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

Re: [PATCH] drm/amd/amdgpu: Allow broadcast on debugfs read (v2)

2016-10-18 Thread Michel Dänzer
On 13/10/16 04:20 PM, Michel Dänzer wrote: > On 13/10/16 12:39 AM, StDenis, Tom wrote: >> It comes from amdgpu_query_gpu_info_init() >> >> >> for (i = 0; i < (int)dev->info.num_shader_engines; i++) { >> unsigned instance

[PATCH xf86-video-amdgpu] Order unique chipsets according to first appearance in ati_pciids.csv

2016-10-19 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Instead of lexically. This makes it more likely for similar generations to be close to each other in the list of unique chipsets. (Ported from radeon commit 1ce1b1656acc6211deb2091ff7f28d51b6daf86b, plus change $numunique++ => ++$numuniq

[PATCH xf86-video-ati 1/2] Order unique chipsets according to first appearance in ati_pciids.csv

2016-10-18 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Instead of lexically. This makes it more likely for similar generations to be close to each other in the list of unique chipsets. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/pcidb/parse_pci_ids.pl | 9 +- src/radeon_c

Re: mm: fix cache mode tracking in vm_insert_mixed() breaks AMDGPU [was: Re: Latest testing with drm-next-4.9-wip and latest LLVM/mesa stack - Regression in PowerPlay/DPM on CIK?]

2016-10-19 Thread Michel Dänzer
e still do WC VRAM writes for uploads. > > WC VRAM for uploads is better than WC GART IMO. It's not a simple choice I'm afraid. While writing directly to WC VRAM can be faster than writing to WC GART and then DMA'ing to VRAM, doing so increa

Re: [PATCH xf86-video-amdgpu] amdgpu_probe: Do not close server managed drm fds

2016-10-19 Thread Michel Dänzer
On 19/10/16 06:49 PM, Hans de Goede wrote: > On 19-10-16 04:42, Michel Dänzer wrote: >> On 18/10/16 11:48 PM, Hans de Goede wrote: >>> This fixes the xserver only seeing AMD/ATI devices supported by the >>> amdgpu >>> driver, as by the time xf86-video-ati get

[PATCH xf86-video-ati] Sayōnara, AM_MAINTAINER_MODE!

2016-10-20 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> If --enable-maintainer-mode got lost from config.status for any reason, builds would fail in mysterious ways after changing between different Git commits. There are more reasons for dropping it in the automake manual: https://www.gnu.org/so

[PATCH] Revert "drm/radeon: fix DP link training issue with second 4K monitor"

2016-10-24 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This reverts commit 1a738347df2ee4977459a8776fe2c62196bdcb1b. It caused at least some Kaveri laptops to incorrectly report DisplayPort connectors as connected. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97857 Cc: sta...@vger.kern

[PATCH xf86-video-ati 2/2] Remove PCI IDs and bus type from ati_pciids.csv

2016-10-18 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This cleans up the list of unique chipsets a little further. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/pcidb/ati_pciids.csv | 282 - src/radeon_chipset

[PATCH xf86-video-ati 0/4] Fixes ported from / inspired by modesetting driver changes

2016-10-25 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Hans de Goede (1): PRIME: Fix swapping of provider sink / source capabilities Michel Dänzer (3): Don't rely on randr_crtc->scanout_pixmap in drmmode_set_scanout_pixmap Always call PixmapStopDirtyTracking in drmmode_set_scanout_pixmap

[PATCH xf86-video-ati 3/4] PRIME: Fix swapping of provider sink / source capabilities

2016-10-25 Thread Michel Dänzer
edhat.com> (Ported from xserver commit 94a1c77259ce39ba59ad87615df39b570ffab435) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/radeon_kms.c b/src/radeon_kms.c index faa1e0f..68c783

Re: [PATCH xf86-video-amdgpu] amdgpu_probe: Do not close server managed drm fds

2016-10-18 Thread Michel Dänzer
tform_probe(DriverPtr pDriver, > pAMDGPUEnt = pPriv->ptr; > pAMDGPUEnt->fd_ref++; > } > - pAMDGPUEnt->platform_dev = dev; > > xf86SetEntityInstanceForScreen(pScrn, pEnt->index, > xf86GetNumEntityInstances(pEnt-> These two hunks aren't really

