Request for reviews (S): 7198084: NPG: addcaps03 looks like it exceeds it's code buffer (original) (raw)
Christian Thalinger christian.thalinger at oracle.com
Thu Sep 27 15:50:58 PDT 2012
- Previous message: Request for reviews (S): 7198084: NPG: addcaps03 looks like it exceeds it's code buffer
- Next message: Request for reviews (S): 7198084: NPG: addcaps03 looks like it exceeds it's code buffer
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Looks good. -- Chris
On Sep 27, 2012, at 3:41 PM, Vladimir Kozlov <vladimir.kozlov at oracle.com> wrote:
http://cr.openjdk.java.net/~kvn/7198084/webrev
We should not use some short branches (T4 SPARC cbcond) in method InterpreterMacroAssembler::testinvocationcounterformdp() because in debug VM with Jvmti events enabled and CodeCache allocation far from runtime functions the code generated by callVM() and by following testbackedgecountforosr() is large. The distance to Lforward label can't be encoded into short branch instructions. Note: bug's synopsis is incorrect, I will change it. Thanks, Vladimir
- Previous message: Request for reviews (S): 7198084: NPG: addcaps03 looks like it exceeds it's code buffer
- Next message: Request for reviews (S): 7198084: NPG: addcaps03 looks like it exceeds it's code buffer
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the hotspot-compiler-dev mailing list