[PATCH] D92398: [AIX][XCOFF] emit traceback table for function in aix

Hubert Tong via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Dec 10 12:01:58 PST 2020


hubert.reinterpretcast added inline comments.


================
Comment at: llvm/lib/BinaryFormat/XCOFF.cpp:107
+
+SmallString<32> XCOFF::parseParmsType(uint32_t Value, unsigned ParmsNum) {
+  SmallString<32> ParmsType;
----------------
When we revisit for vector parameters, we should also revisit this function.


================
Comment at: llvm/lib/BinaryFormat/XCOFF.cpp:127-128
+  }
+  assert(Value == 0u && "The paramter number in the ParamTpye is not equal to "
+                        "argument ParamsNum.");
+  return ParmsType;
----------------



================
Comment at: llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp:1775
+  // There is a lack of information in the IR to assist with determining the
+  // source language, AIX exception handling mechanism would only search for
+  // personality routine and LSDA area when such language supports exception
----------------
Minor nit: Use period instead of comma.


================
Comment at: llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp:1867
+  uint32_t FPRSaved = 0;
+  for (unsigned Reg = PPC::F14; Reg <= PPC::F31; Reg++) {
+    if (MRI.isPhysRegModified(Reg)) {
----------------
Minor nit: preincrement.


================
Comment at: llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp:1889
+
+  for (unsigned Reg = GPRBegin; Reg <= GPREnd; Reg++) {
+    if (MRI.isPhysRegModified(Reg)) {
----------------
Minor nit: same comment re: `++`.


================
Comment at: llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp:1899-1900
+
+  GENBOOLCOMMENT("", SecondHalfOfMandatoryField, HasExtensionTable);
+  GENBOOLCOMMENT(", ", SecondHalfOfMandatoryField, HasVectorInfo);
+  GENVALUECOMMENT(", NumOfGPRsSaved", SecondHalfOfMandatoryField, GPRSaved);
----------------
The values of the constants reflect the XL order but the output order doesn't.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D92398/new/

https://reviews.llvm.org/D92398



More information about the llvm-commits mailing list