Displaying 13 results from an estimated 13 matches for "nvlink".
Did you mean:
nlink
2017 Aug 16
3
CUDA separate compilation
Clang currently doesn't support CUDA separate compilation and thus extern
__device__ functions and variables cannot be used.
Could someone give me any pointers where to look or what has to be done to
support this? If at all possible, I'd like to see what's missing and
possibly try to tackle it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
2016 Jan 20
4
Executing OpenMP 4.0 code on Nvidia's GPU
...i Arpith,
That is exactly what it is :).
My bad, I thought I copied over the libraries to where LIBRARY_PATH
pointing but apparently it was copied to a wrong destination.
Thanks a lot.
On Wed, Jan 20, 2016 at 4:51 AM, Arpith C Jacob <acjacob at us.ibm.com> wrote:
> Hi Ahmed,
>
> nvlink is unable to find the GPU OMP runtime library in its path. Does
> LIBRARY_PATH point to the right location? You could try passing the "-v"
> option to clang to get more information.
>
> Regards,
> Arpith
>
-------------- next part --------------
An HTML attachment was sc...
2018 May 16
0
kernel-lt from ELRepo vs. GRUB: define default boot kernel
...dia,videodev
% dmesg | grep nvidia
[ 10.515333] nvidia: loading out-of-tree module taints kernel.
[ 10.515339] nvidia: module license 'NVIDIA' taints kernel.
[ 10.572111] nvidia: module verification failed: signature and/or
required key missing - tainting kernel
[ 10.630900] nvidia-nvlink: Nvlink Core is being initialized, major
device number 239
[ 10.665785] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver
for UNIX platforms 390.30 Wed Jan 31 21:32:48 PST 2018
[ 10.672168] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 10.672171] [drm] Initialized nvidia-d...
2020 Mar 17
1
[PATCH 2/2] mm: remove device private page support from hmm_range_fault
...rds" is you set the flag on input and never get it on
output, clear the flag in input and maybe get it on output.
Compare with valid or write which don't work that way.
> Also, a device like Nvidia's GPUs may have an alternate path for copying
> one GPU's memory to another (nvlink) without going through system memory
> so getting a device private struct page/PFN from hmm_range_fault() that isn't
> "owned" by the faulting GPU is useful.
> I agree that the current code isn't well tested or thought out for multiple devices
> (rdma, NVMe drives, GPU...
2020 Mar 16
2
[PATCH 2/2] mm: remove device private page support from hmm_range_fault
On Mon, Mar 16, 2020 at 07:49:35PM +0100, Christoph Hellwig wrote:
> On Mon, Mar 16, 2020 at 11:42:19AM -0700, Ralph Campbell wrote:
> >
> > On 3/16/20 10:52 AM, Christoph Hellwig wrote:
> >> No driver has actually used properly wire up and support this feature.
> >> There is various code related to it in nouveau, but as far as I can tell
> >> it never
2018 May 16
4
kernel-lt from ELRepo vs. GRUB: define default boot kernel
Hi,
After upgrading my workstation to CentOS 7.5 (1804), I had to upgrade my
kernel from vanilla to kernel-lt from ELRepo. My NVidia GeForce 210
would only work with the driver provided by NVidia, which in turn
required a more recent kernel than 3.0.10. Anyway.
Right now here's all the kernels that I have on my workstation:
[root at alphamule:~] # rpm -qa | grep -i kernel
2020 Apr 03
2
acr: failed to load firmware with Kernel 5.6. Kernel 5.5 works just fine.
On Fri, Apr 3, 2020 at 1:59 PM Zeno Davatz <zdavatz at gmail.com> wrote:
>
> On Fri, Apr 3, 2020 at 7:23 PM Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> >
> > On Fri, Apr 3, 2020 at 1:21 PM Zeno Davatz <zdavatz at gmail.com> wrote:
> > >
> > > On Fri, Apr 3, 2020 at 6:59 PM Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> > >
2016 Feb 08
0
PROBLEM: mmiotracing issue with nvidia kernel module
...miotrace: ioremap_*(0xf6000000, 0x1000) =
ffffc90000044000
Feb 07 22:29:34 kernel: mmiotrace: Unmapping ffffc90000044000.