Re: [PATCH xf86-video-amdgpu] Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available

2016-11-23 Thread Michel Dänzer
On 23/11/16 02:01 AM, Emil Velikov wrote: > Hi Michel, > > On 22 November 2016 at 07:58, Michel Dänzer <mic...@daenzer.net> wrote: >> From: Michel Dänzer <michel.daen...@amd.com> >> >> (Ported from radeon commit 1106b2f773ad0611c729b27f4c192a26b43ef

[PATCH] drm/radeon: Ensure vblank interrupt is enabled on DPMS transition to on

2016-11-22 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes the vblank interrupt being disabled when it should be on, which can cause at least the following symptoms: * Hangs when running 'xset dpms force off' in a GNOME session with gnome-shell using DRI2. * RandR 1.4 slave outputs fr

Re: gnome-shell is frozen upon wakeup from DPMS (bisected)

2016-11-22 Thread Michel Dänzer
On 21/11/16 07:15 PM, Max Staudt wrote: > On 11/17/2016 09:27 AM, Michel Dänzer wrote: >>> So... what about DRI2? >> >> Does the attached kernel patch help with pre-4.6 kernels? > > This patch works. Thanks! Great, thanks for testing! > Any chance for this t

Re: gnome-shell is frozen upon wakeup from DPMS (bisected)

2016-11-22 Thread Michel Dänzer
On 22/11/16 06:36 PM, Max Staudt wrote: > On 11/22/2016 09:35 AM, Michel Dänzer wrote: >> On 21/11/16 07:15 PM, Max Staudt wrote: >>> On 11/17/2016 09:27 AM, Michel Dänzer wrote: >>>>> So... what about DRI2? >>>> >>>> Does the attached k

[PATCH xf86-video-amdgpu] Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available

2016-11-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> (Ported from radeon commit 1106b2f773ad0611c729b27f4c192a26b43ef1e7) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_dri2.c | 3 +- src/amdgpu_kms.c | 16 - src/amdgpu_present.c | 5 +-- src/am

[PATCH xf86-video-ati] Pass fb_id into drmmode_page_flip_target_absolute/relative

2016-11-24 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> drmmode->fb_id isn't what we need in the TearFree case. Fixes TearFree freezing with (WW) RADEON(0): flip queue failed in radeon_scanout_flip: No such file or directory in the log file. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi

Re: [PATCH] xf86-video-ati: Fix build for xserver < 1.13

2016-11-25 Thread Michel Dänzer
On 25/11/16 03:00 AM, Jochen Rollwagen wrote: > Subject: [PATCH] fix build for xserver < 1.13 > > same procedure every few patches. Sorry about that. I pushed your patch (with the indentation fixed up), thanks. -- Earthling Michel Dänzer | http:/

[PATCH xf86-video-ati] Add radeon_is_gpu_screen helper

2016-11-25 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This will hopefully decrease the chance of accidentally breaking the build against xserver < 1.13 in the future. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 5 + src/radeon.h

[PATCH xf86-video-amdgpu] Make libdrm >= 2.4.72 requirement explicit

2016-11-25 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> And drop compatibility code for older versions. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- configure.ac | 2 +- src/amdgpu_dri2.c | 13 - src/drmmode_display.h | 4 3 files changed, 1 inse

[PATCH xf86-video-amdgpu] Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available

2016-11-25 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> (Ported from radeon commits 1106b2f773ad0611c729b27f4c192a26b43ef1e7 and 5fea5ef2f07eee4a0f94baab427010b936f1d4b4) v2: * Squash in radeon fix for TearFree regression * Remove preprocessor guards for compatibility with libdrm < 2.4.72 (Emi

Re: amdgpu support for Polaris11, Radeon Pro 400 series graphics

2016-11-27 Thread Michel Dänzer
On 26/11/16 10:27 PM, Bernhard Froemel wrote: > >> Device: AMD POLARIS11 (DRM 3.8.0 / 4.9.0-rc6, LLVM 3.8.1) (0x67ef) It might be worth trying Mesa 13 compiled against LLVM 3.9. -- Earthling Michel Dänzer | http://www.amd.com Libre software e

