Re: [Freedreno] linux-4.7-rc3/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c: 2 * pointless tests ?

2016-06-13 Thread Rob Clark
yup, looks like we can drop the two pipe<0 checks. Care to send a patch? BR, -R On Mon, Jun 13, 2016 at 10:51 AM, David Binderman wrote: > Hello there, > > 1. > > linux-4.7-rc3/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c:545]: (style) > Checking if unsigned variable

[Freedreno] [PATCH 09/10] drm/msm/rd: split out snapshot_buf helper

2016-06-16 Thread Rob Clark
(reduce the noise in next patch) Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_rd.c | 36 ++-- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 3

[Freedreno] [PATCH 07/10] drm/msm: deal with arbitrary # of cmd buffers

2016-06-16 Thread Rob Clark
For some optimizations coming on the userspace side, splitting larger draw or gmem cmds into multiple cmdstream buffers, we need to support much more than the previous small/arbitrary limit. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_gem.h

[Freedreno] [PATCH 01/10] drm/msm: use mutex_lock_interruptible for submit ioctl

2016-06-16 Thread Rob Clark
Be kinder to things that do lots of signal handling (ie. Xorg) Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_gem_submit.c | 13 + 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/d

[Freedreno] [PATCH 06/10] drm/msm: wire up vmap shrinker

2016-06-16 Thread Rob Clark
Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_drv.h | 2 ++ drivers/gpu/drm/msm/msm_gem.c | 25 - drivers/gpu/drm/msm/msm_gem.h | 10 + drivers/gpu/drm/msm/msm_gem_shrinker.

[Freedreno] [PATCH 10/10] drm/msm/rd: add module param to dump all bo's

2016-06-16 Thread Rob Clark
By default, if using $debugfs/.../rd to log cmdstream, only the cmdstream buffers themselves are logged. But in some cases we want to capture other buffers in the submit (to see VBO's or shaders). So add a mod-param knob to control this. Signed-off-by: Rob Clark <robdcl...@gmail.

Re: [Freedreno] [Mesa-dev] tilers and out-of-order rendering..

2016-05-20 Thread Rob Clark
On Fri, May 20, 2016 at 3:35 AM, Jose Fonseca <jfons...@vmware.com> wrote: > On 20/05/16 00:34, Rob Clark wrote: >> >> On Thu, May 19, 2016 at 6:21 PM, Eric Anholt <e...@anholt.net> wrote: >>> >>> Rob Clark <robdcl...@gmail.com> writes: >>

Re: [Freedreno] OpenCL - Adreno GPU(a4xx)

2016-08-01 Thread Rob Clark
Hi Lucas Franco, There is no compute support yet, although at least a good chunk of the reverse engineering work is already done. It is probably the most requested feature. Not sure that I'll have time to work on compute myself in the next few months, but if no one beats me to it I'll do it

Re: [Freedreno] OpenCL - Adreno GPU(a4xx)

2016-08-01 Thread Rob Clark
t; > If I want to run OpenCL on linaro, I'll have to change my SBC, isn't it? > > Thanks > Lucas Franco > > 2016-08-01 9:22 GMT-03:00 Rob Clark <robdcl...@gmail.com>: >> >> Hi Lucas Franco, >> >> There is no compute support yet, although at least a good chun

[Freedreno] [PATCH 05/12] freedreno: push resource tracking down into batch

2016-07-02 Thread Rob Clark
From: Rob Clark <robcl...@freedesktop.org> Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/freedreno/freedreno_batch.c| 35 ++ src/gallium/drivers/freedreno/freedreno_batch.h| 7 + src/gallium/drivers/freedreno/freedreno_con

[Freedreno] [PATCH 07/12] freedreno: move more batch related tracking to fd_batch

2016-07-02 Thread Rob Clark
). The alternative would be having to re-calculate GMEM layout on every batch, even if the dimensions of the render targets are the same. Note: This opens up the possibility of pushing gmem/submit into a helper thread. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/fre

[Freedreno] [PATCH 11/12] freedreno: shadow textures if possible to avoid stall/flush

2016-07-02 Thread Rob Clark
reference is held by cmdstream) as a shadow. This is done by blitting the remaining other levels (and whatever part of current level that is not discarded) from the old/shadow buffer to the new one. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/fre

Re: [Freedreno] Use of copy_from_user in msm_gem_submit.c while holding a spin_lock

2016-08-17 Thread Rob Clark
On Wed, Aug 17, 2016 at 7:40 AM, Vaishali Thakkar wrote: > Hello, > > I was wondering about the call to copy_from_user in function > submit_lookup_objects for drive > /gpu/drm/msm/msm_gem_submit.c It calls copy_from_user[1] in a spin_lock, > which is not normally >

Re: [Freedreno] [PATCH 2/4] drm/msm: drop qcom,chipid

2017-02-01 Thread Rob Clark
On Wed, Feb 1, 2017 at 12:09 PM, Rob Herring <r...@kernel.org> wrote: > On Mon, Jan 30, 2017 at 11:49:19AM -0500, Rob Clark wrote: >> The original way we determined the gpu version was based on downstream >> bindings from android kernel. A cleaner way is to get the version f

