Bump NXP package versions

---
 package/freescale-imx/firmware-imx/firmware-imx.hash | 2 +-
 package/freescale-imx/firmware-imx/firmware-imx.mk   | 5 ++---
 2 files changed, 3 insertions(+), 4 deletions(-)

package/freescale-imx/firmware-imx/firmware-imx.hash +1 -1

index a5bc82c..a6e5bd5 100644
@@ -1 +1 @@
-sha256 fd14c3666eb3c12f4f2665d95caf383bb1d0277bb12e2de28fefc8d1230604f2  firmware-imx-5.3.bin
+sha256 c5bd4bff48cce9715a5d6d2c190ff3cd2262c7196f7facb9b0eda231c92cc223  firmware-imx-5.4.bin

package/freescale-imx/firmware-imx/firmware-imx.mk +2 -3

index 8f81c57..1893929 100644
@@ -4,12 +4,11 @@
 #
 ################################################################################
 
-FIRMWARE_IMX_VERSION = 5.3
+FIRMWARE_IMX_VERSION = 5.4
 FIRMWARE_IMX_SITE = $(FREESCALE_IMX_SITE)
 FIRMWARE_IMX_SOURCE = firmware-imx-$(FIRMWARE_IMX_VERSION).bin
 
-FIRMWARE_IMX_LICENSE = Freescale Semiconductor Software License Agreement, \
-	Atheros license (ath6k)
+FIRMWARE_IMX_LICENSE = NXP Semiconductor Software License Agreement
 FIRMWARE_IMX_LICENSE_FILES = EULA COPYING
 FIRMWARE_IMX_REDISTRIBUTE = NO

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:32 UTC #2 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Andrew, All,

Quoted 7 lines · Andrew Webster

Changelog:
- Upgrade to FW v3.1.1_r46072
- v3.1.1_r46070 plus 2 patches 391/392
- FSLARIK-391 [Decoder/H.264] VPU is always busy when decoding an erroneous H264 bitstream
- FSLARIK-392 [Decoder/VC1] VPU outputs only 1 picture
- Change to the format of Linux coda driver firmware

Hmm, that's cool because it means the firmware can now be used by mainline kernel seamlessly, but it also means that it won't work on any kernel < 4.1.15. Do you confirm you've tried it on 4.1.15 kernel only?

This means that updating defconfig should also be part of this series in my opinion,@least for the NXP platforms.

Quoted 2 lines

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773

Please share how you've tested the VPU firmware for instance.

Regards,
Gary

Andrew Webster

author

2016-06-06 17:12 UTC #3 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Quoted 19 lines

-----Original Message-----
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Monday, June 06, 2016 10:33 AM
To: Andrew Webster
Cc: buildroot at buildroot.org
Subject: Re: [Buildroot] [PATCH 01/10] firmware-imx: bump to version 5.4
Andrew, All,
Quoted 7 lines · Andrew Webster
Changelog:
- Upgrade to FW v3.1.1_r46072
- v3.1.1_r46070 plus 2 patches 391/392
- FSLARIK-391 [Decoder/H.264] VPU is always busy when decoding an erroneous H264 bitstream
- FSLARIK-392 [Decoder/VC1] VPU outputs only 1 picture
- Change to the format of Linux coda driver firmware

Quoted 3 lines

Hmm, that's cool because it means the firmware can now be used by mainline kernel seamlessly, but it also means that it won't work on any kernel < 4.1.15. Do you confirm you've tried it on 4.1.15 kernel only?

Correct, tested against the 4.1.15 kernel only.

Quoted 3 lines

This means that updating defconfig should also be part of this series in my opinion, at least for the NXP platforms.

What would be the verification plan for defconfig updates? Build each one and check for build errors?

Quoted 5 lines

Quoted 3 lines
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773
Please share how you've tested the VPU firmware for instance.

I followed the same tests from your previous version bumps (e.g. gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! matroskademux ! imxvpudec ! imxipuvideosink). I can add the test(s) performed to the commit messages if you like.

Quoted 3 lines

Regards,
Gary

Gary Bisson

2016-06-07 07:36 UTC #4 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Andrew, All,

Quoted 19 lines · Andrew Webster

Quoted 19 lines
-----Original Message-----
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Monday, June 06, 2016 10:33 AM
To: Andrew Webster
Cc: buildroot at buildroot.org
Subject: Re: [Buildroot] [PATCH 01/10] firmware-imx: bump to version 5.4
Andrew, All,
Quoted 7 lines · Andrew Webster
Changelog:
- Upgrade to FW v3.1.1_r46072
- v3.1.1_r46070 plus 2 patches 391/392
- FSLARIK-391 [Decoder/H.264] VPU is always busy when decoding an erroneous H264 bitstream
- FSLARIK-392 [Decoder/VC1] VPU outputs only 1 picture
- Change to the format of Linux coda driver firmware

Quoted 5 lines

Quoted 3 lines
Hmm, that's cool because it means the firmware can now be used by mainline kernel seamlessly, but it also means that it won't work on any kernel < 4.1.15. Do you confirm you've tried it on 4.1.15 kernel only?
Correct, tested against the 4.1.15 kernel only.

I'll try but I'm pretty sure this won't work on 3.14.52.

Quoted 4 lines

Quoted 2 lines
This means that updating defconfig should also be part of this series in my opinion, at least for the NXP platforms.
What would be the verification plan for defconfig updates? Build each one and check for build errors?

Yes a build check is sufficient, that is what I've been doing before.

Quoted 6 lines

Quoted 4 lines
Quoted 2 lines
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773
Please share how you've tested the VPU firmware for instance.
I followed the same tests from your previous version bumps (e.g. gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! matroskademux ! imxvpudec ! imxipuvideosink). I can add the test(s) performed to the commit messages if you like.

Yes please, at least in the cover letter so anyone knows how to make sure the package actually works.

Thanks,
Gary

Andrew Webster

author

2016-06-08 17:26 UTC #5 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Gary,

Quoted 10 lines

From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Tuesday, June 07, 2016 3:37 AM
Quoted 5 lines
Quoted 3 lines
Hmm, that's cool because it means the firmware can now be used by mainline kernel seamlessly, but it also means that it won't work on any kernel < 4.1.15. Do you confirm you've tried it on 4.1.15 kernel only?
Correct, tested against the 4.1.15 kernel only.
I'll try but I'm pretty sure this won't work on 3.14.52.

There will be a similar problem with the imx-gpu-viv package. I believe that the kernel version and the gpu library need to match (correct me if I'm wrong). This would mean that if either the kernel or imx-gpu-viv are updated without the other, things may not work properly at run-time. I expect that the X-windows driver also needs to match the others. So, no matter the order of the patches, run-time will be broken between some patches until others have been applied.

I am thinking to re-order the patches as follows:

1. defconfig kernel version bump [NEW]
2. kernel-module-imx-gpu-viv [NEW]
3. imx-gpu-viv bump
4. xdriver_xf86-video-viv bump
5. firmware-imx bump
6. gst1-imx bump
7. imx-vpu bump
8. imx-codec bump/rename
9. imx-parser bump/rename
10. imx-vpuwrap bump/rename
11. libimxvpuapi bump
12. imx-lib bump
13. imx-kobs bump [NEW]

Does this make sense to you?

Thanks,
Andrew

Gary Bisson

2016-06-08 18:20 UTC #6 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Andrew, All,

Le 8 juin 2016 7:33 PM, "Andrew Webster" <[email protected]> a ?crit :

Quoted 9 lines

Gary,
Quoted 6 lines
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Tuesday, June 07, 2016 3:37 AM
Quoted 3 lines
Quoted 3 lines
Hmm, that's cool because it means the firmware can now be used by mainline kernel seamlessly, but it also means that it won't work on any kernel < 4.1.15. Do you confirm you've tried it on 4.1.15 kernel

only?

Quoted 6 lines

Quoted 4 lines
Quoted 2 lines
Correct, tested against the 4.1.15 kernel only.
I'll try but I'm pretty sure this won't work on 3.14.52.
There will be a similar problem with the imx-gpu-viv package. I believe

that the kernel version and the gpu library need to match (correct me if I'm wrong). This would mean that if either the kernel or imx-gpu-viv are updated without the other, things may not work properly at run-time. I expect that the X-windows driver also needs to match the others. So, no matter the order of the patches, run-time will be broken between some patches until others have been applied.

The difference is that there is nothing you can do about the VPU whereas the graphics libraries always match the kernel module which can be built against any kernel >= 3.10.53.

Also I don't mind the order that much, it is just that it makes more sense to have graphics patches together and multimedia one on their own as they are not dependent on each other.

Quoted 16 lines

I am thinking to re-order the patches as follows:
1. defconfig kernel version bump [NEW]
2. kernel-module-imx-gpu-viv [NEW]
3. imx-gpu-viv bump
4. xdriver_xf86-video-viv bump
5. firmware-imx bump
6. gst1-imx bump
7. imx-vpu bump
8. imx-codec bump/rename
9. imx-parser bump/rename
10. imx-vpuwrap bump/rename
11. libimxvpuapi bump
12. imx-lib bump
13. imx-kobs bump [NEW]

Quoted 1 line

Does this make sense to you?

Yes that looks good to me.

Regards,
Gary
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20160608/c253546c/attachment.html>

Andrew Webster

author

2016-06-09 21:18 UTC #7 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Gary,

Quoted 6 lines

From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Tuesday, June 07, 2016 3:37 AM
Quoted 3 lines
Quoted 3 lines
Quoted 2 lines
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773

Quoted 7 lines

Quoted 3 lines
Quoted 1 line
Please share how you've tested the VPU firmware for instance.
I followed the same tests from your previous version bumps (e.g. gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm
! matroskademux ! imxvpudec ! imxipuvideosink). I can add the test(s) performed to the commit messages if you like.
Yes please, at least in the cover letter so anyone knows how to make sure the package actually works.

Given that the gstreamer-0.10 plugins from NXP are no longer being updated, and don't seem to work with the new packages (or maybe I didn't configure things properly), do you know of another way to test the imx-codec, imx-parser, and imx-vpuwrap packages?

Andrew

Gary Bisson

2016-06-10 08:46 UTC #8 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Andrew, All,

Quoted 11 lines · Andrew Webster

Gary,
Quoted 8 lines
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Tuesday, June 07, 2016 3:37 AM
Quoted 5 lines
Quoted 4 lines
Quoted 2 lines
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773
Please share how you've tested the VPU firmware for instance.

Quoted 7 lines

Quoted 5 lines
Quoted 1 line
I followed the same tests from your previous version bumps (e.g. gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm
! matroskademux ! imxvpudec ! imxipuvideosink). I can add the test(s) performed to the commit messages if you like.
Yes please, at least in the cover letter so anyone knows how to make sure the package actually works.
Given that the gstreamer-0.10 plugins from NXP are no longer being updated, and don't seem to work with the new packages (or maybe I didn't configure things properly), do you know of another way to test the imx-codec, imx-parser, and imx-vpuwrap packages?

Why do you say it doesn't seem to work with new packages? Up until now it was still working, do you have a build failure? If so, we need to deprecate the gst-fsl-plugins of Gstreamer0.10.

Regarding the packages, yes right now they are only used by the Gstreamer0.10 plugins. There's another Gstreamer1.0 plugin provided by NXP that makes use of those packages but I've always preferred the clean and open-source Gstreamer-imx implementation so I never bothered integrating it (and don't plan to). If we have to deprecate the Gstreamer0.10 FSL plugins, then we can ask ourselves the question as to whether or not we should keep those packages.

Regards,
Gary

Andrew Webster

author

2016-06-10 14:39 UTC #9 lore raw

PATCH 1/10 firmware-imx: bump to version 5.4

Quoted 12 lines

From: buildroot [mailto:buildroot-bounces at busybox.net] On Behalf Of Gary Bisson
Sent: Friday, June 10, 2016 4:47 AM
Quoted 9 lines
Quoted 9 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773
Please share how you've tested the VPU firmware for instance.
I followed the same tests from your previous version bumps (e.g. gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm
! matroskademux ! imxvpudec ! imxipuvideosink). I can add the test(s) performed to the commit messages if you like.

Quoted 8 lines

Quoted 4 lines
Quoted 2 lines
Yes please, at least in the cover letter so anyone knows how to make sure the package actually works.
Given that the gstreamer-0.10 plugins from NXP are no longer being updated, and don't seem to work with the new packages (or
maybe I didn't configure things properly), do you know of another way to test the imx-codec, imx-parser, and imx-vpuwrap packages?
Why do you say it doesn't seem to work with new packages? Up until now it was still working, do you have a build failure? If so, we need to deprecate the gst-fsl-plugins of Gstreamer0.10.

The build was successful, but I was having run-time issues. gst-launch-0.10 would start and claim to be playing, but I saw nothing. There were VPU errors in the kernel log. This may have been a build failure of some kind, because after a full re-build, it now seems to be working. I am still suspicious that there may be problem.

However, given that it builds, I guess there is no immediate need remove it. Note that support was removed from Yocto in https://github.com/Freescale/meta-fsl-arm/commit/83b240e.

Quoted 5 lines

Regarding the packages, yes right now they are only used by the
Gstreamer0.10 plugins. There's another Gstreamer1.0 plugin provided by NXP that makes use of those packages but I've always preferred the clean and open-source Gstreamer-imx implementation so I never bothered integrating it (and don't plan to). If we have to deprecate the
Gstreamer0.10 FSL plugins, then we can ask ourselves the question as to whether or not we should keep those packages.

I would think that these could be useful even without the gst0.10 plugins (e.g. for Kodi), but they would become more difficult to run-time test.

Andrew

Andrew Webster

author

2016-06-03 13:10 UTC #10 lore raw

PATCH 2/10 gst1-imx: bump to version 0.12.1

Changelog:
* compositor: Rename ImxCompositor(Pad) to ImxVideoCompositor(Pad) This prevents conflicts if both the NXP plugins and the gstreamer-imx are installed in the same rootfs

* Add switches to wscript configuration to enable/disable plugins Until now, all plugins were built (unless their dependencies were missing). On some i.MX machines, some features are missing. For example, the i.MX7 has no VPU. To ensure deterministic builds, it is useful to selectively disable plugins in the build.

* ipu+pxp: : Fix deadlocks and incorrect logging in case of init error This was causing a freeze if for example one tries to use PxP elements on an i.MX6 Dual/Quad

* imxv4l2videosrc fixes and enhancements:
* fixes to interlacing support (incl. workarounds for analog TV input)
* remove format restriction in get_caps function
this removes the limitation on I420 formats
* simplified determination of fps (necessary due to driver limitations)

* eglvivsink: wayland: Do not commit before buffers are swapped This fixes eglSwapBuffer swaps with damaged frames

* Documentation additions
Adds GPU memory leak notess to the FAQ

* vpu: encoder: Expose motion estimation range as property

* vpu: decoder: Drop output frame if its buffer cannot be allocated

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/cec4c47

---
 package/gstreamer1/gst1-imx/gst1-imx.hash | 2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/gstreamer1/gst1-imx/gst1-imx.hash +1 -1

index 03e1729..ca47776 100644
@@ -1,2 +1,2 @@
 # locally computed hash
-sha256 df34f080e24ac8f1619780aa5ad4e6ab1333f822cd2bf5fcb380b28c97248810  gst1-imx-0.12.0.tar.gz
+sha256 d8a7357bc575e22d0d19d6d7cf547bd69cad5fc47adb93f09ead41b0124353fa  gst1-imx-0.12.1.tar.gz

package/gstreamer1/gst1-imx/gst1-imx.mk +1 -1

index 650e097..8838110 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-GST1_IMX_VERSION = 0.12.0
+GST1_IMX_VERSION = 0.12.1
 GST1_IMX_SITE = $(call github,Freescale,gstreamer-imx,$(GST1_IMX_VERSION))
 
 GST1_IMX_LICENSE = LGPLv2+

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:33 UTC #11 lore raw

PATCH 2/10 gst1-imx: bump to version 0.12.1

Andrew, All,

Quoted 10 lines · Andrew Webster

Changelog:
* compositor: Rename ImxCompositor(Pad) to ImxVideoCompositor(Pad) This prevents conflicts if both the NXP plugins and the gstreamer-imx are installed in the same rootfs
* Add switches to wscript configuration to enable/disable plugins Until now, all plugins were built (unless their dependencies were missing). On some i.MX machines, some features are missing. For example, the i.MX7 has no VPU. To ensure deterministic builds, it is useful to selectively disable plugins in the build.

This update is actually very important, we can now explicitly select which element we want to be built. As said to Baruch[1], I think this package should be reworked more deeply in order to have a menu allowing to select the elements to be built.

Since the switches to wscript configuration are brought by this update, my opinion is that it should be done in one patch. I can take care of it if you don't want to.

Regards,
Gary

[1]: http://patchwork.ozlabs.org/patch/611642/

Andrew Webster

author

2016-06-06 18:19 UTC #12 lore raw

PATCH 2/10 gst1-imx: bump to version 0.12.1

Quoted 22 lines

-----Original Message-----
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Monday, June 06, 2016 10:33 AM
To: Andrew Webster
Cc: buildroot at buildroot.org
Subject: Re: [Buildroot] [PATCH 02/10] gst1-imx: bump to version 0.12.1
Andrew, All,
Quoted 10 lines · Andrew Webster
Changelog:
* compositor: Rename ImxCompositor(Pad) to ImxVideoCompositor(Pad) This prevents conflicts if both the NXP plugins and the gstreamer-imx are installed in the same rootfs
* Add switches to wscript configuration to enable/disable plugins Until now, all plugins were built (unless their dependencies were missing). On some i.MX machines, some features are missing. For example, the i.MX7 has no VPU. To ensure deterministic builds, it is useful to selectively disable plugins in the build.

Quoted 8 lines

This update is actually very important, we can now explicitly select which element we want to be built. As said to Baruch[1], I think this package should be reworked more deeply in order to have a menu allowing to select the elements to be built.
Since the switches to wscript configuration are brought by this update, my opinion is that it should be done in one patch. I can take care of it if you don't want to.

I'll take a look to see what's required here and get back to you.

Quoted 5 lines

Regards,
Gary
[1]: http://patchwork.ozlabs.org/patch/611642/

Thomas Petazzoni

2016-06-06 20:28 UTC #13 lore raw

PATCH 2/10 gst1-imx: bump to version 0.12.1

Hello,

Quoted 3 lines · Gary Bisson

Since the switches to wscript configuration are brought by this update, my opinion is that it should be done in one patch. I can take care of it if you don't want to.

Well, I would suggest to do the opposite: do the bump, and then make improvements to the package that are made possible by the bump.

Of course, if it's impossible for some reason to separate these in two patches, it should be done in only one patch.

But if the bump can be done without these additional improvement, they should be separate changes.

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Gary Bisson

2016-06-06 20:37 UTC #14 lore raw

PATCH 2/10 gst1-imx: bump to version 0.12.1

Thomas, All,

Quoted 8 lines · Thomas Petazzoni

Hello,
Quoted 3 lines · Gary Bisson
Since the switches to wscript configuration are brought by this update, my opinion is that it should be done in one patch. I can take care of it if you don't want to.

Quoted 8 lines

Well, I would suggest to do the opposite: do the bump, and then make improvements to the package that are made possible by the bump.
Of course, if it's impossible for some reason to separate these in two patches, it should be done in only one patch.
But if the bump can be done without these additional improvement, they should be separate changes.

It is possible, nothing prevents it I just thought it'd be better to have everything in one patch.

In that case, I think the patch from Baruch should also be merged before the rework.

Andrew, you can keep your patch as-is.

