On 2013.07.01 at 17:01 -0400, alexdeucher at gmail.com wrote:
> From: Alex Deucher
>
> Hi Dave,
>
> A few more patches for 3.11:
> - add debugfs interface to check current DPM state
> - Fix a bug that caused problems with DPM on BTC+ asics.
>
> The following changes since commit f7d452f4fd5d86f
On 07/01/2013 10:30 PM, Daniel Drake wrote:
> Here is a new patch which should incorporate all your previous feedback.
> Now each variant passes clock info to the main driver via a new
> armada_clk_info structure.
>
> A helper function in the core lets each variant find the best clock.
> As you sug
The rv6xx_clocks_per_unit() function pretends it can set flags in a u64
bitfield but really because "1" is an int it doesn't work for more than
32 bits. The only caller truncates the high bits away anyway. I've
just changed it to be a u32.
Signed-off-by: Dan Carpenter
diff --git a/drivers/gpu/
These checks should be ">=" instead of ">". j is used as an offset into
the table->mc_reg_address[] array and that has
SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE (16) elements.
Signed-off-by: Dan Carpenter
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 6918f07..ea
On Mon, Jul 1, 2013 at 10:46 PM, Ben Widawsky wrote:
>> > drm_mm_remove_node() does unlink the node but not remove it. Btw., I
>> > have these fixes in my series, too. I will push it later and write the
>> > git-link to #dri-devel.
>>
>> We have patches in-flight to convert over to embedded drm_mm
Hi guys,
I recently got the rMBP 10,1 model, it has two graphic cards:
00:02.0 VGA compatible controller: Intel Corporation 3rd Gen Core processor
Graphics Controller (rev 09)
01:00.0 VGA compatible controller: NVIDIA Corporation GK107M [GeForce GT 650M
Mac Edition] (rev a1)
I'm running linux
On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote:
> Hi
>
> On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter
> wrote:
> > When converting to the preallocated drm_mm_node interfaces in
> >
> > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> > Author: Chris Wilson
> > Date: Fri Dec
Every other place properly checks whether we've managed to set
up the stolen allocator at boot-up properly, with the exception
of the cleanup code. Which results in an ugly
*ERROR* Memory manager not clean. Delaying takedown
at module unload time since the drm_mm isn't initialized at all.
v2: Wh
On Mon, Jul 01, 2013 at 09:16:56PM +0100, Chris Wilson wrote:
> On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote:
> > When converting to the preallocated drm_mm_node interfaces in
> >
> > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> > Author: Chris Wilson
> > Date: Fri Dec 7 2
Hi
On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter
wrote:
> When converting to the preallocated drm_mm_node interfaces in
>
> commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> Author: Chris Wilson
> Date: Fri Dec 7 20:37:07 2012 +
>
> drm/i915: Preallocate the drm_mm_node prior to mani
On Mon, Jul 01, 2013 at 09:54:17PM +0200, David Herrmann wrote:
> On Mon, Jul 1, 2013 at 9:42 PM, Daniel Vetter wrote:
> > On Mon, Jul 01, 2013 at 08:33:00PM +0200, David Herrmann wrote:
> >> +/** drm_vma_offset_manager_destroy()
> >> + *
> >> + * Destroy an object manager which was previously cre
When converting to the preallocated drm_mm_node interfaces in
commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
Author: Chris Wilson
Date: Fri Dec 7 20:37:07 2012 +
drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm
only the allocation side was converted, but not
drm/i915 is the only user of the color allocation handling and
switched to insert_node a while ago. So we can ditch this.
Signed-off-by: Daniel Vetter
---
include/drm/drm_mm.h | 31 ---
1 file changed, 31 deletions(-)
diff --git a/include/drm/drm_mm.h b/include/drm/d
> -Original Message-
> From: Markus Trippelsdorf [mailto:markus at trippelsdorf.de]
> Sent: Monday, July 01, 2013 5:58 PM
> To: alexdeucher at gmail.com
> Cc: dri-devel at lists.freedesktop.org; airlied at gmail.com; Deucher,
> Alexander
> Subject: Re: [pull] radeon drm-next-3.11
>
> On 2
The usual drm driver has tons of different drm_mm memory managers so the drm
error message in dmesg is pretty useless. WARN instead so that we have the full
backtrace.