Re: [PATCH xf86-video-ati] Replace loop with clz to calculate log base 2 on non-x86 platforms in radeon.h

2016-11-28 Thread Michel Dänzer
using __builtin_clz on x86 as well? AFAICT both gcc and clang seem to generate more or less the same code with that as with the inline assembly. -- 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

[PATCH xf86-video-amdgpu] Add radeon_is_gpu_screen helper

2016-11-28 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This will hopefully decrease the chance of accidentally breaking the build against xserver < 1.13 in the future. (Ported from radeon commit f130b10e63f7526360b41aa0918b4940f63f662a) Signed-off-by: Michel Dänzer <michel.daen...@amd.c

[PATCH xf86-video-amdgpu v2] Add amdgpu_is_gpu_screen helper

2016-11-28 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This will hopefully decrease the chance of accidentally breaking the build against xserver < 1.13 in the future. (Ported from radeon commit f130b10e63f7526360b41aa0918b4940f63f662a) Signed-off-by: Michel Dänzer <michel.daen...@amd.

Re: [PATCH xf86-video-ati] Use finer-grained pointer types in mem copying functions

2016-11-28 Thread Michel Dänzer
for (; nwords > 0; --nwords, ++d, ++s) BTW, this hunk wouldn't apply to current master, looks like it's on top of your previous patch removing the RADEON_HOST_DATA_SWAP_HDW case. Please always either make sure patches you submit apply to current master, or explicitly state depen

[PATCH xf86-video-amdgpu] Don't install Flush/EventCallback for GPU screens

2016-11-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Their purpose is to flush GPU rendering commands corresponding to damage events, but there can be no damage events corresponding to GPU screen rendering operations. (Ported from radeon commit 13c6bc5e382765fe567091e1c616c0a26eec04ca) Sign

Re: gnome-shell is frozen upon wakeup from DPMS (bisected)

2016-11-16 Thread Michel Dänzer
On 15/11/16 06:41 PM, Max Staudt wrote: > On 11/15/2016 10:23 AM, Michel Dänzer wrote: >> Does it still happen with DRI2? You can either disable DRI3 completely with >> >> Option "DRI" "2" > > > Yup, still happens with this option. Feel fr

Re: gnome-shell is frozen upon wakeup from DPMS (bisected)

2016-11-17 Thread Michel Dänzer
On 14/11/16 11:12 PM, Max Staudt wrote: > On 11/14/2016 02:46 AM, Michel Dänzer wrote: >> On 11/11/16 02:21 AM, Max Staudt wrote: >>> Hi, >>> >>> I have bisected a commit in v4.6 that fixes a freeze of the screen on >>> DPMS sleep: >>> &

[ANNOUNCE] xf86-video-ati 7.8.0

2016-11-16 Thread Michel Dänzer
of provider sink / source capabilities Keith Packard (1): Adapt to video API 22 Michel Dänzer (65): Post 7.7.0 release version bump Add support for async flips to radeon_do_pageflip present: Support async flips Handle Zaphod mode correctly in radeon_mode_hotplug

[ANNOUNCE] xf86-video-amdgpu 1.2.0

2016-11-16 Thread Michel Dänzer
Hans de Goede (3): Remove unnecessary fb addition from drmmode_xf86crtc_resize Only add main fb if necessary PRIME: Fix swapping of provider sink / source capabilities Keith Packard (1): Use NotifyFd for drm fd Marek Olšák (1): Fix cursor size for SI Michel Dänzer (65):

Re: [AMDGPU][CIK] PRIME with DRI3 + stalls rendering

2016-11-13 Thread Michel Dänzer
n > running the AMDGPU as dedicated. Please elaborate on what "frame stalls" means exactly. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___

Re: gnome-shell is frozen upon wakeup from DPMS (bisected)

2016-11-13 Thread Michel Dänzer
l still works, > albeit the screen will still be frozen. Run "xterm" and the screen will > unfreeze. Running "xlogo" does not unfreeze it. Does this still happen with current xf86-video-ati Git master? -- Earthling Michel Dänzer | http://www

Re: Bug#839381: xserver-xorg-video-amdgpu: very long Xorg initilization time in recent kernels