Re: [Freedreno] WebProcess crash on DB410c

2017-02-02 Thread Rob Clark
On Thu, Feb 2, 2017 at 2:13 AM, Sivasubramanian Patchaiperumal wrote: > Hi, > I'm trying to port WPE on DB410c with Westeros compositor, but the > webprocess crashes due to null sharingcontext. Webprocess fails to create gl > context as eglChooseConfig

Re: [Freedreno] WebProcess crash on DB410c

2017-02-02 Thread Rob Clark
wrote: > Westeros code uses EGL window surface only, but the WPE code (at > https://github.com/Metrological/WebKitForWayland/) which uses pbuffer works > on HiKey and RPI as mentioned. > > On 2 February 2017 at 17:38, Rob Clark <robdcl...@gmail.com> wrote: >>

Re: [Freedreno] WebProcess crash on DB410c

2017-02-02 Thread Rob Clark
so I think the answer is don't use pbuffers. BR, -R On Thu, Feb 2, 2017 at 9:50 AM, Rob Clark <robdcl...@gmail.com> wrote: > hmm, tons of older stuff uses pbuffers w/ x11.. although a quick look > at mesa/demos.git and it doesn't look like any of them that build for > wayland do.

Re: [Freedreno] WebProcess crash on DB410c

2017-02-02 Thread Rob Clark
on DB410c. Any client that uses > pbuffer surfaces will crash I suspect. Is there is any simple egl > application that uses pixel buffer to verify and confirm? > > On 2 February 2017 at 19:00, Rob Clark <robdcl...@gmail.com> wrote: >> >> hmm, ok, so it is a *client*

Re: [Freedreno] [PATCH 03/11] drm/msm: Add hint to DRM_IOCTL_MSM_GEM_INFO to return an object IOVA

2017-02-06 Thread Rob Clark
On Mon, Feb 6, 2017 at 2:20 PM, Emil Velikov wrote: > Hi Jordan, > > On 6 February 2017 at 17:39, Jordan Crouse wrote: >> Modify the 'pad' member of struct drm_msm_gem_info to 'hint'. If the >> user sets 'hint' to non-zero it means that they want

Re: [Freedreno] [Intel-gfx] [PATCH 00/11] drm/msm: A5XX preemption

2017-02-06 Thread Rob Clark
On Mon, Feb 6, 2017 at 1:23 PM, Daniel Stone wrote: > Hi, > > On 6 February 2017 at 17:59, Daniel Vetter wrote: >> On Mon, Feb 06, 2017 at 10:39:28AM -0700, Jordan Crouse wrote: >>> This initial series implements 4 ringbuffers to give sufficient coverage

[Freedreno] [pull] drm/msm: msm-next for 4.11

2017-02-06 Thread Rob Clark
msm/dsi: Return more timings from PHY to host drm/msm/dsi: Pass down use case to PHY drm/msm/dsi: Move PHY operations out of host drm/msm/dsi: Add new method to calculate 14nm PHY timings Rob Clark (5): drm/msm: remove qcom,gpu-pwrlevels bindings drm/msm: drop qcom,chi

Re: [Freedreno] WebProcess crash on DB410c

2017-02-03 Thread Rob Clark
gt;> reference now. I'm trying to write a simple egl app that uses pbuffer to >> confirm the support with Mesa. Does it sounds correct or you have any >> suggestions? >> >> On 3 February 2017 at 02:06, Rob Clark <robdcl...@gmail.com> wrote: >>> >&g

Re: [Freedreno] [RFC] drm/msm/adreno: clean up gpu bindings

2017-01-24 Thread Rob Clark
On Tue, Jan 24, 2017 at 1:01 PM, Jordan Crouse <jcro...@codeaurora.org> wrote: > On Tue, Jan 24, 2017 at 12:11:32PM -0500, Rob Clark wrote: >> So, cleaning up the GPU bindings is something that has been on my TODO >> list for a while, but always $bigger_fires. Existing bindi

Re: [Freedreno] [RFC] drm/msm/adreno: clean up gpu bindings

2017-01-26 Thread Rob Clark
On Thu, Jan 26, 2017 at 2:11 PM, Rob Herring <r...@kernel.org> wrote: > On Tue, Jan 24, 2017 at 11:11 AM, Rob Clark <robdcl...@gmail.com> wrote: >> So, cleaning up the GPU bindings is something that has been on my TODO >> list for a while, but always $bigger_fires.

[Freedreno] [PATCH 1/4] drm/msm: remove qcom, gpu-pwrlevels bindings

2017-01-30 Thread Rob Clark
with downstream android dt files. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- Documentation/devicetree/bindings/display/msm/gpu.txt | 15 --- drivers/gpu/drm/msm/adreno/adreno_device.c| 6 -- 2 files changed, 4 insertions(+), 17 deletions(-) diff

[Freedreno] [PATCH 4/4] drm/msm: drop _clk suffix from clk names

2017-01-30 Thread Rob Clark
Suggested by Rob Herring. We still support the old names for compatibility with downstream android dt files. Cc: Rob Herring <r...@kernel.org> Signed-off-by: Rob Clark <robdcl...@gmail.com> --- Documentation/devicetree/bindings/display/msm/gpu.txt | 12 ++-- drivers

