Re: [hwloc-users] lstopo and GPus

2012-08-29 Thread Samuel Thibault
Gabriele Fatigati, le Wed 29 Aug 2012 16:56:30 +0200, a écrit : > The id showed in the GPU box from lstopo, is the same device_id CUDA > numeration > used in some function like setDevice() for example? Exactly. Samuel

Re: [hwloc-users] lstopo and GPus

2012-08-29 Thread Gabriele Fatigati
Just another things: The id showed in the GPU box from lstopo, is the same device_id CUDA numeration used in some function like setDevice() for example? More better: gpu 1 from lstopo = ? gpu 1 for CUDA runtime? Thanks. 2012/8/29 Gabriele Fatigati > Good! > > Now it works well. > > Many ta

Re: [hwloc-users] lstopo and GPus

2012-08-29 Thread Gabriele Fatigati
Good! Now it works well. Many tanks! 2012/8/28 Samuel Thibault > Gabriele Fatigati, le Tue 28 Aug 2012 18:10:41 +0200, a écrit : > > How can cuda branch help me? lstopo output of that branch is the same of > the > > trunk. > > You need to make sure that hwloc found the cuda runtime during > ./

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Samuel Thibault
Gabriele Fatigati, le Tue 28 Aug 2012 18:10:41 +0200, a écrit : > How can cuda branch help me? lstopo output of that branch is the same of the > trunk. You need to make sure that hwloc found the cuda runtime during ./configure. You will then get new "cuda" devices. > Another question: the GPU IDs

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Gabriele Fatigati
Hi, thanks for the reply. How can cuda branch help me? lstopo output of that branch is the same of the trunk. Another question: the GPU IDs are the same (10de: 06d2). How is it possible? Thanks. 2012/8/28 Samuel Thibault > Brice Goglin, le Tue 28 Aug 2012 14:43:53 +0200, a écrit : > > > $ ls

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Samuel Thibault
Brice Goglin, le Tue 28 Aug 2012 14:43:53 +0200, a écrit : > > $ lstopo > > Socket #0 > > Socket #1 > > PCI... > > (connected to socket #1) > > > > vs > > > > $ lstopo > > Socket #0 > > Socket #1 > > PCI... > > (connected to both sockets) > > Fortunately, this won't occur in most cas

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Brice Goglin
Le 28/08/2012 14:23, Samuel Thibault a écrit : > Gabriele Fatigati, le Tue 28 Aug 2012 14:19:44 +0200, a écrit : >> I'm using hwloc 1.5. I would to see how GPUs are connected with the processor >> socket using lstopo command. > About connexion with the socket, there is indeed no real graphical > d

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Samuel Thibault
Gabriele Fatigati, le Tue 28 Aug 2012 14:19:44 +0200, a écrit : > I'm using hwloc 1.5. I would to see how GPUs are connected with the processor > socket using lstopo command.  About connexion with the socket, there is indeed no real graphical difference between "connected to socket #1" and "connec

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Samuel Thibault
Gabriele Fatigati, le Tue 28 Aug 2012 14:19:44 +0200, a écrit : > I attach the figure. The system has two GPUs, but I don't understand how to > find that information from PCI boxes. The 10de: devices are the GPUs (or alternatively, use the cuda branch from the svn repository). The figure show

Re: [hwloc-users] lstopo and GPus

2012-08-28 Thread Brice Goglin
Hello, For now, you have to look at PCI ids. NVIDIA GPUs have "10de:" as vendor/device ids, that's what is shown in your boxes on the right. We should have better GPU support in the future. Right now, we only use what Linux knows, and it knows pretty much nothing about NVIDIA GPUs because of th

[hwloc-users] lstopo and GPus

2012-08-28 Thread Gabriele Fatigati
Dear hwloc user, I'm using hwloc 1.5. I would to see how GPUs are connected with the processor socket using lstopo command. I attach the figure. The system has two GPUs, but I don't understand how to find that information from PCI boxes. Thanks in advance. -- Ing. Gabriele Fatigati HPC spec