[PATCH] D91370: [OPENMP]Fix PR48076: Check map types array before accessing its front.

[Joseph Huber via Phabricator via cfe-commits]

jhuber6 added a comment.

In D91370#2392319 <https://reviews.llvm.org/D91370#2392319>, @ABataev wrote:

> From libomptarget. It must be different. Could you copy it here?

Yeah, I think there's some extra information looking at it again.

  Target CUDA RTL --> Start initializing CUDA
  Libomptarget --> Successfully loaded library 'libomptarget.rtl.cuda.so'!
  Libomptarget --> Registering RTL libomptarget.rtl.cuda.so supporting 1 devices!
  Libomptarget --> Loading library 'libomptarget.rtl.aarch64.so'...
  Libomptarget --> Unable to load library 'libomptarget.rtl.aarch64.so': libomptarget.rtl.aarch64.so: cannot open shared object file: No such file or directory!
  Libomptarget --> Loading library 'libomptarget.rtl.ve.so'...
  Libomptarget --> Unable to load library 'libomptarget.rtl.ve.so': libomptarget.rtl.ve.so: cannot open shared object file: No such file or directory!
  Libomptarget --> Loading library 'libomptarget.rtl.amdgpu.so'...
  Libomptarget --> Unable to load library 'libomptarget.rtl.amdgpu.so': libomptarget.rtl.amdgpu.so: cannot open shared object file: No such file or directory!
  Libomptarget --> RTLs loaded!
  Libomptarget --> Image 0x0000000000400a30 is NOT compatible with RTL libomptarget.rtl.x86_64.so!
  Libomptarget --> Image 0x0000000000400a30 is compatible with RTL libomptarget.rtl.cuda.so!
  Libomptarget --> RTL 0x00000000008334c0 has index 0!
  Libomptarget --> Registering image 0x0000000000400a30 with RTL libomptarget.rtl.cuda.so!
  Libomptarget --> Done registering entries!
  Libomptarget --> Call to omp_get_num_devices returning 1
  Libomptarget --> Default TARGET OFFLOAD policy is now mandatory (devices were found)
  Libomptarget --> Entering target region with entry point 0x00000000004009e0 and device Id -1
  Libomptarget --> Checking whether device 0 is ready.
  Libomptarget --> Is the device 0 (local ID 0) initialized? 0
  Target CUDA RTL --> Init requires flags to 1
  Target CUDA RTL --> Getting device 0
  Target CUDA RTL --> The primary context is inactive, set its flags to CU_CTX_SCHED_BLOCKING_SYNC
  Target CUDA RTL --> Max CUDA blocks per grid 2147483647 exceeds the hard team limit 65536, capping at the hard limit
  Target CUDA RTL --> Using 1024 CUDA threads per block
  Target CUDA RTL --> Using warp size 32
  Target CUDA RTL --> Device supports up to 65536 CUDA blocks and 1024 threads with a warp size of 32
  Target CUDA RTL --> Default number of teams set according to library's default 128
  Target CUDA RTL --> Default number of threads set according to library's default 128
  Libomptarget --> Device 0 is ready to use.
  Target CUDA RTL --> Load data from image 0x0000000000400a30
  Target CUDA RTL --> CUDA module successfully loaded!
  Target CUDA RTL --> Entry point 0x0000000000000000 maps to __omp_offloading_fd02_267be70_main_l10 (0x00000000011d0ef0)
  Target CUDA RTL --> Sending global device environment data 4 bytes
  Libomptarget --> Entry  0: Base=0x00007ffc7e77c2a8, Begin=0x00007ffc7e77c370, Size=8, Type=0x20
  Libomptarget --> Entry  1: Base=0x00007ffc7e77c2a8, Begin=0x00007ffc7e77c2a8, Size=208, Type=0x1000000000203
  Libomptarget --> Looking up mapping(HstPtrBegin=0x00007ffc7e77c370, Size=8)...
  Libomptarget --> MemoryManagerTy::allocate: size 8 with host pointer 0x00007ffc7e77c370.
  Libomptarget --> findBucket: Size 8 is floored to 8.
  Libomptarget --> Cannot find a node in the FreeLists. Allocate on device.
  Libomptarget --> Node address 0x00000000011b8ba8, target pointer 0x00007fa792400000, size 8
  Libomptarget --> Creating new map entry: HstBase=0x00007ffc7e77c2a8, HstBegin=0x00007ffc7e77c370, HstEnd=0x00007ffc7e77c378, TgtBegin=0x00007fa792400000
  Libomptarget --> There are 8 bytes allocated at target address 0x00007fa792400000 - is new
  Libomptarget --> Looking up mapping(HstPtrBegin=0x00007ffc7e77c2a8, Size=208)...
  Libomptarget --> WARNING: Pointer is not mapped but section extends into already mapped data
  Libomptarget --> Mapping exists (implicit) with HstPtrBegin=0x00007ffc7e77c2a8, TgtPtrBegin=0x00007fa7923fff38, Size=208, RefCount=1
  Libomptarget --> There are 208 bytes allocated at target address 0x00007fa7923fff38 - is not new
  Libomptarget --> DeviceTy::getMapEntry: requested entry found
  Libomptarget --> Moving 208 bytes (hst:0x00007ffc7e77c2a8) -> (tgt:0x00007fa7923fff38)
  Target CUDA RTL --> Error when copying data from host to device. Pointers: host = 0x00007ffc7e77c2a8, device = 0x00007fa7923fff38, size = 208
  Target CUDA RTL --> CUDA error is: invalid argument
  Libomptarget --> Copying data to device failed.
  Libomptarget --> Call to targetDataBegin failed, abort target.
  Libomptarget --> Failed to process data before launching the kernel.
  Libomptarget error: run with env LIBOMPTARGET_INFO>1 to dump host-targetpointer maps
  Libomptarget fatal error 1: failure of target construct while offloading is mandatory
  Aborted (core dumped)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D91370/new/

https://reviews.llvm.org/D91370