This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.

Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] TLS support for s390*.


On Thu, Jan 30, 2003 at 05:08:27PM +0100, Martin Schwidefsky wrote:
> 
> > With linuxthreads current CVS no threaded programs work with
> > --with-tls --without-__thread.
> Never tested this combination so far. I'll give it a try.

--enable-addons=linuxthreads FYI.

BTW: It seems to work just fine if stack_addr passed to clone
is at least 8*16 bytes below end of particular mmaped area
(this is on s390x). My wild guess is that the register save
area is needed, which means clone.S should subtract 8*16
resp. 8*8 (on s390-32) bytes from stack_pointer before doing
the syscall.

> > From my current debugging it seems the problem is that
> > clone() is passed end of some VM region as stack_addr
> > and segfaults somewhere in the kernel when accessing it.
> > Were there some recent kernel fixes in this area?
> I have a patch for sys_clone/copy_thread that implements
> CLONE_SETTLS which is not yet in the mainstream kernel.
> Does --without-__thread need the CLONE_SETTLS semantic ?

No, linuxthreads doesn't use any of the new CLONE flags.
It needs to run on old kernels too.

	Jakub


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]