[ANNOUNCE] uClibc-0.9.30.2 released

Khem Raj raj.khem at gmail.com
Fri Jan 22 03:02:05 UTC 2010


On Thu, Jan 21, 2010 at 5:13 PM, Rob Landley <rob at landley.net> wrote:
> On Thursday 21 January 2010 12:42:02 Rob Landley wrote:
>> On Sunday 17 January 2010 16:36:06 Rob Landley wrote:
>> > On Wednesday 13 January 2010 14:48:46 Bernhard Reutner-Fischer wrote:
>> > > The uClibc team is pleased to announce the release of
>> > > uClibc-0.9.30.2
>> > > a bugfix release to the stable 0_9_30 branch that contains bugfixes as
>> > > mentioned in ¹) as well as a feature-backport of AT_FILE support off
>> > > master.
>> >
>> > WOOT!
>> >
>> > Testing now...
>>
>> Mips is horked, and it was horked by this commit:
>>
>> 416d6bbbb8835202a802fcf1600df77b7264dc13 is the first bad commit
>> commit 416d6bbbb8835202a802fcf1600df77b7264dc13
>> Author: Khem Raj <raj.khem at gmail.com>
>> Date:   Mon Jul 20 13:41:28 2009 -0700
>>
>>     It fixes the following warnings seen on mips build.
>>
>>     libc/sysdeps/linux/mips/clone.S: Assembler messages:
>>     libc/sysdeps/linux/mips/clone.S:122: Warning: No .frame pseudo-op used
>> in PIC code
>>     libc/sysdeps/linux/mips/clone.S:75: Warning: Pretending global symbol
>> used as branch target is local.
>>
>>     libc/sysdeps/linux/mips/crt1.S: Assembler messages:
>>     libc/sysdeps/linux/mips/crt1.S:134: Warning: No .cprestore pseudo-op
>> used in PIC code
>>
>>     Signed-off-by: Khem Raj <raj.khem at gmail.com>
>>     Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop at gmail.com>
>>
>>
>> The result builds just fine, but panics failing to run init.  (QEMU can run
>> a statically linked hello world, the problem might be dynamic linker.  Not
>> sure yet.)
>>
>> Rob
>
> I should cc: the patch's author on this.
>
> Dude: you broke mips for me.  It compiles fine, but won't run dynamically
> linked executables.  Can you reproduce this problem, or do you need my build
> environment?
>
> (I thought about trying to track down which hunk it was, but the changes are
> all tangled up.  And to confirm, reverting the whole patch fixes the issue I'm
> seeing.)

Could you apply hunks related to  crt1.S from
a45777479c8c2950aed0b0fe1ced74b6c4d45057
and see if that helps.

>
> Thanks,
>
> Rob
> --
> Latency is more important than throughput. It's that simple. - Linus Torvalds
>


More information about the uClibc mailing list