Closed
Bug 534749
Opened 15 years ago
Closed 13 years ago
TM: we try to overflow guard loads
Categories
(Core :: JavaScript Engine, defect)
Tracking
()
RESOLVED
FIXED
People
(Reporter: gal, Assigned: gal)
References
Details
Attachments
(1 file)
|
2.56 KB,
patch
|
Details | Diff | Splinter Review |
Do I really need to explain why thats a really bad idea?
|
Assignee | |
Comment 1•15 years ago
|
||
LIR_ov checks for an arithmetic overflow. If we peephole optimized away the arithmetic op, we check the overflow flag in the processor and it might be set to god knows what. This is a performance bug. No correctness implications.
Assignee: general → gal
|
|
Assignee | |
Updated•15 years ago
|
Attachment #417571 -
Flags: review?(dvander)
|
||
Comment 2•15 years ago
|
||
make sure the nanojit parts are checked into the right repo.
|
|
Assignee | |
Comment 3•15 years ago
|
||
yeah, thanks for the reminder
|
||
Updated•13 years ago
|
Attachment #417571 -
Flags: review?(dvander)
|
|
||
Comment 4•13 years ago
|
||
The addition of addxovi/subxovi/mulxovi made it impossible for this problem to occur.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
You need to log in
before you can comment on or make changes to this bug.
Description
•