search

MagLev / maglev

GemStone Maglev Ruby Repository
http://maglev.github.io
518 stars 43 forks source link

possible bad code generation for < compares #124

Open AllenOtis opened 12 years ago

AllenOtis commented 12 years ago

In working on Gemstone/S 64bit v3.1, I find that bad code may be emitted for the bytecode Bc_SEND_SPECIAL_LT_u1_u32 . That bytecode was added in v3.0.x but not used by smalltalk IR graphs. I'm not sure if that bytecode is present in the Ruby IR graphs.

Fix will be in the v3.1 smalltalk VM in a week or two. I will put this GIt issue number in the SVN history of the VM .

Monty commented 12 years ago

Thanks!

----- Original Message ----- From: "Allen Otis" reply@reply.github.com To: "Monty Williams" monty.williams@gemstone.com Sent: Friday, March 16, 2012 9:51:28 AM Subject: [maglev] possible bad code generation for < compares (#124)

In working on Gemstone/S 64bit v3.1, I find that bad code may be emitted for the bytecode Bc_SEND_SPECIAL_LT_u1_u32 . That bytecode was added in v3.0.x but not used by smalltalk IR graphs. I'm not sure if that bytecode is present in the Ruby IR graphs.

Fix will be in the v3.1 smalltalk VM in a week or two. I will put this GIt issue number in the SVN history of the VM .

Reply to this email directly or view it on GitHub: https://github.com/MagLev/maglev/issues/124

jc00ke commented 12 years ago

Is this still an issue?