[Freedreno] [PATCH 2/4] drm/msm: drop qcom,chipid

2017-01-30 Thread Rob Clark
), so that we are still compatible with the gpu dt node from android device kernels. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- .../devicetree/bindings/display/msm/gpu.txt| 11 +++--- drivers/gpu/drm/msm/adreno/adreno_device.c | 43 +- drivers/gpu/d

[Freedreno] [PATCH 0/4] drm/msm: cleanup gpu bindings

2017-01-30 Thread Rob Clark
for some devices, it is time to clean things up so we can start adding this missing gpu nodes. Note that these patches preserve compatibility with downstream dt files because, at this point, it is easy and convenient to not break people's patchsets for upstream support of various devices. Rob

[Freedreno] [PATCH 3/4] drm/msm: drop quirks binding

2017-01-30 Thread Rob Clark
This was never documented or used in upstream dtb. It is used by downstream bindings from android device kernels. But the quirks are a property of the gpu revision, and as such are redundant to be listed separately in dt. Instead, move the quirks to the device table. Signed-off-by: Rob Clark

Re: [Freedreno] [PATCH 1/4] drm/msm: remove qcom, gpu-pwrlevels bindings

2017-01-30 Thread Rob Clark
On Mon, Jan 30, 2017 at 1:21 PM, Eric Anholt <e...@anholt.net> wrote: > Rob Clark <robdcl...@gmail.com> writes: > >> The plan is to use the OPP bindings. For now, remove the documentation >> for qcom,gpu-pwrlevels, and make the driver fall back to a safe low >&

Re: [Freedreno] [PATCH 2/4] drm/msm: drop qcom,chipid

2017-01-30 Thread Rob Clark
On Mon, Jan 30, 2017 at 1:09 PM, Eric Anholt <e...@anholt.net> wrote: > Rob Clark <robdcl...@gmail.com> writes: > >> The original way we determined the gpu version was based on downstream >> bindings from android kernel. A cleaner way is to get the version

Re: [Freedreno] [PATCH 4/4] drm/msm: drop _clk suffix from clk names

2017-01-30 Thread Rob Clark
On Mon, Jan 30, 2017 at 1:15 PM, Eric Anholt <e...@anholt.net> wrote: > Rob Clark <robdcl...@gmail.com> writes: > >> Suggested by Rob Herring. We still support the old names for >> compatibility with downstream android dt files. >> >> Cc: Rob Herring <

Re: [Freedreno] [RFC] drm/msm/adreno: clean up gpu bindings

2017-01-26 Thread Rob Clark
On Thu, Jan 26, 2017 at 4:09 PM, Rob Herring <r...@kernel.org> wrote: > On Thu, Jan 26, 2017 at 1:51 PM, Rob Clark <robdcl...@gmail.com> wrote: >> On Thu, Jan 26, 2017 at 2:11 PM, Rob Herring <r...@kernel.org> wrote: >>> On Tue, Jan 24, 2017 at 11:11 AM, R

[Freedreno] [PATCH] drm/msm: move submit fence wait out of struct_mutex

2017-02-25 Thread Rob Clark
outside of struct_mutex as well. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_gem_submit.c | 39 +--- 1 file changed, 18 insertions(+), 21 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/d

Re: [Freedreno] [PATCH] drm/msm: move submit fence wait out of struct_mutex

2017-02-25 Thread Rob Clark
On Sat, Feb 25, 2017 at 12:36 PM, Daniel Vetter <dan...@ffwll.ch> wrote: > On Sat, Feb 25, 2017 at 4:43 PM, Rob Clark <robdcl...@gmail.com> wrote: >> Probably a symptom of needing finer grained locking, but if we wait on >> the incoming fence-fd (which could come from

Re: [Freedreno] Use of copy_from_user in msm_gem_submit.c while holding a spin_lock

2016-08-17 Thread Rob Clark
On Wed, Aug 17, 2016 at 3:15 PM, Al Viro <v...@zeniv.linux.org.uk> wrote: > On Wed, Aug 17, 2016 at 02:49:32PM -0400, Rob Clark wrote: > >> I'm not saying that I shouldn't fix it (although not quite sure how >> yet.. taking/dropping the spinlock inside the loop is n

Re: [Freedreno] Use of copy_from_user in msm_gem_submit.c while holding a spin_lock

2016-08-17 Thread Rob Clark
On Wed, Aug 17, 2016 at 2:49 PM, Rob Clark <robdcl...@gmail.com> wrote: > On Wed, Aug 17, 2016 at 1:08 PM, Al Viro <v...@zeniv.linux.org.uk> wrote: >> On Wed, Aug 17, 2016 at 11:08:46AM -0400, Rob Clark wrote: >>> On Wed, Aug 17, 2016 at 7:40 AM, Vaishali Thakkar &g

Re: [Freedreno] [patch v2] drm/msm/dsi: free first element on error

2017-02-27 Thread Rob Clark
M +0200, Jani Nikula wrote: >>>> On Thu, 16 Feb 2017, Rob Clark <robdcl...@gmail.com> wrote: >>>> > On Thu, Feb 16, 2017 at 7:00 AM, Dan Carpenter >>>> > <dan.carpen...@oracle.com> wrote: >>>> >> We're off by one here.

