PTX level as needed for that CUDA version. Can we abandon this flag?

Sep 30 2018, 2:46 AM
Hahnfeld added a comment to D52438: [CUDA] Add basic support for CUDA-10.0.

I think this revision can be closed after rC342924?

Sep 30 2018, 2:38 AM
Hahnfeld removed a reviewer for D52437: [CUDA] Add preliminary support for CUDA 10.0: Hahnfeld.

I think this revision can be closed after rC342924?

Sep 30 2018, 2:38 AM
Hahnfeld committed rOMP343402: [libomptarget-nvptx] Align data sharing stack.
[libomptarget-nvptx] Align data sharing stack
Sep 30 2018, 2:27 AM
Hahnfeld committed rOMP343401: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD).
[libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD)
Sep 30 2018, 2:27 AM
Hahnfeld committed rL343402: [libomptarget-nvptx] Align data sharing stack.
[libomptarget-nvptx] Align data sharing stack
Sep 30 2018, 2:25 AM
Hahnfeld closed D52655: [libomptarget-nvptx] Align data sharing stack.
Sep 30 2018, 2:25 AM
Hahnfeld committed rL343401: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD).
[libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD)
Sep 30 2018, 2:25 AM
Hahnfeld closed D51787: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD).
Sep 30 2018, 2:24 AM

Sep 29 2018

Hahnfeld committed rOMP343381: [libomptarget-nvptx] Ignore calls to dynamic API.
[libomptarget-nvptx] Ignore calls to dynamic API
Sep 29 2018, 9:04 AM
Hahnfeld committed rOMP343382: [libomptarget-nvptx] Add tests for nested parallelism.
[libomptarget-nvptx] Add tests for nested parallelism
Sep 29 2018, 9:04 AM
Hahnfeld committed rOMP343380: [libomptarget-nvptx] Fix number of threads in parallel.
[libomptarget-nvptx] Fix number of threads in parallel
Sep 29 2018, 9:04 AM
Hahnfeld committed rL343381: [libomptarget-nvptx] Ignore calls to dynamic API.
[libomptarget-nvptx] Ignore calls to dynamic API
Sep 29 2018, 9:04 AM
Hahnfeld committed rL343382: [libomptarget-nvptx] Add tests for nested parallelism.
[libomptarget-nvptx] Add tests for nested parallelism
Sep 29 2018, 9:04 AM
Hahnfeld closed D51785: [libomptarget-nvptx] Ignore calls to dynamic API.
Sep 29 2018, 9:04 AM
Hahnfeld closed D51786: [libomptarget-nvptx] Add tests for nested parallelism.
Sep 29 2018, 9:04 AM
Hahnfeld committed rL343380: [libomptarget-nvptx] Fix number of threads in parallel.
[libomptarget-nvptx] Fix number of threads in parallel
Sep 29 2018, 9:04 AM
Hahnfeld closed D51783: [libomptarget-nvptx] Fix number of threads in parallel.
Sep 29 2018, 9:03 AM
Hahnfeld added a comment to D52655: [libomptarget-nvptx] Align data sharing stack.

Interesting. Are there other cases where alignment is needed?

Sep 29 2018, 4:45 AM
Hahnfeld added a comment to D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

You report a slow down which I am not able to reproduce actually. Do you use any additional clauses not present in your previous post?

No, only dist_schedule(static) which is faster. Tested on a Tesla P100 with today's trunk version:

#pragma omp target teams distribute parallel for (new defaults)190 - 250 GB/s
adding clauses for old defaults: schedule(static) dist_schedule(static)30 - 50 GB/s
same directive with only dist_schedule(static) added (fewer registers)320 - 400 GB/s

Which loop size you're using ? What runtime does nvprof report for these kernels?

Sep 29 2018, 4:44 AM

Sep 28 2018

Hahnfeld added a comment to D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

You report a slow down which I am not able to reproduce actually. Do you use any additional clauses not present in your previous post?

Sep 28 2018, 8:33 AM
Hahnfeld created D52655: [libomptarget-nvptx] Align data sharing stack.
Sep 28 2018, 8:23 AM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

Say, last distribute chunk is [L, U]. In the inner for directive it is split into [L,U1], [U1+1, U2], ..., [Un-1 + 1, U]. Distribute marks all these chunks as last, not the last [Un-1 + 1, U].