Signed-off-by: Daniel Vetter
---
drivers/gpu/drm/drm_mm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
Every other place properly checks whether we've managed to set
up the stolen allocator at boot-up properly, with the exception
of the cleanup code. Which results in an ugly
*ERROR* Memory manager not clean. Delaying takedown
at module unload time since the drm_mm isn't initialized at all.
v2: Wh
In
commit 3a359f0b21ab218c1bf7a6a1b638b6fd143d0b99
Author: Daniel Vetter
Date: Sat Apr 20 12:08:11 2013 +0200
drm/mm: fix dump table BUG
I've failed to fix both instances of the regression introduced in
commit 9e8944ab564f2e3dde90a518cd32048c58918608
Author: Chris Wilson
Date: Thu Nov
for the bug.
-- next part --
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130701/a8a8df9b/attachment.html>
Hi
On Mon, Jul 1, 2013 at 9:42 PM, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 08:33:00PM +0200, David Herrmann wrote:
>> If we want to map GPU memory into user-space, we need to linearize the
>> addresses to not confuse mm-core. Currently, GEM and TTM both implement
>> their own offset-manage
On Mon, Jul 01, 2013 at 08:32:57PM +0200, David Herrmann wrote:
> Hi
>
> I picked up the initial work from Dave [1], fixed several bugs, rewrote the
> drm_mm node handling and adjusted the different drivers.
> The series tries to replace the VMA-offset managers from GEM and TTM with a
> single uni
On Mon, Jul 01, 2013 at 08:33:03PM +0200, David Herrmann wrote:
> Instead of unmapping the nodes in TTM and GEM users manually, we provide
> a generic wrapper which does the correct thing for all vma-nodes.
>
> Signed-off-by: David Herrmann
Nice. One nitpick below, otherwise:
Reviewed-by: Danie
On Mon, Jul 01, 2013 at 08:33:00PM +0200, David Herrmann wrote:
> If we want to map GPU memory into user-space, we need to linearize the
> addresses to not confuse mm-core. Currently, GEM and TTM both implement
> their own offset-managers to assign a pgoff to each object for user-space
> CPU access
lated to this and it doesn't appear in the 3.9 kernels.
--
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130701/a84a225c/attachment.html>
On Tue, Jul 2, 2013 at 5:22 AM, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 08:32:58PM +0200, David Herrmann wrote:
>> There is no reason to return "int" as this function never fails.
>> Furthermore, several drivers (ast, sis) already depend on this.
>>
>> Signed-off-by: David Herrmann
>
> Bac
On Mon, Jul 01, 2013 at 08:32:59PM +0200, David Herrmann wrote:
> This helper tests whether a given node is currently linked into a drm_mm
> manager. We use the node->mm pointer for that as it is set for all linked
> objects.
>
> We also reset node->mm whenever a node is removed. All such access i
On Mon, Jul 01, 2013 at 08:32:58PM +0200, David Herrmann wrote:
> There is no reason to return "int" as this function never fails.
> Furthermore, several drivers (ast, sis) already depend on this.
>
> Signed-off-by: David Herrmann
Back when I've reworked drm_mm I was still a rookie and didn't wa
On Mon, Jul 01, 2013 at 10:05:53PM +0200, Daniel Vetter wrote:
> drm/i915 is the only user of the color allocation handling and
> switched to insert_node a while ago. So we can ditch this.
>
> Signed-off-by: Daniel Vetter
If you build it, they will come. Lies!
Reviewed-by: Chris Wilson
-Chris
On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote:
> When converting to the preallocated drm_mm_node interfaces in
>
> commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> Author: Chris Wilson
> Date: Fri Dec 7 20:37:07 2012 +
>
> drm/i915: Preallocate the drm_mm_node prior to
On Mon, Jul 01, 2013 at 10:01:03PM +0200, Daniel Vetter wrote:
> Every other place properly checks whether we've managed to set
> up the stolen allocator at boot-up properly, with the exception
> of the cleanup code. Which results in an ugly
>
> *ERROR* Memory manager not clean. Delaying takedown
; --
> Chris Wilson, Intel Open Source Technology Centre
> ___
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
-- next part --
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130701/ac7b09d9/attachment.html>
Instead of unmapping the nodes in TTM and GEM users manually, we provide
a generic wrapper which does the correct thing for all vma-nodes.