Re: [Freedreno] [PATCH] drm/msm/dsi: Fix the releasing of resources in error path in 'dsi_bus_clk_enable()'

2017-02-26 Thread Rob Clark
ered the error, which is a no-op, and leave 'msm_host->bus_clks[0]' > untouched. > > Shift by one the index array to free resources correctly. > > Signed-off-by: Christophe JAILLET <christophe.jail...@wanadoo.fr> Reviewed-by: Rob Clark <robdcl...@gmail.com> > ---

Re: [Freedreno] [PATCH 3/3] gpu: drm: drivers: Convert printk(KERN_ to pr_

2017-02-28 Thread Rob Clark
On Tue, Feb 28, 2017 at 7:55 AM, Joe Perches <j...@perches.com> wrote: > Use a more common logging style. > > Miscellanea: > > o Coalesce formats and realign arguments > o Neaten a few macros now using pr_ > > Signed-off-by: Joe Perches <j...@perches.com> fo

Re: [Freedreno] [pull] drm/msm: msm-next for 4.9

2016-09-20 Thread Rob Clark
On Tue, Sep 20, 2016 at 10:39 AM, Jordan Crouse <jcro...@codeaurora.org> wrote: > So this isn't how I planned on announcing it, but hey, I'm back. > >> Rob Clark (5): >> drm/msm: extend the submit ioctl to pass in flags > > Renaming 'pipe' to 'flags' would break

[Freedreno] [pull] drm/msm: msm-next for 4.9

2016-09-17 Thread Rob Clark
: mark symbols static where possible Chris Wilson (1): drm/msm: Remove call to reservation_object_test_signaled_rcu before wait Rob Clark (5): drm/msm: extend the submit ioctl to pass in flags drm/msm: submit support for in-fences drm/msm: move fence allocation out

Re: [Freedreno] [PATCH 04/16] drm: msm: Flush the cache immediately after allocating pages

2016-11-07 Thread Rob Clark
On Mon, Nov 7, 2016 at 3:35 AM, Archit Taneja <arch...@codeaurora.org> wrote: > > > On 11/06/2016 07:45 PM, Rob Clark wrote: >> >> On Fri, Nov 4, 2016 at 6:44 PM, Jordan Crouse <jcro...@codeaurora.org> >> wrote: >>> >>> For

Re: [Freedreno] [PATCH 1/5] drm/msm/mdp5: introduce mdp5_hw_pipe

2016-11-07 Thread Rob Clark
On Mon, Nov 7, 2016 at 5:38 AM, Archit Taneja <arch...@codeaurora.org> wrote: > > > On 11/05/2016 09:55 PM, Rob Clark wrote: >> >> Split out the hardware pipe specifics from mdp5_plane. To start, the hw >> pipes are statically assigned to planes, but next step is

Re: [Freedreno] [PATCH 04/16] drm: msm: Flush the cache immediately after allocating pages

2016-11-06 Thread Rob Clark
On Fri, Nov 4, 2016 at 6:44 PM, Jordan Crouse wrote: > For reasons that are not entirely understood using dma_map_sg() > for nocache/write combine buffers doesn't always successfully flush > the cache after the memory is zeroed somewhere deep in the bowels > of the shmem

[Freedreno] [pull] drm/msm: fixes for 4.9

2016-11-06 Thread Rob Clark
drm/msm: Set CLK_IGNORE_UNUSED flag for PLL clocks drm/msm: Fix error handling crashes seen when VRAM allocation fails Rob Clark (3): drm/msm/mdp5: handle non-fullscreen base plane case drm/msm/mdp5: no scaling support on RGBn pipes for 8x16 drm/msm/mdp5: 8x16 actually has

[Freedreno] [PATCH 5/5] drm/msm/mdp5: handle SMP block allocations "atomically"

2016-11-05 Thread Rob Clark
flush bits and releasing old blocks after vblank. (The SMP registers and SMP allocation is not double-buffered, so newly allocated blocks need to be updated in kms->prepare_commit() released blocks in kms->complete_commit().) Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu

[Freedreno] [PATCH 4/5] drm/msm/mdp5: dynamically assign hw pipes to planes

2016-11-05 Thread Rob Clark
, and we don't want global state to become out of sync with the plane state if an atomic update fails, we hit deadlock/ backoff scenario, etc. The use of state_lock keeps multiple parallel updates which both re-assign hwpipes properly serialized. Signed-off-by: Rob Clark <robdcl...@gmail.

[Freedreno] [PATCH 0/5] drm/msm/mdp5: atomic smp + hwpipe assignment

2016-11-05 Thread Rob Clark
is handled, preserving consistency between global and per-plane state. Rob Clark (5): drm/msm/mdp5: introduce mdp5_hw_pipe drm/msm: subclass drm_atomic_state drm/msm/mdp5: add skeletal mdp5_state drm/msm/mdp5: dynamically assign hw pipes to planes drm/msm/mdp5: handle SMP block a

[Freedreno] [PATCH 3/5] drm/msm/mdp5: add skeletal mdp5_state

2016-11-05 Thread Rob Clark
multiple threads trying to update state. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c | 43 + drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h | 22 + 2 files changed, 65 insertions(+) diff --git a/drivers/g

