Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-11-03 Thread Karl Rupp
Hi, another status update on the 1.6.0 release: Now the device database is almost completely populated with the devices we have direct access to. We will need another run of the nightly tests to make sure some the last-minute changes are in good order. If everything runs well we can release tom

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-30 Thread Karl Rupp
Hi all, quick update to the timeline for the 1.6.0 release: Things are progressing well, yet not fast enough in order to have a release tomorrow. I'm now aiming for a release on Monday. Best regards, Karli On 10/21/2014 09:29 AM, Karl Rupp wrote: > Hi everybody, > > we are finally approaching

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-29 Thread Karl Rupp
Hi Matt, okay, you're good to go starting tonight (EST) ;-) Thanks and best regards, Karli On 10/24/2014 09:40 PM, Matthew Musto wrote: > Will do, just let me know when you want me to start the work. > > Thanks, > -Matt > > On Fri, Oct 24, 2014 at 2:08 PM, Karl Rupp

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-24 Thread Matthew Musto
Will do, just let me know when you want me to start the work. Thanks, -Matt On Fri, Oct 24, 2014 at 2:08 PM, Karl Rupp wrote: > Hi Matt, > > > Seeing as how you are approaching a new release. I quickly ran the > >> latest development branch through my test suite on MSVC 2013. I got the >> foll

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-24 Thread Karl Rupp
Hi Matt, > Seeing as how you are approaching a new release. I quickly ran the > latest development branch through my test suite on MSVC 2013. I got the > following compiler errors. Only a few modules had issues and what I > depend on seems to be quite good! I see, these are mostly from the new

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-24 Thread Matthew Musto
Karli, Seeing as how you are approaching a new release. I quickly ran the latest development branch through my test suite on MSVC 2013. I got the following compiler errors. Only a few modules had issues and what I depend on seems to be quite good! Mostly simple fixes and I can do pull requests

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-23 Thread Karl Rupp
Hi, >> Announcements usually stick in the various queues for a couple of days, >> so as long as ViennaCL and PyViennaCL get released within, say, a week, >> we can still announce both of them together in the various channels. >> Actually, I suggest to release ViennaCL 1.6.0 first (end of next week

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-23 Thread Toby St Clere Smithe
Hej, Karl Rupp writes: >> What do you mean exactly? Releasing on the same day? That's probably >> not necessary, as it increases the chances of some unsync'd changes >> at the last minute. Releasing pyViennaCL a few days after ViennaCL >> is fine with me, but I'd say that this pri

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-23 Thread Karl Rupp
Hey, > I think we already talked about it once. > sparse-opencl fails on my laptop's GT540m, and apparently here too: > https://github.com/viennacl/viennacl-dev/issues/105 Ok, strangely there is no regular failures in the nightly tests. Either way, 'partially working' isn't acceptable, I'll tra

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-23 Thread Philippe Tillet
Hey, I think we already talked about it once. sparse-opencl fails on my laptop's GT540m, and apparently here too: https://github.com/viennacl/viennacl-dev/issues/105 On my laptop: Testing transposed upper triangular solve: compressed_matrix Testing transposed unit lo

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-23 Thread Karl Rupp
Hi, > The GEMM bug is fixed, but a bug with the sparse test remains on some > hardware. I have no clue about viennacl's sparse kernels, so I can't > help much. which hardware? Which operation is failing? > Also, how about synchronizing the next viennacl and pyviennacl release > exactly? What

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-22 Thread Philippe Tillet
Hey, The GEMM bug is fixed, but a bug with the sparse test remains on some hardware. I have no clue about viennacl's sparse kernels, so I can't help much. Also, how about synchronizing the next viennacl and pyviennacl release exactly? 2014-10-21 4:44 GMT-04:00 Toby St Clere Smithe : > Hi Karli,

Re: [ViennaCL-devel] Release 1.6.0 approaching

2014-10-21 Thread Toby St Clere Smithe
Hi Karli, Karl Rupp writes: > If no unexpected delays occur, we should have a release by the end of > next week (~Friday, Oct 31). Once ViennaCL 1.6.0 is out, I'd like to > polish up and release the Benchmark GUI shortly after. > > @Toby: What are your plans regarding the next PyViennaCL releas