Regards,
Gary

Andrew Webster

author

2016-06-03 13:10 UTC #15 lore raw

PATCH 3/10 imx-vpu: bump to version 5.4.33

Changelog:
VPU firmware has been changed to the format of Linux codec driver firmware. Change VPU library to adapt to firmware format change.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/f5e1056

---
 package/freescale-imx/imx-vpu/imx-vpu.hash | 2 +-
 package/freescale-imx/imx-vpu/imx-vpu.mk   | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

package/freescale-imx/imx-vpu/imx-vpu.hash +1 -1

index c828c2b..67ebaf2 100644
@@ -1,2 +1,2 @@
 # Locally computed
-sha256	f11b91dcdb84bfefd3c88f10a9c1c1bc3420e4bbaadc38b9280fb52f7769eca8  imx-vpu-5.4.32.bin
+sha256	2e670ef2337a5dbe6c32fa07128ad9aad32afdbf472c23d44133245ec6a66d46  imx-vpu-5.4.33.bin

package/freescale-imx/imx-vpu/imx-vpu.mk +2 -2

index 949fefa..09ff5a8 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-IMX_VPU_VERSION = 5.4.32
+IMX_VPU_VERSION = 5.4.33
 IMX_VPU_SITE = $(FREESCALE_IMX_SITE)
 IMX_VPU_SOURCE = imx-vpu-$(IMX_VPU_VERSION).bin

@@ -16,7 +16,7 @@ IMX_VPU_MAKE_ENV = \
 	CROSS_COMPILE="$(TARGET_CROSS)" \
 	PLATFORM=$(BR2_PACKAGE_FREESCALE_IMX_PLATFORM)
 
-IMX_VPU_LICENSE = Freescale Semiconductor Software License Agreement
+IMX_VPU_LICENSE = NXP Semiconductor Software License Agreement
 IMX_VPU_LICENSE_FILES = EULA COPYING
 IMX_VPU_REDISTRIBUTE = NO

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:33 UTC #16 lore raw

PATCH 3/10 imx-vpu: bump to version 5.4.33

Andrew, All,

Quoted 8 lines · Andrew Webster

Changelog:
VPU firmware has been changed to the format of Linux codec driver firmware. Change VPU library to adapt to firmware format change.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/f5e1056
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-03 13:10 UTC #17 lore raw

PATCH 4/10 imx-codec: rename from libfslcodec and bump version to 4.0.9

Changelog:
--[aac_dec] refine the code for pure c build
aac decoder c build need float point support, which use some float point function.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/4b91194

---
 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-codec}/Config.in                           |  6 +++---
 package/freescale-imx/imx-codec/imx-codec.hash     |  2 ++
 package/freescale-imx/imx-codec/imx-codec.mk       | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  4 ++--
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/gstreamer1/gst1-imx/Config.in              |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk            |  4 ++--
 package/libfslcodec/libfslcodec.hash               |  2 --
 package/libfslcodec/libfslcodec.mk                 | 23 ----------------------
 12 files changed, 42 insertions(+), 35 deletions(-)
 rename package/{libfslcodec => freescale-imx/imx-codec}/Config.in (75%)
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.hash
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.mk
 delete mode 100644 package/libfslcodec/libfslcodec.hash
 delete mode 100644 package/libfslcodec/libfslcodec.mk

Config.in.legacy +7 -0

index b73603b..54d43f1 100644
@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLCODEC
+	bool "libfslcodec has been renamed to imx-codec"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_CODEC
+	help
+	  The libfslcodec has been renamed to match the renamed package.
+
 ###############################################################################
 comment "Legacy options removed in 2016.08"

package/Config.in +0 -1

index ae74f17..c41e4f6 100644
@@ -1089,7 +1089,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslcodec/Config.in"
 	source "package/libfslparser/Config.in"
 	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"

package/freescale-imx/Config.in +1 -0

index 8c03190..052de5a 100644
@@ -49,6 +49,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 		BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \
 		BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q
 
+source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"

package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in +3 -3

similarity index 75%
rename from package/libfslcodec/Config.in
rename to package/freescale-imx/imx-codec/Config.in
index d2007fe..e161699 100644
@@ -1,5 +1,5 @@
-config BR2_PACKAGE_LIBFSLCODEC
-	bool "libfslcodec"
+config BR2_PACKAGE_IMX_CODEC
+	bool "imx-codec"
 	depends on BR2_arm # Only relevant for i.MX
 	depends on BR2_TOOLCHAIN_USES_GLIBC # prebuilt binaries
 	help

@@ -8,6 +8,6 @@ config BR2_PACKAGE_LIBFSLCODEC
 	  This library is provided by Freescale as-is and doesn't have
 	  an upstream.
 
-comment "libfslcodec needs an (e)glibc toolchain"
+comment "imx-codec needs an (e)glibc toolchain"
 	depends on BR2_arm
 	depends on !BR2_TOOLCHAIN_USES_GLIBC

package/freescale-imx/imx-codec/imx-codec.hash new file +2 -0

new file mode 100644
index 0000000..39bddc1
@@ -0,0 +1,2 @@
+# locally computed
+sha256	62bda182143cde9743cba0eb2a1c315a53cb434294605742f9fc364219958888  imx-codec-4.0.9.bin

package/freescale-imx/imx-codec/imx-codec.mk new file +23 -0

new file mode 100644
index 0000000..67d3390
@@ -0,0 +1,23 @@
+################################################################################
+#
+# imx-codec
+#
+################################################################################
+
+IMX_CODEC_VERSION = 4.0.9
+IMX_CODEC_SITE = $(FREESCALE_IMX_SITE)
+IMX_CODEC_SOURCE = imx-codec-$(IMX_CODEC_VERSION).bin
+IMX_CODEC_INSTALL_STAGING = YES
+
+IMX_CODEC_LICENSE = NXP Semiconductor Software License Agreement, BSD-3c (flac, ogg headers)
+IMX_CODEC_LICENSE_FILES = EULA COPYING
+IMX_CODEC_REDISTRIBUTE = NO
+
+define IMX_CODEC_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_CODEC_SOURCE))
+endef
+
+# FIXME The Makefile installs both the arm9 and arm11 versions of the
+# libraries, but we only need one of them.
+
+$(eval $(autotools-package))

package/gstreamer/gst-fsl-plugins/Config.in +2 -2

index 0407dc5..8c82a4a 100644
@@ -9,12 +9,12 @@ config BR2_PACKAGE_GST_FSL_PLUGINS
 	bool "gst-fsl-plugins"
 	depends on BR2_LINUX_KERNEL
 	depends on BR2_arm # Only relevant for i.MX
-	depends on BR2_TOOLCHAIN_USES_GLIBC # libfslcodec
+	depends on BR2_TOOLCHAIN_USES_GLIBC # imx-codec
 	select BR2_PACKAGE_GST_PLUGINS_BASE
 	select BR2_PACKAGE_LIBFSLVPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 	select BR2_PACKAGE_IMX_LIB
 	select BR2_PACKAGE_LIBFSLPARSER
-	select BR2_PACKAGE_LIBFSLCODEC
+	select BR2_PACKAGE_IMX_CODEC
 	select BR2_PACKAGE_FREESCALE_IMX
 	help
 	  GStreamer plugins for hardware-accelerated audio and video

package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk +1 -1

index fa0e093..50732bf 100644
@@ -16,7 +16,7 @@ GST_FSL_PLUGINS_INSTALL_STAGING = YES
 GST_FSL_PLUGINS_AUTORECONF = YES
 
 GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \
-	imx-lib libfslparser libfslcodec
+	imx-lib libfslparser imx-codec
 
 ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y)
 GST_FSL_PLUGINS_DEPENDENCIES += libfslvpuwrap

package/gstreamer1/gst1-imx/Config.in +1 -1

index 0084483..3512666 100644
@@ -18,6 +18,6 @@ config BR2_PACKAGE_GST1_IMX
 	  The VPU plugin is built when BR2_PACKAGE_LIBIMXVPUAPI is enabled.
 	  The EGL plugin is built when BR2_PACKAGE_IMX_GPU_VIV is enabled.
 	  The G2D plugin is built when BR2_PACKAGE_IMX_GPU_VIV_G2D is enabled.
-	  The MP3 plugin is built when BR2_PACKAGE_LIBFSLCODEC is enabled.
+	  The MP3 plugin is built when BR2_PACKAGE_IMX_CODEC is enabled.
 
 	  https://github.com/Freescale/gstreamer-imx

package/gstreamer1/gst1-imx/gst1-imx.mk +2 -2

index 8838110..f81d8c9 100644
@@ -27,8 +27,8 @@ ifeq ($(BR2_PACKAGE_GST1_PLUGINS_BAD),y)
 GST1_IMX_DEPENDENCIES += gst1-plugins-bad
 endif
 
-ifeq ($(BR2_PACKAGE_LIBFSLCODEC),y)
-GST1_IMX_DEPENDENCIES += libfslcodec
+ifeq ($(BR2_PACKAGE_IMX_CODEC),y)
+GST1_IMX_DEPENDENCIES += imx-codec
 endif
 
 ifeq ($(BR2_PACKAGE_LIBIMXVPUAPI),y)

package/libfslcodec/libfslcodec.hash deleted +0 -2

deleted file mode 100644
index fd934f7..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	544ffc5989bce18ca50c7a826cc03370b1cf8455335e4291ef0c0779b4c33de0  libfslcodec-4.0.8.bin

package/libfslcodec/libfslcodec.mk deleted +0 ~~-23~~

deleted file mode 100644
index 502390f..0000000
@@ -1,23 +0,0 @@
-################################################################################
-#
-# libfslcodec
-#
-################################################################################
-
-LIBFSLCODEC_VERSION = 4.0.8
-LIBFSLCODEC_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLCODEC_SOURCE = libfslcodec-$(LIBFSLCODEC_VERSION).bin
-LIBFSLCODEC_INSTALL_STAGING = YES
-
-LIBFSLCODEC_LICENSE = Freescale Semiconductor Software License Agreement, BSD-3c (flac, ogg headers)
-LIBFSLCODEC_LICENSE_FILES = EULA COPYING
-LIBFSLCODEC_REDISTRIBUTE = NO
-
-define LIBFSLCODEC_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLCODEC_SOURCE))
-endef
-
-# FIXME The Makefile installs both the arm9 and arm11 versions of the
-# libraries, but we only need one of them.
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:33 UTC #18 lore raw

PATCH 4/10 imx-codec: rename from libfslcodec and bump version to 4.0.9

Andrew, All,

Quoted 34 lines · Andrew Webster

Changelog:
--[aac_dec] refine the code for pure c build
aac decoder c build need float point support, which use some float point function.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/4b91194

---
 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-codec}/Config.in                           |  6 +++---
 package/freescale-imx/imx-codec/imx-codec.hash     |  2 ++
 package/freescale-imx/imx-codec/imx-codec.mk       | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  4 ++--
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/gstreamer1/gst1-imx/Config.in              |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk            |  4 ++--
 package/libfslcodec/libfslcodec.hash               |  2 --
 package/libfslcodec/libfslcodec.mk                 | 23 ----------------------
 12 files changed, 42 insertions(+), 35 deletions(-)
 rename package/{libfslcodec => freescale-imx/imx-codec}/Config.in (75%)
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.hash
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.mk
 delete mode 100644 package/libfslcodec/libfslcodec.hash
 delete mode 100644 package/libfslcodec/libfslcodec.mk

Config.in.legacy +0 -0

index b73603b..54d43f1 100644

Config.in.legacy comment "build, or run, in unpredictable ways. " · line 151

Quoted 10 lines

Config.in.legacy +7 -0

@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLCODEC
+       bool "libfslcodec has been renamed to imx-codec"
+       select BR2_LEGACY
+       select BR2_PACKAGE_IMX_CODEC
+       help
+         The libfslcodec has been renamed to match the renamed package.
+

This section should actually be under "Legacy options removed in 2016.08" below.

Quoted 43 lines

Config.in.legacy +0 -0

 ###############################################################################
 comment "Legacy options removed in 2016.08"

package/Config.in +0 -1

index ae74f17..c41e4f6 100644
@@ -1089,7 +1089,6 @@ menu "Multimedia"
        source "package/libdvdnav/Config.in"
        source "package/libdvdread/Config.in"
        source "package/libebml/Config.in"
-       source "package/libfslcodec/Config.in"
        source "package/libfslparser/Config.in"
        source "package/libfslvpuwrap/Config.in"
        source "package/libhdhomerun/Config.in"

package/freescale-imx/Config.in +1 -0

index 8c03190..052de5a 100644
@@ -49,6 +49,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
                BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \
                BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q
 
+source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"

package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in +2 -2

similarity index 75%
rename from package/libfslcodec/Config.in
rename to package/freescale-imx/imx-codec/Config.in
index d2007fe..e161699 100644
@@ -1,5 +1,5 @@
-config BR2_PACKAGE_LIBFSLCODEC
-       bool "libfslcodec"
+config BR2_PACKAGE_IMX_CODEC
+       bool "imx-codec"
        depends on BR2_arm # Only relevant for i.MX
        depends on BR2_TOOLCHAIN_USES_GLIBC # prebuilt binaries
        help

package/freescale-imx/imx-codec/Config.in config BR2_PACKAGE_LIBFSLCODEC · line 9

Quoted 2 lines

package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in +0 -0
@@ -8,6 +8,6 @@ config BR2_PACKAGE_LIBFSLCODEC
          This library is provided by Freescale as-is and doesn't have
          an upstream.

Although it isn't done in the Yocto patch I guess this renaming is a good time to remove the Freescale reference. This library is now provided by NXP as-is, targeting NXP i.MX SoCs.

Regards,
Gary

Andrew Webster

author

2016-06-06 18:27 UTC #19 lore raw

PATCH 4/10 imx-codec: rename from libfslcodec and bump version to 4.0.9

Quoted 51 lines

-----Original Message-----
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Monday, June 06, 2016 10:34 AM
To: Andrew Webster
Cc: buildroot at buildroot.org
Subject: Re: [Buildroot] [PATCH 04/10] imx-codec: rename from libfslcodec and bump version to 4.0.9
Andrew, All,
Quoted 8 lines · Andrew Webster
Changelog:
--[aac_dec] refine the code for pure c build
aac decoder c build need float point support, which use some float point function.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/4b91194
Quoted 8 lines
Signed-off-by Andrew Webster <[email protected]>
---
 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-codec}/Config.in                           |  6 +++---
 package/freescale-imx/imx-codec/imx-codec.hash     |  2 ++
 package/freescale-imx/imx-codec/imx-codec.mk       | 23
++++++++++++++++++++++
Config.in.legacy comment "build, or run, in unpredictable ways.
Quoted 18 lines
 package/gstreamer/gst-fsl-plugins/Config.in        |  4 ++--
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/gstreamer1/gst1-imx/Config.in              |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk            |  4 ++--
 package/libfslcodec/libfslcodec.hash               |  2 --
 package/libfslcodec/libfslcodec.mk                 | 23 ----------------------
 12 files changed, 42 insertions(+), 35 deletions(-)  rename
package/{libfslcodec => freescale-imx/imx-codec}/Config.in (75%)
create mode 100644 package/freescale-imx/imx-codec/imx-codec.hash
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.mk
 delete mode 100644 package/libfslcodec/libfslcodec.hash
 delete mode 100644 package/libfslcodec/libfslcodec.mk
Config.in.legacy index renamed from Config.in.legacy +0 -0
b73603b..54d43f1 100644
Config.in.legacy +0 -0
@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.
"
Quoted 3 lines
Config.in.legacy +0 -0
 comment "----------------------------------------------------"
 endif
 

Quoted 10 lines

Config.in.legacy comment "build, or run, in unpredictable ways. " · line 151
Quoted 7 lines
Config.in.legacy +7 -0
+config BR2_PACKAGE_LIBFSLCODEC
+       bool "libfslcodec has been renamed to imx-codec"
+       select BR2_LEGACY
+       select BR2_PACKAGE_IMX_CODEC
+       help
+         The libfslcodec has been renamed to match the renamed package.
+
This section should actually be under "Legacy options removed in 2016.08" below.

I will move. This snuck in when I rebased and I missed it.

Quoted 57 lines

Quoted 3 lines
##########################################################
############
Quoted 43 lines
######### comment "Legacy options removed in 2016.08"
package/Config.in index renamed from package/Config.in +0 -0
ae74f17..c41e4f6 100644
package/Config.in +0 -1
@@ -1089,7 +1089,6 @@ menu "Multimedia"
        source "package/libdvdnav/Config.in"
        source "package/libdvdread/Config.in"
        source "package/libebml/Config.in"
-       source "package/libfslcodec/Config.in"
        source "package/libfslparser/Config.in"
        source "package/libfslvpuwrap/Config.in"
        source "package/libhdhomerun/Config.in"
(quoted diff) +0 -0
b/package/freescale-imx/Config.in index 8c03190..052de5a 100644
package/freescale-imx/Config.in +1 -0
@@ -49,6 +49,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
                BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \
                BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q
 
+source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in rename to +0 -0
similarity index 75%
rename from package/libfslcodec/Config.in rename to
package/freescale-imx/imx-codec/Config.in
index d2007fe..e161699 100644
package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in +2 -2
@@ -1,5 +1,5 @@
-config BR2_PACKAGE_LIBFSLCODEC
-       bool "libfslcodec"
+config BR2_PACKAGE_IMX_CODEC
+       bool "imx-codec"
        depends on BR2_arm # Only relevant for i.MX
        depends on BR2_TOOLCHAIN_USES_GLIBC # prebuilt binaries
        help
package/freescale-imx/imx-codec/Config.in config BR2_PACKAGE_LIBFSLCODEC · line 9
Quoted 2 lines
package/freescale-imx/imx-codec/Config.in renamed from package/libfslcodec/Config.in +0 -0
@@ -8,6 +8,6 @@ config BR2_PACKAGE_LIBFSLCODEC
          This library is provided by Freescale as-is and doesn't have
          an upstream.
Although it isn't done in the Yocto patch I guess this renaming is a good time to remove the Freescale reference. This library is now provided by NXP as-is, targeting NXP i.MX SoCs.
Regards,
Gary

Andrew Webster

author

2016-06-03 13:12 UTC #20 lore raw

PATCH 5/10 imx-parser: rename from libfslparser and bump version to 4.0.9

Changelog:
--Add aac latm parser in parser_libs, support aac latm in mpeg2 parser

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/e2c6f10

---
 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-parser}/Config.in                          |  4 ++--
 package/freescale-imx/imx-parser/imx-parser.hash   |  2 ++
 package/freescale-imx/imx-parser/imx-parser.mk     | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  2 +-
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/libfslparser/libfslparser.hash             |  2 --
 package/libfslparser/libfslparser.mk               | 23 ----------------------
 10 files changed, 37 insertions(+), 30 deletions(-)
 rename package/{libfslparser => freescale-imx/imx-parser}/Config.in (78%)
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.hash
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.mk
 delete mode 100644 package/libfslparser/libfslparser.hash
 delete mode 100644 package/libfslparser/libfslparser.mk

Config.in.legacy +7 -0

index 54d43f1..dab8505 100644
@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLPARSER
+	bool "libfslparser has been renamed to imx-parser"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_PARSER
+	help
+	  The libfslparser has been renamed to match the renamed package.
+
 config BR2_PACKAGE_LIBFSLCODEC
 	bool "libfslcodec has been renamed to imx-codec"
 	select BR2_LEGACY

package/Config.in +0 -1

index c41e4f6..8c3c6b8 100644
@@ -1089,7 +1089,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslparser/Config.in"
 	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"
 	source "package/libimxvpuapi/Config.in"

package/freescale-imx/Config.in +1 -0

index 052de5a..868e51c 100644
@@ -52,6 +52,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
+source "package/freescale-imx/imx-parser/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
 source "package/freescale-imx/imx-vpu/Config.in"
 source "package/freescale-imx/firmware-imx/Config.in"

package/freescale-imx/imx-parser/Config.in renamed from package/libfslparser/Config.in +2 -2

similarity index 78%
rename from package/libfslparser/Config.in
rename to package/freescale-imx/imx-parser/Config.in
index e3ab980..d7048a7 100644
@@ -1,5 +1,5 @@
-config BR2_PACKAGE_LIBFSLPARSER
-	bool "libfslparser"
+config BR2_PACKAGE_IMX_PARSER
+	bool "imx-parser"
 	depends on BR2_arm # Only relevant for i.MX
 	help
 	  Binary parser libraries for the Freescale i.MX SoCs.