[Freedreno] [PATCH 2/5] drm/msm: subclass drm_atomic_state

2016-11-05 Thread Rob Clark
This will give the kms backends a slot to stash their own hw specific global state. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_atomic.c | 31 +++ drivers/gpu/drm/msm/msm_drv.c| 3 +++ drivers/gpu/drm/msm/msm_drv.h

Re: [Freedreno] [PATCH 13/16] drm/msm: gpu: Add support for the GPMU

2016-11-07 Thread Rob Clark
On Mon, Nov 7, 2016 at 7:58 AM, Stanimir Varbanov wrote: >> + /* Turn off protected mode for this operation */ >> + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); > > This looks wrong, shouldn't it be? no, it is actually correct, the 3rd arg is # of dwords of

Re: [Freedreno] [PATCH 09/16] drm/msm: gpu Add new gpu register read/write functions

2016-11-07 Thread Rob Clark
On Fri, Nov 4, 2016 at 6:44 PM, Jordan Crouse wrote: > Add some new functions to manipulate GPU registers. gpu_read64 and > gpu_write64 can read/write a 64 bit value to two 32 bit registers. > For 4XX and older these are normally perfcounter registers, but > future

Re: [Freedreno] [PATCH 05/16] drm/msm: gpu: Return error on hw_init failure

2016-11-07 Thread Rob Clark
On Fri, Nov 4, 2016 at 6:44 PM, Jordan Crouse wrote: > When the GPU hardware init function fails (like say, ME_INIT timed > out) return error instead of blindly continuing on. This gives us > a small chance of saving the system before it goes boom. seems like a good

Re: [Freedreno] [PATCH 07/16] drm/msm: Add adreno_gpu_write64()

2016-11-07 Thread Rob Clark
On Fri, Nov 4, 2016 at 6:44 PM, Jordan Crouse wrote: > Add a new generic function to write a "64" bit value. This isn't > actually a 64 bit operation, it just writes the upper and lower > 32 bit of a 64 bit value to a specified LO and HI register. If > a particular target

Re: [Freedreno] [PATCH 3/5] drm/msm/mdp5: add skeletal mdp5_state

2016-11-07 Thread Rob Clark
On Mon, Nov 7, 2016 at 5:29 AM, Archit Taneja <arch...@codeaurora.org> wrote: > > > On 11/05/2016 09:55 PM, Rob Clark wrote: >> >> Add basic state duplication/apply mechanism. Following commits will >> move actual global hw state into this. >> >> The

[Freedreno] [PATCH] drm/msm: update uapi header license

2016-10-19 Thread Rob Clark
Velikov <emil.l.veli...@gmail.com> Cc: Gabriel Laskar <gabr...@lse.epita.fr> Cc: Mikko Rapeli <mikko.rap...@iki.fi> Signed-off-by: Rob Clark <robdcl...@gmail.com> --- include/uapi/drm/msm_drm.h | 25 - 1 file changed, 16 insertions(+), 9 deletions(-) d

[Freedreno] [PATCH 2/6] drm: add helper for printing to log or seq_file

2016-10-14 Thread Rob Clark
Sometimes it is nice not to duplicate equivalent printk() and seq_printf() code. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/Makefile| 3 ++- drivers/gpu/drm/drm_print.c | 54 +++ include/drm/drm_print.h

[Freedreno] [PATCH 1/6] drm: helper macros to print composite types

2016-10-14 Thread Rob Clark
I'll want to print things in a similar way in a later patch. This will make it easier. TODO drm_rect_debug_print() doesn't have many call sites, and is kind of unnecessary now. Should we just drop it? Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/drm_modes.

[Freedreno] [PATCH 3/6] drm: add helpers to go from plane state to drm_rect

2016-10-14 Thread Rob Clark
Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/drm_simple_kms_helper.c | 14 ++ drivers/gpu/drm/i915/intel_atomic_plane.c | 10 ++ drivers/gpu/drm/mediatek/mtk_drm_plane.c| 15 ++- drivers/gpu/drm/rockchip/rockchip_drm_vop.

Re: [Freedreno] [PATCH] drm/msm/mdp5: handle non-fullscreen base plane case

2016-10-15 Thread Rob Clark
On Sat, Oct 15, 2016 at 5:39 AM, Archit Taneja <arch...@codeaurora.org> wrote: > > On 10/13/2016 10:18 PM, Rob Clark wrote: >> >> If the bottom-most layer is not fullscreen, we need to use the BASE >> mixer stage for solid fill (ie. MDP5_CTL_BLEND_OP_FLAG_BORDER_OU

Re: [Freedreno] [PATCH 2/6] drm: add helper for printing to log or seq_file

2016-10-24 Thread Rob Clark
On Mon, Oct 17, 2016 at 1:00 PM, Sean Paul <seanp...@chromium.org> wrote: > On Fri, Oct 14, 2016 at 7:55 PM, Rob Clark <robdcl...@gmail.com> wrote: >> Sometimes it is nice not to duplicate equivalent printk() and >> seq_printf() code. >> >> Signed

[Freedreno] [PATCH 2/7] freedreno/a4xx: use new shader linkage helper

2016-11-28 Thread Rob Clark
Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/freedreno/a4xx/fd4_program.c | 43 +--- 1 file changed, 16 insertions(+), 27 deletions(-) diff --git a/src/gallium/drivers/freedreno/a4xx/fd4_program.c b/src/gallium/drivers/freedreno/a4xx/fd4_pro

