Re: Non-Exec stack patches

From: David Mosberger
Date: Wed Mar 24 2004 - 13:03:38 EST

>>>>> On Wed, 24 Mar 2004 12:24:54 -0500, Jakub Jelinek <jakub@xxxxxxxxxx> said:

>> Huh? Only one mprotect() call is needed to make the entire stack
>> executable.

Jakub> Nope. Think about multithreaded apps.

I said one mprotect() to make the entire stack executable. Obviously,
if you have multiple stacks, you need one call per stack. Big deal.

Jakub> Furthermore, getting the exact extents of the particular
Jakub> stack is difficult to find for applications, but e.g. the
Jakub> threading library has to know such things.

And how is this a kernel problem? There are other reasons why
applications need to know the stack extents (think garbage collector)
and it's entirely glibc's failing if it's difficult to determine the
stack extent.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at