On Thursday 28 July 2005 19:26, Bodo Stroesser wrote:
> Blaisorblade wrote:
> > On Thursday 28 July 2005 16:25, Jeff Dike wrote:
> Yes, you are right.
> But from a discussion on #uml I know, Jeff is using a Fedora, which seems
> to have another mechanism for vsyscall implementation. The base address of
> vsyscall is 0xa6b000 there.
<complaining>
Interesting, why is Fedora playing so much with the kernel? Having fun?
</complaining>
> That is below TASK_SIZE, and user can unmap it!
> So we have to check for that case in elf_aux.c.
So, assuming vsyscall_ehdr is page aligned, which is reasonable since of this
host definition, it's something like:
#define VSYSCALL_BASE (__fix_to_virt(FIX_VSYSCALL))
#define VSYSCALL_EHDR ((const struct elfhdr *) VSYSCALL_BASE)
#define VSYSCALL_ENTRY ((unsigned long) &__kernel_vsyscall)
- if ( ! __kernel_vsyscall || ! vsyscall_ehdr ||
- ! elf_aux_hwcap || ! elf_aux_platform ||
- ! page_size || (vsyscall_ehdr % page_size) ) {
+ if ( ! __kernel_vsyscall || ! vsyscall_ehdr ||
+ ! elf_aux_hwcap || ! elf_aux_platform ||
+ ! page_size || (vsyscall_ehdr % page_size) || (vsyscall_ehdr <
TASK_SIZE) {
The above change is critical enough to go in 2.6.13.
It does not affect existing users (i.e. no FC4); the problem is whether the
problem Jeff caught will.
I'd like to get the complete patch CC'ed, since I'm testing with TLS glibc
(and set_thread_area() now basically works, and I can even run some simple
test programs) and having problem which I'm unable to diagnose. Don't know
whether this is related...
> Such variant of vsyscall
> should not be made visible to UML-user. Since it is unmapped by UML
> implicitly, we simply should avoid to insert vsyscall info into auxv in
> this case.
Clear...
--
Inform me of my mistakes, so I can keep imitating Homer Simpson's "Doh!".
Paolo Giarrusso, aka Blaisorblade (Skype ID "PaoloGiarrusso", ICQ 215621894)
http://www.user-mode-linux.org/~blaisorblade
___________________________________
Yahoo! Mail: gratis 1GB per i messaggi e allegati da 10MB
http://mail.yahoo.it
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO September
19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
