[Openmp-commits] [openmp] 3b8ac7a - [OpenMP] Fix various alignment issues (#142376)

[via Openmp-commits]

Author: Rainer Orth
Date: 2025-06-25T21:32:14+02:00
New Revision: 3b8ac7a22c649d42e748b93d67ac237a85524328

URL: https://github.com/llvm/llvm-project/commit/3b8ac7a22c649d42e748b93d67ac237a85524328
DIFF: https://github.com/llvm/llvm-project/commit/3b8ac7a22c649d42e748b93d67ac237a85524328.diff

LOG: [OpenMP] Fix various alignment issues (#142376)

When running the `openmp` testsuite on 32-bit SPARC, several tests
`FAIL` apparently randomly, but always with the same kind of error:
```
# error: command failed with exit status: -11
```
The tests die with `SIGBUS`, as can be seen in `truss` output:
```
26461/1:            Incurred fault #5, FLTACCESS  %pc = 0x00010EAC
26461/1:              siginfo: SIGBUS BUS_ADRALN addr=0x0013D12C
26461/1:            Received signal #10, SIGBUS [default]
26461/1:              siginfo: SIGBUS BUS_ADRALN addr=0x0013D12C
```
i.e. the code is trying an unaligned access which cannot work on SPARC,
a strict-alignment target which enforces natural alignment on access.
This explains the apparent randomness of the failures: if the memory
happens to be aligned appropriately, the tests work, but fail if not.

A `Debug` build reveals much more:

- `__kmp_alloc` currently aligns to `sizeof(void *)`, which isn't enough
on strict-alignment targets when the data are accessed as types
requiring larger alignment. Therefore, this patch increases `alignment`
to `SizeQuant`.

- 32-bit Solaris/sparc `libc` guarantees 8-byte alignment from `malloc`,
so this patch adjusts `SizeQuant` to match.

- There's a `SIGBUS` in
  ```
  __kmpc_fork_teams (loc=0x112f8, argc=0, 
microtask=0x16cc8
<__omp_offloading_ffbc020a_4b1abe_main_l9_debug__.omp_outlined>)
      at openmp/runtime/src/kmp_csupport.cpp:573
  573	  *(kmp_int64 *)(&this_thr->th.th_teams_size) = 0L;
  ```
Casting to a pointer to a type requiring 64-bit alignment when that
isn't guaranteed is wrong. Instead, this patch uses `memset` instead.

- There's another `SIGBUS` in
  ```
0xfef8cb9c in __kmp_taskloop_recur (loc=0x10cb8, gtid=0, task=0x23cd00,
lb=0x23cd18, ub=0x23cd20, st=1, ub_glob=499, num_tasks=100, grainsize=5,
      extras=0, last_chunk=0, tc=500, num_t_min=20, 
codeptr_ra=0xfef8dbc8 <__kmpc_taskloop(ident_t*, int, kmp_task_t*, int,
kmp_uint64*, kmp_uint64*, kmp_int64, int, int, kmp_uint64, void*)+240>,
      task_dup=0x0)
      at openmp/runtime/src/kmp_tasking.cpp:5147
  5147	  p->st = st;
  ```
`p->st` doesn't currently guarantee the 8-byte alignment required by
`kmp_int64 st`. `p` is set in
  ```
   __taskloop_params_t *p = (__taskloop_params_t *)new_task->shareds;
  ```
but `shareds_offset` is currently aligned to `sizeof(void *)` only.
Increasing it to `sizeof(kmp_uint64)` to match its use fixes the
`SIGBUS`.

With these fixes I get clean `openmp` test results on 32-bit SPARC (both
Solaris and Linux), with one unrelated exception.

Tested on `sparc-sun-solaris2.11`, `sparcv9-sun-solaris2.11`,
`sparc-unknown-linux-gnu`, `sparc64-unknown-linux-gnu`,
`i386-pc-solaris2.11`, `amd64-pc-solaris2.11`, `i686-pc-linux-gnu`, and
`x86_64-pc-linux-gnu`.

