diff mbox

add support for AM572x in the DDX

Message ID 1417790523-4384-1-git-send-email-robertcnelson@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Robert Nelson Dec. 5, 2014, 2:42 p.m. UTC
Signed-off-by: Robert Nelson <robertcnelson@gmail.com>
---
 README            | 1 +
 src/omap_driver.c | 3 +++
 src/omap_driver.h | 2 +-
 3 files changed, 5 insertions(+), 1 deletion(-)

Comments

Emil Velikov Dec. 5, 2014, 4:57 p.m. UTC | #1
Adding the freedreno ML to the Cc list :-)

-Emil

On 5 December 2014 at 14:42, Robert Nelson <robertcnelson@gmail.com> wrote:
> Signed-off-by: Robert Nelson <robertcnelson@gmail.com>
> ---
>  README            | 1 +
>  src/omap_driver.c | 3 +++
>  src/omap_driver.h | 2 +-
>  3 files changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/README b/README
> index 059f258..d4afd89 100644
> --- a/README
> +++ b/README
> @@ -9,6 +9,7 @@ the following chipsets:
>    + OMAP4460
>    + OMAP5430
>    + OMAP5432
> +  + AM572x
>
>
>  NOTE: this driver is work in progress..  you probably don't want to try
> diff --git a/src/omap_driver.c b/src/omap_driver.c
> index 1fc188d..b9d90c5 100644
> --- a/src/omap_driver.c
> +++ b/src/omap_driver.c
> @@ -95,6 +95,7 @@ static SymTabRec OMAPChipsets[] = {
>                 /*    { 4470, "OMAP4470 with <redacted> ;-)" }, */
>                 { 0x5430, "OMAP5430 with PowerVR SGX544 MP" },
>                 { 0x5432, "OMAP5432 with PowerVR SGX544 MP" },
> +               { 0x0752, "AM572x with PowerVR SGX544 MP" },
>                 {-1, NULL }
>  };
>
> @@ -620,6 +621,7 @@ OMAPPreInit(ScrnInfoPtr pScrn, int flags)
>         case 0x4460:
>         case 0x5430:
>         case 0x5432:
> +       case 0x0752:
>                 if (xf86LoadSubModule(pScrn, SUB_MODULE_PVR)) {
>                         INFO_MSG("Loaded the %s sub-module", SUB_MODULE_PVR);
>                 } else {
> @@ -663,6 +665,7 @@ OMAPAccelInit(ScreenPtr pScreen)
>                 case 0x4460:
>                 case 0x5430:
>                 case 0x5432:
> +               case 0x0752:
>                         INFO_MSG("Initializing the \"%s\" sub-module ...", SUB_MODULE_PVR);
>                         pOMAP->pOMAPEXA = InitPowerVREXA(pScreen, pScrn, pOMAP->drmFD);
>                         if (pOMAP->pOMAPEXA) {
> diff --git a/src/omap_driver.h b/src/omap_driver.h
> index 6d2517e..8277e25 100644
> --- a/src/omap_driver.h
> +++ b/src/omap_driver.h
> @@ -184,7 +184,7 @@ static inline Bool has_video(OMAPPtr pOMAP)
>
>  static inline Bool has_dmm(OMAPPtr pOMAP)
>  {
> -       return pOMAP->chipset >= 0x4430;
> +       return pOMAP->chipset >= 0x4430 || pOMAP->chipset == 0x0752;
>  }
>
>  static inline Bool has_rotation(OMAPPtr pOMAP)
> --
> 2.1.3
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
Robert Nelson Dec. 5, 2014, 5:01 p.m. UTC | #2
On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
> Adding the freedreno ML to the Cc list :-)

Sorry Emil,

underneath this is omapdrm + PowerVR, so i should have prefixed it as:
"xf86-video-omap"...