[Freedreno] [PATCH 1/7] freedreno/ir3: add new helper for shader linkage

2016-11-28 Thread Rob Clark
Helps simplify things on a5xx, where pos/psize get added to the vs-out map. And anyways, simplifies a3xx and a4xx. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/freedreno/ir3/ir3_shader.h | 44 ++ 1 file changed, 44 insertions(+) diff

[Freedreno] [PATCH 4/7] freedreno/ir3: don't offset inloc by 8

2016-11-28 Thread Rob Clark
On a3xx/a4xx, the SP_VS_VPC_DST_REG.OUTLOCn is offset by 8, so we used to add this offset into fs->inputs[n].inloc. But a5xx drops this extra offset-by-8. So instead make inloc zero based and add the offset when we emit OUTLOCn values (for the gen's that need the offset). Signed-off-by:

[Freedreno] [PATCH 5/7] freedreno: make gmem tile size alignment configurable

2016-11-28 Thread Rob Clark
a5xx seems to prefer 64 pixel alignment, in at least some cases. Make this configurable per generation. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- src/gallium/drivers/freedreno/freedreno_gmem.c | 18 ++ src/gallium/drivers/freedreno/freedreno_screen.c | 6 +

[Freedreno] [PATCH 0/7] freedreno: initial a5xx support

2016-11-28 Thread Rob Clark
://lists.freedesktop.org/archives/freedreno/2016-November/000805.html Rob Clark (7): freedreno/ir3: add new helper for shader linkage freedreno/a4xx: use new shader linkage helper freedreno/a3xx: use new shader linkage helper freedreno/ir3: don't offset inloc by 8 freedreno: make gmem tile

Re: [Freedreno] [PATCH 2/6] drm: add helper for printing to log or seq_file

2016-10-17 Thread Rob Clark
On Mon, Oct 17, 2016 at 2:38 AM, Daniel Vetter <dan...@ffwll.ch> wrote: > On Fri, Oct 14, 2016 at 07:55:49PM -0400, Rob Clark wrote: >> Sometimes it is nice not to duplicate equivalent printk() and >> seq_printf() code. >> >> Signed-off-by: Rob Clark <robdcl.

Re: [Freedreno] [PATCH 12/12] drm/msm: gpu: Use the zap shader on 5XX if we can

2016-12-06 Thread Rob Clark
On Tue, Dec 6, 2016 at 10:35 AM, Jordan Crouse wrote: > On Mon, Dec 05, 2016 at 11:57:12AM -0800, Bjorn Andersson wrote: >> On Mon 28 Nov 11:28 PST 2016, Jordan Crouse wrote: >> >> > The A5XX GPU powers on in "secure" mode. In secure mode the GPU can >> > only render to

Re: [Freedreno] [PATCH] drm/msm: adreno: fix build error without debugfs

2016-11-30 Thread Rob Clark
On Wed, Nov 30, 2016 at 4:06 PM, Arnd Bergmann wrote: > The newly added a5xx support fails to build when debugfs is diabled: > > drivers/gpu/drm/msm/adreno/a5xx_gpu.c:849:4: error: 'struct msm_gpu_funcs' > has no member named 'show' > drivers/gpu/drm/msm/adreno/a5xx_gpu.c:849:11:

[Freedreno] [pull] drm/msm: msm-next for 4.10

2016-11-29 Thread Rob Clark
ort drm/msm: gpu: Add support for the GPMU Rob Clark (26): drm/msm/mdp5: handle non-fullscreen base plane case drm/msm/mdp5: no scaling support on RGBn pipes for 8x16 drm/msm: update uapi header license drm/msm/mdp5: 8x16 actually has 8 mixer stages

[Freedreno] [pull] drm/msm: msm-fixes for 4.10-rc2

2016-12-30 Thread Rob Clark
Hi Dave, A few fixes for 4.10.. the first fixes a long-standing logic bug, that by luck (ie. size of packets written into RB for a submit) wasn't hit on a3xx/a4xx but was causing intermittent GPU lockups on a5xx. And a couple other robustness issues that Jordan noticed. The following changes

Re: [Freedreno] [RFC] freedreno: valgrind support

2017-03-22 Thread Rob Clark
On Wed, Mar 22, 2017 at 7:54 AM, Rob Clark <robdcl...@gmail.com> wrote: > From: Rob Clark <robcl...@freedesktop.org> > > --- > This is mostly an attempt at teaching valgrind about the bo cache pool, > so it would not think that gem objects returned to the bo cache we

Re: [Freedreno] [PATCH] drm/msm: for array in-fences, check if all backing fences are from our own context before waiting

2017-03-18 Thread Rob Clark
On Fri, Mar 17, 2017 at 2:38 PM, Philipp Zabel <p.za...@pengutronix.de> wrote: > Use the dma_fence_match_context helper to check if all backing fences > are from our own context, in which case we don't have to wait. > > Signed-off-by: Philipp Zabel <p.za...@pengutronix.de>

