Re: [mpich2-core] Fwd: [mpich2-maint] #509: Cannot build mpich2-1.0.8p1 (nemesis) with PGI 8.0-4 on Linux x86_64
The simplest thing to do is add a line "+pgi" to your $HOME/.soft. and then logout/login. That has the advantage that everything got setup correctly including MANPATH (It took few emails with rick to get this corrected) If you want to do it yourself, add these env variables: PGI=/soft/com/packages/pgi-7.1-6 LD_LIBRARY_PATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/lib MANPATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/man:: PATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/bin A.Chan ----- "Dave Goodell" <[email protected]> wrote:
Anthony, do you know how to set the right license environment variables for our PGI installation?
-Dave
octagon% cat foo.c
int main(int argc, char **argv) { return 0; }
octagon% pgcc foo.c Your PGI license doesn't appear to be working properly.
1. If you installed a trial version, please generate and then install your trial keys at:
http://www.pgroup.com/license/trial
To generate your trial keys, you will need your host ID.
Your hostid is 00304857ff48
If you installed a trial version and your keys have expired, you can request a trial extension by contacting [email protected]
2. If you purchased a license, but have not yet installed your permanent license keys, please generate and then install your permanent keys at:
To log in you'll need your PGI PIN and password; these were e- mailed to the person who purchased this software. To generate your permanent keys, you'll also need your host ID.
Your hostid is 00304857ff48
Please contact [email protected] for purchase related inquiries.
3. If you have installed your license keys and you are still receiving this message, further diagnosis is required. For help on license debugging, please check these resources:
a. Review the PGI installation notes at:
b. Visit the PGI installation FAQ page at:
http://www.pgroup.com/support/install.htm
c. Send mail with a description to PGI licensing support at:
mailto: [email protected]
On Apr 1, 2009, at 11:34 AM, Anthony Chan wrote:
Hi Dave,
Have you tried the PGI-7.1.6 installed in the MCS machines (I checked octagon and octopus) ?
/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/bin/pgcc
A.Chan
----- "Dave Goodell" <[email protected]> wrote:
Does anyone have a modern version of the PGI compilers somewhere?
Or
do we know anyone at PGI we could ask for a copy in order to fix compatibility issues like this?
-Dave
Begin forwarded message:
From: "mpich2" <[email protected]> Date: April 1, 2009 Apr 1 10:07:00 AM CDT To: undisclosed-recipients: ; Subject: [mpich2-maint] #509: Cannot build mpich2-1.0.8p1
(nemesis)
with PGI 8.0-4 on Linux x86_64 Reply-To: [email protected]
--------------------- +------------------------------------------------------ Reporter: goodell | Type: bug Status: new | Priority: major Milestone: | Component: mpich2 --------------------- +------------------------------------------------------
{{{
Begin forwarded message:
From: Dave Goodell <[email protected]> Date: April 1, 2009 Apr 1 10:06:09 AM CDT To: [email protected] Subject: Re: [mpich-discuss] Cannot build mpich2-1.0.8p1
(nemesis)
with PGI 8.0-4 on Linux x86_64 Reply-To: [email protected]
On Mar 31, 2009, at 4:21 PM, Gus Correa wrote:
I decompressed the tarball on a fresh new subdirectory, configured again, ran make again. However, I still get the same exact error.
Moreover, the preprocessor macro HAVE_GCC_AND_X86_64_ASM continues not to be defined in the src/include/mpichconf.h file, and most likely continues to be the source for the error.
By contrast, configure.in is identical to my patched version (with the pgcc_atomics.patch file).
I will send you the output of configure, config.log and src/include/mpichconf.h off list.
And here is the "diff" between the configure script from your latest tarball and the configure script from the 1.2.8p1 after I patched it with pgcc_atomics.patch:
[build_pgi-8.0-4]% diff ../configure /home/swinst/mpich2/1.0.8p1/ mpich2-1.0.8p1/configure29449,29451c29449
Did I do anything wrong?
No, I don't think that you did anything wrong. There are just other bugs wrt to PGI in our configure logic that I haven't spotted yet.
This would all go a lot easier if I had a modern PGI compiler available for testing. In the mean time, I would recommend just editing the generated mpichconf.h after the configure step and before the make step to define HAVE_GCC_AND_X86_64_ASM to 1. That ought to fix your immediate problem, although there may be others that will appear once you are over this hurdle.
I'm not sure that there's much more I can do for you in the short- term until I get my hands on a PGI compiler installation. In the medium-term I'll see if I can get this resolved for the 1.1.0 stable release (likely some time in May).