Regards,
Emil Velikov Dec. 5, 2014, 5:19 p.m. UTC | #3
On 5 December 2014 at 17:01, Robert Nelson <robertcnelson@gmail.com> wrote:
> On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
>> Adding the freedreno ML to the Cc list :-)
>
> Sorry Emil,
>
> underneath this is omapdrm + PowerVR, so i should have prefixed it as:
> "xf86-video-omap"...
>
Hi Robert,

My sleep deprived brain read omap, but registered freedreno. Or
perhaps I'm a bit excited about the contributions from the
CodeAurora/QCom guys in the latter case, which clouded my judgement
:-)

Cheers,
Emil
Robert Nelson Dec. 5, 2014, 5:27 p.m. UTC | #4
On Fri, Dec 5, 2014 at 11:19 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
> On 5 December 2014 at 17:01, Robert Nelson <robertcnelson@gmail.com> wrote:
>> On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
>>> Adding the freedreno ML to the Cc list :-)
>>
>> Sorry Emil,
>>
>> underneath this is omapdrm + PowerVR, so i should have prefixed it as:
>> "xf86-video-omap"...
>>
> Hi Robert,
>
> My sleep deprived brain read omap, but registered freedreno. Or
> perhaps I'm a bit excited about the contributions from the
> CodeAurora/QCom guys in the latter case, which clouded my judgement
> :-)

We are all excited about the CodeAurora/QCom guys contributions. ;) I
make sure to point that out daily in my imgtec dealings. ;)

Regards,
Rob Clark Dec. 5, 2014, 5:34 p.m. UTC | #5
On Fri, Dec 5, 2014 at 12:01 PM, Robert Nelson <robertcnelson@gmail.com> wrote:
> On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
>> Adding the freedreno ML to the Cc list :-)
>
> Sorry Emil,
>
> underneath this is omapdrm + PowerVR, so i should have prefixed it as:
> "xf86-video-omap"...

Hey Robert,

I can update xf86-video-omap..  where there some other patches
wanted/desired (has_dmm() iirc?)

BR,
-R

> Regards,
>
> --
> Robert Nelson
> http://www.rcn-ee.com/
Robert Nelson Dec. 5, 2014, 5:40 p.m. UTC | #6
On Fri, Dec 5, 2014 at 11:34 AM, Rob Clark <robdclark@gmail.com> wrote:
> On Fri, Dec 5, 2014 at 12:01 PM, Robert Nelson <robertcnelson@gmail.com> wrote:
>> On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
>>> Adding the freedreno ML to the Cc list :-)
>>
>> Sorry Emil,
>>
>> underneath this is omapdrm + PowerVR, so i should have prefixed it as:
>> "xf86-video-omap"...
>
> Hey Robert,
>
> I can update xf86-video-omap..  where there some other patches
> wanted/desired (has_dmm() iirc?)

Hi Rob,

The version i posted today has the "has_dmm" change merged in, which
other then the README update is the only change I made.

static inline Bool has_dmm(OMAPPtr pOMAP)
 {
-       return pOMAP->chipset >= 0x4430;
+       return pOMAP->chipset >= 0x4430 || pOMAP->chipset == 0x0752;
 }

Haven't really heard back from TI about all the other random changes
they made to their fork of your xf86-video-omap tree:

http://git.ti.com/gitweb/?p=glsdk/xf86-video-omap.git;a=summary

Regards,
Rob Clark Dec. 6, 2014, 4:43 p.m. UTC | #7
On Fri, Dec 5, 2014 at 12:40 PM, Robert Nelson <robertcnelson@gmail.com> wrote:
> On Fri, Dec 5, 2014 at 11:34 AM, Rob Clark <robdclark@gmail.com> wrote:
>> On Fri, Dec 5, 2014 at 12:01 PM, Robert Nelson <robertcnelson@gmail.com> wrote:
>>> On Fri, Dec 5, 2014 at 10:57 AM, Emil Velikov <emil.l.velikov@gmail.com> wrote:
>>>> Adding the freedreno ML to the Cc list :-)
>>>
>>> Sorry Emil,
>>>
>>> underneath this is omapdrm + PowerVR, so i should have prefixed it as:
>>> "xf86-video-omap"...
>>
>> Hey Robert,
>>
>> I can update xf86-video-omap..  where there some other patches
>> wanted/desired (has_dmm() iirc?)
>
> Hi Rob,
>
> The version i posted today has the "has_dmm" change merged in, which
> other then the README update is the only change I made.
>
> static inline Bool has_dmm(OMAPPtr pOMAP)
>  {
> -       return pOMAP->chipset >= 0x4430;
> +       return pOMAP->chipset >= 0x4430 || pOMAP->chipset == 0x0752;
>  }
>
> Haven't really heard back from TI about all the other random changes
> they made to their fork of your xf86-video-omap tree:
>
> http://git.ti.com/gitweb/?p=glsdk/xf86-video-omap.git;a=summary

gotcha, well I've pushed your patch.  I don't really have the hw
unpacked and setup to test these days, but if someone confirm latest
master is good then I suppose I should spin a release tag for distro's
to pick up..


