Thorsten,
Is there another thread supposed to join in and break the deadlock? Lock
initialized wrong?
I guess yes at the first one, no at the second one, I didn’t see much
m68k-specific code there, and things appear to work on the other arches
(plus, Qt4 has similar issues). As I said earl
On 03.08.2012 16:33, Damien Raude-Morvan wrote:
> Le 03/08/2012 14:36, Thorsten Glaser a écrit :
>> Hi,
>
> Hi Thorsten,
>
>> after doko recommended it to me *way* back then, but only now I’ve
>> got all of its Build-Depends installable, I’m currently trying to
>> build src:openjdk-6 (with all re
schmitz dixit:
> Is there another thread supposed to join in and break the deadlock? Lock
> initialized wrong?
I guess yes at the first one, no at the second one, I didn’t see much
m68k-specific code there, and things appear to work on the other arches
(plus, Qt4 has similar issues). As I said ea
Thorsten,
I guess we should let it use __sync_val_compare_and_swap instead then?
That one’s interesting: it makes an idle-spin instead of busy-spin:
Is there something wrong with threads (whatever façette of it)?
Is there another thread supposed to join in and break the
Dixi quod…
>I guess we should let it use __sync_val_compare_and_swap instead then?
That one’s interesting: it makes an idle-spin instead of busy-spin:
top - 07:26:10 up 11 days, 17:52, 0 users, load average: 0.01, 0.03, 0.05
Tasks: 63 total, 1 running, 62 sleeping, 0 stopped, 0 zombie
Andreas Schwab dixit:
>Thorsten Glaser writes:
>
>>> relocation truncated to fit: R_68K_GOT16O against `.LC1'
>
>Compile with -fPIC.
Thanks, that helped – a bit. I mean, I knew Java™ was slow,
but not *that* slow:
17634 pts/1Sl+ 4662:03
/tmp/buildd/openjdk-6-6b24-1.11.3/build/bootstrap/jd
Thorsten Glaser writes:
>> relocation truncated to fit: R_68K_GOT16O against `.LC1'
Compile with -fPIC.
Andreas.
--
Andreas Schwab, sch...@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
--
To UNSUBSCRIBE
Damien Raude-Morvan dixit:
> This "Too many levels of symbolic links" is not root cause of failure. In
> fact,
> openjdk $(MAKE) is called twice in case of failure [1] (I don't remember exact
> justification for this...).
Ouch, okay! A *lot* of thanks for spotting that!
> In your log file I can
Niels Thykier dixit:
>I am not how or why, but I think your issue is:
>
> [ -f libjvm.so ] || { ln -s libjvm.so libjvm.so;
> ^^^
> ln -s libjvm.so.1 libjvm.so.1; }; \
> ^^^
Le 03/08/2012 14:36, Thorsten Glaser a écrit :
Hi,
Hi Thorsten,
after doko recommended it to me *way* back then, but only now I’ve
got all of its Build-Depends installable, I’m currently trying to
build src:openjdk-6 (with all references to gcj-4.7 changed back
to gcj-4.6, and, for consistenc
On 2012-08-03 14:36, Thorsten Glaser wrote:
> Hi,
>
> after doko recommended it to me *way* back then, but only now I’ve
> got all of its Build-Depends installable, I’m currently trying to
> build src:openjdk-6 (with all references to gcj-4.7 changed back
> to gcj-4.6, and, for consistency, the sa
11 matches
Mail list logo