package/freescale-imx/imx-parser/imx-parser.hash new file +2 -0

new file mode 100644
index 0000000..2193c66
@@ -0,0 +1,2 @@
+# locally computed
+sha256	1bec0101843db2e95d61e3dbe9e35172d874e54be19cc5d8bf72a735d5e3236f  imx-parser-4.0.9.bin

package/freescale-imx/imx-parser/imx-parser.mk new file +23 -0

new file mode 100644
index 0000000..a207d6e
@@ -0,0 +1,23 @@
+################################################################################
+#
+# imx-parser
+#
+################################################################################
+
+IMX_PARSER_VERSION = 4.0.9
+IMX_PARSER_SITE = $(FREESCALE_IMX_SITE)
+IMX_PARSER_SOURCE = imx-parser-$(IMX_PARSER_VERSION).bin
+IMX_PARSER_INSTALL_STAGING = YES
+
+IMX_PARSER_LICENSE = NXP Semiconductor Software License Agreement
+IMX_PARSER_LICENSE_FILES = EULA COPYING
+IMX_PARSER_REDISTRIBUTE = NO
+
+define IMX_PARSER_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_PARSER_SOURCE))
+endef
+
+# The Makefile installs several versions of the libraries, but we only
+# need one of them, depending on the platform.
+
+$(eval $(autotools-package))

package/gstreamer/gst-fsl-plugins/Config.in +1 -1

index 8c82a4a..38c8a47 100644
@@ -13,7 +13,7 @@ config BR2_PACKAGE_GST_FSL_PLUGINS
 	select BR2_PACKAGE_GST_PLUGINS_BASE
 	select BR2_PACKAGE_LIBFSLVPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 	select BR2_PACKAGE_IMX_LIB
-	select BR2_PACKAGE_LIBFSLPARSER
+	select BR2_PACKAGE_IMX_PARSER
 	select BR2_PACKAGE_IMX_CODEC
 	select BR2_PACKAGE_FREESCALE_IMX
 	help

package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk +1 -1

index 50732bf..8db8d1b 100644
@@ -16,7 +16,7 @@ GST_FSL_PLUGINS_INSTALL_STAGING = YES
 GST_FSL_PLUGINS_AUTORECONF = YES
 
 GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \
-	imx-lib libfslparser imx-codec
+	imx-lib imx-parser imx-codec
 
 ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y)
 GST_FSL_PLUGINS_DEPENDENCIES += libfslvpuwrap

package/libfslparser/libfslparser.hash deleted +0 -2

deleted file mode 100644
index c49345f..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	a04621783c84a9776216caff9563c7f840fddd584b0b2d27738d6ca6d2c77f32  libfslparser-4.0.8.bin

package/libfslparser/libfslparser.mk deleted +0 ~~-23~~

deleted file mode 100644
index 44e6039..0000000
@@ -1,23 +0,0 @@
-################################################################################
-#
-# libfslparser
-#
-################################################################################
-
-LIBFSLPARSER_VERSION = 4.0.8
-LIBFSLPARSER_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLPARSER_SOURCE = libfslparser-$(LIBFSLPARSER_VERSION).bin
-LIBFSLPARSER_INSTALL_STAGING = YES
-
-LIBFSLPARSER_LICENSE = Freescale Semiconductor Software License Agreement
-LIBFSLPARSER_LICENSE_FILES = EULA COPYING
-LIBFSLPARSER_REDISTRIBUTE = NO
-
-define LIBFSLPARSER_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLPARSER_SOURCE))
-endef
-
-# The Makefile installs several versions of the libraries, but we only
-# need one of them, depending on the platform.
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:33 UTC #21 lore raw

PATCH 5/10 imx-parser: rename from libfslparser and bump version to 4.0.9

Andrew, All,

Quoted 30 lines · Andrew Webster

Changelog:
--Add aac latm parser in parser_libs, support aac latm in mpeg2 parser

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/e2c6f10

---
 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-parser}/Config.in                          |  4 ++--
 package/freescale-imx/imx-parser/imx-parser.hash   |  2 ++
 package/freescale-imx/imx-parser/imx-parser.mk     | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  2 +-
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/libfslparser/libfslparser.hash             |  2 --
 package/libfslparser/libfslparser.mk               | 23 ----------------------
 10 files changed, 37 insertions(+), 30 deletions(-)
 rename package/{libfslparser => freescale-imx/imx-parser}/Config.in (78%)
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.hash
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.mk
 delete mode 100644 package/libfslparser/libfslparser.hash
 delete mode 100644 package/libfslparser/libfslparser.mk

Config.in.legacy +0 -0

index 54d43f1..dab8505 100644

Config.in.legacy comment "build, or run, in unpredictable ways. " · line 151

Quoted 10 lines

Config.in.legacy +7 -0

@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLPARSER
+       bool "libfslparser has been renamed to imx-parser"
+       select BR2_LEGACY
+       select BR2_PACKAGE_IMX_PARSER
+       help
+         The libfslparser has been renamed to match the renamed package.
+

Same as imx-codec, I believe this should be under the 2016.08 section.

Quoted 35 lines

Config.in.legacy +0 -0

 config BR2_PACKAGE_LIBFSLCODEC
        bool "libfslcodec has been renamed to imx-codec"
        select BR2_LEGACY

package/Config.in +0 -1

index c41e4f6..8c3c6b8 100644
@@ -1089,7 +1089,6 @@ menu "Multimedia"
        source "package/libdvdnav/Config.in"
        source "package/libdvdread/Config.in"
        source "package/libebml/Config.in"
-       source "package/libfslparser/Config.in"
        source "package/libfslvpuwrap/Config.in"
        source "package/libhdhomerun/Config.in"
        source "package/libimxvpuapi/Config.in"

package/freescale-imx/Config.in +1 -0

index 052de5a..868e51c 100644
@@ -52,6 +52,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
+source "package/freescale-imx/imx-parser/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
 source "package/freescale-imx/imx-vpu/Config.in"
 source "package/freescale-imx/firmware-imx/Config.in"

package/freescale-imx/imx-parser/Config.in renamed from package/libfslparser/Config.in +0 -0

similarity index 78%
rename from package/libfslparser/Config.in
rename to package/freescale-imx/imx-parser/Config.in
index e3ab980..d7048a7 100644

package/freescale-imx/imx-parser/Config.in line 5

Quoted 7 lines

package/freescale-imx/imx-parser/Config.in renamed from package/libfslparser/Config.in +2 -2

@@ -1,5 +1,5 @@
-config BR2_PACKAGE_LIBFSLPARSER
-       bool "libfslparser"
+config BR2_PACKAGE_IMX_PARSER
+       bool "imx-parser"
        depends on BR2_arm # Only relevant for i.MX
        help
          Binary parser libraries for the Freescale i.MX SoCs.

Same as imx-codec, I think we should get rid of those Freescale references right now for the packages that get renamed.

Other than that it looks good.

Regards,
Gary

Andrew Webster

author

2016-06-03 13:12 UTC #22 lore raw

PATCH 6/10 imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65

Changelog:
--Update 'nLastFrameEndPosPhy' in seqinit stage to avoid feed
the same SPS to vpu repeatedly.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/d6b9490

---
 Config.in.legacy                                    |  7 +++++++
 package/Config.in                                   |  1 -
 package/freescale-imx/Config.in                     |  1 +
 .../imx-vpuwrap}/Config.in                          |  8 ++++----
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash  |  2 ++
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk    | 21 +++++++++++++++++++++
 package/libfslvpuwrap/libfslvpuwrap.hash            |  2 --
 package/libfslvpuwrap/libfslvpuwrap.mk              | 21 ---------------------
 8 files changed, 35 insertions(+), 28 deletions(-)
 rename package/{libfslvpuwrap => freescale-imx/imx-vpuwrap}/Config.in (69%)
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.hash
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.mk

Config.in.legacy +7 -0

index dab8505..0b0e9e7 100644
@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLVPUWRAP
+	bool "libfslvpuwrap has been renamed to imx-vpuwrap"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_VPUWRAP
+	help
+	  The libfslvpuwrap has been renamed to match the renamed package.
+
 config BR2_PACKAGE_LIBFSLPARSER
 	bool "libfslparser has been renamed to imx-parser"
 	select BR2_LEGACY

package/Config.in +0 -1

index 8c3c6b8..696f30d 100644
@@ -1089,7 +1089,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"
 	source "package/libimxvpuapi/Config.in"
 	source "package/libmatroska/Config.in"

package/freescale-imx/Config.in +1 -0

index 868e51c..056728f 100644
@@ -55,6 +55,7 @@ source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-parser/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
 source "package/freescale-imx/imx-vpu/Config.in"
+source "package/freescale-imx/imx-vpuwrap/Config.in"
 source "package/freescale-imx/firmware-imx/Config.in"
 if (BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX51 || BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53)
 source "package/freescale-imx/gpu-amd-bin-mx51/Config.in"

package/freescale-imx/imx-vpuwrap/Config.in renamed from package/libfslvpuwrap/Config.in +4 -4

similarity index 69%
rename from package/libfslvpuwrap/Config.in
rename to package/freescale-imx/imx-vpuwrap/Config.in
index 72f7126..40ebabf 100644
@@ -1,13 +1,13 @@
-comment "libfslvpuwrap needs an imx-specific Linux kernel to be built"
+comment "imx-vpuwrap needs an imx-specific Linux kernel to be built"
 	depends on BR2_arm
 	depends on !BR2_LINUX_KERNEL
 
-comment "libfslvpuwrap needs an i.MX platform with VPU support"
+comment "imx-vpuwrap needs an i.MX platform with VPU support"
 	depends on BR2_arm
 	depends on BR2_LINUX_KERNEL && !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 
-config BR2_PACKAGE_LIBFSLVPUWRAP
-	bool "libfslvpuwrap"
+config BR2_PACKAGE_IMX_VPUWRAP
+	bool "imx-vpuwrap"
 	depends on BR2_LINUX_KERNEL
 	depends on BR2_arm # Only relevant for i.MX
 	depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU

package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash new file +2 -0

new file mode 100644
index 0000000..42b97a1
@@ -0,0 +1,2 @@
+# locally computed
+sha256	ddeacb7d9cb26c8864374df6d3f090aad47cd2ae409c2b9c817b773447252d61  imx-vpuwrap-1.0.65.bin

package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk new file +21 -0

new file mode 100644
index 0000000..da0e32c
@@ -0,0 +1,21 @@
+################################################################################
+#
+# imx-vpuwrap
+#
+################################################################################
+
+IMX_VPUWRAP_VERSION = 1.0.65
+IMX_VPUWRAP_SITE = $(FREESCALE_IMX_SITE)
+IMX_VPUWRAP_SOURCE = imx-vpuwrap-$(IMX_VPUWRAP_VERSION).bin
+IMX_VPUWRAP_DEPENDENCIES = imx-vpu
+IMX_VPUWRAP_INSTALL_STAGING = YES
+
+IMX_VPUWRAP_LICENSE = NXP Semiconductor Software License Agreement
+IMX_VPUWRAP_LICENSE_FILES = EULA COPYING
+IMX_VPUWRAP_REDISTRIBUTE = NO
+
+define IMX_VPUWRAP_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_VPUWRAP_SOURCE))
+endef
+
+$(eval $(autotools-package))

package/libfslvpuwrap/libfslvpuwrap.hash deleted +0 -2

deleted file mode 100644
index e987077..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	0717faccb5413dc95ce9ad919400095ecf722320478c924366793a6cc450ae16  libfslvpuwrap-1.0.62.bin

package/libfslvpuwrap/libfslvpuwrap.mk deleted +0 ~~-21~~

deleted file mode 100644
index 12a4e86..0000000
@@ -1,21 +0,0 @@
-################################################################################
-#
-# libfslvpuwrap
-#
-################################################################################
-
-LIBFSLVPUWRAP_VERSION = 1.0.62
-LIBFSLVPUWRAP_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLVPUWRAP_SOURCE = libfslvpuwrap-$(LIBFSLVPUWRAP_VERSION).bin
-LIBFSLVPUWRAP_DEPENDENCIES = imx-vpu
-LIBFSLVPUWRAP_INSTALL_STAGING = YES
-
-LIBFSLVPUWRAP_LICENSE = Freescale Semiconductor Software License Agreement
-LIBFSLVPUWRAP_LICENSE_FILES = EULA COPYING
-LIBFSLVPUWRAP_REDISTRIBUTE = NO
-
-define LIBFSLVPUWRAP_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLVPUWRAP_SOURCE))
-endef
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:34 UTC #23 lore raw

PATCH 6/10 imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65

Andrew, All,

Quoted 29 lines · Andrew Webster

Changelog:
--Update 'nLastFrameEndPosPhy' in seqinit stage to avoid feed
the same SPS to vpu repeatedly.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/d6b9490

---
 Config.in.legacy                                    |  7 +++++++
 package/Config.in                                   |  1 -
 package/freescale-imx/Config.in                     |  1 +
 .../imx-vpuwrap}/Config.in                          |  8 ++++----
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash  |  2 ++
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk    | 21 +++++++++++++++++++++
 package/libfslvpuwrap/libfslvpuwrap.hash            |  2 --
 package/libfslvpuwrap/libfslvpuwrap.mk              | 21 ---------------------
 8 files changed, 35 insertions(+), 28 deletions(-)
 rename package/{libfslvpuwrap => freescale-imx/imx-vpuwrap}/Config.in (69%)
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.hash
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.mk

Config.in.legacy +0 -0

index dab8505..0b0e9e7 100644

Config.in.legacy comment "build, or run, in unpredictable ways. " · line 151

Quoted 10 lines

Config.in.legacy +7 -0

@@ -142,6 +142,13 @@ comment "build, or run, in unpredictable ways.               "
 comment "----------------------------------------------------"
 endif
 
+config BR2_PACKAGE_LIBFSLVPUWRAP
+       bool "libfslvpuwrap has been renamed to imx-vpuwrap"
+       select BR2_LEGACY
+       select BR2_PACKAGE_IMX_VPUWRAP
+       help
+         The libfslvpuwrap has been renamed to match the renamed package.
+

Same as imx-codec and imx-parser, I think this section is misplaced.

Also, it seems that packages depending on this package haven't been updated (gst-fsl-plugins and kodi).

Regards,
Gary

Andrew Webster

author

2016-06-03 13:12 UTC #24 lore raw

PATCH 7/10 imx-gpu-viv: bump version to 5.0.11-p8.3

https://www.nxp.com/webapp/Download?colCode=L4.1.15_1.0.0_LINUX_DOCS&location=null&fpsp=1&WT_TYPE=Supporting%20Information&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=gz&WT_ASSET=Documentation&fileExt=.gz&Parent_nodeId=1337637154535695831062&Parent_pageType=product

Changelog:
Software floating point is no longer supported.

For a detailed list of changes in this update, see the Graphics section of the i.MX Linux Release Notes contained in the i.MX Linux BSP documentation here:

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/b667ed4

---
 package/freescale-imx/imx-gpu-viv/Config.in        |  6 +++++-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash |  4 +---
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk   | 10 +++-------
 3 files changed, 9 insertions(+), 11 deletions(-)

package/freescale-imx/imx-gpu-viv/Config.in +5 -1

index 20759b1..9a7cf56 100644
@@ -2,12 +2,16 @@ comment "imx-gpu-viv needs an (e)glibc toolchain"
 	depends on BR2_arm
 	depends on !BR2_TOOLCHAIN_USES_GLIBC
 
+comment "imx-gpu-viv needs EABIhf"
+	depends on BR2_arm
+	depends on !BR2_ARM_EABIHF
+
 config BR2_PACKAGE_IMX_GPU_VIV
 	bool "imx-gpu-viv"
 	select BR2_PACKAGE_HAS_LIBEGL
 	select BR2_PACKAGE_HAS_LIBGLES
 	select BR2_PACKAGE_HAS_LIBOPENVG
-	depends on BR2_arm # Only relevant for i.MX6
+	depends on BR2_ARM_EABIHF # Only relevant for EABIhf i.MX6
 	# Library binaries are linked against libc.so.6
 	depends on BR2_TOOLCHAIN_USES_GLIBC
 	help

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash +1 -3

index 4119388..4fce3f8 100644
@@ -1,3 +1 @@
-# locally computed
-sha256	3ea201f91ce9b56455e4983ab1055e8175415fbbf2800dc097d47e78ceefba34  imx-gpu-viv-5.0.11.p7.4-sfp.bin
-sha256	b80838104f2f2e3b0baf2f1aacdca263c1c92589ce919b3be40aaad23f686b7f  imx-gpu-viv-5.0.11.p7.4-hfp.bin
+sha256	0400f19dce7223e046ca693cae8578201904c8e7bdcf0c444a6114afa14f09fe  imx-gpu-viv-5.0.11.p8.3-hfp.bin

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +3 -7

index b51a7af..513dcf0 100644
@@ -4,19 +4,15 @@
 #
 ################################################################################
 
-IMX_GPU_VIV_BASE_VERSION = 5.0.11.p7.4
-ifeq ($(BR2_ARM_EABIHF),y)
+IMX_GPU_VIV_BASE_VERSION = 5.0.11.p8.3
 IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-hfp
-else
-IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-sfp
-endif
 IMX_GPU_VIV_SITE = $(FREESCALE_IMX_SITE)
 IMX_GPU_VIV_SOURCE = imx-gpu-viv-$(IMX_GPU_VIV_VERSION).bin
 
 IMX_GPU_VIV_INSTALL_STAGING = YES
 
-IMX_GPU_VIV_LICENSE = Freescale Semiconductor Software License Agreement
-IMX_GPU_VIV_LICENSE_FILES = EULA
+IMX_GPU_VIV_LICENSE = NXP Semiconductor Software License Agreement
+IMX_GPU_VIV_LICENSE_FILES = EULA COPYING
 IMX_GPU_VIV_REDISTRIBUTE = NO
 
 IMX_GPU_VIV_PROVIDES = libegl libgles libopenvg

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:34 UTC #25 lore raw

PATCH 7/10 imx-gpu-viv: bump version to 5.0.11-p8.3

Andrew, All,

Quoted 11 lines · Andrew Webster

Changelog:
Software floating point is no longer supported.
For a detailed list of changes in this update, see the Graphics section of the i.MX Linux Release Notes contained in the i.MX Linux BSP documentation here:
https://www.nxp.com/webapp/Download?colCode=L4.1.15_1.0.0_LINUX_DOCS&location=null&fpsp=1&WT_TYPE=Supporting%20Information&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=gz&WT_ASSET=Documentation&fileExt=.gz&Parent_nodeId=1337637154535695831062&Parent_pageType=product
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/b667ed4

I actually recommend going straight to version p8.4 since it brings its fair share of bug fixes:
http://git.freescale.com/git/cgit.cgi/imx/fsl-arm-yocto-bsp.git/tree/GraphicsChangeLog?h=imx-3.14.52-1.1.0_ga
https://github.com/Freescale/meta-fsl-arm/commit/af5b3a6c

Also, as said in the cover letter, this patch should be followed by the kernel module update.
https://github.com/Freescale/meta-fsl-arm/commit/c9c30c56

Regards,
Gary

Andrew Webster

author

2016-06-03 13:12 UTC #26 lore raw

PATCH 8/10 libimxvpuapi: bump version to 0.10.2

Changelog:
- fix AUD NAL positioning in h.264 encoder output
(SPS/PPS/AUD were ordered incorrectly - AUD has to come first, not last)
- fix build error with examples when --enable-static is used
- pass quality factor in simplified JPEG encoder interface correctly
- add functions for querying and setting header data
useful for modifying headers, like VUI data in the SPS RBSP
- documentation updates

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c999668

---
 package/libimxvpuapi/libimxvpuapi.hash | 2 +-
 package/libimxvpuapi/libimxvpuapi.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/libimxvpuapi/libimxvpuapi.hash +1 -1

index fd90a20..c8cabcf 100644
@@ -1,2 +1,2 @@
 # locally computed hash
-sha256 8d7aeed88c06fda44318cef9565ae47b86461bd309908b1103a7974ed0822a8d  libimxvpuapi-0.10.1.tar.gz
+sha256 8e6dc4105db6164c0de72fd60eb73f4ba1632f7ba5052b1abc834b462b566da2  libimxvpuapi-0.10.2.tar.gz

package/libimxvpuapi/libimxvpuapi.mk +1 -1

index e0d8cc9..df84806 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-LIBIMXVPUAPI_VERSION = 0.10.1
+LIBIMXVPUAPI_VERSION = 0.10.2
 LIBIMXVPUAPI_SITE = $(call github,Freescale,libimxvpuapi,$(LIBIMXVPUAPI_VERSION))
 LIBIMXVPUAPI_LICENSE = LGPLv2.1+
 LIBIMXVPUAPI_LICENSE_FILES = LICENSE

