Closed Bug 412615 Opened 17 years ago Closed 14 years ago

optimize LIR_fld with sse2 enabled

Categories

(Core Graveyard :: Nanojit, defect)

Product:
Core Graveyard
Component:
Nanojit
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED FIXED
Milestone:
Future

People

(Reporter: edwsmith, Unassigned)

References

Details

this was spotted in a trace:

     301 fld   299(10)
         00A86A9C  movsd xmm0,-16(ebx)                  ebx(10)
         00A86AA1  movq xmm1,-16(ebx)                   ebx(10) xmm0(301)
         00A86AA6  movq -16(ebp),xmm1                   ebx(10) xmm0(301)

the code could have been simpler:

         movq xmm0,-16(ebx)
         movq -16(ebp),xmm0

movsd and movq do the same thing in this situation.  the change eliminates an extra load.
Component: Virtual Machine → Tracing Virtual Machine
QA Contact: vm → tracing-vm
Priority: -- → P2
Summary: TT: optimize LIR_fld with sse2 enabled → optimize LIR_fld with sse2 enabled
Component: Tracing Virtual Machine → JIT Compiler (NanoJIT)
QA Contact: tracing-vm → nanojit
almost certianly a dupe of 520714, which is more comprehensive.
Status: NEW → RESOLVED
Closed: 15 years ago
Resolution: --- → DUPLICATE
leaving this open to track for Tamarin, work is all in 520714.
Status: RESOLVED → REOPENED
Depends on: 520714
OS: Windows XP → All
Priority: P2 → --
Hardware: x86 → All
Resolution: DUPLICATE → ---
Target Milestone: --- → Future
Component: JIT Compiler (NanoJIT) → Nanojit
Product: Tamarin → Core
QA Contact: nanojit → nanojit
bug 520714 is fixed, marking this one fixed too.
Status: REOPENED → RESOLVED
Closed: 15 years ago → 14 years ago
Resolution: --- → FIXED
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.