Feb 07 22:29:34 kernel: mmiotrace: ioremap_*(0xf6000000, 0x1000) =
ffffc90000046000
Feb 07 22:29:34 kernel: mmiotrace: Unmapping ffffc90000046000.
Feb 07 22:29:34 kernel: nvidia-nvlink: Nvlink Core is being initialized, major
device number 246
Feb 07 22:29:34 kernel: [drm] Initialized nvidia-drm 0.0.0 20150116 for
0000:01:00.0 on minor 1
Feb 07 22:29:34 kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module 361.18
Sat Jan 9 21:27:18 PST 2016
Feb 07 22:29:34 kernel: mmiotrace:...
2015 Jun 09
2
[LLVMdev] Supporting heterogeneous computing in llvm.
Hi Sergos and Samuel,
Thanks for the links, I've got it mostly working now.
I still have a problem with linking the code. It seems that the clang
driver doesn't pass its library search path to nvlink when linking the
generated cuda code to the target library, resulting in it not correctly
finding libtarget-nvptx.a. Is there some flag or environment variable
that I should set here? Manually providing nvlink with a -L flag
pointing to the appropriate path seems to work for the linking step....
2019 Jul 15
2
[PATCH v2 2/4] gpio: fail if gpu external power is missing
Please add a config override to skip this, since we'll invariably get
it wrong for some setup, and should be able to provide users with
workarounds while the issue is being worked out.
On Mon, Jul 15, 2019 at 5:43 AM Mark Menzynski <mmenzyns at redhat.com> wrote:
>
> Currently, nouveau doesn't check if GPU is missing power. This
> patch makes nouveau fail when this happens
2020 Mar 16
0
[PATCH 2/2] mm: remove device private page support from hmm_range_fault
...ant
the device private page to be faulted back to system memory, it wants to
get the device private struct page so it can update its page table to point
to the memory already in the device.
Also, a device like Nvidia's GPUs may have an alternate path for copying
one GPU's memory to another (nvlink) without going through system memory
so getting a device private struct page/PFN from hmm_range_fault() that isn't
"owned" by the faulting GPU is useful.
I agree that the current code isn't well tested or thought out for multiple devices
(rdma, NVMe drives, GPUs, etc.) but it also...
2017 Jun 09
1
NVPTX Back-end: relocatable device code support for dynamic parallelism
...e1.cpp" > "/tmp/tmpxft_00007040_00000000-15_cuda_id_test.ii"
#$ gcc -std=c++11 -c -x c++ "-I/opt/cuda-8.0/bin/..//include" -fpreprocessed -m64 -o "/tmp/tmpxft_00007040_00000000-16_cuda_id_test.o" "/tmp/tmpxft_00007040_00000000-15_cuda_id_test.ii"
#$ nvlink --arch=sm_35 --register-link-binaries="/tmp/tmpxft_00007040_00000000-7_id_test_dlink.reg.c" -m64 "-L/opt/cuda-8.0/bin/..//lib64/stubs" "-L/opt/cuda-8.0/bin/..//lib64" -cpu-arch=X86_64 "/tmp/tmpxft_00007040_00000000-16_cuda_id_test.o" -lcudadevrt -o "...
2015 Jun 08
2
[LLVMdev] Supporting heterogeneous computing in llvm.
Roel,
You have to checkout and build llvm/clang as usual.
For runtime support you'll have to build the libomptarget and make a
plugin for your target. Samuel can help you some more.
As for the OpenMP examples I can recommend you the
http://openmp.org/mp-documents/OpenMP4.0.0.Examples.pdf
look into the target constructs.
Sergos
On Mon, Jun 8, 2015 at 6:13 PM, Roel Jordans <r.jordans at