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:34 UTC #27 lore raw

PATCH 8/10 libimxvpuapi: bump version to 0.10.2

Andrew, All,

Quoted 9 lines · Andrew Webster

Changelog:
- fix AUD NAL positioning in h.264 encoder output
(SPS/PPS/AUD were ordered incorrectly - AUD has to come first, not last)
- fix build error with examples when --enable-static is used
- pass quality factor in simplified JPEG encoder interface correctly
- add functions for querying and setting header data
useful for modifying headers, like VUI data in the SPS RBSP
- documentation updates

Quoted 4 lines

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c999668
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-03 13:12 UTC #28 lore raw

PATCH 9/10 imx-lib: bump version to 5.4

Changelog:
- Remove platform check in hdmi-cec and imx-lib - simplify to support builds on multiple machines
- For ipu use the dummy defaul object to fix build issues on non-ipu hardware

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/7946533

---
 package/freescale-imx/imx-lib/imx-lib.hash | 2 +-
 package/freescale-imx/imx-lib/imx-lib.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/freescale-imx/imx-lib/imx-lib.hash +1 -1

index 8e4bd73..76dcb23 100644
@@ -1,2 +1,2 @@
 # locally computed
-sha256 2b4547ffb1de328cf8997a09a423c6f015d808493efd11a98a5d42c6497e3445  imx-lib-5.1.tar.gz
+sha256 8d12c4a0a44d29569f1ead3a36adc28402b9c4036b53e11dd3ebf2295e3ce1d0  imx-lib-5.4.tar.gz

package/freescale-imx/imx-lib/imx-lib.mk +1 -1

index b49357b..d3603ca 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-IMX_LIB_VERSION = 5.1
+IMX_LIB_VERSION = 5.4
 IMX_LIB_SITE = $(FREESCALE_IMX_SITE)
 IMX_LIB_LICENSE = LGPLv2.1+
 IMX_LIB_LICENSE_FILES = COPYING-LGPL-2.1

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:34 UTC #29 lore raw

PATCH 9/10 imx-lib: bump version to 5.4

Andrew, All,

Quoted 9 lines · Andrew Webster

Changelog:
- Remove platform check in hdmi-cec and imx-lib - simplify to support builds on multiple machines
- For ipu use the dummy defaul object to fix build issues on non-ipu hardware
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/7946533
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-03 13:13 UTC #30 lore raw

PATCH 10/10 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.3

Remove upstreamed patch.
Add patch from Yocto to not use git for versioning.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/06a0574

---
 .../0001-Remove-dix-internal-header-usage.patch    | 31 --------
 ...001-Stop-using-Git-to-write-local-version.patch | 87 ++++++++++++++++++++++
 .../xdriver_xf86-video-imx-viv.hash                |  3 +-
 .../xdriver_xf86-video-imx-viv.mk                  |  2 +-
 4 files changed, 89 insertions(+), 34 deletions(-)
 delete mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch
 create mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 ~~-31~~

deleted file mode 100644
index f146ba3..0000000
@@ -1,31 +0,0 @@
-From 80c419dddf6483873503d90694c4b4ae34ffa5e4 Mon Sep 17 00:00:00 2001
-From: Otavio Salvador <[email protected]>
-Date: Wed, 2 Dec 2015 13:50:57 +0000
-Subject: [PATCH] Remove 'dix' internal header usage
-Organization: O.S. Systems Software LTDA.
-
-The 'swaprep.h' is an internal Xorg header and is not being installed
-anymore. The swap features are provided by the 'misc.h' header now.
-
-Upstream-Status: Pending
-
-Signed-off-by: Otavio Salvador <[email protected]>
----
- EXA/src/vivante_extension/vivante_ext.c | 1 -
- 1 file changed, 1 deletion(-)
-
-diff --git a/EXA/src/vivante_extension/vivante_ext.c b/EXA/src/vivante_extension/vivante_ext.c
-index 0d1085b..1ffbb1b 100644
---- a/EXA/src/vivante_extension/vivante_ext.c
-+++ b/EXA/src/vivante_extension/vivante_ext.c
-@@ -40,7 +40,6 @@
- #include "cursorstr.h"
- #include "scrnintstr.h"
- #include "servermd.h"
--#include "swaprep.h"
- #include "drm.h"
- #include "xf86Module.h"
- #include "globals.h"
--- 
-2.1.4
-

package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch new file +87 -0

87 lines not on this page · raw message

package/x11r7/xdriver_xf86-video-imx-viv/xdriver_xf86-video-imx-viv.hash +1 -2

index 9a16569..9b85e0c 100644
@@ -1,2 +1 @@
-# locally computed
-sha256	3effaf42bde64da8fad2dfe0dcef12d1cfbbbe189511dbd4612442129fd832cc  xserver-xorg-video-imx-viv-5.0.11.p7.4.tar.gz
+sha256	00b4925819601461c0aa293a25024709410ab8d669accfa71ffd3177ff3cdeb2  xserver-xorg-video-imx-viv-5.0.11.p8.3.tar.gz

package/x11r7/xdriver_xf86-video-imx-viv/xdriver_xf86-video-imx-viv.mk +1 -1

index a1b32b1..1f1bc0c 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-XDRIVER_XF86_VIDEO_IMX_VIV_VERSION = 5.0.11.p7.4
+XDRIVER_XF86_VIDEO_IMX_VIV_VERSION = 5.0.11.p8.3
 XDRIVER_XF86_VIDEO_IMX_VIV_SITE = $(FREESCALE_IMX_SITE)
 XDRIVER_XF86_VIDEO_IMX_VIV_SOURCE = xserver-xorg-video-imx-viv-$(XDRIVER_XF86_VIDEO_IMX_VIV_VERSION).tar.gz
 XDRIVER_XF86_VIDEO_IMX_VIV_DEPENDENCIES = imx-gpu-viv xserver_xorg-server \

Signature · 1 line

1.9.1

Gary Bisson

2016-06-06 14:36 UTC #31 lore raw

PATCH 10/10 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.3

Andrew, All,

Quoted 5 lines · Andrew Webster

Remove upstreamed patch.
Add patch from Yocto to not use git for versioning.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/06a0574

Same as imx-gpu-viv package, I think we should go straight to version p8.4. Also, can you re-order your patches to have this one after imx-gpu-viv, just trying to have multimedia updates on one side and then graphics update. Especially now that graphics can be updated separately.

Finally, I forgot to mention it in my cover letter answer, please cc me to the v2 series.

Regards,
Gary

Gary Bisson

2016-06-06 14:32 UTC #32 lore raw

cover Bump NXP package versions

Andrew, All,

First of all, thank you for taking care of this update. It's good to have someone else looking at those ;-)

Quoted 8 lines · Andrew Webster

The following packages were renamed:
libfslcodec -> imx-codec
libfslparser -> imx-parser
libfslvpuwrap -> imx-vpuwrap
In addition to renaming, I also moved them into the freescale-imx directory which seems consistent with other similarily named packages (although that directory may be due for a rename now too). Not sure if this was appropriate or not.

I'm ok moving things to freescale-imx, I think it makes it easier for people looking at i.MX-specific packages instead of browsing through both hardware handling packages and multimedia libraries. However I guess Thomas and/or Peter will have ack this change.

As for changing the 'freescale-imx' name, I agree this will need to happen but there's no rush in my opinion. Especially since the Yocto layer is still called meta-*fsl*-arm.

Quoted 11 lines

Andrew Webster (10):
firmware-imx: bump to version 5.4
gst1-imx: bump to version 0.12.1
imx-vpu: bump to version 5.4.33
imx-codec: rename from libfslcodec and bump version to 4.0.9
imx-parser: rename from libfslparser and bump version to 4.0.9
imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65
imx-gpu-viv: bump version to 5.0.11-p8.3
libimxvpuapi: bump version to 0.10.2
imx-lib: bump version to 5.4 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.3

I will comment on each individual patch above but I have a couple of questions:
- Why isn't there the kernel-module-imx-gpu-viv update? It is very important to keep it in line with imx-gpu-viv.
- What about imx-kobs?
- Can you share your testing setup (platform name + defconfig)?

Regards,
Gary

Andrew Webster

author

2016-06-06 15:57 UTC #33 lore raw

cover Bump NXP package versions

Quoted 11 lines

-----Original Message-----
From: Gary Bisson [mailto:gary.bisson at boundarydevices.com]
Sent: Monday, June 06, 2016 10:32 AM
To: Andrew Webster
Cc: buildroot at buildroot.org
Subject: Re: [Buildroot] [PATCH 00/10] Bump NXP package versions
Andrew, All,
First of all, thank you for taking care of this update. It's good to have someone else looking at those ;-)

Glad to help. Thanks for all your work on these packages.

Quoted 33 lines

Quoted 8 lines · Andrew Webster
The following packages were renamed:
libfslcodec -> imx-codec
libfslparser -> imx-parser
libfslvpuwrap -> imx-vpuwrap
In addition to renaming, I also moved them into the freescale-imx directory which seems consistent with other similarily named packages (although that directory may be due for a rename now too). Not sure if this was appropriate or not.
I'm ok moving things to freescale-imx, I think it makes it easier for people looking at i.MX-specific packages instead of browsing through both hardware handling packages and multimedia libraries.
However I guess Thomas and/or Peter will have ack this change.
As for changing the 'freescale-imx' name, I agree this will need to happen but there's no rush in my opinion. Especially since the Yocto layer is still called meta-*fsl*-arm.
Quoted 11 lines
Andrew Webster (10):
firmware-imx: bump to version 5.4
gst1-imx: bump to version 0.12.1
imx-vpu: bump to version 5.4.33
imx-codec: rename from libfslcodec and bump version to 4.0.9
imx-parser: rename from libfslparser and bump version to 4.0.9
imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65
imx-gpu-viv: bump version to 5.0.11-p8.3
libimxvpuapi: bump version to 0.10.2
imx-lib: bump version to 5.4 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.3

Quoted 4 lines

I will comment on each individual patch above but I have a couple of questions:
- Why isn't there the kernel-module-imx-gpu-viv update? It is very important to keep it in line with imx-gpu-viv.

We don't use this (yet), so I didn't want to update it without testing it.

Quoted 1 line

- What about imx-kobs?

We also don't use this (no NAND on the board). As above, I didn't want to update without testing.

I can certainly update these versions and confirm that they build, if someone else can test.

Quoted 1 line

- Can you share your testing setup (platform name + defconfig)?

Tested on a custom i.MX6Q board. I used the freescale_imx6qsabresd defconfig, but with a different kernel and toolchain.

We are in the process of updating to the latest 4.1.15 kernel release. I will update the gpu packages to p8.4 after that and resubmit.

Quoted 3 lines

Regards,
Gary

Andrew

Gary Bisson

2016-06-07 07:45 UTC #34 lore raw

cover Bump NXP package versions

Andrew, All,

Quoted 7 lines · Andrew Webster

Quoted 5 lines
<snip>
I will comment on each individual patch above but I have a couple of questions:
- Why isn't there the kernel-module-imx-gpu-viv update? It is very important to keep it in line with imx-gpu-viv.
We don't use this (yet), so I didn't want to update it without testing it.

Ok, but you can test it on any kernel though, it doesn't require any modification. If you enable the option, it will remove the built-in driver dynamically from your kernel configuration and use the module instead.

Quoted 5 lines

Quoted 1 line
- What about imx-kobs?
We also don't use this (no NAND on the board). As above, I didn't want to update without testing.
I can certainly update these versions and confirm that they build, if someone else can test.

I never tested it either since I don't have any NAND on board, making sure it builds fine is sufficient. Plus I've heard people using it just fine so I'm not worried and it's still nice to have.

As a FYI, I think SabreAuto is the only one that offers NAND.

Quoted 5 lines

Quoted 1 line
- Can you share your testing setup (platform name + defconfig)?
Tested on a custom i.MX6Q board. I used the freescale_imx6qsabresd defconfig, but with a different kernel and toolchain.
We are in the process of updating to the latest 4.1.15 kernel release. I will update the gpu packages to p8.4 after that and resubmit.

Ok good, thanks for the details.

Thanks,
Gary

Thomas Petazzoni

2016-06-06 20:33 UTC #35 lore raw

cover Bump NXP package versions

Hello,

Quoted 2 lines · Gary Bisson

First of all, thank you for taking care of this update. It's good to have someone else looking at those ;-)

And thanks to you for reviewing!

Quoted 4 lines

I'm ok moving things to freescale-imx, I think it makes it easier for people looking at i.MX-specific packages instead of browsing through both hardware handling packages and multimedia libraries. However I guess Thomas and/or Peter will have ack this change.

We normally want all packages to be directly under package/, precisely because it is easier to find them this way.

The only case where we accept to have sub-directories for groups of packages is when several packages need to share some variables, like package/qt5/qt5.mk.

For freescale-imx stuff, it seems like the only variable left shared by various packages in FREESCALE_IMX_SITE, and in fact it's already used outside of package/freescale-imx/.

So I would almost suggest to get rid of the sub-directory, and move this variable, as well as the other macro defined in freescale-imx.mk to another file, though I'm not sure which one.

Peter?

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Thomas Petazzoni

2016-06-06 21:23 UTC #36 lore raw

cover Bump NXP package versions

Hello,

Quoted 11 lines · Andrew Webster

Andrew Webster (10):
firmware-imx: bump to version 5.4
gst1-imx: bump to version 0.12.1
imx-vpu: bump to version 5.4.33
imx-codec: rename from libfslcodec and bump version to 4.0.9
imx-parser: rename from libfslparser and bump version to 4.0.9
imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65
imx-gpu-viv: bump version to 5.0.11-p8.3
libimxvpuapi: bump version to 0.10.2
imx-lib: bump version to 5.4 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.3

Thanks a lot for this work! Since you've received several comments from Gary, I've marked your patch series as "Changes Requested" in our patch tracking system (which helps us to not lose and forget patches).

Could you send an updated version of your patch series after reworking it to take into account Gary's comments?

Thanks a lot!

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Andrew Webster

author

2016-06-10 18:09 UTC #37 lore raw

v2 cover Bump NXP package versions

The two patches for nxp defconfigs and the patch for imx-kobs have been build tested but not run-time tested.

Changes v1 -> v2:
- Reordered patches to keep gpu related patches together and multimedia patches together.
- Added patches for bumping the kernel/u-boot versions of NXP defconfigs
- Added patch to bump version of imx-kobs
- Added patch to bump version of kernel-module-imx-gpu-viv

Andrew Webster (14):
nxp defconfigs: bump u-boot version to 4.1.15_1.1.0_ga nxp defconfigs: bump kernel version to 4.1.15
kernel-module-imx-gpu-viv: bump version to that of the 5.0.11.p8.4
release
imx-gpu-viv: bump version to 5.0.11-p8.4 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.4
firmware-imx: bump to version 5.4
gst1-imx: bump to version 0.12.1
imx-vpu: bump to version 5.4.33
imx-codec: rename from libfslcodec and bump version to 4.0.9
imx-parser: rename from libfslparser and bump version to 4.0.9
imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65
libimxvpuapi: bump version to 0.10.2
imx-lib: bump version to 5.4
imx-kobs: bump version to 5.4

 Config.in.legacy                                   | 21 ++++++
 configs/freescale_imx6dlsabreauto_defconfig        |  8 +-
 configs/freescale_imx6dlsabresd_defconfig          |  8 +-
 configs/freescale_imx6qsabreauto_defconfig         |  8 +-
 configs/freescale_imx6qsabresd_defconfig           |  8 +-
 configs/freescale_imx6sololiteevk_defconfig        |  8 +-
 configs/freescale_imx6sxsabresd_defconfig          |  8 +-
 configs/freescale_imx6ulevk_defconfig              |  8 +-
 package/Config.in                                  |  3 -
 package/freescale-imx/Config.in                    |  3 +
 .../freescale-imx/firmware-imx/firmware-imx.hash   |  2 +-
 package/freescale-imx/firmware-imx/firmware-imx.mk |  5 +-
 .../imx-codec}/Config.in                           |  6 +-
 package/freescale-imx/imx-codec/imx-codec.hash     |  2 +
 package/freescale-imx/imx-codec/imx-codec.mk       | 23 ++++++
 package/freescale-imx/imx-gpu-viv/Config.in        |  6 +-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash |  4 +-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk   | 10 +--
 package/freescale-imx/imx-kobs/imx-kobs.hash       |  3 +-
 package/freescale-imx/imx-kobs/imx-kobs.mk         |  2 +-
 package/freescale-imx/imx-lib/imx-lib.hash         |  2 +-
 package/freescale-imx/imx-lib/imx-lib.mk           |  2 +-
 .../imx-parser}/Config.in                          |  4 +-
 package/freescale-imx/imx-parser/imx-parser.hash   |  2 +
 package/freescale-imx/imx-parser/imx-parser.mk     | 23 ++++++
 package/freescale-imx/imx-vpu/imx-vpu.hash         |  2 +-
 package/freescale-imx/imx-vpu/imx-vpu.mk           |  4 +-
 .../imx-vpuwrap}/Config.in                         |  8 +-
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash |  2 +
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk   | 21 ++++++
 .../kernel-module-imx-gpu-viv.hash                 |  2 +-
 .../kernel-module-imx-gpu-viv.mk                   |  2 +-
 package/gstreamer/gst-fsl-plugins/Config.in        |  8 +-
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  4 +-
 package/gstreamer1/gst1-imx/Config.in              |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.hash          |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk            |  6 +-
 package/kodi/kodi.mk                               |  4 +-
 package/libfslcodec/libfslcodec.hash               |  2 -
 package/libfslcodec/libfslcodec.mk                 | 23 ------
 package/libfslparser/libfslparser.hash             |  2 -
 package/libfslparser/libfslparser.mk               | 23 ------
 package/libfslvpuwrap/libfslvpuwrap.hash           |  2 -
 package/libfslvpuwrap/libfslvpuwrap.mk             | 21 ------
 package/libimxvpuapi/libimxvpuapi.hash             |  2 +-
 package/libimxvpuapi/libimxvpuapi.mk               |  2 +-
 .../0001-Remove-dix-internal-header-usage.patch    | 31 --------
 ...001-Stop-using-Git-to-write-local-version.patch | 87 ++++++++++++++++++++++
 .../xdriver_xf86-video-imx-viv.hash                |  3 +-
 .../xdriver_xf86-video-imx-viv.mk                  |  2 +-
 50 files changed, 259 insertions(+), 187 deletions(-)
 rename package/{libfslcodec => freescale-imx/imx-codec}/Config.in (75%)
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.hash
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.mk
 rename package/{libfslparser => freescale-imx/imx-parser}/Config.in (78%)
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.hash
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.mk
 rename package/{libfslvpuwrap => freescale-imx/imx-vpuwrap}/Config.in (69%)
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk
 delete mode 100644 package/libfslcodec/libfslcodec.hash
 delete mode 100644 package/libfslcodec/libfslcodec.mk
 delete mode 100644 package/libfslparser/libfslparser.hash
 delete mode 100644 package/libfslparser/libfslparser.mk
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.hash
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.mk
 delete mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch
 create mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch

Signature · 1 line

1.9.1

Andrew Webster

author

2016-06-10 18:09 UTC #38 lore raw

PATCH v2 1/14 nxp defconfigs: bump u-boot version to 4.1.15_1.1.0_ga

---
Changes v1 -> v2:
- added to patch set

 configs/freescale_imx6dlsabreauto_defconfig | 2 +-
 configs/freescale_imx6dlsabresd_defconfig   | 2 +-
 configs/freescale_imx6qsabreauto_defconfig  | 2 +-
 configs/freescale_imx6qsabresd_defconfig    | 2 +-
 configs/freescale_imx6sololiteevk_defconfig | 2 +-
 configs/freescale_imx6sxsabresd_defconfig   | 2 +-
 configs/freescale_imx6ulevk_defconfig       | 2 +-
 7 files changed, 7 insertions(+), 7 deletions(-)

configs/freescale_imx6dlsabreauto_defconfig +1 -1

index 699e684..f55d4fa 100644
@@ -31,4 +31,4 @@ BR2_TARGET_UBOOT_BOARDNAME="mx6dlsabreauto"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"

configs/freescale_imx6dlsabresd_defconfig +1 -1

index b6d31e0..bec80b5 100644
@@ -31,4 +31,4 @@ BR2_TARGET_UBOOT_BOARDNAME="mx6dlsabresd"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"

configs/freescale_imx6qsabreauto_defconfig +1 -1

index 23ca2dc..2b89541 100644
@@ -31,4 +31,4 @@ BR2_TARGET_UBOOT_BOARDNAME="mx6qsabreauto"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"

configs/freescale_imx6qsabresd_defconfig +1 -1

index bf446a3..ec8f265 100644
@@ -31,4 +31,4 @@ BR2_TARGET_UBOOT_BOARDNAME="mx6qsabresd"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"

configs/freescale_imx6sololiteevk_defconfig +1 -1

index 72a1e2d..db68d79 100644
@@ -28,4 +28,4 @@ BR2_TARGET_UBOOT_BOARDNAME="mx6slevk"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"

configs/freescale_imx6sxsabresd_defconfig +1 -1

index 943528b..7437b63 100644
@@ -28,5 +28,5 @@ BR2_TARGET_UBOOT=y
 BR2_TARGET_UBOOT_BOARDNAME="mx6sxsabresd"
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_TARGET_UBOOT_FORMAT_IMX=y

configs/freescale_imx6ulevk_defconfig +1 -1

index fbb420e..cd27335 100644
@@ -22,7 +22,7 @@ BR2_TARGET_UBOOT=y
 BR2_TARGET_UBOOT_BOARDNAME="mx6ul_14x14_evk"
 BR2_TARGET_UBOOT_CUSTOM_GIT=y
 BR2_TARGET_UBOOT_CUSTOM_REPO_URL="git://git.freescale.com/imx/uboot-imx.git"
-BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_TARGET_UBOOT_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_TARGET_UBOOT_FORMAT_IMX=y
 
 # required tools to create the microSD image

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:18 UTC #39 lore raw

PATCH v2 1/14 nxp defconfigs: bump u-boot version to 4.1.15_1.1.0_ga

Andrew, All,

Quoted 1 line · Andrew Webster

Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:09 UTC #40 lore raw

PATCH v2 2/14 nxp defconfigs: bump kernel version to 4.1.15

There is a 1.1.0_ga release, and a 1.1.1_patch release. The patch release appears to be only for i.MX6UL, so it is bumped to 1.1.1_patch, while the other are 1.1.0_ga.

