All, The MPICH2 trunk is now frozen for any more commits, except for Jayesh who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow. Thanks, -- Pavan PS: I have not made the trunk RO since Jayesh might need to commit some stuff in. -- Pavan Balaji http://www.mcs.anl.gov/~balaji
The mpich2-1.2.1rc1 tarball is available here: http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downloa... I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine. -- Pavan On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits, except for Jayesh who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
The nightly tests didn't bring up any new surprises. I'm trying out the default build of the tarball now. Can others try out other configurations too? Thanks, -- Pavan On 11/09/2009 08:18 PM, Pavan Balaji wrote:
The mpich2-1.2.1rc1 tarball is available here: http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downloa...
I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine.
-- Pavan
On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits, except for Jayesh who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
The nightly tests look fine. It is good enough for an rc at least. Whenever you are satisfied with it, I think you can release it. Rajeev
-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Pavan Balaji Sent: Tuesday, November 10, 2009 11:03 AM To: [email protected] Subject: Re: [mpich2-core] Code freeze for mpich2-1.2.1rc1
The nightly tests didn't bring up any new surprises. I'm trying out the default build of the tarball now.
Can others try out other configurations too?
Thanks,
-- Pavan
On 11/09/2009 08:18 PM, Pavan Balaji wrote:
The mpich2-1.2.1rc1 tarball is available here:
http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/ mpich2/downloads/tarballs/1.2.1rc1/mpich2-1.2.1rc1.tar.gz
I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine.
-- Pavan
On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits,
except for Jayesh
who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
Both the default and the build with gcc/ifort work fine. ----- "Pavan Balaji" <[email protected]> wrote:
The nightly tests didn't bring up any new surprises. I'm trying out the default build of the tarball now.
Can others try out other configurations too?
Thanks,
-- Pavan
On 11/09/2009 08:18 PM, Pavan Balaji wrote:
The mpich2-1.2.1rc1 tarball is available here:
http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downloa...
I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine.
-- Pavan
On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits, except for
Jayesh
who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
smpd+sock looks fine... -jayesh ----- Original Message ----- From: "Anthony Chan" <[email protected]> To: [email protected] Sent: Tuesday, November 10, 2009 11:23:00 AM GMT -06:00 US/Canada Central Subject: Re: [mpich2-core] Code freeze for mpich2-1.2.1rc1 Both the default and the build with gcc/ifort work fine. ----- "Pavan Balaji" <[email protected]> wrote:
The nightly tests didn't bring up any new surprises. I'm trying out the default build of the tarball now.
Can others try out other configurations too?
Thanks,
-- Pavan
On 11/09/2009 08:18 PM, Pavan Balaji wrote:
The mpich2-1.2.1rc1 tarball is available here:
http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downloa...
I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine.
-- Pavan
On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits, except for
Jayesh
who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
Thanks. Looks good on my side too. Sending out the announcement. -- Pavan On 11/10/2009 11:25 AM, Jayesh Krishna wrote:
smpd+sock looks fine...
-jayesh ----- Original Message ----- From: "Anthony Chan" <[email protected]> To: [email protected] Sent: Tuesday, November 10, 2009 11:23:00 AM GMT -06:00 US/Canada Central Subject: Re: [mpich2-core] Code freeze for mpich2-1.2.1rc1
Both the default and the build with gcc/ifort work fine.
----- "Pavan Balaji" <[email protected]> wrote:
The nightly tests didn't bring up any new surprises. I'm trying out the default build of the tarball now.
Can others try out other configurations too?
Thanks,
-- Pavan
On 11/09/2009 08:18 PM, Pavan Balaji wrote:
The mpich2-1.2.1rc1 tarball is available here:
http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downloa...
I'm planning to wait till tomorrow morning to make sure the nightly tests go through fine.
-- Pavan
On 11/09/2009 01:05 PM, Pavan Balaji wrote:
All,
The MPICH2 trunk is now frozen for any more commits, except for Jayesh who is testing out the newly committed stuff on Windows. We will be running a final set of tests before releasing 1.2.1rc1 either today or tomorrow.
Thanks,
-- Pavan
PS: I have not made the trunk RO since Jayesh might need to commit some stuff in.
