Notes |
(0013483 - 51 - 51 - 51 - 51 - 51 - 51)
lewis
01-16-10 15:29
|
Yes, it crashes with a seg fault. Definitely a bug. |
|
(0013658 - 116 - 122 - 122 - 122 - 122 - 122)
laza
04-08-10 12:26
|
It crashes my image too, but I'm using an older linux vm.
Dave, is this fixed with the new and shiny 4.0.2-2172 vm? |
|
(0013676 - 326 - 338 - 338 - 338 - 338 - 338)
lewis
04-09-10 02:22
|
No, we do not have a fix for this. On 4.0.2-2172 I do not get a seg fault, but the VM locks up 100% cpu. So we see various symptoms, but no fix has been proposed and it is still definitely a bug.
I do not see this issue as a concern for Squeak 4.1 release. It is a pre-existing VM condition, unrelated to the image version. |
|
(0013831 - 60 - 60 - 60 - 60 - 60 - 60)
leves
08-01-10 14:03
|
The CogVM doesn't have this issue, so it probably has a fix. |
|
(0013986 - 263 - 297 - 297 - 297 - 297 - 297)
lewis
12-16-10 18:50
|
The underlying issue seems to be that we should get a DNU for this:
1 value: (CompiledMethod allInstances anyOne)
Cog correctly gives a DNU and the interpreter VM does not. It either evaluates to the argument (i.e. a CompiledMethod), or it crashes the VM. |
|