On Wed, Oct 16, 2024 at 08:27:51AM +0200, Thomas Hellström wrote:
> On Wed, 2024-10-16 at 03:18 +, Matthew Brost wrote:
> > On Wed, Oct 09, 2024 at 12:50:42PM +0200, Thomas Hellström wrote:
> > > Hi, Matthew.
> > >
> > > Some comments below around migrating to SRAM.
> > >
> > >
> > > On Tue,
On Wed, 2024-10-16 at 03:18 +, Matthew Brost wrote:
> On Wed, Oct 09, 2024 at 12:50:42PM +0200, Thomas Hellström wrote:
> > Hi, Matthew.
> >
> > Some comments below around migrating to SRAM.
> >
> >
> > On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > > This patch introduces suppo
On Wed, Oct 09, 2024 at 12:50:42PM +0200, Thomas Hellström wrote:
> Hi, Matthew.
>
> Some comments below around migrating to SRAM.
>
>
> On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > This patch introduces support for GPU Shared Virtual Memory (SVM) in
> > the
> > Direct Rendering M
Hi, Matthew.
Some comments below around migrating to SRAM.
On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> This patch introduces support for GPU Shared Virtual Memory (SVM) in
> the
> Direct Rendering Manager (DRM) subsystem. SVM allows for seamless
> sharing of memory between the CPU
On Tue, Sep 24, 2024 at 04:30:06PM +, Matthew Brost wrote:
> On Tue, Sep 24, 2024 at 12:42:56PM +0200, Thomas Hellström wrote:
> > Hi, Matt,
> >
> > Some random review comments on this patch I came across while looking
> > at multi-device.
> >
> > Thanks,
> > Thomas
> >
> >
> > On Tue, 2024
; dan...@ffwll.ch; airl...@gmail.com;
> christian.koe...@amd.com
> Subject: Re: [RFC PATCH 05/28] drm/gpusvm: Add support for GPU
> Shared Virtual Memory
>
> On Fri, Sep 06, 2024 at 06:41:18PM +, Zeng, Oak wrote:
> > There are fundamental design conflicts with what we have a
On Tue, Sep 24, 2024 at 12:42:56PM +0200, Thomas Hellström wrote:
> Hi, Matt,
>
> Some random review comments on this patch I came across while looking
> at multi-device.
>
> Thanks,
> Thomas
>
>
> On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > This patch introduces support for GPU
Hi, Matt,
Some random review comments on this patch I came across while looking
at multi-device.
Thanks,
Thomas
On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> This patch introduces support for GPU Shared Virtual Memory (SVM) in
> the
> Direct Rendering Manager (DRM) subsystem. SVM al
7, 2024 10:49 PM
> > To: intel...@lists.freedesktop.org; dri-devel@lists.freedesktop.org
> > Cc: airl...@gmail.com; christian.koe...@amd.com;
> > thomas.hellst...@linux.intel.com; Auld, Matthew
> > ; dan...@ffwll.ch
> > Subject: [RFC PATCH 05/28] drm/gpusvm: Add support for GPU
> >
On Wed, Sep 04, 2024 at 02:27:15PM +0200, Thomas Hellström wrote:
> Hi, Sima,
>
> On Mon, 2024-09-02 at 14:33 +0200, Daniel Vetter wrote:
> > Jumping in here in the middle, since I think it's a solid place to
> > drop my
> > idea of "align with core mm" gpusvm locking ...
> >
> > On Thu, Aug 29,
On Mon, Sep 02, 2024 at 05:03:20PM +, Matthew Brost wrote:
> On Mon, Sep 02, 2024 at 01:53:14PM +0200, Daniel Vetter wrote:
> > On Thu, Aug 29, 2024 at 05:27:13PM +, Matthew Brost wrote:
> > > On Thu, Aug 29, 2024 at 11:45:08AM +0200, Daniel Vetter wrote:
> > > > On Tue, Aug 27, 2024 at 07:
airl...@gmail.com; christian.koe...@amd.com;
> thomas.hellst...@linux.intel.com; Auld, Matthew
> ; dan...@ffwll.ch
> Subject: [RFC PATCH 05/28] drm/gpusvm: Add support for GPU
> Shared Virtual Memory
>
> This patch introduces support for GPU Shared Virtual Memory (SVM)
> in the
>
Hi, Sima,
On Mon, 2024-09-02 at 14:33 +0200, Daniel Vetter wrote:
> Jumping in here in the middle, since I think it's a solid place to
> drop my
> idea of "align with core mm" gpusvm locking ...
>
> On Thu, Aug 29, 2024 at 08:56:23PM +, Matthew Brost wrote:
> > On Thu, Aug 29, 2024 at 09:18:2
On Mon, Sep 02, 2024 at 01:53:14PM +0200, Daniel Vetter wrote:
> On Thu, Aug 29, 2024 at 05:27:13PM +, Matthew Brost wrote:
> > On Thu, Aug 29, 2024 at 11:45:08AM +0200, Daniel Vetter wrote:
> > > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > > This patch introduces suppo
Jumping in here in the middle, since I think it's a solid place to drop my
idea of "align with core mm" gpusvm locking ...
