[PATCH] drm/amdgpu/si: Call amdgpu_ttm_set_active_vram_size from si_dma_start/stop

2016-09-06 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Without this, we were only ever using the first 256MB of VRAM. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97593 Tested-by: Konstantin A. Lepikhov <lakos...@altlinux.org> Tested-by: Arek Ruśniak <arek.r...@gmail.com> Sig

[PATCH v2 xf86-video-ati 1/2] Only copy from screen pixmap to shared pixmap on demand for slave scanout

2016-09-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Only copy once for each time we update the corresponding scanout pixmap. This can significantly reduce the bandwidth usage when there are frequent updates to the screen pixmap. This initial implementation only works when both the master and

Re: [PATCH 01/10] drm/amdgpu: handle runtime pm in drm pre/post close

2016-09-01 Thread Michel Dänzer
On 02/09/16 05:38 AM, Alex Deucher wrote: > On Wed, Aug 31, 2016 at 9:03 PM, Michel Dänzer <mic...@daenzer.net> wrote: >> On 01/09/16 07:08 AM, Alex Deucher wrote: >>> Close was not handled correctly. >>> >>> Signed-off-by: Alex Deucher <alexander.deuc

[PATCH xf86-video-amdgpu 10/12] Move up amdgpu_scanout_extents_intersect

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Will be needed higher up by the following changes. No functional change. (Ported from radeon commit 2f6e5fb15f1a9ce523c85550e493f8bda9d0c00f) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgp

[PATCH xf86-video-amdgpu 11/12] Synchronize scanout pixmaps for TearFree

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Copy the damaged areas which are still valid in the other scanout pixmap from there, then only copy the remaining damaged area from the screen pixmap. This is slightly more efficient (only needs one Damage record instead of two, and only needs t

[PATCH xf86-video-amdgpu 12/12] Make TearFree effective with PRIME slave scanout

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> TearFree can now prevent tearing with any possible display configuration. Note that there may still be inter-GPU tearing if the primary GPU uses a different driver. (Ported from radeon commit 38797a33117222dadbc89e5f21ed8cd5deef9bea) Sign

[PATCH xf86-video-amdgpu 01/12] Add explicit AMDGPU_DRM_QUEUE_ERROR define

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Should make the amdgpu_drm_queue_alloc error handling clearer, and gets rid of a compile warning about it returning NULL. (Ported from radeon commit a37af701768b12d86868a831a79f1e02ee4968cf) Signed-off-by: Michel Dänzer <michel.daen..

[PATCH xf86-video-amdgpu 04/12] Wait for pending flips to complete before turning off an output or CRTC

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> At least with older kernels, the flip may never complete otherwise, which can result in us hanging in drmmode_set_mode_major. Fixes: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/1577170 (Ported from radeon c

[PATCH xf86-video-amdgpu 09/12] Factor out transform_region helper

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> (Ported from radeon commit 5a57005178fc13b6f7e513458ca6dae72a3e5783) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_kms.c | 74 1 file changed, 43 inser

[PATCH xf86-video-amdgpu 07/12] Track damage accurately for RandR 1.4 slave scanout

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This further reduces the PCIe bandwidth usage. (Ported from radeon commit b0867063abb197b9134166706d99fcbe5f204bb5, plus leak fix from 5a57005178fc13b6f7e513458ca6dae72a3e5783) Signed-off-by: Michel Dänzer <michel.daen...@amd.com&

[PATCH xf86-video-amdgpu 03/12] Keep track of damage event related flushes per-client

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This further reduces the compositing slowdown due to flushing overhead, by only flushing when the X server actually sends XDamageNotify events to a client, and there hasn't been a flush yet in the meantime. (Ported from radeon

[PATCH xf86-video-amdgpu 08/12] Only copy from screen pixmap to shared pixmap on demand for slave scanout

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Only copy once for each time we update the corresponding scanout pixmap. This can significantly reduce the bandwidth usage when there are frequent updates to the screen pixmap. This initial implementation only works when both the master and

[PATCH xf86-video-amdgpu 05/12] Use drmmode_crtc_scanout_* helpers for RandR 1.4 scanout pixmaps

2016-09-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This should allow using multiple CRTCs via RandR 1.4 even with xserver < 1.17. It also simplifies the code a little, and paves the way for following changes. (Ported from radeon commits 4cfa4615f79f64062e5e771cd45dd70