I got that. This is why the outer distribute only passes the global address for its last chunk. Then the inner for decides which thread executes [Un-1 + 1, U] and writes the lastprivate value.

Yes, that's right! You got it.

Sep 28 2018, 8:23 AM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I don't see why the distribute loop cares which thread actually executes the last iteration of the for loop, that's only relevant in the outlined parallel region.

Because it marks as lastprivate not the last loop chunk executed by the last thread, but the set of loop chunks executed by the last team. It means that when you try to write the lastprivate value after the distribute loop you will have multiple writes from the different threads with the different values of lastprivates.

Say, last distribute chunk is [L, U]. In the inner for directive it is split into [L,U1], [U1+1, U2], ..., [Un-1 + 1, U]. Distribute marks all these chunks as last, not the last [Un-1 + 1, U].

Sep 28 2018, 8:15 AM
Hahnfeld added a comment to D51787: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD).

Ping

Sep 28 2018, 8:09 AM
Hahnfeld added a comment to D51786: [libomptarget-nvptx] Add tests for nested parallelism.

Ping

Sep 28 2018, 8:08 AM
Hahnfeld added a comment to D51783: [libomptarget-nvptx] Fix number of threads in parallel.

Ping

Sep 28 2018, 8:08 AM
Hahnfeld committed rOMP343324: [libomptarget-nvptx] Add testing infrastructure.
[libomptarget-nvptx] Add testing infrastructure
Sep 28 2018, 8:08 AM
Hahnfeld added a comment to D51785: [libomptarget-nvptx] Ignore calls to dynamic API.

Ping

Sep 28 2018, 8:08 AM
Hahnfeld committed rL343324: [libomptarget-nvptx] Add testing infrastructure.
[libomptarget-nvptx] Add testing infrastructure
Sep 28 2018, 8:07 AM
Hahnfeld closed D51687: [libomptarget-nvptx] Add testing infrastructure.
Sep 28 2018, 8:07 AM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..
  1. It is not how clang works, it is how standard requires.
Sep 28 2018, 8:03 AM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

No, you're not correct here.

void outlined_target_fn(int *last) {
  int *last_ds = /* get data sharing frame from runtime */
  for (/* distribute loop from 0 to 9999 */) {
    outlined_parallel_fn(lb, ub, last_ds);
  }
  if (/* received last chunk */) {
    *last = *last_ds;
  }
}

This code is for the distribute loop. And here you have conflict without the datasharing scheme. The problem here is that this check /* received last chunk */ is true for all inner loop iterations for inner for directive and *last_ds may come not from the last iteration of for loop, but from some other iterations. To solve this problem, we need to share the same last_ds between all the threads in the team.

Sep 28 2018, 6:44 AM
Hahnfeld added a comment to D51687: [libomptarget-nvptx] Add testing infrastructure.

I don't have any comments, just waiting for the --libomptarget-nvptx-path flag to be accepted into the clang driver. Then we can have a look at the rest of this stack.

Sep 28 2018, 6:25 AM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I already described it - it breaks the compatibility with other outlined regions and breaks the whole design of the OpenMP implementation.

[...]

Yes, Alex Eichenberger tries to invent something, that will allow us to use something similar to ibm-devel but without breaking the design of OpenMP in the compiler. But it requires some time. But I'd like to have something working, at least.

Sep 28 2018, 6:13 AM
Hahnfeld added a comment to D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

One big problem your code has is that the trip count is incredibly small, especially for STREAM and especially on GPUs. You need a much larger loop size otherwise the timings will be dominated by OpenMP setups costs.

Sep 28 2018, 5:34 AM
Hahnfeld added a comment to D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

Just tested this and got very weird results for register usage:

void func(double *a) {
  #pragma omp target teams distribute parallel for map(a[0:100]) // dist_schedule(static)
  for (int i = 0; i < 100; i++) {
    a[i]++;
  }
}
Sep 28 2018, 1:57 AM

