[PATCH] D27338: [globalisel] Tablegen-erate current Register Bank Information
Diana Picus via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Dec 6 08:58:56 PST 2016
rovka added a comment.
Just a few nitpicks, other than that LGTM (but you should probably wait for someone more experienced to approve).
================
Comment at: lib/Target/AArch64/AArch64RegisterBankInfo.h:19
+#define TBLGEN_REGISTERBANKINFO_H
+#include "AArch64GenRegisterBank.inc"
----------------
I think it's nicer to use a macro that describes what we're trying to include rather than where we're including it, e.g. GET_REGBANK_DECLS or something of the sort. This also seems to be the style used in GenRegisterInfo.inc, GenSubtargetInfo.inc etc.
================
Comment at: utils/TableGen/RegisterBankEmitter.cpp:202
+/// is a member.
+static void visitSelfAndSubclassesAndClassesWithSubRegs(
+ CodeGenRegBank &RegisterClassHierarchy, const RegisterBank &Bank,
----------------
How about visitRegisterBankClasses? The comment is clear enough on what something like that would mean :)
================
Comment at: utils/TableGen/RegisterBankEmitter.cpp:233
+ StringRef Enumeration, StringRef EnumeratorPrefix) {
+ OS << " assert(";
+ for (RegisterBank::PartialMappingsTy::const_iterator I = Begin;
----------------
I think it would be cleaner to generate a comment instead of a moot assertion if Begin == End or Begin + 1 == End (and then you can also remove the I != End check in the loop).
================
Comment at: utils/TableGen/RegisterBankEmitter.cpp:304
+ } else {
+ OS << " assert(" << PartialMappingEnumerationName.str() << "::PMI_"
+ << Bank.getName() << Bank.getRCWithSmallestRegsSize()->SpillSize
----------------
Are the calls to .str() really necessary?
================
Comment at: utils/TableGen/RegisterBankEmitter.cpp:309
+ << "\"" << Bank.getName()
+ << Bank.getRCWithSmallestRegsSize()->SpillSize << " not last in the "
+ << Bank.getName() << " list\");\n";
----------------
" not first "
https://reviews.llvm.org/D27338
More information about the llvm-commits
mailing list