[Freedreno] [RFC] freedreno: valgrind support

2017-03-22 Thread Rob Clark
From: Rob Clark <robcl...@freedesktop.org> --- This is mostly an attempt at teaching valgrind about the bo cache pool, so it would not think that gem objects returned to the bo cache were leaked. Unfortunately the list head node in the gem bo is used to store the bo in the pool. This is

[Freedreno] [pull] drm/msm: msm-next for 4.12

2017-04-09 Thread Rob Clark
MU during GPU cleanup drm/msm: Reference count address spaces drm/msm: Add MSM_PARAM_GMEM_BASE drm/msm: Hard code the GPU "slow frequency" drm/msm: gpu: Use OPP tables if we can msm/drm: gpu: Dynamically locate the clocks from the device tree Rob Clark (6):

Re: [Freedreno] [RESEND PATCH] drm/msm: adreno: fix build error without debugfs

2017-03-13 Thread Rob Clark
On Mon, Mar 13, 2017 at 12:43 PM, Arnd Bergmann wrote: > The newly added a5xx support fails to build when debugfs is diabled: > > drivers/gpu/drm/msm/adreno/a5xx_gpu.c:849:4: error: 'struct msm_gpu_funcs' > has no member named 'show' > drivers/gpu/drm/msm/adreno/a5xx_gpu.c:849:11:

Re: [Freedreno] Whether A200 driver is supported by Linux Mainline Kernel

2017-07-14 Thread Rob Clark
On Fri, Jul 14, 2017 at 6:52 AM, abhijit wrote: > Hi all, > > Running xinit load freedreno_drv.so file after adding configuration in > /etc/X11/xorg.conf, > > Section "Server Flags" > Option "IgnoreABI" > EndSection > > Section "Device" >

Re: [Freedreno] Whether A200 driver is supported by Linux Mainline Kernel

2017-07-09 Thread Rob Clark
The most obvious thing to check is make sure you use --enable-freedreno-kgsl when building libdrm.. if you have that, then try: LIBGL_DEBUG=verbose glxinfo BR, -R On Sat, Jul 8, 2017 at 2:07 AM, abhijit wrote: > Hi, > > I needed some help with respect to binding of

Re: [Freedreno] [PATCH] drm/msm/mdp5: Fix compilation warnings

2017-07-18 Thread Rob Clark
2017-07-18 2:34 GMT-04:00 Viresh Kumar : > On 29-06-17, 14:49, Viresh Kumar wrote: >> Following compilation warnings were observed for these files: >> >> CC [M] drivers/gpu/drm/msm/mdp/mdp5/mdp5_mdss.o >> drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c: In function

[Freedreno] [pull] drm/msm: msm-fixes for 4.13

2017-08-08 Thread Rob Clark
ser from __u64 data types Rob Clark (1): drm/msm/mdp5: fix unclocked register access in _cursor_set() Viresh Kumar (1): drm/msm/mdp5: Fix compilation warnings drivers/gpu/drm/msm/Kconfig | 2 +- drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 181 ++

[Freedreno] [pull] drm/msm: msm-next for 4.14

2017-08-23 Thread Rob Clark
buffer allocations Rob Clark (7): drm/msm: remove unused define drm/msm/mdp5: add tracking for clk enable-count drm/msm: add modeset module param drm/msm: don't track fbdev's gem object separately drm/msm: add helper to allocate stolen fb drm/msm: make

Re: [Freedreno] [PATCH] drm/msm: gpu: Enable zap shader for A5XX

2017-05-10 Thread Rob Clark
On Wed, May 10, 2017 at 7:28 PM, Bjorn Andersson wrote: > On Mon 08 May 13:28 PDT 2017, Jordan Crouse wrote: > >> The A5XX GPU powers on in "secure" mode. In secure mode the GPU can >> only render to buffers that are marked as secure and inaccessible >> to the kernel

[Freedreno] [pull] drm/msm: msm-next for 4.13

2017-06-20 Thread Rob Clark
drm/msm: Add a struct to pass configuration to msm_gpu_init() drm/msm: Remove memptrs->wptr Kasin Li (1): drm/msm: Fix potential buffer overflow issue Liviu Dudau (1): drm/msm/hdmi: Use bitwise operators when building register values Rob Clark (9): drm/msm: fix lock

Re: [Freedreno] [PATCH] drm: remove NULL pointer check for clk_disable_unprepare

2017-05-20 Thread Rob Clark
On Sat, May 20, 2017 at 3:04 PM, Masahiro Yamada wrote: > 2017-05-21 2:58 GMT+09:00 Masahiro Yamada : >> After long term efforts of fixing non-common clock implementations, >> clk_disable() is a no-op for a NULL pointer input, and this

Re: [Freedreno] [PATCH 10/13] drm/msm: Support multiple ringbuffers

2017-05-28 Thread Rob Clark
On Mon, May 8, 2017 at 4:35 PM, Jordan Crouse wrote: > Add the infrastructure to support the idea of multiple ringbuffers. > Assign each ringbuffer an id and use that as an index for the various > ring specific operations. > > The biggest delta is to support legacy fences.