Sorry, -Dave
}}}
-- Ticket URL: <https://trac.mcs.anl.gov/projects/mpich2/ticket/509>
I have added a test with PGI compilers to the old nightly tests. We had stopped testing them a while back because builds would often fail because the compiler couldn't get license from the server. Let's see if it works now. Rajeev
-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Anthony Chan Sent: Wednesday, April 01, 2009 1:12 PM To: [email protected] Subject: Re: [mpich2-core] Fwd: [mpich2-maint] #509: Cannot build mpich2-1.0.8p1 (nemesis) with PGI 8.0-4 on Linux x86_64
The simplest thing to do is add a line "+pgi" to your $HOME/.soft. and then logout/login. That has the advantage that everything got setup correctly including MANPATH (It took few emails with rick to get this corrected)
If you want to do it yourself, add these env variables:
PGI=/soft/com/packages/pgi-7.1-6 LD_LIBRARY_PATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/lib MANPATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/man:: PATH=....:/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/bin
A.Chan
----- "Dave Goodell" <[email protected]> wrote:
Anthony, do you know how to set the right license environment variables for our PGI installation?
-Dave
octagon% cat foo.c
int main(int argc, char **argv) { return 0; }
octagon% pgcc foo.c Your PGI license doesn't appear to be working properly.
1. If you installed a trial version, please generate and then install your trial keys at:
http://www.pgroup.com/license/trial
To generate your trial keys, you will need your host ID.
Your hostid is 00304857ff48
If you installed a trial version and your keys have expired, you can request a trial extension by contacting [email protected]
2. If you purchased a license, but have not yet installed your permanent license keys, please generate and then install your permanent keys at:
To log in you'll need your PGI PIN and password; these were e- mailed to the person who purchased this software. To generate your permanent keys, you'll also need your host ID.
Your hostid is 00304857ff48
Please contact [email protected] for purchase related inquiries.
3. If you have installed your license keys and you are still receiving this message, further diagnosis is required. For help on license debugging, please check these resources:
a. Review the PGI installation notes at:
b. Visit the PGI installation FAQ page at:
http://www.pgroup.com/support/install.htm
c. Send mail with a description to PGI licensing support at:
mailto: [email protected]
On Apr 1, 2009, at 11:34 AM, Anthony Chan wrote:
Hi Dave,
Have you tried the PGI-7.1.6 installed in the MCS machines (I checked octagon and octopus) ?
/soft/com/packages/pgi-7.1-6/linux86-64/7.1-6/bin/pgcc
A.Chan
----- "Dave Goodell" <[email protected]> wrote:
Does anyone have a modern version of the PGI compilers
somewhere? Or
do we know anyone at PGI we could ask for a copy in order to fix compatibility issues like this?
-Dave
Begin forwarded message:
From: "mpich2" <[email protected]> Date: April 1, 2009 Apr 1 10:07:00 AM CDT To: undisclosed-recipients: ; Subject: [mpich2-maint] #509: Cannot build mpich2-1.0.8p1
(nemesis)
with PGI 8.0-4 on Linux x86_64 Reply-To: [email protected]
--------------------- +------------------------------------------------------ Reporter: goodell | Type: bug Status: new | Priority: major Milestone: | Component: mpich2 --------------------- +------------------------------------------------------
{{{
Begin forwarded message:
From: Dave Goodell <[email protected]> Date: April 1, 2009 Apr 1 10:06:09 AM CDT To: [email protected] Subject: Re: [mpich-discuss] Cannot build mpich2-1.0.8p1
(nemesis)
with PGI 8.0-4 on Linux x86_64 Reply-To: [email protected]
On Mar 31, 2009, at 4:21 PM, Gus Correa wrote:
I decompressed the tarball on a fresh new subdirectory, configured again, ran make again. However, I still get the same exact error.
Moreover, the preprocessor macro HAVE_GCC_AND_X86_64_ASM continues not to be defined in the src/include/mpichconf.h file, and most likely continues to be the source for the error.
By contrast, configure.in is identical to my patched version (with the pgcc_atomics.patch file).
I will send you the output of configure, config.log and src/include/mpichconf.h off list.
And here is the "diff" between the configure script from your latest tarball and the configure script from the 1.2.8p1 after I patched it with pgcc_atomics.patch:
[build_pgi-8.0-4]% diff ../configure /home/swinst/mpich2/1.0.8p1/ mpich2-1.0.8p1/configure29449,29451c29449
Did I do anything wrong?
No, I don't think that you did anything wrong. There are just other bugs wrt to PGI in our configure logic that I haven't spotted yet.
This would all go a lot easier if I had a modern PGI compiler available for testing. In the mean time, I would recommend just editing the generated mpichconf.h after the configure step and before the make step to define HAVE_GCC_AND_X86_64_ASM to 1. That ought to fix your immediate problem, although there may be others that will appear once you are over this hurdle.
I'm not sure that there's much more I can do for you in the short- term until I get my hands on a PGI compiler installation. In the medium-term I'll see if I can get this resolved for the 1.1.0 stable release (likely some time in May).
Sorry, -Dave
}}}
-- Ticket URL: <https://trac.mcs.anl.gov/projects/mpich2/ticket/509>
participants (2)
-
Anthony Chan -
Rajeev Thakur