Signed-off-by: David Herrmann
---
drivers/gpu/drm/i915/i915_gem.c | 6 +-
drivers/gpu/drm/ttm/ttm_bo.c| 8 +---
include/drm/drm_vma_manager.h
Use the new vma-manager infrastructure. This doesn't change any
implementation details as the vma-offset-manager is nearly copied 1-to-1
from TTM.
Even though the vma-manager uses its own locks, we still need bo->vm_lock
to prevent bos from being destroyed before we can get a reference during
look
Use the new vma manager instead of the old hashtable. Also convert all
drivers to use the new convenience helpers. This drops all the
(map_list.hash.key << PAGE_SHIFT) non-sense.
Locking and access-management is exactly the same as before with an
additional lock inside of the vma-manager, which st
If we want to map GPU memory into user-space, we need to linearize the
addresses to not confuse mm-core. Currently, GEM and TTM both implement
their own offset-managers to assign a pgoff to each object for user-space
CPU access. GEM uses a hash-table, TTM uses an rbtree.
This patch provides a unif
This helper tests whether a given node is currently linked into a drm_mm
manager. We use the node->mm pointer for that as it is set for all linked
objects.
We also reset node->mm whenever a node is removed. All such access is
currently safe as everyone calls kfree() on the object directly after
re
There is no reason to return "int" as this function never fails.
Furthermore, several drivers (ast, sis) already depend on this.
Signed-off-by: David Herrmann
---
drivers/gpu/drm/drm_gem.c| 8 ++--
drivers/gpu/drm/drm_mm.c | 4 +---
drivers/gpu/drm/ttm/ttm_bo.c
Hi
I picked up the initial work from Dave [1], fixed several bugs, rewrote the
drm_mm node handling and adjusted the different drivers.
The series tries to replace the VMA-offset managers from GEM and TTM with a
single unified implementation. It uses the TTM RBTree idea to allow sub-mappings
(whic
On Mon, 1 Jul 2013 22:01:04 +0200
Daniel Vetter wrote:
> The usual drm driver has tons of different drm_mm memory managers so
> the drm error message in dmesg is pretty useless. WARN instead so
> that we have the full backtrace.
>
> Signed-off-by: Daniel Vetter
I've written this patch myself
Hello Chris,
On 2013? 07? 01? 19:57, Chris Wilson wrote:
> On Mon, Jul 01, 2013 at 07:49:10PM +0900, Seung-Woo Kim wrote:
>> +
>> +out_close:
>> +if (dev->driver->postclose)
>> +dev->driver->postclose(dev, priv);
>> +out_free:
>> kfree(priv);
>> filp->private_data = NULL;
On Mon, 1 Jul 2013 22:01:04 +0200
Daniel Vetter wrote:
> The usual drm driver has tons of different drm_mm memory managers so
> the drm error message in dmesg is pretty useless. WARN instead so
> that we have the full backtrace.
>
> Signed-off-by: Daniel Vetter
I've written this patch myself
On Mon, Jul 1, 2013 at 3:48 PM, Sebastian Hesselbarth
wrote:
> I guess "extclk0" and "extclk1" should be sufficient for clock names.
> Also, they are not dedicated as you can have CRTC0 and CRTC1 use e.g.
> extclk0 simultaneously. See below for .is_dedicated in general.
Maybe we can find better t
From: YoungJun Cho
There are wrong cases to handle error in drm_open_helper().
The priv->minor, assigned by idr_find() which can return NULL,
should be checked whether it is NULL or not before referencing it.
And if an error occurs after executing dev->driver->open() which
allocates driver specif
seq of a trace point is unsigned int but print format was %d. So
it fixes the format as %u.
Signed-off-by: Seung-Woo Kim
Signed-off-by: Kyungmin Park
---
change from v1
- remove wrong commit messageas Chris commented
drivers/gpu/drm/drm_trace.h |6 +++---
1 files changed, 3 insertions(+),
The error path does this:
for (--i; i >= 0; --i) {
which is a forever loop because "i" is unsigned.