2016-11-13 Thread Michel Dänzer
ing a patch which > re enables the boot parameter. > > Furthermore i'm not sure what causes the problem with my GPU in the > first place. If there is anything i can do please let me know. Adding the amd-gfx mailing list to Cc. This bug report should be reassigned to the kernel packa

[PATCH xf86-video-ati] Don't install Flush/EventCallback for GPU screens

2016-11-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Their purpose is to flush GPU rendering commands corresponding to damage events, but there can be no damage events corresponding to GPU screen rendering operations. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/rade

[PATCH xf86-video-ati] Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available

2016-11-21 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 100 -- src/drmmode_display.h | 11 +- src/radeon_dri2.c | 3 +- src/radeon_kms.c | 15 +

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

2016-11-01 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 c7d27c94cb65 ("Keep track of damage event related flushes per-client"). (Ported from radeon commit 5df36de39952c3a26cb2fbc125f298139a9dd

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

2016-11-01 Thread Michel Dänzer
(t->rid == dev->info.pci_rev_id)) Wrong indentation of the second line. With these issues fixed, Reviewed-by: Michel Dänzer <michel.daen...@amd.com> Would be nice to land this in master soon, as we're going to use it in xf86-video-amdgpu:

[PATCH xf86-video-amdgpu 7/7] Remove generated header files

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> No longer used. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/Makefile.am | 7 -- src/amdgpu_chipinfo_gen.h | 194 -- src/amdgpu_chipset_gen.h

[PATCH xf86-video-amdgpu 2/7] Use family information from libdrm_amdgpu / kernel

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_drv.h | 2 +- src/amdgpu_kms.c | 18 +- 2 files changed, 6 insertions(+), 14 deletions(-) diff --git a/src/amdgpu_drv.h b/src/amdgpu_drv.h index fc6a

[PATCH xf86-video-amdgpu 3/7] Stop using generated amdgpu_device_match

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Just match on PCI device ID 0x1002. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_probe.c | 9 ++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/src/amdgpu_probe.c b/src/amdgpu_probe.c i

[PATCH xf86-video-amdgpu 6/7] Stop using AMDGPU(Unique)Chipsets

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Use libdrm_amdgpu's amdgpu_get_marketing_name for the chipset name, or "Unknown AMD Radeon GPU" as a fallback. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_kms.c | 12 src/amdgpu_probe

[PATCH xf86-video-amdgpu 4/7] Remove amdpciids.h

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Not useful anymore. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/Makefile.am | 1 - src/amdgpu_drv.h | 3 +-- src/amdgpu_kms.c | 4 +--- src/amdpciids.h | 39 --- 4 files changed,

[PATCH xf86-video-amdgpu 5/7] Stop using AMDGPUPciChipsets

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Not actually used by Xorg. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_probe.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/src/amdgpu_probe.c b/src/amdgpu_probe.c index fb6f8a3..ac

[PATCH xf86-video-amdgpu 0/7] Remove generated header files

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This series removes the reliance on information in generated header files, and removes the generated header files. The main motivation for this is to remove the need to add PCI ID entries for each supported GPU. Patch 6

[PATCH xf86-video-amdgpu 2/4] Always call PixmapStopDirtyTracking in drmmode_set_scanout_pixmap

2016-10-26 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Otherwise, we may leak screen->pixmap_dirty_list entries if drmmode_set_scanout_pixmap is called repatedly with ppix != NULL, which can happen from RRReplaceScanoutPixmap. (Inspired by xserver commit b773a9c8126222e5fed2904d012fbf917a9f22fd

[PATCH xf86-video-amdgpu 1/4] Don't rely on randr_crtc->scanout_pixmap in drmmode_set_scanout_pixmap

2016-10-26 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> RRReplaceScanoutPixmap may set randr_crtc->scanout_pixmap = NULL before we get here. (Inspired by xserver commit f4c377953df1fe0e3196eda452acf0078e61) v2: Always return TRUE in the if (!ppix) block. (Cherry picked from rade

[PATCH xf86-video-amdgpu 3/4] PRIME: Fix swapping of provider sink / source capabilities