[PATCH xf86-video-ati 2/4] Move up radeon_scanout_extents_intersect

2016-09-02 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Will be needed higher up by the following changes. No functional change. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 42 +- 1 file changed, 21 insertions(+),

[PATCH xf86-video-ati 1/4] Factor out transform_region helper

2016-09-02 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> While we're at it, fix leaking the memory allocated for xRectangles. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 73 +--- 1 file changed, 43 inser

[PATCH xf86-video-ati 4/4] Make TearFree effective with PRIME slave scanout

2016-09-02 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> TearFree can now prevent tearing with any possible display configuration. Note that with PRIME slave scanout, there may still be inter-GPU tearing if the primary GPU uses a different driver. Signed-off-by: Michel Dänzer <michel.daen..

[PATCH xf86-video-ati 3/4] Synchronize scanout pixmaps for TearFree

2016-09-02 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Copy the damaged areas which are still valid in the other scanout pixmap from there, then only copy the remaining damaged area from the screen pixmap. This is slightly more efficient, and will be significantly more efficient with the following

Re: [PATCH 7/7] drm/amd/amdgpu: Comment out currently unused SI DPM struct

2016-09-07 Thread Michel Dänzer
On 07/09/16 11:49 PM, StDenis, Tom wrote: > > Would love to but unfortunately I don't have a tahiti_le (or any SI) > part to test it on. There are potential testers on https://bugs.freedesktop.org/show_bug.cgi?id=60879 . -- Earthling Michel Dänzer |

Re: [PATCH 1/5] drm/amdgpu: add AMDGPU_GEM_CREATE_VRAM_LINEAR flag v2

2016-08-29 Thread Michel Dänzer
GUOUS > or the anti-flag AMDGPU_GEM_CREATE_VRAM_SCATTER_GATHER to avoid > confusion with linear in the tiling sense. Seconded. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer _

[PATCH xf86-video-ati 1/2] Only copy from screen pixmap to shared pixmap on demand for slave scanout

2016-08-31 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Only copy once for each time we update the corresponding scanout pixmap. This can significantly reduce the bandwidth usage when there are frequent updates to the screen pixmap. This initial implementation only works when both the master and

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-08 Thread Michel Dänzer
ch mark whether they > should be using shared or exclusive fences. Afaict, in amdgpu they are > all set to shared, the relevant user interface seems to be > amdgpu_bo_list_set(). This all makes sense to me. Christian, why is amdgpu setting only shared fences? Can we fix that? -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH xserver] config: fix GPUDevice fail when AutoAddGPU off + BusID

2016-09-14 Thread Michel Dänzer
{ > @@ -450,9 +463,8 @@ xf86platformProbeDev(DriverPtr drvp) > if (j == xf86_num_platform_devices) > continue; > > -foundScreen = probeSingleDevice(_platform_devices[j], drvp, > devList[i], 0); > - if (!foundScreen) > -co

[PATCH xf86-video-amdgpu] Use --with-xorg-conf-dir=$prefix/share/X11/xorg.conf.d by default

2016-09-14 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We were using the result of `pkg-config --variable=sysconfigdir xorg-server` before, which may not be inside $prefix, so make install might fail for 10-amdgpu.conf . Fixes make distcheck in that case, and possibly also 10-amdgpu.conf seemingly m

[PATCH xf86-video-ati 2/4] Fix radeon_mode_hotplug crash on multi GPU platform.

2016-09-14 Thread Michel Dänzer
From: Qiang Yu <qiang...@amd.com> On multi GPU platform, some screen is created by other GPU DDX. Signed-off-by: Qiang Yu <qiang...@amd.com> (Ported from amdgpu commit 978242977e5dc905e1d5a46b1b0d34b356c7af26) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drm

[PATCH xf86-video-ati 3/4] DRI2: Add interpolated_vblanks in radeon_dri2_get_crtc_msc

2016-09-14 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We need that in radeon_dri2_drawable_crtc as well for priv->vblank_delta to work as intended. radeon_dri2_get_msc was already doing this. Fixes hangs in some cases when using VDPAU via DRI2 and moving the window between CRTCs. (Ported fr

[PATCH xf86-video-ati 4/4] DRI2: Fix radeon_dri2_exchange_buffers width/height copy'n'paste error