BR,
-R


> Regards,
>
> --
> Robert Nelson
> http://www.rcn-ee.com/
Robert Nelson Dec. 6, 2014, 7:19 p.m. UTC | #8
> gotcha, well I've pushed your patch.  I don't really have the hw
> unpacked and setup to test these days, but if someone confirm latest
> master is good then I suppose I should spin a release tag for distro's
> to pick up..

Thanks Rob!

I'll test and re-confirm on my x15 alpha board when i'm back in the
lab on monday.

Regards,
Tomi Valkeinen Dec. 8, 2014, 8:45 a.m. UTC | #9
On 06/12/14 21:19, Robert Nelson wrote:
>> gotcha, well I've pushed your patch.  I don't really have the hw
>> unpacked and setup to test these days, but if someone confirm latest
>> master is good then I suppose I should spin a release tag for distro's
>> to pick up..
> 
> Thanks Rob!
> 
> I'll test and re-confirm on my x15 alpha board when i'm back in the
> lab on monday.

I tested xf86-video-omap HEAD on my dra7-evm, and X starts fine now.

 Tomi
diff mbox

Patch

diff --git a/README b/README
index 059f258..d4afd89 100644
--- a/README
+++ b/README
@@ -9,6 +9,7 @@  the following chipsets:
   + OMAP4460
   + OMAP5430
   + OMAP5432
+  + AM572x
 
 
 NOTE: this driver is work in progress..  you probably don't want to try
diff --git a/src/omap_driver.c b/src/omap_driver.c
index 1fc188d..b9d90c5 100644
--- a/src/omap_driver.c
+++ b/src/omap_driver.c
@@ -95,6 +95,7 @@  static SymTabRec OMAPChipsets[] = {
 		/*    { 4470, "OMAP4470 with <redacted> ;-)" }, */
 		{ 0x5430, "OMAP5430 with PowerVR SGX544 MP" },
 		{ 0x5432, "OMAP5432 with PowerVR SGX544 MP" },
+		{ 0x0752, "AM572x with PowerVR SGX544 MP" },
 		{-1, NULL }
 };
 
@@ -620,6 +621,7 @@  OMAPPreInit(ScrnInfoPtr pScrn, int flags)
 	case 0x4460:
 	case 0x5430:
 	case 0x5432:
+	case 0x0752:
 		if (xf86LoadSubModule(pScrn, SUB_MODULE_PVR)) {
 			INFO_MSG("Loaded the %s sub-module", SUB_MODULE_PVR);
 		} else {
@@ -663,6 +665,7 @@  OMAPAccelInit(ScreenPtr pScreen)
 		case 0x4460:
 		case 0x5430:
 		case 0x5432:
+		case 0x0752:
 			INFO_MSG("Initializing the \"%s\" sub-module ...", SUB_MODULE_PVR);
 			pOMAP->pOMAPEXA = InitPowerVREXA(pScreen, pScrn, pOMAP->drmFD);
 			if (pOMAP->pOMAPEXA) {
diff --git a/src/omap_driver.h b/src/omap_driver.h
index 6d2517e..8277e25 100644
--- a/src/omap_driver.h
+++ b/src/omap_driver.h
@@ -184,7 +184,7 @@  static inline Bool has_video(OMAPPtr pOMAP)
 
 static inline Bool has_dmm(OMAPPtr pOMAP)
 {
-	return pOMAP->chipset >= 0x4430;
+	return pOMAP->chipset >= 0x4430 || pOMAP->chipset == 0x0752;
 }
 
 static inline Bool has_rotation(OMAPPtr pOMAP)