---
Changes v1 -> v2:
- added to patch set

 configs/freescale_imx6dlsabreauto_defconfig | 6 +++---
 configs/freescale_imx6dlsabresd_defconfig   | 6 +++---
 configs/freescale_imx6qsabreauto_defconfig  | 6 +++---
 configs/freescale_imx6qsabresd_defconfig    | 6 +++---
 configs/freescale_imx6sololiteevk_defconfig | 6 +++---
 configs/freescale_imx6sxsabresd_defconfig   | 6 +++---
 configs/freescale_imx6ulevk_defconfig       | 6 +++---
 7 files changed, 21 insertions(+), 21 deletions(-)

configs/freescale_imx6dlsabreauto_defconfig +3 -3

index f55d4fa..a58ab5d 100644
@@ -5,8 +5,8 @@ BR2_cortex_a9=y
 # patches
 BR2_GLOBAL_PATCH_DIR="board/freescale/imx6sabre/patches"
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc3"

@@ -15,7 +15,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc3"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_UIMAGE=y
 BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x10008000"

configs/freescale_imx6dlsabresd_defconfig +3 -3

index bec80b5..ecb5f8b 100644
@@ -5,8 +5,8 @@ BR2_cortex_a9=y
 # patches
 BR2_GLOBAL_PATCH_DIR="board/freescale/imx6sabre/patches"
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"

@@ -15,7 +15,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_UIMAGE=y
 BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x10008000"

configs/freescale_imx6qsabreauto_defconfig +3 -3

index 2b89541..6ed33ee 100644
@@ -5,8 +5,8 @@ BR2_cortex_a9=y
 # patches
 BR2_GLOBAL_PATCH_DIR="board/freescale/imx6sabre/patches"
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc3"

@@ -15,7 +15,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc3"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_UIMAGE=y
 BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x10008000"

configs/freescale_imx6qsabresd_defconfig +3 -3

index ec8f265..bd01f05 100644
@@ -5,8 +5,8 @@ BR2_cortex_a9=y
 # patches
 BR2_GLOBAL_PATCH_DIR="board/freescale/imx6sabre/patches"
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"

@@ -15,7 +15,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_UIMAGE=y
 BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x10008000"

configs/freescale_imx6sololiteevk_defconfig +3 -3

index db68d79..18043e8 100644
@@ -2,8 +2,8 @@
 BR2_arm=y
 BR2_cortex_a9=y
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"

@@ -12,7 +12,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_UIMAGE=y
 BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x80008000"

configs/freescale_imx6sxsabresd_defconfig +3 -3

index 7437b63..2519f49 100644
@@ -5,8 +5,8 @@ BR2_cortex_a9=y
 # patches
 BR2_GLOBAL_PATCH_DIR="board/freescale/imx6sabre/patches"
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"

@@ -15,7 +15,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.0_ga"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_DTS_SUPPORT=y
 BR2_LINUX_KERNEL_INTREE_DTS_NAME="imx6sx-sdb"

configs/freescale_imx6ulevk_defconfig +3 -3

index cd27335..5a18c1d 100644
@@ -2,8 +2,8 @@
 BR2_arm=y
 BR2_cortex_a7=y
 
-# Linux headers same as kernel, a 3.14 series
-BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_14=y
+# Linux headers same as kernel, a 4.1 series
+BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_1=y
 
 # system
 BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"

@@ -12,7 +12,7 @@ BR2_TARGET_GENERIC_GETTY_PORT="ttymxc0"
 BR2_LINUX_KERNEL=y
 BR2_LINUX_KERNEL_CUSTOM_GIT=y
 BR2_LINUX_KERNEL_CUSTOM_REPO_URL="git://git.freescale.com/imx/linux-2.6-imx.git"
-BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_3.14.52_1.1.0_ga"
+BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="rel_imx_4.1.15_1.1.1_patch"
 BR2_LINUX_KERNEL_DEFCONFIG="imx_v7"
 BR2_LINUX_KERNEL_DTS_SUPPORT=y
 BR2_LINUX_KERNEL_INTREE_DTS_NAME="imx6ul-14x14-evk"

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:18 UTC #41 lore raw

PATCH v2 2/14 nxp defconfigs: bump kernel version to 4.1.15

Andrew, All,

Quoted 5 lines · Andrew Webster

There is a 1.1.0_ga release, and a 1.1.1_patch release. The patch release appears to be only for i.MX6UL, so it is bumped to 1.1.1_patch, while the other are 1.1.0_ga.
Signed-off-by Andrew Webster <[email protected]>

Although every variation of the mx6 is using the 1.1.0_ga revision in the Yocto layer, I agree the 1.1.1_patch should be used for the mx6ul evk since it adds three patches required to handle the 6UL CPU frequencies properly.

Regards,
Gary

Andrew Webster

author

2016-06-10 18:09 UTC #42 lore raw

PATCH v2 3/14 kernel-module-imx-gpu-viv: bump version to that of the 5.0.11.p8.4 release

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c9c30c5

This package has been tested with the following commands with the matching imx-gpu-viv library:

 # modprobe galcore
 # cd /usr/share/examples/viv_samples/vdk/
 # ./tutorial7

---
Changes v1 -> v2:
- added to patch set

 .../kernel-module-imx-gpu-viv/kernel-module-imx-gpu-viv.hash            | 2 +-
 .../kernel-module-imx-gpu-viv/kernel-module-imx-gpu-viv.mk              | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/freescale-imx/kernel-module-imx-gpu-viv/kernel-module-imx-gpu-viv.hash +1 -1

index 0842c7a..6d810ea 100644
@@ -1,2 +1,2 @@
 # locally computed
-sha256	971254d204f6c8f739d0a2094055cd0b8001dc7d366345f996b14a495f92459e  kernel-module-imx-gpu-viv-2b3290d1c1ce383e860a0fd6b53074848a848e65.tar.gz
+sha256	6618b36baefd591e94a41167fa50d325c3df473918bceef027ca5a067aa1cffd  kernel-module-imx-gpu-viv-aa83874b916cf5655cb90498b1ea5691537fdbc8.tar.gz

package/freescale-imx/kernel-module-imx-gpu-viv/kernel-module-imx-gpu-viv.mk +1 -1

index aad8e40..11c6768 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-KERNEL_MODULE_IMX_GPU_VIV_VERSION = 2b3290d1c1ce383e860a0fd6b53074848a848e65
+KERNEL_MODULE_IMX_GPU_VIV_VERSION = aa83874b916cf5655cb90498b1ea5691537fdbc8
 KERNEL_MODULE_IMX_GPU_VIV_SITE = \
 	$(call github,Freescale,kernel-module-imx-gpu-viv,$(KERNEL_MODULE_IMX_GPU_VIV_VERSION))
 KERNEL_MODULE_IMX_GPU_VIV_LICENSE = GPLv2

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:19 UTC #43 lore raw

PATCH v2 3/14 kernel-module-imx-gpu-viv: bump version to that of the 5.0.11.p8.4 release

Andrew, All,

Quoted 10 lines · Andrew Webster

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c9c30c5
This package has been tested with the following commands with the matching imx-gpu-viv library:
 # modprobe galcore
 # cd /usr/share/examples/viv_samples/vdk/
 # ./tutorial7
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:09 UTC #44 lore raw

PATCH v2 4/14 imx-gpu-viv: bump version to 5.0.11-p8.4

https://www.nxp.com/webapp/Download?colCode=L4.1.15_1.1.0_LINUX_DOCS&location=null&fpsp=1&WT_TYPE=Supporting%20Information&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=gz&WT_ASSET=Documentation&fileExt=.gz&Parent_nodeId=1337637154535695831062&Parent_pageType=product

Changelog:
Software floating point is no longer supported.

For a detailed list of changes in this update, see the Graphics section of the i.MX Linux Release Notes contained in the i.MX Linux BSP documentation here:

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/b667ed4
https://github.com/Freescale/meta-fsl-arm/commit/af5b3a6

This package has been tested with both X11 and Framebuffer backends:

 # cd /usr/share/examples/viv_samples/vdk/
 # apitrace trace --api egl ./tutorial7
 # gmem_info
 ... display memory use per PID ...
 #?apitrace replay tutorial7.trace
 #?eglretrace tutorial7.trace

---
Changes v1 -> v2:
- added test procedure to comment
- updated from p8.3 to p8.4

 package/freescale-imx/imx-gpu-viv/Config.in        |  6 +++++-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash |  4 +---
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk   | 10 +++-------
 3 files changed, 9 insertions(+), 11 deletions(-)

package/freescale-imx/imx-gpu-viv/Config.in +5 -1

index 20759b1..9a7cf56 100644
@@ -2,12 +2,16 @@ comment "imx-gpu-viv needs an (e)glibc toolchain"
 	depends on BR2_arm
 	depends on !BR2_TOOLCHAIN_USES_GLIBC
 
+comment "imx-gpu-viv needs EABIhf"
+	depends on BR2_arm
+	depends on !BR2_ARM_EABIHF
+
 config BR2_PACKAGE_IMX_GPU_VIV
 	bool "imx-gpu-viv"
 	select BR2_PACKAGE_HAS_LIBEGL
 	select BR2_PACKAGE_HAS_LIBGLES
 	select BR2_PACKAGE_HAS_LIBOPENVG
-	depends on BR2_arm # Only relevant for i.MX6
+	depends on BR2_ARM_EABIHF # Only relevant for EABIhf i.MX6
 	# Library binaries are linked against libc.so.6
 	depends on BR2_TOOLCHAIN_USES_GLIBC
 	help

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash +1 -3

index 4119388..f94d3148 100644
@@ -1,3 +1 @@
-# locally computed
-sha256	3ea201f91ce9b56455e4983ab1055e8175415fbbf2800dc097d47e78ceefba34  imx-gpu-viv-5.0.11.p7.4-sfp.bin
-sha256	b80838104f2f2e3b0baf2f1aacdca263c1c92589ce919b3be40aaad23f686b7f  imx-gpu-viv-5.0.11.p7.4-hfp.bin
+sha256	8f79ce562d8ac1efb0a2af403b405541f98c55ae1631a18a3d7fad3224dfb44b  imx-gpu-viv-5.0.11.p8.4-hfp.bin

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +3 -7

index b51a7af..2a15b55 100644
@@ -4,19 +4,15 @@
 #
 ################################################################################
 
-IMX_GPU_VIV_BASE_VERSION = 5.0.11.p7.4
-ifeq ($(BR2_ARM_EABIHF),y)
+IMX_GPU_VIV_BASE_VERSION = 5.0.11.p8.4
 IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-hfp
-else
-IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-sfp
-endif
 IMX_GPU_VIV_SITE = $(FREESCALE_IMX_SITE)
 IMX_GPU_VIV_SOURCE = imx-gpu-viv-$(IMX_GPU_VIV_VERSION).bin
 
 IMX_GPU_VIV_INSTALL_STAGING = YES
 
-IMX_GPU_VIV_LICENSE = Freescale Semiconductor Software License Agreement
-IMX_GPU_VIV_LICENSE_FILES = EULA
+IMX_GPU_VIV_LICENSE = NXP Semiconductor Software License Agreement
+IMX_GPU_VIV_LICENSE_FILES = EULA COPYING
 IMX_GPU_VIV_REDISTRIBUTE = NO
 
 IMX_GPU_VIV_PROVIDES = libegl libgles libopenvg

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:00 UTC #45 lore raw

PATCH v2 4/14 imx-gpu-viv: bump version to 5.0.11-p8.4

Andrew, All,

A couple of aesthetic remarks.

Quoted 41 lines · Andrew Webster

Changelog:
Software floating point is no longer supported.
For a detailed list of changes in this update, see the Graphics section of the i.MX Linux Release Notes contained in the i.MX Linux BSP documentation here:
https://www.nxp.com/webapp/Download?colCode=L4.1.15_1.1.0_LINUX_DOCS&location=null&fpsp=1&WT_TYPE=Supporting%20Information&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=gz&WT_ASSET=Documentation&fileExt=.gz&Parent_nodeId=1337637154535695831062&Parent_pageType=product
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/b667ed4
https://github.com/Freescale/meta-fsl-arm/commit/af5b3a6
This package has been tested with both X11 and Framebuffer backends:
 # cd /usr/share/examples/viv_samples/vdk/
 # apitrace trace --api egl ./tutorial7
 # gmem_info
 ... display memory use per PID ...
 # apitrace replay tutorial7.trace
 # eglretrace tutorial7.trace
Signed-off-by Andrew Webster <[email protected]>
---
Changes v1 -> v2:
- added test procedure to comment
- updated from p8.3 to p8.4
 package/freescale-imx/imx-gpu-viv/Config.in        |  6 +++++-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash |  4 +---
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk   | 10 +++-------
 3 files changed, 9 insertions(+), 11 deletions(-)
package/freescale-imx/imx-gpu-viv/Config.in +1 -0
index 20759b1..9a7cf56 100644
@@ -2,12 +2,16 @@ comment "imx-gpu-viv needs an (e)glibc toolchain"
        depends on BR2_arm
        depends on !BR2_TOOLCHAIN_USES_GLIBC
 
+comment "imx-gpu-viv needs EABIhf"

package/freescale-imx/imx-gpu-viv/Config.in comment "imx-gpu-viv needs an (e)glibc toolchain" · line 14

Quoted 10 lines

package/freescale-imx/imx-gpu-viv/Config.in +4 -1

+       depends on BR2_arm
+       depends on !BR2_ARM_EABIHF
+
 config BR2_PACKAGE_IMX_GPU_VIV
        bool "imx-gpu-viv"
        select BR2_PACKAGE_HAS_LIBEGL
        select BR2_PACKAGE_HAS_LIBGLES
        select BR2_PACKAGE_HAS_LIBOPENVG
-       depends on BR2_arm # Only relevant for i.MX6
+       depends on BR2_ARM_EABIHF # Only relevant for EABIhf i.MX6

I'm not fond of this comment since "EABIhf i.MX6" are two terms that don't go together IMO. I'd either just keep the previous comment or something like "Only relevant for i.MX6 + EABIhf toolchain".

Thomas, what do you think?

Quoted 19 lines

package/freescale-imx/imx-gpu-viv/Config.in +0 -0

        # Library binaries are linked against libc.so.6
        depends on BR2_TOOLCHAIN_USES_GLIBC
        help

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash +1 -3

index 4119388..f94d3148 100644
@@ -1,3 +1 @@
-# locally computed
-sha256 3ea201f91ce9b56455e4983ab1055e8175415fbbf2800dc097d47e78ceefba34  imx-gpu-viv-5.0.11.p7.4-sfp.bin
-sha256 b80838104f2f2e3b0baf2f1aacdca263c1c92589ce919b3be40aaad23f686b7f  imx-gpu-viv-5.0.11.p7.4-hfp.bin
+sha256 8f79ce562d8ac1efb0a2af403b405541f98c55ae1631a18a3d7fad3224dfb44b  imx-gpu-viv-5.0.11.p8.4-hfp.bin

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +0 -0

index b51a7af..2a15b55 100644
@@ -4,19 +4,15 @@
 #
 ################################################################################

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk line 10

Quoted 10 lines

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +1 -5

 
-IMX_GPU_VIV_BASE_VERSION = 5.0.11.p7.4
-ifeq ($(BR2_ARM_EABIHF),y)
+IMX_GPU_VIV_BASE_VERSION = 5.0.11.p8.4
 IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-hfp
-else
-IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-sfp
-endif
 IMX_GPU_VIV_SITE = $(FREESCALE_IMX_SITE)
 IMX_GPU_VIV_SOURCE = imx-gpu-viv-$(IMX_GPU_VIV_VERSION).bin

I think it is ok to keep the version naming like you did but m

Gary Bisson

2016-06-13 13:02 UTC #46 lore raw

PATCH v2 4/14 imx-gpu-viv: bump version to 5.0.11-p8.4

Andrew, All,

Quoted 83 lines · Gary Bisson

Andrew, All,
A couple of aesthetic remarks.
Quoted 41 lines · Andrew Webster
Changelog:
Software floating point is no longer supported.
For a detailed list of changes in this update, see the Graphics section of the i.MX Linux Release Notes contained in the i.MX Linux BSP documentation here:
https://www.nxp.com/webapp/Download?colCode=L4.1.15_1.1.0_LINUX_DOCS&location=null&fpsp=1&WT_TYPE=Supporting%20Information&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=gz&WT_ASSET=Documentation&fileExt=.gz&Parent_nodeId=1337637154535695831062&Parent_pageType=product
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/b667ed4
https://github.com/Freescale/meta-fsl-arm/commit/af5b3a6
This package has been tested with both X11 and Framebuffer backends:
 # cd /usr/share/examples/viv_samples/vdk/
 # apitrace trace --api egl ./tutorial7
 # gmem_info
 ... display memory use per PID ...
 # apitrace replay tutorial7.trace
 # eglretrace tutorial7.trace
Signed-off-by Andrew Webster <[email protected]>
---
Changes v1 -> v2:
- added test procedure to comment
- updated from p8.3 to p8.4
 package/freescale-imx/imx-gpu-viv/Config.in        |  6 +++++-
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash |  4 +---
 package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk   | 10 +++-------
 3 files changed, 9 insertions(+), 11 deletions(-)
package/freescale-imx/imx-gpu-viv/Config.in +1 -0
index 20759b1..9a7cf56 100644
@@ -2,12 +2,16 @@ comment "imx-gpu-viv needs an (e)glibc toolchain"
        depends on BR2_arm
        depends on !BR2_TOOLCHAIN_USES_GLIBC
 
+comment "imx-gpu-viv needs EABIhf"
package/freescale-imx/imx-gpu-viv/Config.in comment "imx-gpu-viv needs an (e)glibc toolchain" · line 14
Quoted 10 lines
package/freescale-imx/imx-gpu-viv/Config.in +4 -1
+       depends on BR2_arm
+       depends on !BR2_ARM_EABIHF
+
 config BR2_PACKAGE_IMX_GPU_VIV
        bool "imx-gpu-viv"
        select BR2_PACKAGE_HAS_LIBEGL
        select BR2_PACKAGE_HAS_LIBGLES
        select BR2_PACKAGE_HAS_LIBOPENVG
-       depends on BR2_arm # Only relevant for i.MX6
+       depends on BR2_ARM_EABIHF # Only relevant for EABIhf i.MX6
I'm not fond of this comment since "EABIhf i.MX6" are two terms that don't go together IMO. I'd either just keep the previous comment or something like "Only relevant for i.MX6 + EABIhf toolchain".
Thomas, what do you think?
Quoted 20 lines
package/freescale-imx/imx-gpu-viv/Config.in +0 -0
        # Library binaries are linked against libc.so.6
        depends on BR2_TOOLCHAIN_USES_GLIBC
        help
package/freescale-imx/imx-gpu-viv/imx-gpu-viv.hash +1 -3
index 4119388..f94d3148 100644
@@ -1,3 +1 @@
-# locally computed
-sha256 3ea201f91ce9b56455e4983ab1055e8175415fbbf2800dc097d47e78ceefba34  imx-gpu-viv-5.0.11.p7.4-sfp.bin
-sha256 b80838104f2f2e3b0baf2f1aacdca263c1c92589ce919b3be40aaad23f686b7f  imx-gpu-viv-5.0.11.p7.4-hfp.bin
+sha256 8f79ce562d8ac1efb0a2af403b405541f98c55ae1631a18a3d7fad3224dfb44b  imx-gpu-viv-5.0.11.p8.4-hfp.bin
package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +0 -0
index b51a7af..2a15b55 100644
@@ -4,19 +4,15 @@
 #
 ################################################################################
 

Quoted 11 lines

package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk line 10
Quoted 9 lines
package/freescale-imx/imx-gpu-viv/imx-gpu-viv.mk +1 -5
-IMX_GPU_VIV_BASE_VERSION = 5.0.11.p7.4
-ifeq ($(BR2_ARM_EABIHF),y)
+IMX_GPU_VIV_BASE_VERSION = 5.0.11.p8.4
 IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-hfp
-else
-IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-sfp
-endif
 IMX_GPU_VIV_SITE = $(FREESCALE_IMX_SITE)
 IMX_GPU_VIV_SOURCE = imx-gpu-viv-$(IMX_GPU_VIV_VERSION).bin
I think it is ok to keep the version naming like you did but m

Sorry, pressed send too early. I wanted to say that maybe some may prefer if we get rid of IMX_GPU_VIV_BASE_VERSION and have IMX_GPU_VIV_VERSION directly set to 5.0.11.p8.4-hfp.