2016-09-14 Thread Michel Dänzer
From: Qiang Yu <qiang...@amd.com> Signed-off-by: Qiang Yu <qiang...@amd.com> (Ported from amdgpu commit 73c8dc000ad6b2b53ba3aa7155f5e8f6b55623b7) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_dri2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

Re: [PATCH] drm/amdgpu: improve GTT BO alloc speed in OGL

2016-09-12 Thread Michel Dänzer
nd up in the > slow path even when an IOMMU is available. This change will break any cases where SWIOTLB is actually necessary though, won't it? -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast

Re: [PATCH 5/6] drm/amdgpu: use expolicit limit fro VRAM<->GTT moves

2016-09-12 Thread Michel Dänzer
address space allocation. >> >> Signed-off-by: Christian König <christian.koe...@amd.com> > > Typo in the patch title: > s/expolicit/explicit/ and s/fro/for/ -- Earthling Michel Dänzer | http://www.amd.com Lib

[PATCH xf86-video-ati 2/4] Consolidate get_drawable_pixmap helper

2016-09-13 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> There were two static helpers for the same purpose. Consolidate them into a single inline helper which can be used anywhere. (Ported from amdgpu commit 641f4647b7f51dfd2da330376cd10fa9702b6423) Signed-off-by: Michel Dänzer <michel.daen..

Re: [PATCH 2/5] drm/amdgpu: fix a bunch of coding style issues in amdgpu_ttm.c

2016-09-15 Thread Michel Dänzer
that we reduce readability by breaking up a single line statement to multiple lines. It makes more sense when saving a more significant amount of horizontal space or when the statement already spans multiple lines though. -- Earthling Michel Dänzer

[ANNOUNCE] xf86-video-amdgpu 1.1.2

2016-09-16 Thread Michel Dänzer
this release in favour of 1.1.1. Michel Dänzer (2): Fix handling of configure option --with-xorg-conf-dir Bump version for 1.1.2 release git tag: xf86-video-amdgpu-1.1.2 http://xorg.freedesktop.org/archive/individual/driver/xf86-video-amdgpu-1.1.2.tar.bz2 MD5

[PATCH] Fix handling of configure option --with-xorg-conf-dir

2016-09-16 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> There were two problems: I accidentally changed the variable name in the AC_ARG_WITH stanza from configdir to xorgconfigdir, so specifying --with-xorg-conf-dir wouldn't work correctly. Fix this back to configdir. If neither --with-xorg-conf-d

[PATCH xf86-video-amdgpu] Untangle HAS_XORG_CONF_DIR / --with-xorg-conf-dir lines in configure.ac

2016-09-16 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> $sysconfigdir used to be part of the default --with-xorg-conf-dir value, but it no longer is. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- configure.ac | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/

Re: Bonair support Kernel compiling instructions

2016-09-25 Thread Michel Dänzer
time. As I said to you on IRC before, don't expect amdgpu to magically make anything significantly faster. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-25 Thread Michel Dänzer
On 23/09/16 09:09 PM, Daniel Vetter wrote: > On Fri, Sep 23, 2016 at 07:00:25PM +0900, Michel Dänzer wrote: >> On 22/09/16 10:22 PM, Christian König wrote: >>> Am 22.09.2016 um 15:05 schrieb Daniel Vetter: >>>> >>>> But the current approach in amdgpu_syn

[PATCH xf86-video-ati] Rotate and reflect cursor hotspot position for drmModeSetCursor2

2016-10-06 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We were always passing the hotspot position in the X screen coordinate space, but drmModeSetCursor2 needs it in the CRTC coordinate space. The wrong hotspot position would cause the kernel driver to adjust the HW cursor position incorrectl

[PATCH xf86-video-amdgpu] Rotate and reflect cursor hotspot position for drmModeSetCursor2

2016-10-05 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We were always passing the hotspot position in the X screen coordinate space, but drmModeSetCursor2 needs it in the CRTC coordinate space. The wrong hotspot position would cause the kernel driver to adjust the HW cursor position incorrectl

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-21 Thread Michel Dänzer
On 13/09/16 09:52 PM, Christian König wrote: > Am 13.09.2016 um 11:39 schrieb Chris Wilson: >> On Tue, Sep 13, 2016 at 10:44:11AM +0200, Christian König wrote: >>> Am 09.09.2016 um 03:15 schrieb Michel Dänzer: >>>> On 09/09/16 01:23 AM, Chris Wilson wrote: >>&g