Sep 27 2018

Hahnfeld committed rC343240: Fix greedy FileCheck expression in test/Driver/mips-abi.c.
Fix greedy FileCheck expression in test/Driver/mips-abi.c
Sep 27 2018, 10:29 AM
Hahnfeld committed rL343240: Fix greedy FileCheck expression in test/Driver/mips-abi.c.
Fix greedy FileCheck expression in test/Driver/mips-abi.c
Sep 27 2018, 10:29 AM
Hahnfeld closed D52619: Fix greedy FileCheck expression in test/Driver/mips-abi.c.
Sep 27 2018, 10:29 AM
Hahnfeld created D52619: Fix greedy FileCheck expression in test/Driver/mips-abi.c.
Sep 27 2018, 9:31 AM
Hahnfeld committed rC343230: [OpenMP] Improve search for libomptarget-nvptx.
[OpenMP] Improve search for libomptarget-nvptx
Sep 27 2018, 9:14 AM
Hahnfeld committed rL343230: [OpenMP] Improve search for libomptarget-nvptx.
[OpenMP] Improve search for libomptarget-nvptx
Sep 27 2018, 9:14 AM
Hahnfeld closed D51686: [OpenMP] Improve search for libomptarget-nvptx.
Sep 27 2018, 9:14 AM
Hahnfeld accepted D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

That is the intention. I just took out that part from here to not confuse things since I wanted to have that as a separate patch. :)

Sep 27 2018, 8:26 AM
Hahnfeld added a comment to D52434: [OpenMP] Make default distribute schedule for NVPTX target regions in SPMD mode achieve coalescing.

Should we also change the default schedule to static, 1? I know that's not really needed for teams distribute parallel for (because the new default dist_schedule only leaves one iteration per thread), but this doesn't happen for target parallel for. Additionally it would make the intent more explicit and LLVM doesn't need to look through divisions needed to implement static without chunk. Just thinking aloud, not sure if that's worth it.

Sep 27 2018, 8:07 AM
Hahnfeld added a comment to D51687: [libomptarget-nvptx] Add testing infrastructure.

Considering your comment in the description about requiring latest Clang perhaps you should revisit this patch: D46842

Sep 27 2018, 3:58 AM
Hahnfeld updated the diff for D51686: [OpenMP] Improve search for libomptarget-nvptx.

Add const per review comments.

Sep 27 2018, 2:06 AM
Hahnfeld added inline comments to D51686: [OpenMP] Improve search for libomptarget-nvptx.
Sep 27 2018, 2:06 AM
Hahnfeld added a comment to D51937: [OPENMP]Increment iterator when the loop is continued..

(for reference: I missed this in D51623)

Sep 27 2018, 1:17 AM

Sep 26 2018

Hahnfeld added inline comments to D51686: [OpenMP] Improve search for libomptarget-nvptx.
Sep 26 2018, 12:52 PM
Hahnfeld added inline comments to D51786: [libomptarget-nvptx] Add tests for nested parallelism.
Sep 26 2018, 12:45 PM
Hahnfeld added a comment to D51686: [OpenMP] Improve search for libomptarget-nvptx.

Ping

Sep 26 2018, 12:38 PM
Hahnfeld added a comment to D52436: [OpenMP][libomptarget] Add runtime functions for default schedule for distribute.

Do we really need new entry points for this? I think we could avoid code duplication by letting the compiler generated code pass the correct chunk to __kmpc_for_static_init_??. This could either come from a (single) new query function (__kmpc_nvptx_distribute_default_chunk?) or can be hard coded to threadsPerBlock because this is only relevant for SPMD.

Sep 26 2018, 2:43 AM

Sep 21 2018

Hahnfeld added a comment to D52259: [CUDA] Fix two failed test cases using --cuda-path-ignore-env.

IMO the current logic makes sense for the end user, prioritizing the environment guess over default paths. I think the tests should be fixed using --cuda-path-ignore-env, I remember I added this parameter to fix the failing tests when I wrote the code.

Sep 21 2018, 6:07 AM · Restricted Project

Sep 10 2018

Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I already described it - it breaks the compatibility with other outlined regions and breaks the whole design of the OpenMP implementation.

First that's a general statement without any explanation. Second I'm not asking about the scratchpad pointer solution in ibm-devel but rather why we can't pass RequiresDataSharing = true to __kmpc_spmd_kernel_init. Which will give us the data sharing in existing buffers.

First, stop talking like this. I don't owe you anything.

Sep 10 2018, 1:53 PM
Hahnfeld added inline comments to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..
Sep 10 2018, 1:39 PM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I really, really dislike adding even more global buffers. 4096 * 32 * 56 are another 7MiB that are not usable for applications. What's wrong with using the existing ones?

Can you upload the CodeGen patch for reductions somewhere? I thought we need a global scratchpad buffer that is adressable for all teams?