Anyway, tested it and it works fine:

Also, if Thomas agrees to keep the naming like this and doesn't care about the comment above:

Regards,
Gary

Thomas Petazzoni

2016-06-15 20:45 UTC #47 lore raw

PATCH v2 4/14 imx-gpu-viv: bump version to 5.0.11-p8.4

Hello,

Quoted 11 lines · Gary Bisson

package/freescale-imx/imx-gpu-viv/Config.in comment "imx-gpu-viv needs an (e)glibc toolchain" · line 7
Quoted 11 lines
package/freescale-imx/imx-gpu-viv/Config.in +3 -0
index 20759b1..9a7cf56 100644
@@ -2,12 +2,16 @@ comment "imx-gpu-viv needs an (e)glibc toolchain"
        depends on BR2_arm
        depends on !BR2_TOOLCHAIN_USES_GLIBC
 
+comment "imx-gpu-viv needs EABIhf"
+       depends on BR2_arm
+       depends on !BR2_ARM_EABIHF

The mali-t76x had a similar situation, but uses a single comment for the glibc and ARM EABIhf dependency, so I've changed it to do the same:

comment "mali-t76x needs an (e)glibc toolchain with armhf enabled"
depends on BR2_arm
depends on !BR2_TOOLCHAIN_USES_GLIBC || !BR2_ARM_EABIHF

Quoted 8 lines

package/freescale-imx/imx-gpu-viv/Config.in comment "imx-gpu-viv needs an (e)glibc toolchain" · line 14

Quoted 7 lines

package/freescale-imx/imx-gpu-viv/Config.in +1 -1

 config BR2_PACKAGE_IMX_GPU_VIV
        bool "imx-gpu-viv"
        select BR2_PACKAGE_HAS_LIBEGL
        select BR2_PACKAGE_HAS_LIBGLES
        select BR2_PACKAGE_HAS_LIBOPENVG
-       depends on BR2_arm # Only relevant for i.MX6
+       depends on BR2_ARM_EABIHF # Only relevant for EABIhf i.MX6

Quoted 5 lines

I'm not fond of this comment since "EABIhf i.MX6" are two terms that don't go together IMO. I'd either just keep the previous comment or something like "Only relevant for i.MX6 + EABIhf toolchain".
Thomas, what do you think?

Correct, I've changed this to:

# Pre-built binaries only available for ARM EABIhf depends on BR2_ARM_EABIHF

Quoted 11 lines

package/freescale-imx/imx-gpu-viv/Config.in comment "imx-gpu-viv needs an (e)glibc toolchain"
Quoted 9 lines
package/freescale-imx/imx-gpu-viv/Config.in +1 -5
-IMX_GPU_VIV_BASE_VERSION = 5.0.11.p7.4
-ifeq ($(BR2_ARM_EABIHF),y)
+IMX_GPU_VIV_BASE_VERSION = 5.0.11.p8.4
 IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-hfp
-else
-IMX_GPU_VIV_VERSION = $(IMX_GPU_VIV_BASE_VERSION)-sfp
-endif
 IMX_GPU_VIV_SITE = $(FREESCALE_IMX_SITE)
 IMX_GPU_VIV_SOURCE = imx-gpu-viv-$(IMX_GPU_VIV_VERSION).bin  
I think it is ok to keep the version naming like you did but m

I've removed IMX_GPU_VIV_BASE_VERSION entirely, and use directly IMX_GPU_VIV_VERSION.

Thanks!

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Andrew Webster

author

2016-06-10 18:11 UTC #48 lore raw

PATCH v2 5/14 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.4

Remove upstreamed patch.
Add patch from Yocto to not use git for versioning.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/06a0574
https://github.com/Freescale/meta-fsl-arm/commit/076df90

The following xorg.conf was used in order to force the use of the vivante module:
https://github.com/Freescale/meta-fsl-arm/blob/master/recipes-graphics/xorg-xserver/xserver-xf86-config/mx6/xorg.conf

Note that the X server must be started with the "noreset" option to avoid crashes when closing X apps.
Tested with:
# X -noreset &

 # cd /usr/share/examples/viv_samples/vdk/
 # DISPLAY=:0 ./tutorial7

---
Changes v1 -> v2:
- added test procedure to comment
- updated from p8.3 to p8.4

 .../0001-Remove-dix-internal-header-usage.patch    | 31 --------
 ...001-Stop-using-Git-to-write-local-version.patch | 87 ++++++++++++++++++++++
 .../xdriver_xf86-video-imx-viv.hash                |  3 +-
 .../xdriver_xf86-video-imx-viv.mk                  |  2 +-
 4 files changed, 89 insertions(+), 34 deletions(-)
 delete mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch
 create mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 ~~-31~~

deleted file mode 100644
index f146ba3..0000000
@@ -1,31 +0,0 @@
-From 80c419dddf6483873503d90694c4b4ae34ffa5e4 Mon Sep 17 00:00:00 2001
-From: Otavio Salvador <[email protected]>
-Date: Wed, 2 Dec 2015 13:50:57 +0000
-Subject: [PATCH] Remove 'dix' internal header usage
-Organization: O.S. Systems Software LTDA.
-
-The 'swaprep.h' is an internal Xorg header and is not being installed
-anymore. The swap features are provided by the 'misc.h' header now.
-
-Upstream-Status: Pending
-
-Signed-off-by: Otavio Salvador <[email protected]>
----
- EXA/src/vivante_extension/vivante_ext.c | 1 -
- 1 file changed, 1 deletion(-)
-
-diff --git a/EXA/src/vivante_extension/vivante_ext.c b/EXA/src/vivante_extension/vivante_ext.c
-index 0d1085b..1ffbb1b 100644
---- a/EXA/src/vivante_extension/vivante_ext.c
-+++ b/EXA/src/vivante_extension/vivante_ext.c
-@@ -40,7 +40,6 @@
- #include "cursorstr.h"
- #include "scrnintstr.h"
- #include "servermd.h"
--#include "swaprep.h"
- #include "drm.h"
- #include "xf86Module.h"
- #include "globals.h"
--- 
-2.1.4
-

package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch new file +87 -0

87 lines not on this page · raw message

package/x11r7/xdriver_xf86-video-imx-viv/xdriver_xf86-video-imx-viv.hash +1 -2

index 9a16569..2cdf2d0 100644
@@ -1,2 +1 @@
-# locally computed
-sha256	3effaf42bde64da8fad2dfe0dcef12d1cfbbbe189511dbd4612442129fd832cc  xserver-xorg-video-imx-viv-5.0.11.p7.4.tar.gz
+sha256	ddb6de9e00ce0b22f6c905eaf6694424413eade53ef6cd3b36e20ac99dcc0e0f  xserver-xorg-video-imx-viv-5.0.11.p8.4.tar.gz

package/x11r7/xdriver_xf86-video-imx-viv/xdriver_xf86-video-imx-viv.mk +1 -1

index a1b32b1..9ce08ef 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-XDRIVER_XF86_VIDEO_IMX_VIV_VERSION = 5.0.11.p7.4
+XDRIVER_XF86_VIDEO_IMX_VIV_VERSION = 5.0.11.p8.4
 XDRIVER_XF86_VIDEO_IMX_VIV_SITE = $(FREESCALE_IMX_SITE)
 XDRIVER_XF86_VIDEO_IMX_VIV_SOURCE = xserver-xorg-video-imx-viv-$(XDRIVER_XF86_VIDEO_IMX_VIV_VERSION).tar.gz
 XDRIVER_XF86_VIDEO_IMX_VIV_DEPENDENCIES = imx-gpu-viv xserver_xorg-server \

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:20 UTC #49 lore raw

PATCH v2 5/14 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.4

Andrew, All,

Quoted 40 lines · Andrew Webster

Remove upstreamed patch.
Add patch from Yocto to not use git for versioning.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/06a0574
https://github.com/Freescale/meta-fsl-arm/commit/076df90
The following xorg.conf was used in order to force the use of the vivante module:
https://github.com/Freescale/meta-fsl-arm/blob/master/recipes-graphics/xorg-xserver/xserver-xf86-config/mx6/xorg.conf
Note that the X server must be started with the "noreset" option to avoid crashes when closing X apps.
Tested with:
# X -noreset &
 # cd /usr/share/examples/viv_samples/vdk/
 # DISPLAY=:0 ./tutorial7
Signed-off-by Andrew Webster <[email protected]>
---
Changes v1 -> v2:
- added test procedure to comment
- updated from p8.3 to p8.4
 .../0001-Remove-dix-internal-header-usage.patch    | 31 --------
 ...001-Stop-using-Git-to-write-local-version.patch | 87 ++++++++++++++++++++++
 .../xdriver_xf86-video-imx-viv.hash                |  3 +-
 .../xdriver_xf86-video-imx-viv.mk                  |  2 +-
 4 files changed, 89 insertions(+), 34 deletions(-)
 delete mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch
 create mode 100644 package/x11r7/xdriver_xf86-video-imx-viv/0001-Stop-using-Git-to-write-local-version.patch
package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 -2
deleted file mode 100644
index f146ba3..0000000
@@ -1,31 +0,0 @@
-From 80c419dddf6483873503d90694c4b4ae34ffa5e4 Mon Sep 17 00:00:00 2001
-From: Otavio Salvador <[email protected]>

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch line 12

Quoted 10 lines

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 ~~-10~~

-Date: Wed, 2 Dec 2015 13:50:57 +0000
-Subject: [PATCH] Remove 'dix' internal header usage
-Organization: O.S. Systems Software LTDA.
-
-The 'swaprep.h' is an internal Xorg header and is not being installed
-anymore. The swap features are provided by the 'misc.h' header now.
-
-Upstream-Status: Pending
-
-Signed-off-by: Otavio Salvador <[email protected]>

I think you need to add your SOB in the patch like:

sob: Otavio

Downloaded from:
<url from yocto>

sob: Andrew

---

Also, should we add a dependency on ARM_EABIHF since it depends on imx-gpu-viv? I'm never a 100% sure when you need to propagate those dependencies.

Regards,
Gary

Thomas Petazzoni

2016-06-15 20:47 UTC #50 lore raw

PATCH v2 5/14 xdriver_xf86-video-imx-viv: bump version to 5.0.11-p8.4

Hello,

Quoted 7 lines · Gary Bisson

Quoted 7 lines

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 -6

@@ -1,31 +0,0 @@
-From 80c419dddf6483873503d90694c4b4ae34ffa5e4 Mon Sep 17 00:00:00 2001
-From: Otavio Salvador <[email protected]>
-Date: Wed, 2 Dec 2015 13:50:57 +0000
-Subject: [PATCH] Remove 'dix' internal header usage
-Organization: O.S. Systems Software LTDA.
-

Quoted 8 lines

package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch line 12
Quoted 6 lines
package/x11r7/xdriver_xf86-video-imx-viv/0001-Remove-dix-internal-header-usage.patch deleted +0 -6
-The 'swaprep.h' is an internal Xorg header and is not being installed
-anymore. The swap features are provided by the 'misc.h' header now.
-
-Upstream-Status: Pending
-
-Signed-off-by: Otavio Salvador <[email protected]>  
I think you need to add your SOB in the patch like:

You're commenting on the patch that is *removed* so it's not really where a change should be done. However, I've added the reference to the upstream commit in the patch added by Andrew, indeed.

Quoted 3 lines

Also, should we add a dependency on ARM_EABIHF since it depends on imx-gpu-viv? I'm never a 100% sure when you need to propagate those dependencies.

You need to propagate the dependencies when you're doing a select, i.e:

config BR2_PACKAGE_FOO
depends on BR2_ARM_EABIHF

config BR2_PACKAGE_BAR
select BR2_PACKAGE_FOO

is wrong, because BR2_PACKAGE_BAR should replicate the "depends on BR2_ARM_EABIHF".

However, in the case here, xdriver_xf86-video-imx-viv does:

depends on BR2_PACKAGE_IMX_GPU_VIV_OUTPUT_X11

so we're good (i.e it doesn't "select" the imx-gpu-viv package).

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Andrew Webster

author

2016-06-10 18:11 UTC #51 lore raw

PATCH v2 6/14 firmware-imx: bump to version 5.4

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773

Implicitly tested through gstreamer as the plugin relies on it:

 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink

---
Changes v1 -> v2:
- added test procedure to comment

 package/freescale-imx/firmware-imx/firmware-imx.hash | 2 +-
 package/freescale-imx/firmware-imx/firmware-imx.mk   | 5 ++---
 2 files changed, 3 insertions(+), 4 deletions(-)

package/freescale-imx/firmware-imx/firmware-imx.hash +1 -1

index a5bc82c..a6e5bd5 100644
@@ -1 +1 @@
-sha256 fd14c3666eb3c12f4f2665d95caf383bb1d0277bb12e2de28fefc8d1230604f2  firmware-imx-5.3.bin
+sha256 c5bd4bff48cce9715a5d6d2c190ff3cd2262c7196f7facb9b0eda231c92cc223  firmware-imx-5.4.bin

package/freescale-imx/firmware-imx/firmware-imx.mk +2 -3

index 8f81c57..1893929 100644
@@ -4,12 +4,11 @@
 #
 ################################################################################
 
-FIRMWARE_IMX_VERSION = 5.3
+FIRMWARE_IMX_VERSION = 5.4
 FIRMWARE_IMX_SITE = $(FREESCALE_IMX_SITE)
 FIRMWARE_IMX_SOURCE = firmware-imx-$(FIRMWARE_IMX_VERSION).bin
 
-FIRMWARE_IMX_LICENSE = Freescale Semiconductor Software License Agreement, \
-	Atheros license (ath6k)
+FIRMWARE_IMX_LICENSE = NXP Semiconductor Software License Agreement
 FIRMWARE_IMX_LICENSE_FILES = EULA COPYING
 FIRMWARE_IMX_REDISTRIBUTE = NO

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:20 UTC #52 lore raw

PATCH v2 6/14 firmware-imx: bump to version 5.4

Andrew, All,

Quoted 8 lines · Andrew Webster

Changelog:
- Upgrade to FW v3.1.1_r46072
- v3.1.1_r46070 plus 2 patches 391/392
- FSLARIK-391 [Decoder/H.264] VPU is always busy when decoding an erroneous H264 bitstream
- FSLARIK-392 [Decoder/VC1] VPU outputs only 1 picture
- Change to the format of Linux coda driver firmware

Quoted 8 lines

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/dbb8773
Implicitly tested through gstreamer as the plugin relies on it:
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink
Signed-off-by Andrew Webster <[email protected]>

Note that I've tested the VPU firmware on a 3.14.52 kernel just fine. The reason is that this firmware is actually loaded by the user space imx-vpu component, so as long as firmware-imx and imx-vpu are aligned, the endianness of the fw file doesn't matter.

Regards,
Gary

Andrew Webster

author

2016-06-10 18:12 UTC #53 lore raw

PATCH v2 7/14 gst1-imx: bump to version 0.12.1

Changelog:
* compositor: Rename ImxCompositor(Pad) to ImxVideoCompositor(Pad) This prevents conflicts if both the NXP plugins and the gstreamer-imx are installed in the same rootfs

* ipu+pxp: : Fix deadlocks and incorrect logging in case of init error This was causing a freeze if for example one tries to use PxP elements on an i.MX6 Dual/Quad

* eglvivsink: wayland: Do not commit before buffers are swapped This fixes eglSwapBuffer swaps with damaged frames

* Documentation additions
Adds GPU memory leak notess to the FAQ

* vpu: encoder: Expose motion estimation range as property

* vpu: decoder: Drop output frame if its buffer cannot be allocated

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/cec4c47

Tested with the following commands on i.MX6Q (IPU):

 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxeglvivsink

---
Changes v1 -> v2:
- added test procedure to comment

 package/gstreamer1/gst1-imx/gst1-imx.hash | 2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/gstreamer1/gst1-imx/gst1-imx.hash +1 -1

index 03e1729..ca47776 100644
@@ -1,2 +1,2 @@
 # locally computed hash
-sha256 df34f080e24ac8f1619780aa5ad4e6ab1333f822cd2bf5fcb380b28c97248810  gst1-imx-0.12.0.tar.gz
+sha256 d8a7357bc575e22d0d19d6d7cf547bd69cad5fc47adb93f09ead41b0124353fa  gst1-imx-0.12.1.tar.gz

package/gstreamer1/gst1-imx/gst1-imx.mk +1 -1

index 650e097..8838110 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-GST1_IMX_VERSION = 0.12.0
+GST1_IMX_VERSION = 0.12.1
 GST1_IMX_SITE = $(call github,Freescale,gstreamer-imx,$(GST1_IMX_VERSION))
 
 GST1_IMX_LICENSE = LGPLv2+

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:20 UTC #54 lore raw

PATCH v2 7/14 gst1-imx: bump to version 0.12.1

Andrew, All,

Quoted 34 lines · Andrew Webster

Changelog:
* compositor: Rename ImxCompositor(Pad) to ImxVideoCompositor(Pad) This prevents conflicts if both the NXP plugins and the gstreamer-imx are installed in the same rootfs
* Add switches to wscript configuration to enable/disable plugins Until now, all plugins were built (unless their dependencies were missing). On some i.MX machines, some features are missing. For example, the i.MX7 has no VPU. To ensure deterministic builds, it is useful to selectively disable plugins in the build.
* ipu+pxp: : Fix deadlocks and incorrect logging in case of init error This was causing a freeze if for example one tries to use PxP elements on an i.MX6 Dual/Quad
* imxv4l2videosrc fixes and enhancements:
* fixes to interlacing support (incl. workarounds for analog TV input)
* remove format restriction in get_caps function
this removes the limitation on I420 formats
* simplified determination of fps (necessary due to driver limitations)
* eglvivsink: wayland: Do not commit before buffers are swapped This fixes eglSwapBuffer swaps with damaged frames
* Documentation additions
Adds GPU memory leak notess to the FAQ
* vpu: encoder: Expose motion estimation range as property
* vpu: decoder: Drop output frame if its buffer cannot be allocated
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/cec4c47

Quoted 8 lines

Tested with the following commands on i.MX6Q (IPU):

 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxeglvivsink

Once merged, I'll make a follow-up patch that allows to disable components you don't want.

