[PATCH] D146157: [MSAN] Support load and stores of scalable vector types

Philip Reames via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Mar 22 14:10:20 PDT 2023


reames added inline comments.


================
Comment at: llvm/test/Instrumentation/MemorySanitizer/vector-load-store.ll:29
 
 define void @load.v2i32(ptr %p) sanitize_address {
 ; CHECK-LABEL: @load.v2i32(
----------------
vitalybuka wrote:
> it should be sanitize_memory
> can you please fix and regenerate the test separately?
Fixed in 473e9adb84c29764da51e4d2a995fce9b2e2ffa5, thanks for noticing.  


================
Comment at: llvm/test/Instrumentation/MemorySanitizer/vector-load-store.ll:385
+; CALLS-NEXT:    [[TMP4:%.*]] = ptrtoint ptr [[P:%.*]] to i64
+; CALLS-NEXT:    call void @__asan_loadN(i64 [[TMP4]], i64 [[TMP3]])
+; CALLS-NEXT:    [[TMP5:%.*]] = load <vscale x 8 x i32>, ptr [[P]], align 32
----------------
vitalybuka wrote:
> I am confused, how passes=msan created asan calls?
This looks like a copy paste mistake on my part.  update_test_checks.py doesn't strip out check lines which don't correspond to a run line, and I'd copied these from the ASAN test.  Will fix and refresh.


================
Comment at: llvm/test/Instrumentation/MemorySanitizer/vector-load-store.ll:482
+
+define void @store.nxv2i32(ptr %p) sanitize_address {
+; CHECK-LABEL: @store.nxv2i32(
----------------
vitalybuka wrote:
> could you include these cases into precommited tests?
Can't.  We crash on scalable input types.  That's the whole point of the patch.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D146157/new/

https://reviews.llvm.org/D146157



More information about the llvm-commits mailing list