Signed-off-by: Dan Carpenter
diff --git a/drivers/gpu/drm/radeon/radeon_test.c
b/drivers/gpu/drm/radeon/radeon_test.c
index f4d6bce..12e8099 100644
--- a/drivers/gpu/drm/radeon/radeon_t
It should be ">=" instead of ">" here. The table->mc_reg_address[]
array has SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE (16) elements.
Signed-off-by: Dan Carpenter
diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c
index bab0185..55491e7 100644
--- a/drivers/gpu/drm/rade
On Mon, Jul 1, 2013 at 5:26 PM, Dave Airlie wrote:
>> Of course, I share the idea of true, full-blown of_drm_something
>> helpers. But the DT patch, is not an improvement but a real fix as in
>> "make DRM not break on some platforms". From that on, I can start
>> digging into DRM API and improve D
Hello Chris,
Thank you for reviewing.
On 2013? 07? 01? 19:23, Chris Wilson wrote:
> On Mon, Jul 01, 2013 at 07:06:31PM +0900, Seung-Woo Kim wrote:
>> seq of a trace point is unsigned int but print format was %d. So
>> it fixes the format as %u even the format can be not used.
>
> I don't underst
From: YoungJun Cho
There are wrong cases to handle error in drm_open_helper().
The priv->minor, assigned by idr_find() which can return NULL,
should be checked whether it is NULL or not before referencing it.
And if an error occurs after executing dev->driver->open() which
allocates driver specif
If raw_edid is null, it will crash, so checking in bad label is
meaningless.
Signed-off-by: Seung-Woo Kim
Signed-off-by: Kyungmin Park
---
drivers/gpu/drm/drm_edid.c |5 -
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edi
seq of a trace point is unsigned int but print format was %d. So
it fixes the format as %u even the format can be not used.
Signed-off-by: Seung-Woo Kim
Signed-off-by: Kyungmin Park
---
drivers/gpu/drm/drm_trace.h |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/dri
This patch series fixes minor code issues including wrong trace point foramts,
meaningless null checking, and possible resource leak in error cases.
This is based drm-next branch.
Seung-Woo Kim (2):
drm: fix print format of sequence in trace point
drm: move edid null check to the first part
--
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130701/00ac8086/attachment.html>
On Mon, Jul 1, 2013 at 4:52 AM, Sebastian Hesselbarth
wrote:
> On 07/01/13 02:01, Dave Airlie wrote:
>>
>> how about instead of writing:
>> "However, at least I've taken the time to_think_ about what I'm doing
>> and realise that there_is_ scope here for the DRM core to improve,
>>
>> rather tha
From: YoungJun Cho
There are missing parts to handle error in drm_open_helper().
The priv->minor, assigned by idr_find() which can return NULL,
should be checked whether it is NULL or not before referencing it.
put_pid(), drm_gem_release(), and drm_prime_destory_file_private()
should be called wh
If raw_edid of drm_edid_block_vaild() is null, it will crash, so
checking in bad label is removed and instead assertion is added at
the top of the function.
The type of return for the function is bool, so it fixes to return
true and false instead of 1 and 0.
Signed-off-by: Seung-Woo Kim
Signed-of
From: YoungJun Cho
There are missing parts to handle error in drm_open_helper().
The priv->minor, assigned by idr_find() which can return NULL,
should be checked whether it is NULL or not before referencing it.
put_pid(), drm_gem_release(), and drm_prime_destory_file_private()
should be called wh
From: Alex Deucher
Hi Dave,
A few more patches for 3.11:
- add debugfs interface to check current DPM state
- Fix a bug that caused problems with DPM on BTC+ asics.
The following changes since commit f7d452f4fd5d86f764807a1234a407deb5b105ef:
Merge branch 'drm-nouveau-next' of
git://anongit.
From: YoungJun Cho
When the exynos_drm_subdrv_open() returns error, the file_priv
should be released and file->driver_priv set to NULL.