I really, really dislike an implementation in ibm-devel, the scratchpad solution will never be added to the trunk. The existing ones cannot be reused, as they are allocated only if the full runtime is used.

What's the overhead of initializing it? The whole libomptarget-nvptx is already a pretty much mess, see my thread on openmp-dev.

It is not the runtime issue, it is the problem with the compiler itself. It breaks compatibility with the other outlined regions and, thus, it cannot be committed to trunk.

Can you please describe the problems? Again, maybe posting the patch may help.

I already described it - it breaks the compatibility with other outlined regions and breaks the whole design of the OpenMP implementation.

Sep 10 2018, 1:35 PM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I really, really dislike adding even more global buffers. 4096 * 32 * 56 are another 7MiB that are not usable for applications. What's wrong with using the existing ones?

Can you upload the CodeGen patch for reductions somewhere? I thought we need a global scratchpad buffer that is adressable for all teams?

I really, really dislike an implementation in ibm-devel, the scratchpad solution will never be added to the trunk. The existing ones cannot be reused, as they are allocated only if the full runtime is used.

What's the overhead of initializing it? The whole libomptarget-nvptx is already a pretty much mess, see my thread on openmp-dev.

It is not the runtime issue, it is the problem with the compiler itself. It breaks compatibility with the other outlined regions and, thus, it cannot be committed to trunk.

Sep 10 2018, 1:25 PM
Hahnfeld added a comment to D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime..

I really, really dislike adding even more global buffers. 4096 * 32 * 56 are another 7MiB that are not usable for applications. What's wrong with using the existing ones?

Can you upload the CodeGen patch for reductions somewhere? I thought we need a global scratchpad buffer that is adressable for all teams?

I really, really dislike an implementation in ibm-devel, the scratchpad solution will never be added to the trunk. The existing ones cannot be reused, as they are allocated only if the full runtime is used.

Sep 10 2018, 1:16 PM
Hahnfeld added a reviewer for D51875: [OPENMP][NVPTX] Add support for lastprivates/reductions handling in SPMD constructs with lightweight runtime.: Hahnfeld.

I really, really dislike adding even more global buffers. 4096 * 32 * 56 are another 7MiB that are not usable for applications. What's wrong with using the existing ones?

Sep 10 2018, 1:04 PM

Sep 8 2018

Hahnfeld committed rOMP341748: [libomptarget-nvptx] Remove last mentions of __kmpc_print_*.
[libomptarget-nvptx] Remove last mentions of __kmpc_print_*
Sep 8 2018, 5:12 AM
Hahnfeld committed rL341748: [libomptarget-nvptx] Remove last mentions of __kmpc_print_*.
[libomptarget-nvptx] Remove last mentions of __kmpc_print_*
Sep 8 2018, 5:12 AM

Sep 7 2018

Hahnfeld added a dependent revision for D51786: [libomptarget-nvptx] Add tests for nested parallelism: D51787: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD).
Sep 7 2018, 6:35 AM
Hahnfeld added a dependency for D51787: [libomptarget-nvptx] Fix ancestor_thread_num and team_size (non-SPMD): D51786: [libomptarget-nvptx] Add tests for nested parallelism.
Sep 7 2018, 6:35 AM
Hahnfeld added a dependency for D51785: [libomptarget-nvptx] Ignore calls to dynamic API: D51687: [libomptarget-nvptx] Add testing infrastructure.
Sep 7 2018, 6:35 AM
Hahnfeld added a dependency for D51786: [libomptarget-nvptx] Add tests for nested parallelism: D51687: [libomptarget-nvptx] Add testing infrastructure.
Sep 7 2018, 6:35 AM
Hahnfeld added a dependent revision for D51687: [libomptarget-nvptx] Add testing infrastructure: D51786: [libomptarget-nvptx] Add tests for nested parallelism.
Sep 7 2018, 6:35 AM
Hahnfeld added a dependent revision for D51687: [libomptarget-nvptx] Add testing infrastructure: D51783: [libomptarget-nvptx] Fix number of threads in parallel.
Sep 7 2018, 6:35 AM
Hahnfeld created D51786: [libomptarget-nvptx] Add tests for nested parallelism.
Sep 7 2018, 6:35 AM
Hahnfeld added a dependency for D51783: [libomptarget-nvptx] Fix number of threads in parallel: D51687: [libomptarget-nvptx] Add testing infrastructure.
Sep 7 2018, 6:35 AM