2016-10-26 Thread Michel Dänzer
edhat.com> (Ported from xserver commit 94a1c77259ce39ba59ad87615df39b570ffab435) (Ported from radeon commit 82d3c8f5500d2a6fb1495e217a0b79c396f1534c) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_kms.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

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

2016-10-31 Thread Michel Dänzer
iting another frame before the iGPU is done processing the previous one. Christian, does the amdgpu scheduler wait for shared fences of shared BOs to signal before submitting jobs using them to the GPU? -- Earthling Michel Dänzer | http://www.amd.com Libre software ent

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

2016-10-31 Thread Michel Dänzer
On 29/10/16 02:37 AM, Mario Kleiner wrote: > On 10/28/2016 03:34 AM, Michel Dänzer wrote: >> 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 fur

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

2016-10-31 Thread Michel Dänzer
On 31/10/16 05:00 PM, Christian König wrote: > Am 31.10.2016 um 07:44 schrieb Michel Dänzer: >> On 29/10/16 10:58 PM, Mike Lothian wrote: >>> I turned on vsync and everything works great in tomb raider >>> >>> :D >>> >>> Thanks again to every

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

2016-11-01 Thread Michel Dänzer
The first attached patch will result in drmParsePciDeviceInfo always reporting revision 0 on kernels without the second attached patch. Will that be an issue for the amdgpu-pro stack? Please follow up directly to the patch e-mails with any comments on the patches. -- Earthling Michel Dänzer

[PATCH xf86-video-amdgpu] Check Xorg version at runtime instead of build time in two places

2016-11-01 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This means that all possible paths can be handled as intended, no matter which Xorg version the driver happened to be compiled against. (Ported from radeon commit 350a2645a1b127227ff294c0b62d2d0fd48a) Signed-off-by: Michel Dänzer <mi

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

2016-11-01 Thread Michel Dänzer
On 01/11/16 06:34 PM, Michel Dänzer wrote: > On 13/09/16 12:14 AM, Alex Deucher wrote: >> 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 updat

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

2016-10-11 Thread Michel Dänzer
On 11/10/16 09:04 PM, Christian König wrote: > Am 11.10.2016 um 05:58 schrieb Michel Dänzer: >> On 07/10/16 09:34 PM, Mike Lothian wrote: >>> This has discussion has gone a little quiet >>> >>> Was there any agreement about what needed doing to get this worki

Re: [PATCH] drm/amd/amdgpu: Allow broadcast on debugfs read (v2)

2016-10-11 Thread Michel Dänzer
On 11/10/16 09:32 PM, StDenis, Tom wrote: > It's used by the UMD though they read from 0/*/* when reading the > RASTER_CONFIG registers (which may be a bug...) We should probably clarify what userspace is trying to do there, and whether the hardware actually does that. -- Earthling

Re: mm: fix cache mode tracking in vm_insert_mixed() breaks AMDGPU [was: Re: Latest testing with drm-next-4.9-wip and latest LLVM/mesa stack - Regression in PowerPlay/DPM on CIK?]

2016-10-13 Thread Michel Dänzer
ting a mapping with a > not-allowed cache mode, and now you are (rightfully) getting errors?" It would be nice to get some more specific pointers what amdgpu (or maybe ttm, since that calls vm_insert_mixed in ttm_bo_vm_fault) might be doing wrong. -- Earthling Michel Dänzer

Re: [PATCH] drm/amd/amdgpu: Allow broadcast on debugfs read (v2)

2016-10-13 Thread Michel Dänzer
ey're performed from instance 0, or does it combine the values from all instances with logical and/or? -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___

Re: [PATCH 1/2] drm/radeon: fix up dp aux tear down

2016-10-12 Thread Michel Dänzer
t64_t value) I'd drop this hunk. Either way though, the series is Reviewed-and-Tested-by: Michel Dänzer <michel.daen...@amd.com> -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___

[PATCH xf86-video-amdgpu] Enable HW cursor support with PRIME slave output & Xorg > 1.18.99.901

2016-10-13 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Supported since Xorg 1.18.99.2, but buggy until 1.18.99.901. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/drmmode_dis

Re: [PATCH] drm/amd/amdgpu: Allow broadcast on debugfs read (v2)

2016-10-10 Thread Michel Dänzer
d" is supposed to work? I can only see the register specs talking about broadcast in connection with writes. -- 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.