[PATCH xf86-video-ati 05/15] Only add main fb if necessary

2016-09-21 Thread Michel Dänzer
(Ported from amdgpu commit a3ca1500703837cbb8d49c554199a25dea7d5e1e) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 44 1 file changed, 16 insertions(+), 28 deletions(-) diff --git a/src/drmmode_display.c b/src

[PATCH xf86-video-ati 15/15] Only list each unique chipset family once in the log file

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> (Ported from amdgpu commit 6a1ba044c2b71081e6060d0c096917d6238f2145) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/pcidb/parse_pci_ids.pl | 7 + src/radeon_chipset_gen.h | 339 ++

[PATCH xf86-video-ati 07/15] Make sure drmmode_crtc->scanout[] are destroyed when not needed

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We failed to do this when going back to scanning out directly from the screen pixmap. As a bonus, since we now destroy drmmode_crtc->scanout[] after setting the new scanout buffer, we may avoid the CRTC turning off intermittently in

[PATCH xf86-video-ati 04/15] Remove unnecessary fb addition from drmmode_xf86crtc_resize

2016-09-21 Thread Michel Dänzer
m amdgpu commit 9ca1c24235ff5ab2e028333fc326e2eff008c574) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 9 + 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/src/drmmode_display.c b/src/drmmode_display.c index 89922d6..4f25ae7 100644 --- a/src/drmmod

[PATCH xf86-video-ati 09/15] dri2: Don't allow flipping when using a dedicated scanout buffer

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes issues when mixing rotation and page flipping with current xserver Git master. (Ported from amdgpu commit decabd574f90d3df397c80ec931b3fde8a4afb49) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_dri2.c | 5 +

[PATCH xf86-video-ati 02/15] glamor: Fix leak of pixmap private when replacing BO

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Reported-by: Qiang Yu <qiang...@amd.com> (Ported from amdgpu commit 397aedafee437c125b8ac1feafb1c3b466842aeb) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_glamor.c | 3 +-- 1 file changed, 1 insertion(+), 2

[PATCH xf86-video-ati 03/15] Free priv in amdgpu_set_pixmap_bo also if priv->bo == NULL

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes memory leak when destroying pixmaps with priv->bo == NULL. Reported-by: Qiang Yu <qiang...@amd.com> (Ported from amdgpu commit 7f7f9825caf3983902491da27c16d14cd8bf9b7d) Signed-off-by: Michel Dänzer <michel.daen...@amd.com>

[PATCH xf86-video-ati 12/15] Make the dedicated scanout mechanism work with arbitrary transforms

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This makes TearFree work with arbitrary transforms, and makes transforms work better even without TearFree, with xserver >= 1.12. (Ported from amdgpu commit bf000ea7ef91f5ecb59fc3c1ab8ed9eddcc0841d) Signed-off-by: Michel Dänzer <

[PATCH xf86-video-ati 13/15] Remove w/h parameters from radeon_scanout_extents_intersect

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We can use the dimensions of the CRTC's mode instead. (Ported from amdgpu commit ede7f2bcae63be65e05e3029bfe7c742e5978932) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 18 ++ 1 fil

[PATCH xf86-video-ati 08/15] present: Don't allow flipping when using a dedicated scanout buffer

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes issues when mixing rotation and page flipping with current xserver Git master. (Ported from amdgpu commit 3ed28ce7cd26f89969617ba901ff253091d0d469) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_present.c | 3

[PATCH xf86-video-ati 14/15] Clear damage in radeon_scanout_update if it doesn't intersect the CRTC

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> There's no need to test that same damage again. (Ported from amdgpu commit a576430526cbc404de64b30e1377a356644e8024) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 4 +++- 1 file changed, 3 insertions(+)

[PATCH xf86-video-ati 10/15] present: Separate checks for flips vs unflips

2016-09-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> All unflip checks apply to flips as well, but not vice versa. (Ported from amdgpu commit 4d506c23c9a628204fa23607931557b07ada3e31) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_pr

[ANNOUNCE] xf86-video-amdgpu 1.1.1

2016-09-15 Thread Michel Dänzer
Add more Polaris 11 PCI IDs Marek Olšák (1): Fix cursor size for SI Michel Dänzer (14): Handle Zaphod mode correctly in amdgpu_mode_hotplug glamor: Fix amdgpu_glamor_share_pixmap_backing for priv->bo == NULL Remove amdgpu_share_pixmap_backing

