[Bug c/27702] the trampoline code of nested functions depends on executable stacks
--- Comment #5 from pinskia at gcc dot gnu dot org 2006-09-07 15:20 --- (In reply to comment #4) > Andrew, could you elaborate on the FC5 kernel bug and "the rules that are > always used"? This kernel problem has been fixed for a while now, since around June or so. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27702
[Bug c/27702] the trampoline code of nested functions depends on executable stacks
--- Comment #4 from hollis at penguinppc dot org 2006-09-07 05:22 --- Andrew, could you elaborate on the FC5 kernel bug and "the rules that are always used"? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27702
[Bug c/27702] the trampoline code of nested functions depends on executable stacks
--- Comment #3 from pinskia at gcc dot gnu dot org 2006-05-21 20:32 --- yes but you should report the bug to the distros that change the rules and let them work out a solution instead of having the FSF GCC fix their bug. Anyways Ada needs trampolines. Also it is harder to use mmap and you have to unmmap them which gets crazy which is why using the stack makes stuff easier to work with. Now people writting good code should not be punished by the case where people writting bad code that can be exploited. The problem here is about trust, if you cannot trust a program, don't run it. Making disucsions for me, is wrong. I should be able to decide if I want to trust a program or not and not the distro. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27702
[Bug c/27702] the trampoline code of nested functions depends on executable stacks
--- Comment #2 from subdino2004 at yahoo dot fr 2006-05-21 20:25 --- (In reply to comment #1) > This is not a bug in GCC but with your distro changing the rules that are > always used. Why should being unable to use local functions be an acceptable side-effect of prohibiting executable stack if there is known solution ? What would be wrong to provide it as an alternative - if not accepted as the default behaviour ? Or should it become a no-one's fault bug ? Ah, not bug, I mean problematic side-effect. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27702
[Bug c/27702] the trampoline code of nested functions depends on executable stacks
--- Comment #1 from pinskia at gcc dot gnu dot org 2006-05-21 16:01 --- There is a bug in the Redhat's kernel for FC5 that causes the trampoline to fail, someone should report this to them. This is not a bug in GCC but with your distro changing the rules that are always used. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||INVALID http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27702