Signed-off-by: YoungJun Cho
Signed-off-by: Kyungmin Park
---
drivers/gpu/drm/exynos/exynos_drm_drv.c |9 -
1 file changed, 8 insertions(+), 1 del
On Mon, Jul 1, 2013 at 12:21 PM, Chris Wilson
wrote:
> On Mon, Jul 01, 2013 at 07:06:32PM +0900, Seung-Woo Kim wrote:
>> If raw_edid is null, it will crash, so checking in bad label is
>> meaningless.
>
> It would be an error on part of the caller, but the defense looks sane.
> As the function is
Hi Daniel,
On 2013년 07월 01일 23:56, Daniel Vetter wrote:
> On Mon, Jul 1, 2013 at 12:21 PM, Chris Wilson
> wrote:
>> On Mon, Jul 01, 2013 at 07:06:32PM +0900, Seung-Woo Kim wrote:
>>> If raw_edid is null, it will crash, so checking in bad label is
>>> meaningless.
>>
>> It would be an error on pa
On Mon, Jul 1, 2013 at 5:26 PM, Dave Airlie wrote:
>> Of course, I share the idea of true, full-blown of_drm_something
>> helpers. But the DT patch, is not an improvement but a real fix as in
>> "make DRM not break on some platforms". From that on, I can start
>> digging into DRM API and improve D
Hi Sergei,
On Thursday 27 June 2013 17:04:45 Sergei Shtylyov wrote:
> On 27-06-2013 13:49, Laurent Pinchart wrote:
> > Replace the devm_request_mem_region() and devm_ioremap_nocache() calls
> > with devm_ioremap_resource().
> >
> > Signed-off-by: Laurent Pinchart
> >
> > ---
> >
> > drivers/g
Hi Russell,
Here is a new patch which should incorporate all your previous feedback.
Now each variant passes clock info to the main driver via a new
armada_clk_info structure.
A helper function in the core lets each variant find the best clock.
As you suggested we first try external ("dedicated")
Convert drivers/gpu/drm class to use dev_pm_ops for power management and
remove Legacy PM ops hooks. With this change, drm class registers
suspend/resume callbacks via class->pm (dev_pm_ops) instead of Legacy
class->suspend/resume. When __device_suspend() runs call-backs, it will
find class->pm ops
> -Original Message-
> From: Markus Trippelsdorf [mailto:mar...@trippelsdorf.de]
> Sent: Monday, July 01, 2013 5:58 PM
> To: alexdeuc...@gmail.com
> Cc: dri-devel@lists.freedesktop.org; airl...@gmail.com; Deucher, Alexander
> Subject: Re: [pull] radeon drm-next-3.11
>
> On 2013.07.01 at 17
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130701/12346430/attachment.html>
On 2013.07.01 at 17:01 -0400, alexdeuc...@gmail.com wrote:
> From: Alex Deucher
>
> Hi Dave,
>
> A few more patches for 3.11:
> - add debugfs interface to check current DPM state
> - Fix a bug that caused problems with DPM on BTC+ asics.
>
> The following changes since commit f7d452f4fd5d86f764
https://bugs.freedesktop.org/show_bug.cgi?id=66473
Priority: medium
Bug ID: 66473
Assignee: dri-devel@lists.freedesktop.org
Summary: hdmi oops with mem sleep
Severity: normal
Classification: Unclassified
OS: Linux (All)
ttp://lists.freedesktop.org/archives/dri-devel/attachments/20130701/2f3658f8/attachment.html>
On Mon, Jul 1, 2013 at 4:52 AM, Sebastian Hesselbarth
wrote:
> On 07/01/13 02:01, Dave Airlie wrote:
>>
>> how about instead of writing:
>> "However, at least I've taken the time to_think_ about what I'm doing
>> and realise that there_is_ scope here for the DRM core to improve,
>>
>> rather tha
On Mon, Jul 1, 2013 at 2:32 PM, David Herrmann wrote:
> Hi
>
> I picked up the initial work from Dave [1], fixed several bugs, rewrote the
> drm_mm node handling and adjusted the different drivers.
> The series tries to replace the VMA-offset managers from GEM and TTM with a
> single unified imple
https://bugs.freedesktop.org/show_bug.cgi?id=66425
--- Comment #2 from Austin Lund ---
(In reply to comment #1)
> This is a mac?
Yes. Macbookpro8,2
Also, this doesn't happen with a 3.9.7 kernel.