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-21 Thread Michel Dänzer
On 22/09/16 12:21 AM, Christian König wrote: > Am 21.09.2016 um 17:13 schrieb Michel Dänzer: >> On 21/09/16 07:30 PM, Christian König wrote: >>> Am 21.09.2016 um 11:56 schrieb Michel Dänzer: >>>> FWIW, we seem to have the same issue with radeon vs. amdgpu: radeo

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-21 Thread Michel Dänzer
On 22/09/16 12:15 AM, Christian König wrote: > Am 21.09.2016 um 17:07 schrieb Michel Dänzer: >> On 21/09/16 09:56 PM, Daniel Vetter wrote: >>> On Wed, Sep 21, 2016 at 1:19 PM, Christian König >>> <deathsim...@vodafone.de> wrote: >>>> >>>>

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-21 Thread Michel Dänzer
.de> wrote: >>>> Am 21.09.2016 um 11:56 schrieb Michel Dänzer: >>>>> >>>>> >>>>> Looks like there are different interpretations of the semantics of >>>>> exclusive vs. shared fences. Where are these semantics documented? >

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-21 Thread Michel Dänzer
On 21/09/16 07:30 PM, Christian König wrote: > Am 21.09.2016 um 11:56 schrieb Michel Dänzer: >> >> FWIW, we seem to have the same issue with radeon vs. amdgpu: radeon only >> seems to wait for exclusive fences, so e.g. running Xorg on amdgpu and >> using PRIME sla

[ANNOUNCE] xf86-video-ati 7.7.1

2016-09-15 Thread Michel Dänzer
. Thanks to Qiang Yu for his contributions to this release! Michel Dänzer (15): Handle Zaphod mode correctly in radeon_mode_hotplug Explicitly set the fbcon pixmap pitch again Only use RandR APIs if RandR is enabled Don't enable micro-tiling for scanout buffers on pre-R600

Re: [PATCH 0/3] drm/amdgpu: implement raster configuration

2016-09-19 Thread Michel Dänzer
On 18/09/16 11:36 AM, Huang Rui wrote: > Hi all, > > This patch set is to introduce the raster configuration for all gfx > generations. What's the motivation for doing this in the kernel? Mesa will write its own values to these registers anyway. -- Earthling M

Re: [PATCH 4/6] drm/radeon: Provide page_flip_target hook

2016-09-19 Thread Michel Dänzer
should make those old asics work > well as well. Thanks for the testing and patches! -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer

[PATCH xf86-video-ati] Only define transform_region function for XF86_CRTC_VERSION >= 4

2016-09-20 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We're not using it with older xserver. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 4 1 file changed, 4 insertions(+) diff --git a/src/radeon_kms.c b/src/radeon_kms.c index 9bcf657..0d61d49 1006

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-23 Thread Michel Dänzer
is no way I can fix amdgpu in this case. > > The handling clearly needs to be changed on the receiving side of the > reservation objects if I don't completely want to disable concurrent > access to BOs in amdgpu. Anyway, we need a solution for this between radeon and

Re: [PATCH] drm/amdgpu: For virtual_display feature, the vblank_get_counter hook is always return 0 when there's no hardware frame counter which can be used.

2016-08-17 Thread Michel Dänzer
On 17/08/16 02:47 PM, Deng, Emily wrote: >> -Original Message- >> From: Michel Dänzer [mailto:mic...@daenzer.net] >> Sent: Wednesday, August 17, 2016 11:50 AM >> To: Deng, Emily <emily.d...@amd.com> >> Cc: amd-gfx@lists.freedesktop.or

[PATCH xf86-video-amdgpu] Remove unused lut_r/g/b arrays from drmmode_crtc_private_rec

2016-08-17 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes: 1091f28e1fa2 ("Remove drmmode_load_palette") Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.h | 1 - 1 file changed, 1 deletion(-) diff --git a/src/drmmode_display.h b/src/drmmode_display.h

Re: [PATCH xserver 0/6] modesetting: add DRI2 page flip support

2016-08-18 Thread Michel Dänzer
; modesetting: remove redundent pixmap destroy Patches 2, 4 & 5 are Reviewed-by: Michel Dänzer <michel.daen...@amd.com> Patch 6 was already submitted by Hans de Goede: https://patchwork.freedesktop.org/patch/105351/ -- Earthling Michel Dänzer |