2016-10-10 Thread Michel Dänzer
On 07/10/16 09:34 PM, Mike Lothian wrote: > > This has discussion has gone a little quiet > > Was there any agreement about what needed doing to get this working > for i965/amdgpu? Christian, do you see anything which could prevent the solution I outlined from working? -- E

[PATCH xf86-video-ati 2/4] Fix radeon_scanout_extents_intersect for GPU screens

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes incorrect screen updates with TearFree enabled on PRIME slave outputs which are not located at (0, 0). Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 20 +++- 1 file changed, 15 inse

[PATCH xf86-video-ati 1/4] Take current scanout_id into account everywhere involved with TearFree

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes various potential issues with TearFree enabled, e.g. outputs freezing after display configuration changes. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 22 +++--- src/radeon.h

[PATCH xf86-video-ati 3/4] Call ValidateGC after ChangeClip in radeon_sync_scanout_pixmaps

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> The wrong order meant that the clipping region wasn't actually applied, so it always copied the full contents from the other scanout pixmap. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 2 +- 1 fil

[PATCH xf86-video-ati 4/4] Call radeon_drm_abort_entry on failure to flip to a scanout pixmap

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes leaking the corresponding struct radeon_drm_queue list entry in that case. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/radeon_kms.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/radeon_kms.c b/src/

Re: [PATCH 2/2] drm/amdgpu: fix cursor setting of dce6/dce8

2016-12-13 Thread Michel Dänzer
dce_v8_0_cursor_move_locked(crtc, x, y); > > - amdgpu_crtc->cursor_width = width; > - amdgpu_crtc->cursor_height = height; > amdgpu_crtc->cursor_hot_x = hot_x; > amdgpu_crtc->cursor_hot_y = hot_y; >

Re: [RFC] Using DC in amdgpu for upcoming GPU

2016-12-13 Thread Michel Dänzer
amdgpu-pro for > enterprise/workstation customers. The pipe dream is replacing our Xorg drivers with -modesetting. I fully agree with you Daniels when it comes to non-Xorg userspace. > In the end AMD missing that seems just another case of designing

Re: [PATCH] udev_monitor_receive_device() will block when hotplug monitor

2016-12-13 Thread Michel Dänzer
pushed, thanks! P.S. Please run git config --global user.name "Jim Qu" git config email.prefix "PATCH $(basename $PWD)" in your Git tree. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast

[PATCH xf86-video-amdgpu] Simplify drmmode_handle_uevents

2016-12-14 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> No functional change intended. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 14 -- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/src/drmmode_display.c b/src/drmmode_dis

Re: [V3 04/11] drm/amdgpu/virt: use kiq to access registers

2017-01-11 Thread Michel Dänzer
On 12/01/17 12:21 PM, Liu, Monk wrote: > >if (in_interrupt()) > BUG(); Current coding style is BUG_ON(in_interrupt()); according to https://kernelnewbies.org/FAQ/BUG . -- Earthling Michel Dänzer | http://www.amd.com Libre software e

Re: [PATCH] Revert "drm/amdgpu: Only update the CUR_SIZE register when necessary"

2017-01-12 Thread Michel Dänzer
On 12/01/17 02:35 AM, Deucher, Alexander wrote: >> -Original Message- >> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf >> Of Michel Dänzer >> Sent: Wednesday, January 11, 2017 4:36 AM >> To: amd-gfx@lists.freedesktop.org >

[PATCH] Revert "drm/amdgpu: Only update the CUR_SIZE register when necessary"

2017-01-11 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> This reverts commits 7c83d7abc9997cf1efac2c0ce384b5e8453ee870 and a1f49cc179ce6b7b7758ae3ff5cdb138d0ee0f56. They caused the HW cursor to disappear under various circumstances in the wild. I wasn't able to reproduce any of them, and I'm not sure

Re: [PATCH] drm/amd/amdgpu: validate the shadow BO.