Added: 
    

Modified: 
    openmp/runtime/src/kmp_alloc.cpp
    openmp/runtime/src/kmp_csupport.cpp
    openmp/runtime/src/kmp_tasking.cpp

Removed: 
    


################################################################################
diff  --git a/openmp/runtime/src/kmp_alloc.cpp b/openmp/runtime/src/kmp_alloc.cpp
index 801cd06c95502..051f88c5a0996 100644
--- a/openmp/runtime/src/kmp_alloc.cpp
+++ b/openmp/runtime/src/kmp_alloc.cpp
@@ -70,10 +70,10 @@ static void bectl(kmp_info_t *th, bget_compact_t compact,
 /* Buffer allocation size quantum: all buffers allocated are a
    multiple of this size.  This MUST be a power of two. */
 
-/* On IA-32 architecture with  Linux* OS, malloc() does not
-   ensure 16 byte alignment */
+/* On some architectures, malloc() does not ensure 16 byte alignment,
+   Solaris/sparc and x86 among them. */
 
-#if KMP_ARCH_X86 || !KMP_HAVE_QUAD
+#if KMP_ARCH_X86 || KMP_ARCH_SPARC || !KMP_HAVE_QUAD
 
 #define SizeQuant 8
 #define AlignType double
@@ -1861,7 +1861,7 @@ typedef struct kmp_mem_desc { // Memory block descriptor
   void *ptr_align; // Pointer to aligned memory, returned
   kmp_allocator_t *allocator; // allocator
 } kmp_mem_desc_t;
-static int alignment = sizeof(void *); // align to pointer size by default
+constexpr size_t alignment = SizeQuant;
 
 // external interfaces are wrappers over internal implementation
 void *__kmpc_alloc(int gtid, size_t size, omp_allocator_handle_t allocator) {

diff  --git a/openmp/runtime/src/kmp_csupport.cpp b/openmp/runtime/src/kmp_csupport.cpp
index fdbf9ff45e354..3ca32ba583fe2 100644
--- a/openmp/runtime/src/kmp_csupport.cpp
+++ b/openmp/runtime/src/kmp_csupport.cpp
@@ -570,7 +570,7 @@ void __kmpc_fork_teams(ident_t *loc, kmp_int32 argc, kmpc_micro microtask,
 
   this_thr->th.th_teams_microtask = NULL;
   this_thr->th.th_teams_level = 0;
-  *(kmp_int64 *)(&this_thr->th.th_teams_size) = 0L;
+  memset(&this_thr->th.th_teams_size, 0, sizeof(kmp_teams_size_t));
   va_end(ap);
 #if KMP_STATS_ENABLED
   if (previous_state == stats_state_e::SERIAL_REGION) {

diff  --git a/openmp/runtime/src/kmp_tasking.cpp b/openmp/runtime/src/kmp_tasking.cpp
index 3d85a29423540..d7bc4922d54f7 100644
--- a/openmp/runtime/src/kmp_tasking.cpp
+++ b/openmp/runtime/src/kmp_tasking.cpp
@@ -1528,7 +1528,7 @@ kmp_task_t *__kmp_task_alloc(ident_t *loc_ref, kmp_int32 gtid,
   // Calculate shared structure offset including padding after kmp_task_t struct
   // to align pointers in shared struct
   shareds_offset = sizeof(kmp_taskdata_t) + sizeof_kmp_task_t;
-  shareds_offset = __kmp_round_up_to_val(shareds_offset, sizeof(void *));
+  shareds_offset = __kmp_round_up_to_val(shareds_offset, sizeof(kmp_uint64));
 
   // Allocate a kmp_taskdata_t block and a kmp_task_t block.
   KA_TRACE(30, ("__kmp_task_alloc: T#%d First malloc size: %ld\n", gtid,