This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Threads vs hpux10
- From: Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com>
- To: gcc-patches at gcc dot gnu dot org
- Cc: law at redhat dot com
- Date: Mon, 3 Dec 2001 14:41:58 -0600 (CST)
- Subject: Re: Threads vs hpux10
- References: <1223.1007166471@porcupine.cygnus.com>
- Reply-to: rittle at labs dot mot dot com
In article <1223.1007166471@porcupine.cygnus.com>, law@redhat.com writes:
>> This e-mail is just to advise you that I think you solved the same
>> problem AIX had (no weak symbols and thus always needing to link
>> against the threading library) with a different solution. The
>> solution they use with multilibs was to have `gcc -v' report single as
>> the threading model in one multilib configuration and the real thread
>> library in the other.
>> I don't care if they are different (although I personally like the AIX
>> solution better compared to yours for HPUX). I could send you the
>> exact one-line patch they used, but it would not be till Monday.
> Yea, please do send it. I'm more than open to better solutions :-)
Hi Jeff,
Here was the patch that I thought was related (the commentary about a
lack of weak symbols on AIX came from my memory of another thread):
http://gcc.gnu.org/ml/gcc-patches/2001-05/msg00708.html
However, perhaps I was confused since I now see that
gcc/config/pa/pa-hpux10.h already has the AIX "solution" (installed a
few days later, approved by you around May 15, 2001):
S rittle@latour; find . -type f -exec grep THREAD_MODEL_SPEC /dev/null '{}' \;
./pa/pa-hpux10.h:#undef THREAD_MODEL_SPEC
./pa/pa-hpux10.h:#define THREAD_MODEL_SPEC "%{!threads:single}%{threads:dce}"
./rs6000/aix51.h:#define THREAD_MODEL_SPEC "%{pthread:posix}%{!pthread:single}"
./rs6000/aix43.h:#define THREAD_MODEL_SPEC "%{pthread:posix}%{!pthread:single}"
Perhaps you could explain why gthr-dce.h is being used at all to build
the multilib for single-threaded use on your platform. I thought that
the whole point of the THREAD_MODEL_SPEC setting as shown above was to
get the right gthr-X.h file in light of multilibing. There might be a
more general problem that we need to fix (I sure hope not).
Regards,
Loren