Re: [PATCH xserver 1/6] modesetting: make ms_do_pageflip generic for share with DRI2

2016-08-18 Thread Michel Dänzer
long long) seq)); Might be nice to move this debugging output to ms_present_flip instead of removing it completely. With that fixed, this patch is Reviewed-by: Michel Dänzer <michel.daen...@amd.com> -- Earthling Michel Dänzer | http://www.amd.com Libre software e

Re: [PATCH xserver 3/6] modesetting: add DRI2 page flip support

2016-08-18 Thread Michel Dänzer
py and shove it over the USB. > + * also for sw cursors. > + */ SW cursors cannot work correctly with page flipping. For that reason, xf86-video-ati/amdgpu disable page flipping while there's an SW cursor. -- Earthling Michel Dänzer

[PATCH xf86-video-ati] Fix build against xserver < 1.13

2016-08-26 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> pScreen->isGPU was only introduced in 1.13. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97490 Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 5 - src/radeon_kms.c | 5 - 2 f

Re: [PATCH] Re: Mullins support in xf86-video-amdgpu

2016-08-22 Thread Michel Dänzer
On 23/08/16 11:57 AM, Reid Hekman wrote: > On 08/22/2016 08:27 PM, Michel Dänzer wrote: >> On 23/08/16 10:18 AM, Reid Hekman wrote: >>> >>> I was encouraged by a commit I saw today to add Sea Islands PCI ids to >>> xf86-video-amdgpu. However I did not see M

Re: [PATCH 3/3] drm/amdgpu: reset the asic when hibernating v2

2016-08-23 Thread Michel Dänzer
gpu_device_suspend(drm_dev, false, true, true); > } FWIW, using a single flags parameter tends to be more readable than multiple boolean parameters. Just a suggestion for possible further improvement, not a blocker for this patch. -- Earthling Michel Dänzer |

Re: cannot send email to amd-gfx

2016-08-24 Thread Michel Dänzer
On 24/08/16 03:05 PM, Michel Dänzer wrote: > On 24/08/16 01:11 PM, Liu, Monk wrote: >> Hi Alex >> >> >> >> I tried use git send-email to send patch to amd-gfx, but seems it failed, >> >> So I use git send-email to send patch to myself, it’s okay. >

[PATCH xf86-video-amdgpu] Add missing Kaveri PCI ID (1318)

2016-08-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Found by comparing src/pcidb/ati_pciids.csv with xf86-video-ati. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_chipinfo_gen.h | 1 + src/amdgpu_chipset_gen.h | 1 + src/amdgpu_pci_chipset_gen.h

[PATCH xf86-video-amdgpu] Only list each unique chipset family once in the log file

2016-08-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_chipset_gen.h | 21 + src/amdgpu_probe.c | 2 +- src/pcidb/parse_pci_ids.pl | 9 + 3 files changed, 31 insertions(+), 1 delet

[PATCH v2 xf86-video-amdgpu] Only list each unique chipset family once in the log file

2016-08-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- v2: Don't bother making up PCI IDs for the AMDGPUUniqueChipsets array, they're not used anyway. src/amdgpu_chipset_gen.h | 21 + src/amdgpu_probe.c

Re: [PATCH] modetest: Adding amdgpu to module list

2016-08-28 Thread Michel Dänzer
ot;, > "radeon", > "nouveau", > "vmwgfx", > Reviewed-by: Michel Dänzer <michel.daen...@amd.com> -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast |

[PATCH xf86-video-amdgpu] Add Mullins PCI IDs

2016-08-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97472 Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_chipinfo_gen.h | 16 src/amdgpu_chipset_gen.h | 16 +

Re: [PATCH xf86-video-amdgpu] Fix amdgpu_dri2_exchange_buffers width/height miss copy

2016-08-23 Thread Michel Dänzer
region.extents.y2 = front_priv->pixmap->drawable.width; > + region.extents.y2 = front_priv->pixmap->drawable.height; > region.data = NULL; > DamageRegionAppend(_priv->pixmap->drawable, ); > > Reviewed and pushed, thanks! -- Earthlin

[PATCH xf86-video-ati] Also handle disabled CRTCs in drmmode_clear_pending_flip

2016-08-25 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> If disabling a CRTC had to be deferred due to a pending flip in drmmode_crtc_dpms, there may no longer be any outputs associated with the CRTC when we get here. So we have to check for !crtc->enabled and call drmmode_crtc_dpms in that cas