Regards,
Gary

Andrew Webster

author

2016-06-10 18:12 UTC #55 lore raw

PATCH v2 8/14 imx-vpu: bump to version 5.4.33

Changelog:
VPU firmware has been changed to the format of Linux codec driver firmware. Change VPU library to adapt to firmware format change.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/f5e1056

Implicitly tested through gstreamer as the plugins rely on it:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink

---
Changes v1 -> v2:
- added test procedure to comment

 package/freescale-imx/imx-vpu/imx-vpu.hash | 2 +-
 package/freescale-imx/imx-vpu/imx-vpu.mk   | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

package/freescale-imx/imx-vpu/imx-vpu.hash +1 -1

index c828c2b..67ebaf2 100644
@@ -1,2 +1,2 @@
 # Locally computed
-sha256	f11b91dcdb84bfefd3c88f10a9c1c1bc3420e4bbaadc38b9280fb52f7769eca8  imx-vpu-5.4.32.bin
+sha256	2e670ef2337a5dbe6c32fa07128ad9aad32afdbf472c23d44133245ec6a66d46  imx-vpu-5.4.33.bin

package/freescale-imx/imx-vpu/imx-vpu.mk +2 -2

index 949fefa..09ff5a8 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-IMX_VPU_VERSION = 5.4.32
+IMX_VPU_VERSION = 5.4.33
 IMX_VPU_SITE = $(FREESCALE_IMX_SITE)
 IMX_VPU_SOURCE = imx-vpu-$(IMX_VPU_VERSION).bin

@@ -16,7 +16,7 @@ IMX_VPU_MAKE_ENV = \
 	CROSS_COMPILE="$(TARGET_CROSS)" \
 	PLATFORM=$(BR2_PACKAGE_FREESCALE_IMX_PLATFORM)
 
-IMX_VPU_LICENSE = Freescale Semiconductor Software License Agreement
+IMX_VPU_LICENSE = NXP Semiconductor Software License Agreement
 IMX_VPU_LICENSE_FILES = EULA COPYING
 IMX_VPU_REDISTRIBUTE = NO

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:20 UTC #56 lore raw

PATCH v2 8/14 imx-vpu: bump to version 5.4.33

Andrew, All,

Quoted 7 lines · Andrew Webster

Changelog:
VPU firmware has been changed to the format of Linux codec driver firmware. Change VPU library to adapt to firmware format change.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/f5e1056

Quoted 7 lines

Implicitly tested through gstreamer as the plugins rely on it:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink

Also confirm that gstreamer0.10 plugins are still working fine.

Regards,
Gary

Andrew Webster

author

2016-06-10 18:13 UTC #57 lore raw

PATCH v2 9/14 imx-codec: rename from libfslcodec and bump version to 4.0.9

Changelog:
--[aac_dec] refine the code for pure c build
aac decoder c build need float point support, which use some float point function.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/4b91194

Implicitly tested through gstreamer as the 0.10 plugin relies on it:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm

---
Changes v1 -> v2:
- added test procedure to comment
- moved legacy config under 2016.08 section
- renamed Freescale to NXP in Config.in

 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-codec}/Config.in                           |  6 +++---
 package/freescale-imx/imx-codec/imx-codec.hash     |  2 ++
 package/freescale-imx/imx-codec/imx-codec.mk       | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  4 ++--
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/gstreamer1/gst1-imx/Config.in              |  2 +-
 package/gstreamer1/gst1-imx/gst1-imx.mk            |  4 ++--
 package/libfslcodec/libfslcodec.hash               |  2 --
 package/libfslcodec/libfslcodec.mk                 | 23 ----------------------
 12 files changed, 42 insertions(+), 35 deletions(-)
 rename package/{libfslcodec => freescale-imx/imx-codec}/Config.in (75%)
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.hash
 create mode 100644 package/freescale-imx/imx-codec/imx-codec.mk
 delete mode 100644 package/libfslcodec/libfslcodec.hash
 delete mode 100644 package/libfslcodec/libfslcodec.mk

Config.in.legacy +7 -0

index 434aa5f..e5a568d 100644
@@ -145,6 +145,13 @@ endif
 ###############################################################################
 comment "Legacy options removed in 2016.08"
 
+config BR2_PACKAGE_LIBFSLCODEC
+	bool "libfslcodec has been renamed to imx-codec"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_CODEC
+	help
+	  The libfslcodec has been renamed to match the renamed package.
+
 config BR2_PACKAGE_UBOOT_TOOLS_MKIMAGE_FIT_SIGNATURE_SUPPORT
 	bool "FIT support in uboot-tools has been refactored"
 	select BR2_LEGACY

package/Config.in +0 -1

index 38c8527..1cdbaa0 100644
@@ -1098,7 +1098,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslcodec/Config.in"
 	source "package/libfslparser/Config.in"
 	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"

package/freescale-imx/Config.in +1 -0

index 8c03190..052de5a 100644
@@ -49,6 +49,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 		BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \
 		BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q
 
+source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"

package/freescale-imx/imx-codec/Config.in new file +13 -0

new file mode 100644
index 0000000..70d340a
@@ -0,0 +1,13 @@
+config BR2_PACKAGE_IMX_CODEC
+	bool "imx-codec"
+	depends on BR2_arm # Only relevant for i.MX
+	depends on BR2_TOOLCHAIN_USES_GLIBC # prebuilt binaries
+	help
+	  Binary codec libraries for the NXP i.MX SoCs.
+
+	  This library is provided by NXP as-is and doesn't have
+	  an upstream.
+
+comment "imx-codec needs an (e)glibc toolchain"
+	depends on BR2_arm
+	depends on !BR2_TOOLCHAIN_USES_GLIBC

package/freescale-imx/imx-codec/imx-codec.hash new file +2 -0

new file mode 100644
index 0000000..39bddc1
@@ -0,0 +1,2 @@
+# locally computed
+sha256	62bda182143cde9743cba0eb2a1c315a53cb434294605742f9fc364219958888  imx-codec-4.0.9.bin

package/freescale-imx/imx-codec/imx-codec.mk new file +23 -0

new file mode 100644
index 0000000..67d3390
@@ -0,0 +1,23 @@
+################################################################################
+#
+# imx-codec
+#
+################################################################################
+
+IMX_CODEC_VERSION = 4.0.9
+IMX_CODEC_SITE = $(FREESCALE_IMX_SITE)
+IMX_CODEC_SOURCE = imx-codec-$(IMX_CODEC_VERSION).bin
+IMX_CODEC_INSTALL_STAGING = YES
+
+IMX_CODEC_LICENSE = NXP Semiconductor Software License Agreement, BSD-3c (flac, ogg headers)
+IMX_CODEC_LICENSE_FILES = EULA COPYING
+IMX_CODEC_REDISTRIBUTE = NO
+
+define IMX_CODEC_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_CODEC_SOURCE))
+endef
+
+# FIXME The Makefile installs both the arm9 and arm11 versions of the
+# libraries, but we only need one of them.
+
+$(eval $(autotools-package))

package/gstreamer/gst-fsl-plugins/Config.in +2 -2

index 0407dc5..8c82a4a 100644
@@ -9,12 +9,12 @@ config BR2_PACKAGE_GST_FSL_PLUGINS
 	bool "gst-fsl-plugins"
 	depends on BR2_LINUX_KERNEL
 	depends on BR2_arm # Only relevant for i.MX
-	depends on BR2_TOOLCHAIN_USES_GLIBC # libfslcodec
+	depends on BR2_TOOLCHAIN_USES_GLIBC # imx-codec
 	select BR2_PACKAGE_GST_PLUGINS_BASE
 	select BR2_PACKAGE_LIBFSLVPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 	select BR2_PACKAGE_IMX_LIB
 	select BR2_PACKAGE_LIBFSLPARSER
-	select BR2_PACKAGE_LIBFSLCODEC
+	select BR2_PACKAGE_IMX_CODEC
 	select BR2_PACKAGE_FREESCALE_IMX
 	help
 	  GStreamer plugins for hardware-accelerated audio and video

package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk +1 -1

index fa0e093..50732bf 100644
@@ -16,7 +16,7 @@ GST_FSL_PLUGINS_INSTALL_STAGING = YES
 GST_FSL_PLUGINS_AUTORECONF = YES
 
 GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \
-	imx-lib libfslparser libfslcodec
+	imx-lib libfslparser imx-codec
 
 ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y)
 GST_FSL_PLUGINS_DEPENDENCIES += libfslvpuwrap

package/gstreamer1/gst1-imx/Config.in +1 -1

index 0084483..3512666 100644
@@ -18,6 +18,6 @@ config BR2_PACKAGE_GST1_IMX
 	  The VPU plugin is built when BR2_PACKAGE_LIBIMXVPUAPI is enabled.
 	  The EGL plugin is built when BR2_PACKAGE_IMX_GPU_VIV is enabled.
 	  The G2D plugin is built when BR2_PACKAGE_IMX_GPU_VIV_G2D is enabled.
-	  The MP3 plugin is built when BR2_PACKAGE_LIBFSLCODEC is enabled.
+	  The MP3 plugin is built when BR2_PACKAGE_IMX_CODEC is enabled.
 
 	  https://github.com/Freescale/gstreamer-imx

package/gstreamer1/gst1-imx/gst1-imx.mk +2 -2

index 8838110..f81d8c9 100644
@@ -27,8 +27,8 @@ ifeq ($(BR2_PACKAGE_GST1_PLUGINS_BAD),y)
 GST1_IMX_DEPENDENCIES += gst1-plugins-bad
 endif
 
-ifeq ($(BR2_PACKAGE_LIBFSLCODEC),y)
-GST1_IMX_DEPENDENCIES += libfslcodec
+ifeq ($(BR2_PACKAGE_IMX_CODEC),y)
+GST1_IMX_DEPENDENCIES += imx-codec
 endif
 
 ifeq ($(BR2_PACKAGE_LIBIMXVPUAPI),y)

package/libfslcodec/Config.in deleted +0 ~~-13~~

deleted file mode 100644
index d2007fe..0000000
@@ -1,13 +0,0 @@
-config BR2_PACKAGE_LIBFSLCODEC
-	bool "libfslcodec"
-	depends on BR2_arm # Only relevant for i.MX
-	depends on BR2_TOOLCHAIN_USES_GLIBC # prebuilt binaries
-	help
-	  Binary codec libraries for the Freescale i.MX SoCs.
-
-	  This library is provided by Freescale as-is and doesn't have
-	  an upstream.
-
-comment "libfslcodec needs an (e)glibc toolchain"
-	depends on BR2_arm
-	depends on !BR2_TOOLCHAIN_USES_GLIBC

package/libfslcodec/libfslcodec.hash deleted +0 -2

deleted file mode 100644
index fd934f7..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	544ffc5989bce18ca50c7a826cc03370b1cf8455335e4291ef0c0779b4c33de0  libfslcodec-4.0.8.bin

package/libfslcodec/libfslcodec.mk deleted +0 ~~-23~~

deleted file mode 100644
index 502390f..0000000
@@ -1,23 +0,0 @@
-################################################################################
-#
-# libfslcodec
-#
-################################################################################
-
-LIBFSLCODEC_VERSION = 4.0.8
-LIBFSLCODEC_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLCODEC_SOURCE = libfslcodec-$(LIBFSLCODEC_VERSION).bin
-LIBFSLCODEC_INSTALL_STAGING = YES
-
-LIBFSLCODEC_LICENSE = Freescale Semiconductor Software License Agreement, BSD-3c (flac, ogg headers)
-LIBFSLCODEC_LICENSE_FILES = EULA COPYING
-LIBFSLCODEC_REDISTRIBUTE = NO
-
-define LIBFSLCODEC_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLCODEC_SOURCE))
-endef
-
-# FIXME The Makefile installs both the arm9 and arm11 versions of the
-# libraries, but we only need one of them.
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #58 lore raw

PATCH v2 9/14 imx-codec: rename from libfslcodec and bump version to 4.0.9

Andrew, All,

Quoted 5 lines · Andrew Webster

Changelog:
--[aac_dec] refine the code for pure c build
aac decoder c build need float point support, which use some float point function.

Quoted 7 lines

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/4b91194
Implicitly tested through gstreamer as the 0.10 plugin relies on it:
 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:13 UTC #59 lore raw

PATCH v2 10/14 imx-parser: rename from libfslparser and bump version to 4.0.9

Changelog:
--Add aac latm parser in parser_libs, support aac latm in mpeg2 parser

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/e2c6f10

Implicitly tested through gstreamer as the 0.10 plugin relies on it:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm

---
Changes v1 -> v2:
- added test procedure to comment
- moved legacy config under 2016.08 section
- renamed Freescale to NXP in Config.in

 Config.in.legacy                                   |  7 +++++++
 package/Config.in                                  |  1 -
 package/freescale-imx/Config.in                    |  1 +
 .../imx-parser}/Config.in                          |  4 ++--
 package/freescale-imx/imx-parser/imx-parser.hash   |  2 ++
 package/freescale-imx/imx-parser/imx-parser.mk     | 23 ++++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in        |  2 +-
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk   |  2 +-
 package/libfslparser/libfslparser.hash             |  2 --
 package/libfslparser/libfslparser.mk               | 23 ----------------------
 10 files changed, 37 insertions(+), 30 deletions(-)
 rename package/{libfslparser => freescale-imx/imx-parser}/Config.in (78%)
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.hash
 create mode 100644 package/freescale-imx/imx-parser/imx-parser.mk
 delete mode 100644 package/libfslparser/libfslparser.hash
 delete mode 100644 package/libfslparser/libfslparser.mk

Config.in.legacy +7 -0

index e5a568d..12426c9 100644
@@ -145,6 +145,13 @@ endif
 ###############################################################################
 comment "Legacy options removed in 2016.08"
 
+config BR2_PACKAGE_LIBFSLPARSER
+	bool "libfslparser has been renamed to imx-parser"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_PARSER
+	help
+	  The libfslparser has been renamed to match the renamed package.
+
 config BR2_PACKAGE_LIBFSLCODEC
 	bool "libfslcodec has been renamed to imx-codec"
 	select BR2_LEGACY

package/Config.in +0 -1

index 1cdbaa0..6760af0 100644
@@ -1098,7 +1098,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslparser/Config.in"
 	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"
 	source "package/libimxvpuapi/Config.in"

package/freescale-imx/Config.in +1 -0

index 052de5a..868e51c 100644
@@ -52,6 +52,7 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 source "package/freescale-imx/imx-codec/Config.in"
 source "package/freescale-imx/imx-kobs/Config.in"
 source "package/freescale-imx/imx-lib/Config.in"
+source "package/freescale-imx/imx-parser/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
 source "package/freescale-imx/imx-vpu/Config.in"
 source "package/freescale-imx/firmware-imx/Config.in"

package/freescale-imx/imx-parser/Config.in new file +8 -0

new file mode 100644
index 0000000..526f066
@@ -0,0 +1,8 @@
+config BR2_PACKAGE_IMX_PARSER
+	bool "imx-parser"
+	depends on BR2_arm # Only relevant for i.MX
+	help
+	  Binary parser libraries for the NXP i.MX SoCs.
+
+	  This library is provided by NXP as-is and doesn't have
+	  an upstream.

package/freescale-imx/imx-parser/imx-parser.hash new file +2 -0

new file mode 100644
index 0000000..2193c66
@@ -0,0 +1,2 @@
+# locally computed
+sha256	1bec0101843db2e95d61e3dbe9e35172d874e54be19cc5d8bf72a735d5e3236f  imx-parser-4.0.9.bin

package/freescale-imx/imx-parser/imx-parser.mk new file +23 -0

new file mode 100644
index 0000000..a207d6e
@@ -0,0 +1,23 @@
+################################################################################
+#
+# imx-parser
+#
+################################################################################
+
+IMX_PARSER_VERSION = 4.0.9
+IMX_PARSER_SITE = $(FREESCALE_IMX_SITE)
+IMX_PARSER_SOURCE = imx-parser-$(IMX_PARSER_VERSION).bin
+IMX_PARSER_INSTALL_STAGING = YES
+
+IMX_PARSER_LICENSE = NXP Semiconductor Software License Agreement
+IMX_PARSER_LICENSE_FILES = EULA COPYING
+IMX_PARSER_REDISTRIBUTE = NO
+
+define IMX_PARSER_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_PARSER_SOURCE))
+endef
+
+# The Makefile installs several versions of the libraries, but we only
+# need one of them, depending on the platform.
+
+$(eval $(autotools-package))

package/gstreamer/gst-fsl-plugins/Config.in +1 -1

index 8c82a4a..38c8a47 100644
@@ -13,7 +13,7 @@ config BR2_PACKAGE_GST_FSL_PLUGINS
 	select BR2_PACKAGE_GST_PLUGINS_BASE
 	select BR2_PACKAGE_LIBFSLVPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 	select BR2_PACKAGE_IMX_LIB
-	select BR2_PACKAGE_LIBFSLPARSER
+	select BR2_PACKAGE_IMX_PARSER
 	select BR2_PACKAGE_IMX_CODEC
 	select BR2_PACKAGE_FREESCALE_IMX
 	help

package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk +1 -1

index 50732bf..8db8d1b 100644
@@ -16,7 +16,7 @@ GST_FSL_PLUGINS_INSTALL_STAGING = YES
 GST_FSL_PLUGINS_AUTORECONF = YES
 
 GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \
-	imx-lib libfslparser imx-codec
+	imx-lib imx-parser imx-codec
 
 ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y)
 GST_FSL_PLUGINS_DEPENDENCIES += libfslvpuwrap

package/libfslparser/Config.in deleted +0 -8

deleted file mode 100644
index e3ab980..0000000
@@ -1,8 +0,0 @@
-config BR2_PACKAGE_LIBFSLPARSER
-	bool "libfslparser"
-	depends on BR2_arm # Only relevant for i.MX
-	help
-	  Binary parser libraries for the Freescale i.MX SoCs.
-
-	  This library is provided by Freescale as-is and doesn't have
-	  an upstream.

package/libfslparser/libfslparser.hash deleted +0 -2

deleted file mode 100644
index c49345f..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	a04621783c84a9776216caff9563c7f840fddd584b0b2d27738d6ca6d2c77f32  libfslparser-4.0.8.bin

package/libfslparser/libfslparser.mk deleted +0 ~~-23~~

deleted file mode 100644
index 44e6039..0000000
@@ -1,23 +0,0 @@
-################################################################################
-#
-# libfslparser
-#
-################################################################################
-
-LIBFSLPARSER_VERSION = 4.0.8
-LIBFSLPARSER_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLPARSER_SOURCE = libfslparser-$(LIBFSLPARSER_VERSION).bin
-LIBFSLPARSER_INSTALL_STAGING = YES
-
-LIBFSLPARSER_LICENSE = Freescale Semiconductor Software License Agreement
-LIBFSLPARSER_LICENSE_FILES = EULA COPYING
-LIBFSLPARSER_REDISTRIBUTE = NO
-
-define LIBFSLPARSER_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLPARSER_SOURCE))
-endef
-
-# The Makefile installs several versions of the libraries, but we only
-# need one of them, depending on the platform.
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #60 lore raw

PATCH v2 10/14 imx-parser: rename from libfslparser and bump version to 4.0.9

Andrew, All,

Quoted 10 lines · Andrew Webster

Changelog:
--Add aac latm parser in parser_libs, support aac latm in mpeg2 parser
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/e2c6f10
Implicitly tested through gstreamer as the 0.10 plugin relies on it:
 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Thomas Petazzoni

2016-06-15 20:50 UTC #61 lore raw

PATCH v2 10/14 imx-parser: rename from libfslparser and bump version to 4.0.9

Hello,

Quoted 5 lines · Andrew Webster

package/freescale-imx/imx-parser/imx-parser.mk new file +5 -0

+IMX_PARSER_VERSION = 4.0.9
+IMX_PARSER_SITE = $(FREESCALE_IMX_SITE)
+IMX_PARSER_SOURCE = imx-parser-$(IMX_PARSER_VERSION).bin
+IMX_PARSER_INSTALL_STAGING = YES
+

package/freescale-imx/imx-parser/imx-parser.mk line 21

Quoted 10 lines

package/freescale-imx/imx-parser/imx-parser.mk new file +10 -0

+IMX_PARSER_LICENSE = NXP Semiconductor Software License Agreement
+IMX_PARSER_LICENSE_FILES = EULA COPYING
+IMX_PARSER_REDISTRIBUTE = NO
+
+define IMX_PARSER_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_PARSER_SOURCE))
+endef
+
+# The Makefile installs several versions of the libraries, but we only
+# need one of them, depending on the platform.