It seems to be related to the UVD stuff that was added to 3.10. Ring 5 appears
to be related to t
On Mon, Jul 1, 2013 at 8:00 PM, Maarten Lankhorst
wrote:
> This is a bit messed up because chan->cli->mutex is a different class,
> depending on whether it is the global drm client or not. This is
> because the global cli->mutex lock can be taken for eviction,
> so locking it before pinning the bu
> Of course, I share the idea of true, full-blown of_drm_something
> helpers. But the DT patch, is not an improvement but a real fix as in
> "make DRM not break on some platforms". From that on, I can start
> digging into DRM API and improve DT support here and there.
>
> So for the three patches I
From: Dave Airlie
qxl has a feature to allow the userspace driver do arbitrary resizes
when the viewer resizes, this fixes it by removing unnecessary code
from the kernel side.
Signed-off-by: Dave Airlie
---
drivers/gpu/drm/qxl/qxl_display.c | 58 ++-
1 file
From: Alex Deucher
Hi Dave,
A few more patches for 3.11:
- add debugfs interface to check current DPM state
- Fix a bug that caused problems with DPM on BTC+ asics.
The following changes since commit f7d452f4fd5d86f764807a1234a407deb5b105ef:
Merge branch 'drm-nouveau-next' of
git://anongit.
From: Mark Brown
Ensure that all externally accessed functions are correctly prototyped
when defined in each file by making sure the headers with the protoypes
are included in the file with the definition.
Signed-off-by: Mark Brown
---
drivers/gpu/drm/exynos/exynos_drm_connector.c | 1 +
drive
On Mon, Jul 1, 2013 at 10:46 PM, Ben Widawsky wrote:
>> > drm_mm_remove_node() does unlink the node but not remove it. Btw., I
>> > have these fixes in my series, too. I will push it later and write the
>> > git-link to #dri-devel.
>>
>> We have patches in-flight to convert over to embedded drm_mm
On Mon, Jul 01, 2013 at 10:39:03PM +0200, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote:
> > Hi
> >
> > On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter
> > wrote:
> > > When converting to the preallocated drm_mm_node interfaces in
> > >
> > > commit dc9dd7a
On Mon, Jul 01, 2013 at 10:39:03PM +0200, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote:
> > Hi
> >
> > On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter
> > wrote:
> > > When converting to the preallocated drm_mm_node interfaces in
> > >
> > > commit dc9dd7a
On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote:
> Hi
>
> On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter wrote:
> > When converting to the preallocated drm_mm_node interfaces in
> >
> > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> > Author: Chris Wilson
> > Date: Fri Dec 7
Every other place properly checks whether we've managed to set
up the stolen allocator at boot-up properly, with the exception
of the cleanup code. Which results in an ugly
*ERROR* Memory manager not clean. Delaying takedown
at module unload time since the drm_mm isn't initialized at all.
v2: Wh
On Mon, Jul 01, 2013 at 09:16:56PM +0100, Chris Wilson wrote:
> On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote:
> > When converting to the preallocated drm_mm_node interfaces in
> >
> > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> > Author: Chris Wilson
> > Date: Fri Dec 7 2
On Mon, Jul 01, 2013 at 09:54:17PM +0200, David Herrmann wrote:
> On Mon, Jul 1, 2013 at 9:42 PM, Daniel Vetter wrote:
> > On Mon, Jul 01, 2013 at 08:33:00PM +0200, David Herrmann wrote:
> >> +/** drm_vma_offset_manager_destroy()
> >> + *
> >> + * Destroy an object manager which was previously cre
org/archives/dri-devel/attachments/20130701/f6c72282/attachment.html>
Hi
On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter wrote:
> When converting to the preallocated drm_mm_node interfaces in
>
> commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> Author: Chris Wilson
> Date: Fri Dec 7 20:37:07 2012 +
>
> drm/i915: Preallocate the drm_mm_node prior to manip
On Mon, Jul 01, 2013 at 10:05:53PM +0200, Daniel Vetter wrote:
> drm/i915 is the only user of the color allocation handling and
> switched to insert_node a while ago. So we can ditch this.
>
> Signed-off-by: Daniel Vetter
If you build it, they will come. Lies!
