On Mon, 06 Jan 2014, "Lu, Ran" wrote:
> I just got a Retina MBP with Iris Pro and Nvidia GT 750. The iGPU works in
> OSX(Mavericks), but when I boot into Linux, the device was disabled. I
> debugged the kernel a little, apparently when pci_bus_read_dev_vendor_id is
> called for PCI 00:2.0 (the
On Fri, Jan 03, 2014 at 02:19:51PM -0200, Paulo Zanoni wrote:
> 2014/1/3 Jani Nikula :
> > With 20+ module parameters I think referring to them via a struct
> > improves clarity. The downsides are losing static on a couple of
> > variables and not having the initialization and module_param_named()
On Thu, Jan 02, 2014 at 07:50:33PM -1000, Ben Widawsky wrote:
> This will allow mesa to determine if it needs to create a context, or
> can reuse the default context. Reusing the default context saves memory,
> and startup time.
>
> To keep the libdrm interface as dumb as possible, we simply expos
On Fri, Jan 03, 2014 at 02:07:47PM +0200, Mika Kuoppala wrote:
> Ben Widawsky writes:
>
> > It makes all the code which calls into this function way too confusing.
> >
> > v2: Fix destroy IOCTL as well
> >
> > v3: Clarify the other two callers of i915_gem_context_get() to never
> > check for NULL
On Wed, Jan 01, 2014 at 02:00:54PM +, Chris Wilson wrote:
> One side-effect of the introduction of ppgtt was that we needed to
> rebind the object into the appropriate vm (and global gtt in some
> peculiar cases). For simplicity this was done twice for every object on
> every call to execbuffer
On Mon, Dec 30, 2013 at 07:59:49AM +0530, Oravil Nair wrote:
> Hi,
>
> i915_gem_object_pin(), during i915 driver create, seems to write to the
> memory written by BIOS. Where can the start address be specified to
> allocate memory so that the memory written by BIOS is not overwritten at
> initiali
On Mon, Dec 30, 2013 at 01:56:53PM -0200, Paulo Zanoni wrote:
> From: Paulo Zanoni
>
> Looks like we have to do a lot of work just to count the number of
> modes...
>
> Caught by Valgrind.
>
> Signed-off-by: Paulo Zanoni
Nice set of patches, please push. I agree that the mode/output config
ha
On Sat, Dec 28, 2013 at 01:31:49PM -0800, Ben Widawsky wrote:
> The initialized flag is used to specify a context has been initialized
> and it's context is safe to load, ie. the 3d state is setup properly.
> With full PPGTT, we emit the address space loads during context switch
> and this currentl
On Tue, Jan 7, 2014 at 12:31 AM, Jesse Barnes wrote:
> On Thu, 19 Dec 2013 10:48:01 -0800
> Jesse Barnes wrote:
>
>> When fastbooting, we read out the pipe timings early on, and then in a
>> panel fitted config, disable the fitter later. But we weren't updating
>> the pipe src h/w, which meant t
On 01/06/2014 07:02 AM, Damien Lespiau wrote:
> Having one source for registers defines sounds like a good idea. These
> patches,
> written long ago, were a baby step towards that goal by cleaning-up the
> INSTDONE definitons to be like the kernel ones.
>
> Along the way, I removed one tool that
According to Art, we don't have a way to read back the state reliably at
runtime, at least not without risking disabling it again. So drop the
readout and checking on BDW.
References: https://bugs.freedesktop.org/show_bug.cgi?id=71906
Signed-off-by: Jesse Barnes
---
drivers/gpu/drm/i915/intel_d
On Thu, 19 Dec 2013 10:48:01 -0800
Jesse Barnes wrote:
> When fastbooting, we read out the pipe timings early on, and then in a
> panel fitted config, disable the fitter later. But we weren't updating
> the pipe src h/w, which meant the mouse cursor was clipped to the
> pfitted size rather than
Hi
2014/1/6 Damien Lespiau :
> Follow up of the v3:
> http://lists.freedesktop.org/archives/intel-gfx/2013-December/037333.html
>
> The main change is the removal of INTEL_INFO() from the driver (patches 1 to
> 5), a clean-up suggested by Chris: the usage of that macro hides that we go
> from de
Reviewed-by: Mika Kuoppala
Reviewed-by: Ville Syrjälä
Signed-off-by: Damien Lespiau
---
drivers/gpu/drm/i915/i915_drv.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index a0713e2..c289b8f 100644
--- a/d
A lot of the WM functions are only reading from that structure and are
already using const. While converting the code to use dev_priv instead
of dev, I noticed a few places where we can give that hint.
Signed-off-by: Damien Lespiau
---
drivers/gpu/drm/i915/intel_pm.c | 8
1 file changed
We had 2 set of defines for the same register, so make it one.
Reviewed-by: Mika Kuoppala
Reviewed-by: Ville Syrjälä
Signed-off-by: Damien Lespiau
---
drivers/gpu/drm/i915/i915_reg.h | 18 --
drivers/gpu/drm/i915/intel_ddi.c | 2 +-
drivers/gpu/drm/i915/intel_display.
While the longer goal is to have dev_priv == dev by properly subclassing
struct drm_device, quite a few macros in i915_drv.h need the dev
argument. Let's not try to disrupt everything yet and just remove their
usage of INTEL_INFO() as a first step.
Signed-off-by: Damien Lespiau
---
drivers/gpu/d
FUSE_STRAP has a bit to inform us that the display has been fused off.