Re: [Freedreno] [PATCH] fixup! drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-16 Thread Rob Clark
the madv/priv->inactive_list locking cleanup > separately. > > Thanks, > Sushmita > > > > On 2017-06-16 08:22, Rob Clark wrote: >> >> --- >> Ok, 2nd fixup to handle vmap shrinking. >> >> drivers/gpu/drm/msm/msm_gem.c | 44 >> ++

Re: [Freedreno] xinerma - xrandr apq8064 drm fbdev

2017-06-14 Thread Rob Clark
On Wed, Jun 14, 2017 at 7:25 AM, Vinay Simha B N wrote: > hi, > > i am having two encoders dsi1 and dsi2 of same resolution, i want to > configure as single resolution 960x864, > > after configuring as right-of , i got 960x864 resolution, but not able to > use as single fb as

Re: [Freedreno] [PATCH] drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-13 Thread Rob Clark
On Tue, Jun 13, 2017 at 6:52 PM, Sushmita Susheelendra wrote: > Buffer object specific resources like pages, domains, sg list > need not be protected with struct_mutex. They can be protected > with a buffer object level lock. This simplifies locking and > makes it easier

Re: [Freedreno] [PATCH] drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-14 Thread Rob Clark
On Tue, Jun 13, 2017 at 6:52 PM, Sushmita Susheelendra wrote: > Buffer object specific resources like pages, domains, sg list > need not be protected with struct_mutex. They can be protected > with a buffer object level lock. This simplifies locking and > makes it easier

Re: [Freedreno] [PATCH] drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-14 Thread Rob Clark
t_pages/vmap on objects that’ve been marked > as DONTNEED. I’ll send you a patch soon for that. > > Thanks, > Sushmita > > > On Jun 14, 2017, at 10:49 AM, Rob Clark <robdcl...@gmail.com> wrote: > > On Tue, Jun 13, 2017 at 6:52 PM, Sushmita Susheelendra > <ssus

Re: [Freedreno] [PATCH] fixup! drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-15 Thread Rob Clark
drop struct_mutex lock for shrinker. I think this first patch is pretty close to being ready in time to queue up for 4.13 (which I probably need to do this weekend). We should try and tackle the list+madv locks for 4.14, I think, since this is already a pretty big change. BR, -R > Thanks,

[Freedreno] [PATCH] fixup! drm/msm: Separate locking of buffer resources from struct_mutex

2017-06-16 Thread Rob Clark
--- Ok, 2nd fixup to handle vmap shrinking. drivers/gpu/drm/msm/msm_gem.c | 44 +++ 1 file changed, 36 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index f5d1f84..3190e05 100644 ---

[Freedreno] [PATCH 3/6] drm/msm: pass address-space to _get_iova() and friends

2017-06-13 Thread Rob Clark
No functional change, that will come later. But this will make it easier to deal with dynamically created address spaces (ie. per- process pagetables for gpu). Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 8 drivers/gpu/drm/msm/

[Freedreno] [PATCH 0/6] drm/msm: support arbitrary number of vma's per gem bo

2017-06-13 Thread Rob Clark
IOMMU case. Rob Clark (6): drm/msm/mdp5: kill pipe_lock drm/msm/mdp4+5: move aspace/id to base class drm/msm: pass address-space to _get_iova() and friends drm/msm: refactor how we handle vram carveout buffers drm/msm: support for an arbitrary number of address spaces drm/msm: remo

[Freedreno] [PATCH 4/6] drm/msm: refactor how we handle vram carveout buffers

2017-06-13 Thread Rob Clark
. So failing at allocation time is a more sane option. Plus this simplifies things in the next patch. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/msm_gem.c | 48 --- 1 file changed, 27 insertions(+), 21 deletions(-) diff

[Freedreno] [PATCH 1/6] drm/msm/mdp5: kill pipe_lock

2017-06-13 Thread Rob Clark
It serves no purpose, things should be sufficiently synchronized already by atomic framework. And it is somewhat awkward to be holding a spinlock when msm_gem_iova() is going to start needing to grab a mutex. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/md

[Freedreno] [PATCH 6/6] drm/msm: remove address-space id

2017-06-13 Thread Rob Clark
Now that the msm_gem supports an arbitrary number of vma's, we no longer need to assign an id (index) to each address space. So rip out the associated code. Signed-off-by: Rob Clark <robdcl...@gmail.com> --- drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c | 7 --- drivers/gpu/drm/msm/md

Re: [Freedreno] [PATCH v2 1/5] drm/msm: Expose our reservation object when exporting a dmabuf.

2017-05-02 Thread Rob Clark
.@anholt.net> > Reviewed-by: Daniel Vetter <daniel.vet...@ffwll.ch> > Cc: sta...@vger.kernel.org > Cc: Rob Clark <robdcl...@gmail.com> > Cc: linux-arm-...@vger.kernel.org > Cc: freedreno@lists.freedesktop.org Reviewed-by: Rob Clark <robdcl...@gmail.com> (go ahe

Re: [Freedreno] [PATCH 01/13] drm/msm: Take the mutex before calling msm_gem_new_impl

2017-05-08 Thread Rob Clark
On Mon, May 8, 2017 at 4:34 PM, Jordan Crouse wrote: > Amongst its other duties, msm_gem_new_impl adds the newly created > GEM object to the shared inactive list which may also be actively > modifiying the list during submission. All the paths to modify > the list are

  1   2   3   4   5   >