2016-12-01 Thread Michel Dänzer
, this can be simplified to Reported-and-Tested-by: Nicolai Haehnle <nicolai.haeh...@amd.com> :) -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer __

[PATCH xf86-video-amdgpu 4/4] Call amdgpu_drm_abort_entry on failure to flip to a scanout pixmap

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes leaking the corresponding struct amdgpu_drm_queue list entry in that case. (Ported from radeon commit e2942449171fe628a7726e59bcaab65e27d88563) Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/amdgpu_kms.c | 2 ++ 1

[PATCH xf86-video-amdgpu 2/4] Fix amdgpu_scanout_extents_intersect for GPU screens

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes incorrect screen updates with TearFree enabled on PRIME slave outputs which are not located at (0, 0). (Ported from radeon commit a995f5830916a0fee5126263d1bfe48632be3a15) Signed-off-by: Michel Dänzer <michel.daen...@amd.com&

[PATCH xf86-video-amdgpu 1/4] Take current scanout_id into account everywhere involved with TearFree

2016-11-29 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Fixes various potential issues with TearFree enabled, e.g. outputs freezing after display configuration changes. (Ported from radeon commit e543ef3a2fb304cbe3a965fb780632af2e4186f4) Signed-off-by: Michel Dänzer <michel.daen...@amd.com&

Re: [PATCH] drm/amdgpu: don't add files at control minor debugfs directory

2016-12-05 Thread Michel Dänzer
g/show_bug.cgi?id=98915 > Fixes: 8a357d10043c ("drm: Nerf DRM_CONTROL nodes") > Signed-off-by: Alex Deucher <alexander.deuc...@amd.com> Regardless of the above, Reviewed-by: Michel Dänzer <michel.daen...@amd.com> -- Earthlin

[PATCH] drm/amdgpu/dce6: Set MASTER_UPDATE_MODE to 0 in resume_mc_access as well

2016-12-06 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Looks like this was missed when dce_v6_0.c was added. Fixes: e2cdf640cbb5 ("drm/amdgpu: add display controller implementation for si v10") Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- drivers/gpu/drm/amd/amdgpu

Re: [PATCH xf86-video-ati] Replace loop with clz to calculate log base 2 on non-x86 platforms in radeon.h

2016-11-30 Thread Michel Dänzer
On 01/12/16 02:52 AM, Jochen Rollwagen wrote: > Am 29.11.2016 um 08:32 schrieb Michel Dänzer: >> On 29/11/16 03:18 AM, Jochen Rollwagen wrote: >>> This commit replaces the loop for calculating log base 2 for >>> non-x86-platforms in radeon.h with a clz (count leadin

Re: [PATCH] drm/amdgpu: add flag for high priority contexts

2017-01-04 Thread Michel Dänzer
s probably better to hold off on merging this upstream until you have at least a proof of concept showing that it's actually useful in practice. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer

[PATCH xf86-video-amdgpu] Pass pitch from drmmode_crtc_scanout_allocate to drmmode_create_bo_pixmap

2017-01-06 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> Mostly to align with radeon commit ea30d856ba5e7274c8ea499293b8b0e721b8e082, but also gets rid of a gbm_bo_get_stride call. Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_display.c | 22 ++ 1

[PATCH xf86-video-ati] Pass pitch from drmmode_crtc_scanout_allocate to drmmode_create_bo_pixmap

2017-01-06 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> radeon_bo_get_tiling can fail, e.g. for linear BOs. Fixes TearFree not working in those cases. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97873 Signed-off-by: Michel Dänzer <michel.daen...@amd.com> --- src/drmmode_di

[PATCH xf86-video-amdgpu] Don't set modes before AMDGPUWindowExposures_oneshot is called

2017-03-23 Thread Michel Dänzer
From: Michel Dänzer <michel.daen...@amd.com> The root window contents may be undefined before that, so we don't to show anything yet. Fixes a crash on startup with rotation and virtual resolution set in xorg.conf. Bugzilla: https://bugs.freedesktop.org/100276 Signed-off-by: Michel

Re: [PATCH libdrm 2/7] amdgpu: update amdgpu_drm.h for Vega10

2017-03-23 Thread Michel Dänzer
On 22/03/17 07:13 PM, Marek Olšák wrote: > On Mar 22, 2017 2:44 AM, "Michel Dänzer" <mic...@daenzer.net > <mailto:mic...@daenzer.net>> wrote: >> On 22/03/17 06:46 AM, Marek Olšák wrote: >>> On Tue, Mar 21, 2017 at 10:27 PM, Nicolai Hähnle >>