Re: [PATCH xf86-video-amdgpu] Add support for ScreenPtr::SyncSharedPixmap

2016-09-28 Thread Michel Dänzer
On 28/09/16 05:42 PM, Mike Lothian wrote: > Am I right in guessing this is only required for DRI2? It's for PRIME slave output (xrandr --setprovideroutputsource and friends), which isn't directly related to DRI2 or DRI3. -- Earthling Michel Dänzer | h

[PATCH xf86-video-amdgpu] Add support for ScreenPtr::SyncSharedPixmap

2016-09-28 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This allows deferring shared pixmap updates between different drivers. (Ported from radeon commit 53be26b00e83f871f0afd39caa5a7a1d6ec4aea1) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgp

Re: [PATCH 0/3] drm/amdgpu: implement raster configuration

2016-09-27 Thread Michel Dänzer
hat's not necessary. One bump per cycle is OK. > > As I wrote in the other thread as well, bumping once per cycle sucks > when you want to bisect something. > > We are also not short on numbers, so I suggest that we establish that we > always bump when we add a new interface

[PATCH 3/3] drm/radeon: Hide the HW cursor while it's out of bounds

2016-10-27 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes hangs in that case under some circumstances. Bugzilla: https://bugzilla.suse.com/show_bug.cgi?id=1000433 Cc: sta...@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- drivers/gpu/drm/radeon/radeon_c

[PATCH 2/3] drm/radeon: Always store CRTC relative radeon_crtc->cursor_x/y values

2016-10-27 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> We were storing viewport relative coordinates for AVIVO/DCE display engines. However, radeon_crtc_cursor_set2 and radeon_cursor_reset pass radeon_crtc->cursor_x/y as the x/y parameters of radeon_cursor_move_locked, which would break if the C

[PATCH 1/3] drm/radeon: Also call cursor_move_locked when the cursor size changes

2016-10-27 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> The cursor size also affects the register programming. Cc: sta...@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- drivers/gpu/drm/radeon/radeon_cursor.c | 9 + 1 file changed, 5 insertions(+), 4 deleti

[PATCH xf86-video-ati] Require xserver 1.10 or newer

2016-10-26 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> 1.10.0 was released in February 2011. We've been accidentally requiring 1.10 or newer since 121a6de72da5 ("Keep track of damage event related flushes per-client v2"). Signed-off-by: Michel Dänzer <michel.daen...@amd.com

[PATCH xf86-video-amdgpu] present: Check tiling info for flips

2016-10-26 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> The kernel driver doesn't handle flipping between buffers with different tiling parameters correctly. Fixes display corruption with fullscreen apps using different tiling modes (e.g. due to R600_DEBUG=notiling or R600_DEBUG=no2d) via DRI3. Sign

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-10-27 Thread Michel Dänzer
On 27/10/16 10:33 PM, Mike Lothian wrote: > > Just another gentle ping to see where you are with this? I haven't got a chance to look into this any further. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast |

[PATCH xf86-video-ati] Reindent code in radeon_dri2_create_buffer2

2016-11-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes warning about misleading indentation from recent versions of gcc: ../../src/radeon_dri2.c: In function ‘radeon_dri2_create_buffer2’: ../../src/radeon_dri2.c:224:2: warning: this ‘if’ clause does not guard... [-Wmisleading-inden

[PATCH xf86-video-ati 2/2] Use pRADEONEnt to find both screens of a GPU in radeon_mode_hotplug

2016-11-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes misbehaviour when hotplugging DisplayPort connectors on secondary GPUs. Fixes: c801f9f10a5d ("Handle Zaphod mode correctly in radeon_mode_hotplug") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98626 Signed-off-b

[PATCH xf86-video-ati 1/2] Refactor radeon_mode_hotplug

2016-11-08 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Preparation for the next change, no functional change intended. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 58 ++- 1 file changed, 34 insertions(+),

Re: [PATCH] drm/amdgpu: Attach exclusive fence to prime exported bo's. (v3)

2016-11-06 Thread Michel Dänzer
DEBUG_PRIME("Fence wait failed: %li\n", lret); > + return lret; > + } > + > + bo->prime_exported = true; We should probably clear bo->prime_exported in amdgpu_gem_prime_unpin. Also, I think we should set bo->prime_exported (prime_shared?) i