On Thu, Aug 29, 2024 at 08:56:23PM +, Matthew Brost wrote:
> On Thu, Aug 29, 2024 at 09:18:29PM +0200, Thomas Hellström wrote:
> Issues with removing a SVM range:
>
> - X
On Fri, Aug 30, 2024 at 11:16:53AM +0200, Thomas Hellström wrote:
> Hi, Matthew
>
> On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > +/**
> > + * DOC: Overview
> > + *
> > + * GPU Shared Virtual Memory (GPU SVM) layer for the Direct
> > Rendering Manager (DRM)
> > + *
> > + * The GPU SV
On Thu, Aug 29, 2024 at 05:27:13PM +, Matthew Brost wrote:
> On Thu, Aug 29, 2024 at 11:45:08AM +0200, Daniel Vetter wrote:
> > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > This patch introduces support for GPU Shared Virtual Memory (SVM) in the
> > > Direct Rendering Ma
On Thu, Aug 29, 2024 at 04:49:15PM +, Matthew Brost wrote:
> On Wed, Aug 28, 2024 at 08:50:02PM +0200, Daniel Vetter wrote:
> > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > +int drm_gpusvm_migrate_to_sram(struct drm_gpusvm *gpusvm,
> > > +struct d
On Fri, Aug 30, 2024 at 05:00:11AM +, Matthew Brost wrote:
> On Wed, Aug 28, 2024 at 04:31:19PM +0200, Daniel Vetter wrote:
> > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > + if (!ctx->mmap_locked) {
> > > + /*
> > > + * XXX: HMM
On Thu, Aug 29, 2024 at 04:40:47PM +, Matthew Brost wrote:
> On Wed, Aug 28, 2024 at 06:25:18PM +0200, Daniel Vetter wrote:
> > On Wed, Aug 28, 2024 at 03:43:48PM +, Matthew Brost wrote:
> > > On Wed, Aug 28, 2024 at 04:46:24PM +0200, Christian König wrote:
> > > > Am 28.08.24 um 16:31 schr
Hi, Matt
On Fri, 2024-08-30 at 13:58 +, Matthew Brost wrote:
> >
> > So I specifically asked Jason about the performance problem about
> > using
> > many notifiers vs using a single one, and he responded that the
> > problem
> > is slowing down the core mm on invalidations, if the RB tree get
Hi, Matt.
On Fri, 2024-08-30 at 13:47 +, Matthew Brost wrote:
> On Fri, Aug 30, 2024 at 11:57:33AM +0200, Thomas Hellström wrote:
> > Hi, Matthew,
> >
> > Agreed the below might not be important just now, but some ideas:
> >
> > On Thu, 2024-08-29 at 20:56 +, Matthew Brost wrote:
> > > I
On Fri, Aug 30, 2024 at 10:18:58AM +0200, Thomas Hellström wrote:
> Hi, Matthew,
>
> On Thu, 2024-08-29 at 20:56 +, Matthew Brost wrote:
> > On Thu, Aug 29, 2024 at 09:18:29PM +0200, Thomas Hellström wrote:
> > > Hi, Matthew,
> > >
> > > On Thu, 2024-08-29 at 17:45 +, Matthew Brost wrote:
On Fri, Aug 30, 2024 at 11:57:33AM +0200, Thomas Hellström wrote:
> Hi, Matthew,
>
> Agreed the below might not be important just now, but some ideas:
>
> On Thu, 2024-08-29 at 20:56 +, Matthew Brost wrote:
> > Issues with removing a SVM range:
> >
> > - Xe bind code stores invalidation / pr
Hi, Matthew,
Agreed the below might not be important just now, but some ideas:
On Thu, 2024-08-29 at 20:56 +, Matthew Brost wrote:
> Issues with removing a SVM range:
>
> - Xe bind code stores invalidation / present state in VMA, this would
> need to be moved to the radix tree. I have Jira
Hi, Matthew
On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> +/**
> + * DOC: Overview
> + *
> + * GPU Shared Virtual Memory (GPU SVM) layer for the Direct
> Rendering Manager (DRM)
> + *
> + * The GPU SVM layer is a component of the DRM framework designed to
> manage shared
> + * virtual
Hi, Matthew,
On Thu, 2024-08-29 at 20:56 +, Matthew Brost wrote:
> On Thu, Aug 29, 2024 at 09:18:29PM +0200, Thomas Hellström wrote:
> > Hi, Matthew,
> >
> > On Thu, 2024-08-29 at 17:45 +, Matthew Brost wrote:
> > > On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> > > >
On Wed, Aug 28, 2024 at 04:31:19PM +0200, Daniel Vetter wrote:
> On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > + if (!ctx->mmap_locked) {
> > + /*
> > +* XXX: HMM locking document indicates only a read-lock
> > +
On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> Hi, Matt.