Use it to setup the definitive number of pipes at run-time.
v2: actually tweak num_pipes, not num_planes
v3: also tests SFUSE_STRAP bit 7
v4: rebase on top of the INTEL_INFO() removal
Reviewed-by: Mika Kuoppala (for v3)
Revie
A pointer to the struct drm_device is not needed in a few of the WM
functions, so just pass a pointer to dev_priv.
Signed-off-by: Damien Lespiau
---
drivers/gpu/drm/i915/intel_pm.c | 50 +
1 file changed, 26 insertions(+), 24 deletions(-)
diff --git a/dri
And rename it to num_sprites as this value doesn't count the primary plane.
This limit lives with num_pipes really, and now that dev_priv->info is writable
we can put it there instead.
While at it, introduce a intel_device_info_runtime_init() where we'll be able
to gather the device info fields a
We now read out the FUSE_STRAP and SFUSE_STRAP registers, looking for
configurations with display fused off. Let's remove the Quanta special
case and rely on the programmed fuses to set num_pipes to 0.
This patch is untested and could use a good soul with such a device to
give it a go.
Cc: Ben Wi
Finally, we reach the goal of the last few patches: the removal of
INTEL_INFO() in favor of a direct dev_priv usage.
Signed-off-by: Damien Lespiau
---
drivers/gpu/drm/i915/i915_drv.h | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_
The last occurences of INTEL_INFO() that weren't caught by the sed in:
drm/i915: Mass replace INTEL_INFO() by dev_priv->info
Nothing particularly noticeable except, maybe, the nice feeling of
removing a few back and forth between *dev and *dev_priv.
Signed-off-by: Damien Lespiau
---
drivers/g
Follow up of the v3:
http://lists.freedesktop.org/archives/intel-gfx/2013-December/037333.html
The main change is the removal of INTEL_INFO() from the driver (patches 1 to
5), a clean-up suggested by Chris: the usage of that macro hides that we go
from dev to dev_priv while we could use the dev_
Having one source for registers defines sounds like a good idea. These patches,
written long ago, were a baby step towards that goal by cleaning-up the
INSTDONE definitons to be like the kernel ones.
Along the way, I removed one tool that isn't even compiled and another unusued
one.
--
Damien
_
The GEN6_INSTDONE_1 and GEN6_INSTDONE_2 registers are just the old
INSTDONE_I965 and INSTDONE_1 registers but renamed. Let's use the old
names (this is what the kernel does).
Signed-off-by: Damien Lespiau
---
lib/instdone.c | 6 +++---
tools/intel_reg_dumper.c | 4 ++--
2 files changed
Signed-off-by: Damien Lespiau
---
lib/instdone.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/instdone.c b/lib/instdone.c
index b4d51f0..1399df7 100644
--- a/lib/instdone.c
+++ b/lib/instdone.c
@@ -37,6 +37,7 @@ int num_instdone_bits = 0;
static void
add_instdone_bit(uint32_t reg, ui
This is the only place where they are used and we've even started using
1 << n constants with gen 7.
Signed-off-by: Damien Lespiau
---
lib/instdone.c | 242
lib/intel_reg.h | 237 --
2 f
This tool only supports ILK. I take the fact that nobody has felt the
need to update for later platform a sign it's not very useful.
Signed-off-by: Damien Lespiau
---
tools/.gitignore | 1 -
tools/Makefile.sources | 1 -
tools/intel_disable_clock_gating.c | 71 ---
That's how the registers are named in the kernel defines.
Signed-off-by: Damien Lespiau
---
lib/instdone.c | 6 +++---
lib/intel_reg.h| 6 +++---
tools/intel_error_decode.c | 2 +-
tools/intel_gpu_top.c | 8
tools/intel_reg_dumper.c | 2 +-
5 files changed
Also intel_iosf_read() does not exist, and would need a bit more
arguments.
Signed-off-by: Damien Lespiau
---
tools/intel_iosf_read.c | 70 -
1 file changed, 70 deletions(-)
delete mode 100644 tools/intel_iosf_read.c
diff --git a/tools/intel_iosf
On Mon, Dec 16, 2013 at 12:02:53PM +, Damien Lespiau wrote:
> 4 pieces of code were looking very similar. Let's factor out a common
> function in the not so unlikely case we need to tweak that code.
>
> Signed-off-by: Damien Lespiau
Pushed.
--
Damien
> ---
> tools/intel_error_decode.c |
On Sun, Jan 05, 2014 at 10:08:42AM +0900, Tetsuo Handa wrote:
> Chris Wilson wrote:
> > > Then, this file->pid in &dev->filelist keeps at least one reference.
> > > OK. Updated patch follows.
> >
> > Looks good to me, and the comment is great.
> > -Chris
>
> Please pick up this patch via your git
Am Dienstag, 31. Dezember 2013, 13:52:05 schrieb Martin Steigerwald:
> Am Dienstag, 31. Dezember 2013, 13:41:22 schrieb Martin Steigerwald:
> > Am Samstag, 30. November 2013, 14:53:51 schrieb Martin Steigerwald:
> > > Just added linux-mmc. And I might git-bisect that at some time, but I do
> > > no
Am Samstag, 30. November 2013, 14:53:51 schrieb Martin Steigerwald:
> Just added linux-mmc. And I might git-bisect that at some time, but I do not
> intend to do it during my precious weekend. The chances of me bisecting it
> increase with workable suggestions on how to cut down the amount of
> ite
Hi guys,
I just got a Retina MBP with Iris Pro and Nvidia GT 750. The iGPU works in
OSX(Mavericks), but when I boot into Linux, the device was disabled. I
debugged the kernel a little, apparently when pci_bus_read_dev_vendor_id is
called for PCI 00:2.0 (the intel gpu), it returns -1 which indic
36 matches
Mail list logo
