This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] PowerPC thread pointer bias requires powerpc/lowlevelrobustlock.sym
- From: Fabio Massimo Di Nitto <fabbione at ubuntu dot com>
- To: Benjamin Herrenschmidt <benh at kernel dot crashing dot org>
- Cc: "Ryan S. Arnold" <rsa at us dot ibm dot com>, libc-alpha at sourceware dot org, Ben Collins <ben dot collins at ubuntu dot com>, Jeff Bailey <jbailey at ubuntu dot com>, Paul Mackerras <paulus at samba dot org>, Steve Munroe <sjmunroe at us dot ibm dot com>
- Date: Thu, 02 Nov 2006 06:54:44 +0100
- Subject: Re: [PATCH] PowerPC thread pointer bias requires powerpc/lowlevelrobustlock.sym
- References: <1162426488.28331.0.camel@ryanarn-desktop> <email@example.com>
Benjamin Herrenschmidt wrote:
> On Wed, 2006-11-01 at 18:14 -0600, Ryan S. Arnold wrote:
>> I've been sitting on this patch for way too long.
>> The TID implementation in the default lowlevelrobustlock.sym is wrong
>> for PowerPC. The thread point on PowerPC is biased by -0x7000 to extend
>> the size of the FAST TLS to 64KB. This patch provides a powerpc
>> implementation to take this into account.
> Fabio, Jeff: That should fix the root cause of the errors on 2.6.18 and
> later for which my patch corrected the kernel oops (you would still have
> gotten an error from the syscall due to that glibc problem).
Perfect. I will apply in the next upload.
I'm going to make him an offer he can't refuse.