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
The announcement has gone out. Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch? We might have another bug fix release for the 1.2 series, depending on when 1.3 GA goes out, but most of our pending changes are for 1.3 anyway. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji
Btw, the trunk is open for commits again. -- Pavan On 11/10/2009 11:56 AM, Pavan Balaji wrote:
The announcement has gone out. Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch? We might have another bug fix release for the 1.2 series, depending on when 1.3 GA goes out, but most of our pending changes are for 1.3 anyway.
-- Pavan
-- Pavan Balaji http://www.mcs.anl.gov/~balaji
On Nov 10, 2009, at 11:56 AM, Pavan Balaji wrote:
Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch?
Either way is fine by me, both approaches have their pros and cons. I'd probably lean in favor of treating trunk as 1.3 and creating the 1.2.x branch lazily if/when we decide to have a 1.2.2 release. Release wrangling and creation has non-negligible overhead, and we should avoid unnecessary releases. -Dave
When are we planning the 1.2.1 release ? If that is going to take time, I vote for creating a branch so that we can make changes (intrusive) to the trunk. -Jayesh ----- Original Message ----- From: "Dave Goodell" <[email protected]> To: [email protected] Sent: Tuesday, November 10, 2009 12:09:10 PM GMT -06:00 US/Canada Central Subject: Re: [mpich2-core] Code freeze for mpich2-1.2.1rc1 On Nov 10, 2009, at 11:56 AM, Pavan Balaji wrote:
Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch?
Either way is fine by me, both approaches have their pros and cons. I'd probably lean in favor of treating trunk as 1.3 and creating the 1.2.x branch lazily if/when we decide to have a 1.2.2 release. Release wrangling and creation has non-negligible overhead, and we should avoid unnecessary releases. -Dave
On 11/10/2009 01:01 PM, Jayesh Krishna wrote:
When are we planning the 1.2.1 release ? If that is going to take time, I vote for creating a branch so that we can make changes (intrusive) to the trunk.
There's already a branch for 1.2.1. We are referring to 1.2.x. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji
On 11/10/2009 12:09 PM, Dave Goodell wrote:
On Nov 10, 2009, at 11:56 AM, Pavan Balaji wrote:
Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch?
Either way is fine by me, both approaches have their pros and cons. I'd probably lean in favor of treating trunk as 1.3 and creating the 1.2.x branch lazily if/when we decide to have a 1.2.2 release. Release wrangling and creation has non-negligible overhead, and we should avoid unnecessary releases.
That was my preference too, but then we will need to avoid committing any 1.3 specific stuff. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji
On Nov 10, 2009, at 1:14 PM, Pavan Balaji wrote:
On 11/10/2009 12:09 PM, Dave Goodell wrote:
On Nov 10, 2009, at 11:56 AM, Pavan Balaji wrote:
Should we create a branch for the 1.2.x series and make trunk the active 1.3 branch?
Either way is fine by me, both approaches have their pros and cons. I'd probably lean in favor of treating trunk as 1.3 and creating the 1.2.x branch lazily if/when we decide to have a 1.2.2 release. Release wrangling and creation has non-negligible overhead, and we should avoid unnecessary releases.
That was my preference too, but then we will need to avoid committing any 1.3 specific stuff.
Not if we cut the 1.2.x branch from the 1.2.1 branch (or at least use the same root commit). It's just a question of do we cut the branch and merge bug fix commits to it from the trunk as we make them (this requires discipline), or do we cut the 1.2.x branch later on and sift backwards through the history for bug fix commits to cherry-pick (this potentially requires lots of sifting, good commit messages, and a good memory). Neither one is perfect, but neither choice requires us to hold off on 1.3-specific changes. My preference is to be optimistic and hope that there isn't a need for a 1.2.2 release at all, which also requires less work up front. -Dave
On 11/10/2009 01:25 PM, Dave Goodell wrote:
Not if we cut the 1.2.x branch from the 1.2.1 branch (or at least use the same root commit). It's just a question of do we cut the branch and merge bug fix commits to it from the trunk as we make them (this requires discipline), or do we cut the 1.2.x branch later on and sift backwards through the history for bug fix commits to cherry-pick (this potentially requires lots of sifting, good commit messages, and a good memory). Neither one is perfect, but neither choice requires us to hold off on 1.3-specific changes.
My preference is to be optimistic and hope that there isn't a need for a 1.2.2 release at all, which also requires less work up front.
Ok, I can buy that. If this is OK with everyone, we can stick to it. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji
participants (5)
-
Anthony Chan -
Dave Goodell -
Jayesh Krishna -
Pavan Balaji -
Rajeev Thakur