Use new factory functions on x86 JIT for code readability#23940
Use new factory functions on x86 JIT for code readability#239400xdaryl wants to merge 4 commits into
Conversation
Signed-off-by: Daryl Maier <maier@ca.ibm.com>
Signed-off-by: Daryl Maier <maier@ca.ibm.com>
Signed-off-by: Daryl Maier <maier@ca.ibm.com>
Signed-off-by: Daryl Maier <maier@ca.ibm.com>
|
Jenkins test sanity xlinux,win,osx jdk21 |
|
Jenkins test sanity.functional xlinux,win jdk21 |
|
Jenkins test sanity.functional xlinux jdk21 |
|
Jenkins test sanity.functional xlinux,osx,win jdk21 |
|
Jenkins test sanity.functional win jdk21 |
|
It looks like there are already merge conflicts. Ignore that comment about the length of the commit comments. I was looking at the wrong commits. |
hzongaro
left a comment
hzongaro
left a comment
There was a problem hiding this comment.
I think this looks good. I just had a couple of questions about specific instances that weren't changed.
| @@ -3442,7 +3372,7 @@ TR::Register *J9::X86::TreeEvaluator::ArrayStoreCHKEvaluator(TR::Node *node, TR: | |||
| TR::Instruction *pachable | |||
| = generateVirtualGuardNOPInstruction(node, virtualGuard->addNOPSite(), NULL, oolASCLabel, cg); | |||
There was a problem hiding this comment.
Should generateVirtualGuardNOPInstruction become Inst_VirtualGuardNOPInstruction here?
| @@ -1912,15 +1893,15 @@ bool J9::X86::PrivateLinkage::buildVirtualGuard(TR::X86CallSite &site, TR::Label | |||
| = generateVirtualGuardNOPInstruction(callNode, virtualGuard->addNOPSite(), NULL, revirtualizeLabel, cg()); | |||
There was a problem hiding this comment.
Should generateVirtualGuardNOPInstruction become Inst_VirtualGuardNOPInstruction here?
| TR_VirtualGuard *HCRGuard | ||
| = TR_VirtualGuard::createGuardedDevirtualizationGuard(TR_HCRGuard, comp(), callNode); | ||
| TR::Instruction *HCRpatchable | ||
| = generateVirtualGuardNOPInstruction(callNode, HCRGuard->addNOPSite(), NULL, revirtualizeLabel, cg()); |
There was a problem hiding this comment.
Should generateVirtualGuardNOPInstruction become Inst_VirtualGuardNOPInstruction here?
| return cursor; | ||
| } | ||
|
|
||
| TR::Instruction *J9::X86::AMD64::PrivateLinkage::generateFlushInstruction(TR::Instruction *prev, |
There was a problem hiding this comment.
Should generateFlushInstruction and the references to it become Inst_FlushInstruction?
2 participants
Inst_XXXfamily of functions instead ofgenerateXXXInstructionRegDepsin place ofgenerateRegisterDependencyConditionsMRef_family of functions in place ofgenerateX86MemoryReferenceOP::typedef instead ofTR::InstOpCode::