This package seems to have several variants of the libraries, including some that are enabled/disabled with --enable-fsw / --enable-fhw. It would be good to understand what those variants are, and if they are related to ARM EABIhf or not.

Indeed, it is a bit strange that some parts of the i.MX6 blobs are EABIhf only (imx-gpu-viv), while some other parts don't see to be EABIhf. Or maybe imx-parser has both EABI and EABIhf variants ?

In any case, a clarification would be good. I've nonetheless applied, since it was already something that wasn't clear in the pre-existing libfslparser package.

Thanks!

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Andrew Webster

author

2016-06-10 18:13 UTC #62 lore raw

PATCH v2 11/14 imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65

Changelog:
--Update 'nLastFrameEndPosPhy' in seqinit stage to avoid feed
the same SPS to vpu repeatedly.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/d6b9490

Implicitly tested through gstreamer as the 0.10 plugin relies on it:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm

---
Changes v1 -> v2:
- added test procedure to comment
- moved legacy config under 2016.08 section
- renamed Freescale to NXP in Config.in
- updated references in kodi and gst-fsl-plugins

 Config.in.legacy                                    |  7 +++++++
 package/Config.in                                   |  1 -
 package/freescale-imx/Config.in                     |  1 +
 .../imx-vpuwrap}/Config.in                          |  8 ++++----
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash  |  2 ++
 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk    | 21 +++++++++++++++++++++
 package/gstreamer/gst-fsl-plugins/Config.in         |  2 +-
 .../gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk    |  2 +-
 package/kodi/kodi.mk                                |  4 ++--
 package/libfslvpuwrap/libfslvpuwrap.hash            |  2 --
 package/libfslvpuwrap/libfslvpuwrap.mk              | 21 ---------------------
 11 files changed, 39 insertions(+), 32 deletions(-)
 rename package/{libfslvpuwrap => freescale-imx/imx-vpuwrap}/Config.in (69%)
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash
 create mode 100644 package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.hash
 delete mode 100644 package/libfslvpuwrap/libfslvpuwrap.mk

Config.in.legacy +7 -0

index 12426c9..bb12d7b 100644
@@ -145,6 +145,13 @@ endif
 ###############################################################################
 comment "Legacy options removed in 2016.08"
 
+config BR2_PACKAGE_LIBFSLVPUWRAP
+	bool "libfslvpuwrap has been renamed to imx-vpuwrap"
+	select BR2_LEGACY
+	select BR2_PACKAGE_IMX_VPUWRAP
+	help
+	  The libfslvpuwrap has been renamed to match the renamed package.
+
 config BR2_PACKAGE_LIBFSLPARSER
 	bool "libfslparser has been renamed to imx-parser"
 	select BR2_LEGACY

package/Config.in +0 -1

index 6760af0..617bfde 100644
@@ -1098,7 +1098,6 @@ menu "Multimedia"
 	source "package/libdvdnav/Config.in"
 	source "package/libdvdread/Config.in"
 	source "package/libebml/Config.in"
-	source "package/libfslvpuwrap/Config.in"
 	source "package/libhdhomerun/Config.in"
 	source "package/libimxvpuapi/Config.in"
 	source "package/libmatroska/Config.in"

package/freescale-imx/Config.in +1 -0

index 868e51c..056728f 100644
@@ -55,6 +55,7 @@ source "package/freescale-imx/imx-lib/Config.in"
 source "package/freescale-imx/imx-parser/Config.in"
 source "package/freescale-imx/imx-uuc/Config.in"
 source "package/freescale-imx/imx-vpu/Config.in"
+source "package/freescale-imx/imx-vpuwrap/Config.in"
 source "package/freescale-imx/firmware-imx/Config.in"
 if (BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX51 || BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53)
 source "package/freescale-imx/gpu-amd-bin-mx51/Config.in"

package/freescale-imx/imx-vpuwrap/Config.in renamed from package/libfslvpuwrap/Config.in +5 -5

similarity index 59%
rename from package/libfslvpuwrap/Config.in
rename to package/freescale-imx/imx-vpuwrap/Config.in
index 72f7126..3509595 100644
@@ -1,13 +1,13 @@
-comment "libfslvpuwrap needs an imx-specific Linux kernel to be built"
+comment "imx-vpuwrap needs an imx-specific Linux kernel to be built"
 	depends on BR2_arm
 	depends on !BR2_LINUX_KERNEL
 
-comment "libfslvpuwrap needs an i.MX platform with VPU support"
+comment "imx-vpuwrap needs an i.MX platform with VPU support"
 	depends on BR2_arm
 	depends on BR2_LINUX_KERNEL && !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 
-config BR2_PACKAGE_LIBFSLVPUWRAP
-	bool "libfslvpuwrap"
+config BR2_PACKAGE_IMX_VPUWRAP
+	bool "imx-vpuwrap"
 	depends on BR2_LINUX_KERNEL
 	depends on BR2_arm # Only relevant for i.MX
 	depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU

@@ -15,5 +15,5 @@ config BR2_PACKAGE_LIBFSLVPUWRAP
 	help
 	  Wrapper library for the vpu library, giving it a different API.
 
-	  This library is provided by Freescale as-is and doesn't have
+	  This library is provided by NXP as-is and doesn't have
 	  an upstream.

package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash new file +2 -0

new file mode 100644
index 0000000..42b97a1
@@ -0,0 +1,2 @@
+# locally computed
+sha256	ddeacb7d9cb26c8864374df6d3f090aad47cd2ae409c2b9c817b773447252d61  imx-vpuwrap-1.0.65.bin

package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk new file +21 -0

new file mode 100644
index 0000000..da0e32c
@@ -0,0 +1,21 @@
+################################################################################
+#
+# imx-vpuwrap
+#
+################################################################################
+
+IMX_VPUWRAP_VERSION = 1.0.65
+IMX_VPUWRAP_SITE = $(FREESCALE_IMX_SITE)
+IMX_VPUWRAP_SOURCE = imx-vpuwrap-$(IMX_VPUWRAP_VERSION).bin
+IMX_VPUWRAP_DEPENDENCIES = imx-vpu
+IMX_VPUWRAP_INSTALL_STAGING = YES
+
+IMX_VPUWRAP_LICENSE = NXP Semiconductor Software License Agreement
+IMX_VPUWRAP_LICENSE_FILES = EULA COPYING
+IMX_VPUWRAP_REDISTRIBUTE = NO
+
+define IMX_VPUWRAP_EXTRACT_CMDS
+	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(IMX_VPUWRAP_SOURCE))
+endef
+
+$(eval $(autotools-package))

package/gstreamer/gst-fsl-plugins/Config.in +1 -1

index 38c8a47..f34a067 100644
@@ -11,7 +11,7 @@ config BR2_PACKAGE_GST_FSL_PLUGINS
 	depends on BR2_arm # Only relevant for i.MX
 	depends on BR2_TOOLCHAIN_USES_GLIBC # imx-codec
 	select BR2_PACKAGE_GST_PLUGINS_BASE
-	select BR2_PACKAGE_LIBFSLVPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
+	select BR2_PACKAGE_IMX_VPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
 	select BR2_PACKAGE_IMX_LIB
 	select BR2_PACKAGE_IMX_PARSER
 	select BR2_PACKAGE_IMX_CODEC

package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk +1 -1

index 8db8d1b..65993a2 100644
@@ -19,7 +19,7 @@ GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \
 	imx-lib imx-parser imx-codec
 
 ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y)
-GST_FSL_PLUGINS_DEPENDENCIES += libfslvpuwrap
+GST_FSL_PLUGINS_DEPENDENCIES += imx-vpuwrap
 endif
 
 GST_FSL_PLUGINS_CONF_ENV = \

package/kodi/kodi.mk +2 -2

index e163c24..90c50c3 100644
@@ -53,8 +53,8 @@ KODI_CONF_ENV += INCLUDES="-I$(STAGING_DIR)/usr/include/interface/vcos/pthreads
 	LIBS="-lvcos -lvchostif"
 endif
 
-ifeq ($(BR2_PACKAGE_LIBFSLVPUWRAP),y)
-KODI_DEPENDENCIES += libfslvpuwrap
+ifeq ($(BR2_PACKAGE_IMX_VPUWRAP),y)
+KODI_DEPENDENCIES += imx-vpuwrap
 KODI_CONF_OPTS += --enable-codec=imxvpu
 endif

package/libfslvpuwrap/libfslvpuwrap.hash deleted +0 -2

deleted file mode 100644
index e987077..0000000
@@ -1,2 +0,0 @@
-# locally computed
-sha256	0717faccb5413dc95ce9ad919400095ecf722320478c924366793a6cc450ae16  libfslvpuwrap-1.0.62.bin

package/libfslvpuwrap/libfslvpuwrap.mk deleted +0 ~~-21~~

deleted file mode 100644
index 12a4e86..0000000
@@ -1,21 +0,0 @@
-################################################################################
-#
-# libfslvpuwrap
-#
-################################################################################
-
-LIBFSLVPUWRAP_VERSION = 1.0.62
-LIBFSLVPUWRAP_SITE = $(FREESCALE_IMX_SITE)
-LIBFSLVPUWRAP_SOURCE = libfslvpuwrap-$(LIBFSLVPUWRAP_VERSION).bin
-LIBFSLVPUWRAP_DEPENDENCIES = imx-vpu
-LIBFSLVPUWRAP_INSTALL_STAGING = YES
-
-LIBFSLVPUWRAP_LICENSE = Freescale Semiconductor Software License Agreement
-LIBFSLVPUWRAP_LICENSE_FILES = EULA COPYING
-LIBFSLVPUWRAP_REDISTRIBUTE = NO
-
-define LIBFSLVPUWRAP_EXTRACT_CMDS
-	$(call FREESCALE_IMX_EXTRACT_HELPER,$(DL_DIR)/$(LIBFSLVPUWRAP_SOURCE))
-endef
-
-$(eval $(autotools-package))

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #63 lore raw

PATCH v2 11/14 imx-vpuwrap: rename from libfslvpuwrap and bump version to 1.0.65

Andrew, All,

Quoted 11 lines · Andrew Webster

Changelog:
--Update 'nLastFrameEndPosPhy' in seqinit stage to avoid feed
the same SPS to vpu repeatedly.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/d6b9490
Implicitly tested through gstreamer as the 0.10 plugin relies on it:
 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:13 UTC #64 lore raw

PATCH v2 12/14 libimxvpuapi: bump version to 0.10.2

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c999668

Implicitly tested through gstreamer as the plugin relies on it:

 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm

---
Changes v1 -> v2:
- added test procedure to comment

 package/libimxvpuapi/libimxvpuapi.hash | 2 +-
 package/libimxvpuapi/libimxvpuapi.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/libimxvpuapi/libimxvpuapi.hash +1 -1

index fd90a20..c8cabcf 100644
@@ -1,2 +1,2 @@
 # locally computed hash
-sha256 8d7aeed88c06fda44318cef9565ae47b86461bd309908b1103a7974ed0822a8d  libimxvpuapi-0.10.1.tar.gz
+sha256 8e6dc4105db6164c0de72fd60eb73f4ba1632f7ba5052b1abc834b462b566da2  libimxvpuapi-0.10.2.tar.gz

package/libimxvpuapi/libimxvpuapi.mk +1 -1

index e0d8cc9..df84806 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-LIBIMXVPUAPI_VERSION = 0.10.1
+LIBIMXVPUAPI_VERSION = 0.10.2
 LIBIMXVPUAPI_SITE = $(call github,Freescale,libimxvpuapi,$(LIBIMXVPUAPI_VERSION))
 LIBIMXVPUAPI_LICENSE = LGPLv2.1+
 LIBIMXVPUAPI_LICENSE_FILES = LICENSE

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #65 lore raw

PATCH v2 12/14 libimxvpuapi: bump version to 0.10.2

Andrew, All,

Quoted 9 lines · Andrew Webster

Changelog:
- fix AUD NAL positioning in h.264 encoder output
(SPS/PPS/AUD were ordered incorrectly - AUD has to come first, not last)
- fix build error with examples when --enable-static is used
- pass quality factor in simplified JPEG encoder interface correctly
- add functions for querying and setting header data
useful for modifying headers, like VUI data in the SPS RBSP
- documentation updates

Quoted 7 lines

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/c999668
Implicitly tested through gstreamer as the plugin relies on it:
 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:13 UTC #66 lore raw

PATCH v2 13/14 imx-lib: bump version to 5.4

Changelog:
- Remove platform check in hdmi-cec and imx-lib - simplify to support builds on multiple machines
- For ipu use the dummy defaul object to fix build issues on non-ipu hardware

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/7946533

Implicitly tested through gstreamer as the
plugins rely on them for vpu decoding, e.g.:

 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink

---
Changes v1 -> v2:
- added test procedure to comment

 package/freescale-imx/imx-lib/imx-lib.hash | 2 +-
 package/freescale-imx/imx-lib/imx-lib.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

package/freescale-imx/imx-lib/imx-lib.hash +1 -1

index 8e4bd73..76dcb23 100644
@@ -1,2 +1,2 @@
 # locally computed
-sha256 2b4547ffb1de328cf8997a09a423c6f015d808493efd11a98a5d42c6497e3445  imx-lib-5.1.tar.gz
+sha256 8d12c4a0a44d29569f1ead3a36adc28402b9c4036b53e11dd3ebf2295e3ce1d0  imx-lib-5.4.tar.gz

package/freescale-imx/imx-lib/imx-lib.mk +1 -1

index b49357b..d3603ca 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-IMX_LIB_VERSION = 5.1
+IMX_LIB_VERSION = 5.4
 IMX_LIB_SITE = $(FREESCALE_IMX_SITE)
 IMX_LIB_LICENSE = LGPLv2.1+
 IMX_LIB_LICENSE_FILES = COPYING-LGPL-2.1

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #67 lore raw

PATCH v2 13/14 imx-lib: bump version to 5.4

Andrew, All,

Quoted 8 lines · Andrew Webster

Changelog:
- Remove platform check in hdmi-cec and imx-lib - simplify to support builds on multiple machines
- For ipu use the dummy defaul object to fix build issues on non-ipu hardware
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/7946533

Quoted 8 lines

Implicitly tested through gstreamer as the
plugins rely on them for vpu decoding, e.g.:
 # gst-launch-0.10 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 playbin uri=file:///root/tears_of_steel_1080p.webm
 # gst-launch-1.0 filesrc location=/root/tears_of_steel_1080p.webm ! \
   matroskademux ! imxvpudec ! imxipuvideosink
Signed-off-by Andrew Webster <[email protected]>

Regards,
Gary

Andrew Webster

author

2016-06-10 18:14 UTC #68 lore raw

PATCH v2 14/14 imx-kobs: bump version to 5.4

Changelog:
- Check raw mode flag for backward compatbility
- Fix membership name in bch geometry structure.
- Fix the previous implementation that may ruin the BBM
- Adapt raw page accesses to match the new raw_read/write implementation

The old raw access implementation (in GPMI driver) was considering that data and OOB data were separated in their respective regions (the data area and the OOB area of the page), which is not true.
They are actually interleaved this way:

METADATA + ((DATA + ECCBYTES) * N)

The new raw access implementation (in the GPMI driver) is hiding this weird
layout to MTD users by exposing a more common layout:

DATA + METADATA + (N * ECCBYTES)

Here METADATA + (N * ECCBYTES) are exposed as if they were stored in the OOB area.

Unfortunately kobs-ng rely on this weird layout when accessing the NAND in raw mode.

This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/a7545b1

---
Changes v1 -> v2:
- added to patch set

 package/freescale-imx/imx-kobs/imx-kobs.hash | 3 +--
 package/freescale-imx/imx-kobs/imx-kobs.mk   | 2 +-
 2 files changed, 2 insertions(+), 3 deletions(-)

package/freescale-imx/imx-kobs/imx-kobs.hash +1 -2

index 09d42ed..b942bd3 100644
@@ -1,2 +1 @@
-# locally computed
-sha256	45f729fc2b49556f1ca9df778f52bf5cc749cfe53664c8206daab29991c5f6c1  imx-kobs-5.3.tar.gz
+sha256	85171b46068ac47c42fedb8104167bf9afd33dd9527ed127e1ca2eb29d7a86bf  imx-kobs-5.4.tar.gz

package/freescale-imx/imx-kobs/imx-kobs.mk +1 -1

index cc4721f..b80d5bd 100644
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-IMX_KOBS_VERSION = 5.3
+IMX_KOBS_VERSION = 5.4
 IMX_KOBS_SITE = $(FREESCALE_IMX_SITE)
 IMX_KOBS_LICENSE = GPLv2+
 IMX_KOBS_LICENSE_FILES = COPYING

Signature · 1 line

1.9.1

Gary Bisson

2016-06-13 13:21 UTC #69 lore raw

PATCH v2 14/14 imx-kobs: bump version to 5.4

Andrew, All,

Quoted 22 lines · Andrew Webster

Changelog:
- Check raw mode flag for backward compatbility
- Fix membership name in bch geometry structure.
- Fix the previous implementation that may ruin the BBM
- Adapt raw page accesses to match the new raw_read/write implementation
The old raw access implementation (in GPMI driver) was considering that data and OOB data were separated in their respective regions (the data area and the OOB area of the page), which is not true.
They are actually interleaved this way:
METADATA + ((DATA + ECCBYTES) * N)
The new raw access implementation (in the GPMI driver) is hiding this weird
layout to MTD users by exposing a more common layout:
DATA + METADATA + (N * ECCBYTES)
Here METADATA + (N * ECCBYTES) are exposed as if they were stored in the OOB area.

Quoted 7 lines

Unfortunately kobs-ng rely on this weird layout when accessing the NAND in raw mode.
This patch is based on the Yocto equivalent:
https://github.com/Freescale/meta-fsl-arm/commit/a7545b1
Signed-off-by Andrew Webster <[email protected]>

Cannot test it since I don't own a platform with NAND but the build is fine.

Regards,
Gary

Thomas Petazzoni

2016-06-11 13:20 UTC #70 lore raw

v2 cover Bump NXP package versions

Hello,

Quoted 6 lines · Andrew Webster

Changes v1 -> v2:
- Reordered patches to keep gpu related patches together and multimedia patches together.
- Added patches for bumping the kernel/u-boot versions of NXP defconfigs
- Added patch to bump version of imx-kobs
- Added patch to bump version of kernel-module-imx-gpu-viv

Thanks Andrew for this new version! I'll wait for Gary's feedback on the series.

Thanks,

Thomas

Signature · 3 lines

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Gary Bisson

2016-06-13 13:18 UTC #71 lore raw

v2 cover Bump NXP package versions

Thomas, Andrew, All,

Quoted 11 lines · Thomas Petazzoni

Hello,
Quoted 6 lines · Andrew Webster
Changes v1 -> v2:
- Reordered patches to keep gpu related patches together and multimedia patches together.
- Added patches for bumping the kernel/u-boot versions of NXP defconfigs
- Added patch to bump version of imx-kobs
- Added patch to bump version of kernel-module-imx-gpu-viv

Quoted 2 lines

Thanks Andrew for this new version! I'll wait for Gary's feedback on the series.

I'm going to reply to each patch but overall I'm very happy with the series.

I've tested most of them using a Nitrogen6x + 3.14.52 kernel.

Thanks,
Gary

Thomas Petazzoni

2016-06-15 20:44 UTC #72 lore raw

v2 cover Bump NXP package versions