Re: [PATCH] Revert "drm/radeon: Try evicting from CPU accessible to inaccessible VRAM first"

2017-03-23 Thread Michel Dänzer
solution. -- 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] Revert "drm/radeon: Try evicting from CPU accessible to inaccessible VRAM first"

2017-03-24 Thread Michel Dänzer
ned at the end of his reply. I'll follow up to his post about that. > But in general, isn't "radeon_lockup_timeout" supposed to detect this > situation ? No, it's for detecting GPU hangs, whereas this is a CPU "hang" (infinite loop). -- Earthling Michel Dänzer

Re: [PATCH] Revert "drm/radeon: Try evicting from CPU accessible to inaccessible VRAM first"

2017-03-24 Thread Michel Dänzer
&& bo->mem.start < (rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT)) { unsigned fpfn = rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT; int i; -- Earthling Miche

Re: [PATCH] Revert "drm/radeon: Try evicting from CPU accessible to inaccessible VRAM first"

2017-03-28 Thread Michel Dänzer
2.), our test reproduces the hard > lockup which one does not give any info in kern.log (sometimes some NUL > ^@ characters but not always). What exactly does "hard lockup" mean? What are the symptoms? -- Earthling Michel Dänzer | http://www.amd.c

Re: Plan: BO move throttling for visible VRAM evictions

2017-03-28 Thread Michel Dänzer
On 28/03/17 05:29 PM, Christian König wrote: > Am 28.03.2017 um 08:00 schrieb Michel Dänzer: >> On 28/03/17 12:50 PM, zhoucm1 wrote: >>> On 2017年03月28日 10:40, Michel Dänzer wrote: >>>> On 27/03/17 04:53 PM, Zhou, David(ChunMing) wrote: >>>>> For APU sp

Re: [PATCH] Revert "drm/radeon: Try evicting from CPU accessible to inaccessible VRAM first"

2017-03-26 Thread Michel Dänzer
On 25/03/17 03:59 AM, Julien Isorce wrote: > Hi Michel, > > I double checked and you are right, the change 0 -> i works. Thanks for testing, fix submitted for review. -- Earthling Michel Dänzer | http://www.amd.com Libre softwar

Re: Plan: BO move throttling for visible VRAM evictions

2017-03-27 Thread Michel Dänzer
ciple in some cases (e.g. for all BOs except those with AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) or even always. Implementing eviction from CPU visible to CPU invisible VRAM, similar to how it's done in radeon. Note that there's potential for userspace triggering an infinite loop i

Re: [PATCH] Fix IB va_start+ib_bytes range check on 32Bit systems

2017-03-29 Thread Michel Dänzer
(chunk_ib->va_start + chunk_ib->ib_bytes) > > (u64)(m->it.last + 1) * AMDGPU_GPU_PAGE_SIZE) { That won't work correctly if m->it.last == 0x ? Or is that not possible? -- 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 3/3] drm/amdgpu: Fix multi-level page table bugs for large BOs

2017-03-29 Thread Michel Dänzer
but it won't cause crazy log spam if the bad condition is consistently hit. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer ___ amd-gfx mailing list amd-gfx@lists

Re: [PATCH 3/3] drm/amdgpu: do not block gpu resume when cp ring test failed.

2017-03-30 Thread Michel Dänzer
e, but at least the commit log would need to be changed not to talk about the KIQ related random failures but just about that gfx_v8_0_kiq_resume would only return failure if the last ring test failed anyway. -- Earthling Michel Dänzer | http://www.amd.com Libre soft

Re: Plan: BO move throttling for visible VRAM evictions

2017-03-30 Thread Michel Dänzer
attached proof-of-concept patch (on top of Christian's "CPU mapping of split VRAM buffers" series, ported from radeon) results in 145.05 fps on my Tonga. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast |

Re: CPU mapping of split VRAM buffers

2017-03-30 Thread Michel Dänzer
understand all changes in there, so I only feel comfortable giving Acked-by: Michel Dänzer <michel.daen...@amd.com> for them. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast |

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