>
> Some initial design comments / questions:
>
Hi, Thomas. Missed one question in initial reply.
> On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > This patch introduces support for GPU Shared Virtual Mem
On Thu, Aug 29, 2024 at 09:18:29PM +0200, Thomas Hellström wrote:
> Hi, Matthew,
>
> On Thu, 2024-08-29 at 17:45 +, Matthew Brost wrote:
> > On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> > > Hi, Matt.
> > >
> > > Some initial design comments / questions:
> > >
> > > On
Hi, Matthew,
On Thu, 2024-08-29 at 17:45 +, Matthew Brost wrote:
> On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> > Hi, Matt.
> >
> > Some initial design comments / questions:
> >
> > On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > > This patch introduces su
On Thu, Aug 29, 2024 at 05:45:07PM +, Matthew Brost wrote:
> On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> > Hi, Matt.
> >
> > Some initial design comments / questions:
> >
> > On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > > This patch introduces support f
On Thu, Aug 29, 2024 at 11:16:49AM +0200, Thomas Hellström wrote:
> Hi, Matt.
>
> Some initial design comments / questions:
>
> On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> > This patch introduces support for GPU Shared Virtual Memory (SVM) in
> > the
> > Direct Rendering Manager (D
On Thu, Aug 29, 2024 at 11:45:08AM +0200, Daniel Vetter wrote:
> On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > This patch introduces support for GPU Shared Virtual Memory (SVM) in the
> > Direct Rendering Manager (DRM) subsystem. SVM allows for seamless
> > sharing of memory be
On Wed, Aug 28, 2024 at 08:50:02PM +0200, Daniel Vetter wrote:
> On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > +int drm_gpusvm_migrate_to_sram(struct drm_gpusvm *gpusvm,
> > + struct drm_gpusvm_range *range,
> > + const struct d
On Wed, Aug 28, 2024 at 06:25:18PM +0200, Daniel Vetter wrote:
> On Wed, Aug 28, 2024 at 03:43:48PM +, Matthew Brost wrote:
> > On Wed, Aug 28, 2024 at 04:46:24PM +0200, Christian König wrote:
> > > Am 28.08.24 um 16:31 schrieb Daniel Vetter:
> > > > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Ma
On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> This patch introduces support for GPU Shared Virtual Memory (SVM) in the
> Direct Rendering Manager (DRM) subsystem. SVM allows for seamless
> sharing of memory between the CPU and GPU, enhancing performance and
> flexibility in GPU c
Hi, Matt.
Some initial design comments / questions:
On Tue, 2024-08-27 at 19:48 -0700, Matthew Brost wrote:
> This patch introduces support for GPU Shared Virtual Memory (SVM) in
> the
> Direct Rendering Manager (DRM) subsystem. SVM allows for seamless
> sharing of memory between the CPU and GPU
On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> +int drm_gpusvm_migrate_to_sram(struct drm_gpusvm *gpusvm,
> +struct drm_gpusvm_range *range,
> +const struct drm_gpusvm_ctx *ctx)
> +{
> + u64 start = range->va.start, end =
On Wed, Aug 28, 2024 at 03:43:48PM +, Matthew Brost wrote:
> On Wed, Aug 28, 2024 at 04:46:24PM +0200, Christian König wrote:
> > Am 28.08.24 um 16:31 schrieb Daniel Vetter:
> > > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > > + if (!ctx->mmap_locked) {
> >
On Wed, Aug 28, 2024 at 11:53 AM Matthew Brost wrote:
>
> On Wed, Aug 28, 2024 at 04:46:24PM +0200, Christian König wrote:
> > Am 28.08.24 um 16:31 schrieb Daniel Vetter:
> > > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > > + if (!ctx->mmap_locked) {
> > > > +
On Wed, Aug 28, 2024 at 04:46:24PM +0200, Christian König wrote:
> Am 28.08.24 um 16:31 schrieb Daniel Vetter:
> > On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> > > + if (!ctx->mmap_locked) {
> > > + /*
> > > + * XXX: HMM locking document
Am 28.08.24 um 16:31 schrieb Daniel Vetter:
On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
+ if (!ctx->mmap_locked) {
+ /*
+* XXX: HMM locking document indicates only a read-lock
+* is required bu
On Tue, Aug 27, 2024 at 07:48:38PM -0700, Matthew Brost wrote:
> + if (!ctx->mmap_locked) {
> + /*
> + * XXX: HMM locking document indicates only a read-lock
> + * is required but there apears to be a window between
> +
This patch introduces support for GPU Shared Virtual Memory (SVM) in the
Direct Rendering Manager (DRM) subsystem. SVM allows for seamless
sharing of memory between the CPU and GPU, enhancing performance and
flexibility in GPU computing tasks.
The patch adds the necessary infrastructure for SVM, i
45 matches
Mail list logo