Re: [PATCH] Remove RADEON_HOST_DATA_SWAP_HDW case from RadeonCopySwap

2016-11-06 Thread Michel Dänzer
;Unsupported swap value RADEON_HOST_DATA_SWAP_HDW\n"); so that if somebody makes a change which causes this value to be passed in, there's an obvious failure (Xorg aborts with the message passed to FatalError) instead of a subtle one (RADEONCopySwap uses different byte swapping than intended

Re: [PATCH 2/2] drm/amdgpu: remove extra placement for AMDGPU_GEM_CREATE_NO_CPU_ACCESS

2016-11-06 Thread Michel Dänzer
c++; > - } > - > places[c].fpfn = 0; > places[c].lpfn = lpfn; > places[c].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED | > Reviewed-by: Michel Dänzer <michel.da

[PATCH xf86-video-amdgpu 2/2] Use pAMDGPUEnt to find both screens of a GPU in amdgpu_mode_hotplug

2016-11-09 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes misbehaviour when hotplugging DisplayPort connectors on secondary GPUs. Fixes: 14606e127f4b ("Handle Zaphod mode correctly in amdgpu_mode_hotplug") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98626 (Ported fr

[PATCH xf86-video-amdgpu 1/2] Refactor amdgpu_mode_hotplug

2016-11-09 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Preparation for the next change, no functional change intended. (Cherry picked from radeon commit 35bec4937d89b48a79acfcb4f814b7370cb631b2) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_di

Re: [PATCH 2/2] drm/amdgpu: access stolen VRAM directly on CZ

2016-11-07 Thread Michel Dänzer
d writes complete before any access to UC memory (such as writing to an MMIO register or doorbell to submit a GPU operation). -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer

Re: [PATCH libdrm] amdgpu: add the function to get the marketing name (v3)

2016-11-08 Thread Michel Dänzer
On 08/11/16 01:02 AM, Emil Velikov wrote: > On 7 November 2016 at 11:43, Emil Velikov <emil.l.veli...@gmail.com> wrote: >> On 7 November 2016 at 09:09, Michel Dänzer <mic...@daenzer.net> wrote: >> >>> +static struct amdgpu_asic_id_table_t { >>> +

Re: [PATCH 1/2] drm/amdgpu: access stolen VRAM directly on KV/KB

2016-11-08 Thread Michel Dänzer
On 08/11/16 06:23 PM, Christian König wrote: > Am 08.11.2016 um 08:31 schrieb Michel Dänzer: >> On 07/11/16 06:33 PM, Christian König wrote: >>> Am 07.11.2016 um 10:29 schrieb Michel Dänzer: >>>> On 07/11/16 06:21 PM, Christian König wrote: >>>>> Fr

Re: [1/4] amdgpu: add the function to get the marketing name (v2)

2016-11-07 Thread Michel Dänzer
On 02/11/16 10:48 PM, Deucher, Alexander wrote: >> -Original Message- >> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf >> Of Michel Dänzer >> Sent: Tuesday, November 01, 2016 11:51 PM >> To: Alex Deucher >> Cc: Zhang, Jerry; amd-

Re: [PATCH 1/2] drm/amdgpu: access stolen VRAM directly on KV/KB

2016-11-07 Thread Michel Dänzer
lt;christian.koe...@amd.com> The series is Reviewed-by: Michel Dänzer <michel.daen...@amd.com> We could also drop the HDP flushes / invalidates in this case, right? -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast

[PATCH libdrm] amdgpu: add the function to get the marketing name (v3)

2016-11-07 Thread Michel Dänzer
From: Junwei Zhang <jerry.zh...@amd.com> This function is used to look up the marking name for a specific board. v2: agd: Squash in subsequent updates to the table. v3: [Michel Dänzer] * Make amdgpu_asic_id_table static, so it's not exported from libdrm_amdgpu.so.1

Re: [PATCH libdrm] xf86drm: Parse the separate files to retrieve the vendor/device info

2016-11-07 Thread Michel Dänzer
On 05/11/16 03:14 AM, Emil Velikov wrote: > On 2 November 2016 at 03:07, Michel Dänzer <mic...@daenzer.net> wrote: >> >> The first attached patch will result in drmParsePciDeviceInfo always >> reporting revision 0 on kernels without the second attached patc

<    1   2   3   4   5   6   7   8   9   10   >