Reviewed-by: Chris Wilson
-Chris
On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote:
> When converting to the preallocated drm_mm_node interfaces in
>
> commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
> Author: Chris Wilson
> Date: Fri Dec 7 20:37:07 2012 +
>
> drm/i915: Preallocate the drm_mm_node prior to
On Mon, Jul 01, 2013 at 10:01:03PM +0200, Daniel Vetter wrote:
> Every other place properly checks whether we've managed to set
> up the stolen allocator at boot-up properly, with the exception
> of the cleanup code. Which results in an ugly
>
> *ERROR* Memory manager not clean. Delaying takedown
When converting to the preallocated drm_mm_node interfaces in
commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d
Author: Chris Wilson
Date: Fri Dec 7 20:37:07 2012 +
drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm
only the allocation side was converted, but not
drm/i915 is the only user of the color allocation handling and
switched to insert_node a while ago. So we can ditch this.
Signed-off-by: Daniel Vetter
---
include/drm/drm_mm.h | 31 ---
1 file changed, 31 deletions(-)
diff --git a/include/drm/drm_mm.h b/include/drm/d
The usual drm driver has tons of different drm_mm memory managers so the drm
error message in dmesg is pretty useless. WARN instead so that we have the full
backtrace.
Signed-off-by: Daniel Vetter
---
drivers/gpu/drm/drm_mm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
Every other place properly checks whether we've managed to set
up the stolen allocator at boot-up properly, with the exception
of the cleanup code. Which results in an ugly
*ERROR* Memory manager not clean. Delaying takedown
at module unload time since the drm_mm isn't initialized at all.
v2: Wh
In
commit 3a359f0b21ab218c1bf7a6a1b638b6fd143d0b99
Author: Daniel Vetter
Date: Sat Apr 20 12:08:11 2013 +0200
drm/mm: fix dump table BUG
I've failed to fix both instances of the regression introduced in
commit 9e8944ab564f2e3dde90a518cd32048c58918608
Author: Chris Wilson
Date: Thu Nov
From: YoungJun Cho
The buf_num in vp_video_buffer() should be 1 or 2, but it is not
initialized, and only set to 2 in NV12M or NV12MT cases.
So this patch initializes the buf_num with 1 as default.
Signed-off-by: YoungJun Cho
Signed-off-by: Seung-Woo Kim
Signed-off-by: Kyungmin Park
---
driv
From: YoungJun Cho
The type of input parameter enable is bool, so it does not need
to check whether true or false.
Signed-off-by: YoungJun Cho
Signed-off-by: Kyungmin Park
---
drivers/gpu/drm/exynos/exynos_drm_vidi.c |3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/gpu/drm/exy
Hi
On Mon, Jul 1, 2013 at 9:42 PM, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 08:33:00PM +0200, David Herrmann wrote:
>> If we want to map GPU memory into user-space, we need to linearize the
>> addresses to not confuse mm-core. Currently, GEM and TTM both implement
>> their own offset-manage
On Mon, Jul 01, 2013 at 08:14:42PM +0900, Seung-Woo Kim wrote:
> Hello Chris,
>
> On 2013? 07? 01? 19:57, Chris Wilson wrote:
> > On Mon, Jul 01, 2013 at 07:49:10PM +0900, Seung-Woo Kim wrote:
> >> +
> >> +out_close:
> >> + if (dev->driver->postclose)
> >> + dev->driver->postclose(dev, p
On 07/01/13 11:42, Daniel Vetter wrote:
> On Mon, Jul 01, 2013 at 10:52:03AM +0200, Sebastian Hesselbarth wrote:
>> at least on this point I do share Russell's impression. I've sent
>> bunch of patches improving TDA998x and DRM+DT:
>> - TDA998x irq handling - ignored
>> - TDA998x sync fix - ignored
On Mon, Jul 01, 2013 at 08:32:57PM +0200, David Herrmann wrote:
> Hi
>
> I picked up the initial work from Dave [1], fixed several bugs, rewrote the
> drm_mm node handling and adjusted the different drivers.
> The series tries to replace the VMA-offset managers from GEM and TTM with a
> single uni
1 - 100 of 174 matches
Mail list logo
