search for: apus

...ristos, We would be very interested in learning more about this. In my group, we (Prakalp Srivastava, Maria Kotsifakou and I) have been working on LLVM extensions to make it easier to target a wide range of accelerators in a heterogeneous mobile device, such as Qualcomm's Snapdragon and other APUs. Our approach has been to (a) add better abstractions of parallelism to the LLVM instruction set that can be mapped down to a wide range of parallel hardware accelerators; and (b) to develop optimizing "back-end" translators to generate efficient code for the accelerators from the extend...

...ning more about this. >> >> In my group, we (Prakalp Srivastava, Maria Kotsifakou and I) have been >> working on LLVM extensions to make it easier to target a wide range of >> accelerators in a heterogeneous mobile device, such as Qualcomm's Snapdragon >> and other APUs. Our approach has been to (a) add better abstractions of >> parallelism to the LLVM instruction set that can be mapped down to a wide >> range of parallel hardware accelerators; and (b) to develop optimizing >> "back-end" translators to generate efficient code for the a...

Regarding the topic of the thread I think they're not that different from each other nowadays. I've been using an ATi graphics card for over a year and it's running pretty well. NVIDIA graphics cards are better. But AMD graphics cards nowadays aren't as bad as they sound. They're nearly equal.

...d is > willing to do some testing, then please contact me. It shouldn't be > difficult. You will likely have to buy a working DP-to-HDMI adapter. > See https://hverkuil.home.xs4all.nl/cec-status.txt for a (sadly very > short) list of working adapters. Actually you added support for APUs as well. In amdgpu, there are two sets of modesetting code, an older less featured version (amd/amdgpu/dce*.c) and the newer more featured code (amd/display/*). Newer asics (vega and raven) are only supported by DC. Older asics are supported by both. Eventually we'd like to remove the older...

...n my group, we (Prakalp Srivastava, Maria Kotsifakou and I) have been >>>> working on LLVM extensions to make it easier to target a wide range of >>>> accelerators in a heterogeneous mobile device, such as Qualcomm's >>>> Snapdragon >>>> and other APUs. Our approach has been to (a) add better abstractions of >>>> parallelism to the LLVM instruction set that can be mapped down to a >>>> wide >>>> range of parallel hardware accelerators; and (b) to develop optimizing >>>> "back-end" transla...

On Mon, Feb 19, 2018 at 03:05:53PM +0100, Daniel Vetter wrote: > On Mon, Feb 19, 2018 at 12:58:17PM +0100, Lukas Wunner wrote: > > On Mon, Feb 19, 2018 at 12:34:43PM +0100, Daniel Vetter wrote: > > > Well, userspace expects hotplug events, even when we runtime suspend > > > stuff. Hence waking shit up with polling. Imo ignoring hotplug events is a > > > pretty

...) > have been > >>>> working on LLVM extensions to make it easier to target a wide > range of > >>>> accelerators in a heterogeneous mobile device, such as Qualcomm's > >>>> Snapdragon > >>>> and other APUs. Our approach has been to (a) add better > abstractions of > >>>> parallelism to the LLVM instruction set that can be mapped > down to a > >>>> wide > >>>> range of parallel hardware accelerators; and (b) to develop >...

On Tue, Jul 07, 2015 at 11:29:38AM -0400, Ilia Mirkin wrote: > On Mon, Jul 6, 2015 at 8:42 PM, Andrew Chew <achew at nvidia.com> wrote: > > Hello, > > > > I am currently looking into ways to support fixed virtual address allocations > > and sparse mappings in nouveau, as a step towards supporting CUDA. > > > > CUDA requires that the GPU virtual address

From: Hans Verkuil <hans.verkuil at cisco.com> Now that the DisplayPort CEC-Tunneling-over-AUX drm+i915 support has been merged in the mainline kernel it is time to roll this out to nouveau and amdgpu as well. I combined both in the same patch series since both depend on the same first patch, the comments in this cover letter apply to both and the implementation is also very similar (and

Hi, I did a master-failed test with 3 PCs.All of the PCs are installed with XCP V1. What I am going to do is that how to shutdown a slave when master of pool failed. I set up a pool included the PCs,and I hard poweroff the master.I typed "poweroff" in CLI of slaves.But the slaves didn''t shutdown.They sticked in "Storing network configuration". I just want the slave