imx: update multimedia packages to 4.9.88_2.0.0_ga
cover imx: update multimedia packages to 4.9.88_2.0.0_ga
Hi,
This series updates the i.MX multimedia packages to be in line with the NXP 4.9.88-2.0.0_ga release [1].
The main benefit of this release is that it works for both 32-bit i.MX6/7 processors as well as 64-bit i.MX8MQ.
That means that new packages had to be added to support the i.MX8MQ VPU IP (from Hantro).
In Yocto, the imx-vpu package for the Chips&Media VPU present on i.MX6 wasn't renamed. But packages such as imx-vpuwrap now depend on virtual/imxvpu which can either be provided by imx-vpu (Chips&Media) or imx-vpu-hantro (Hantro) [2][3].
As I thought it'd be confusing to have a virtual imxvpu virtual package with an imx-vpu provider, I renamed imx-vpu to imx-vpu-cnm and added an imx-vpu virtual package.
Each patch for this imx-vpu change modifies several packages (those depending on it) so that it doesn't break the build in case someone bisects the tree.
Finally, I've verified those packages on both i.MX6Q and i.MX8MQ. But in the case of i.MX8MQ, the VPU testing was limited to the package unit tests since the open-source imx plugin doesn't support Hantro VPU (yet). Note that I don't plan on integrating NXP plugin since it implies changing all the standard plugins to be "NXP-compliant" [4].
Let me know if you are ok with this approach. As usual, comments are welcome.
Regards,
Gary
[1] https://community.nxp.com/docs/DOC-340805
[2] https://github.com/Freescale/meta-freescale/blob/master/recipes-bsp/imx-vpu/imx-vpu_5.4.38.bb#L11
[3] https://github.com/Freescale/meta-freescale/blob/master/recipes-bsp/imx-vpu-hantro/imx-vpu-hantro_1.6.0.bb#L9
[4] https://github.com/Freescale/meta-freescale/tree/master/recipes-multimedia/gstreamer
Gary Bisson (8):
firmware-imx: bump to version 7.5
imx-vpu: rename package to imx-vpu-cnm
imx-vpu-cnm: bump version to 5.4.38
imx-vpu-hantro: new package
imx-vpu: new virtual package
imx-vpuwrap: bump version to 4.3.5
imx-codec: bump version to 4.3.5
imx-parser: bump version to 4.3.5
package/freescale-imx/Config.in | 5 ++- .../firmware-imx/firmware-imx.hash | 2 +- .../firmware-imx/firmware-imx.mk | 2 +- .../freescale-imx/imx-codec/imx-codec.hash | 3 +- package/freescale-imx/imx-codec/imx-codec.mk | 4 +- .../freescale-imx/imx-parser/imx-parser.hash | 3 +- .../freescale-imx/imx-parser/imx-parser.mk | 2 +- package/freescale-imx/imx-vpu-cnm/Config.in | 25 +++++++++++ .../imx-vpu-cnm/imx-vpu-cnm.hash | 3 ++ .../freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk | 39 ++++++++++++++++ ...on.h-header-inclusion-to-be-standard.patch | 44 +++++++++++++++++++ .../freescale-imx/imx-vpu-hantro/Config.in | 23 ++++++++++ .../imx-vpu-hantro/imx-vpu-hantro.hash | 2 + .../imx-vpu-hantro/imx-vpu-hantro.mk | 42 ++++++++++++++++++ package/freescale-imx/imx-vpu/Config.in | 21 +++------ package/freescale-imx/imx-vpu/imx-vpu.hash | 2 - package/freescale-imx/imx-vpu/imx-vpu.mk | 34 +------------- package/freescale-imx/imx-vpuwrap/Config.in | 11 +++-- .../imx-vpuwrap/imx-vpuwrap.hash | 3 +- .../freescale-imx/imx-vpuwrap/imx-vpuwrap.mk | 2 +- package/gstreamer/gst-fsl-plugins/Config.in | 2 +- .../gst-fsl-plugins/gst-fsl-plugins.mk | 2 +- package/libimxvpuapi/Config.in | 7 ++- 23 files changed, 210 insertions(+), 73 deletions(-) create mode 100644 package/freescale-imx/imx-vpu-cnm/Config.in create mode 100644 package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.hash create mode 100644 package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk create mode 100644 package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch create mode 100644 package/freescale-imx/imx-vpu-hantro/Config.in create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk delete mode 100644 package/freescale-imx/imx-vpu/imx-vpu.hash
Signature · 1 line
2.18.0
PATCH 1/8 firmware-imx: bump to version 7.5
This new package includes new binaries for i.MX8QXP.
--- package/freescale-imx/firmware-imx/firmware-imx.hash | 2 +- package/freescale-imx/firmware-imx/firmware-imx.mk | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
package/freescale-imx/firmware-imx/firmware-imx.hash +1 -1
@@ -1,4 +1,4 @@ # Locally calculated -sha256 8a8ec439d014e928f42c49bab1c382914ec5b4f1041df28e5d687eb194bd07ab firmware-imx-7.4.bin +sha256 a8f099bdf786b2da1e8b43094950c033ccdbf93f1b8a93caffb912e1500cd735 firmware-imx-7.5.bin sha256 faf01d10e484879247963eb97d96622a980232e22a35e487dfe53b13708b686a EULA sha256 0f34f6175247762e2e1c38319aadf657a53f00ce124e569dfc61b30451549e7a COPYING
package/freescale-imx/firmware-imx/firmware-imx.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -FIRMWARE_IMX_VERSION = 7.4 +FIRMWARE_IMX_VERSION = 7.5 FIRMWARE_IMX_SITE = $(FREESCALE_IMX_SITE) FIRMWARE_IMX_SOURCE = firmware-imx-$(FIRMWARE_IMX_VERSION).bin
Signature · 1 line
2.18.0
PATCH 1/8 firmware-imx: bump to version 7.5
Quoted 3 lines · Gary Bisson
This new package includes new binaries for i.MX8QXP.
Signed-off-by Gary Bisson <[email protected]>
I've applied this patch to master, the rest is changes requested.
Regards,
Arnout
Quoted 29 lines
--- package/freescale-imx/firmware-imx/firmware-imx.hash | 2 +- package/freescale-imx/firmware-imx/firmware-imx.mk | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)package/freescale-imx/firmware-imx/firmware-imx.hash +1
-1@@ -1,4 +1,4 @@ # Locally calculated -sha256 8a8ec439d014e928f42c49bab1c382914ec5b4f1041df28e5d687eb194bd07ab firmware-imx-7.4.bin +sha256 a8f099bdf786b2da1e8b43094950c033ccdbf93f1b8a93caffb912e1500cd735 firmware-imx-7.5.bin sha256 faf01d10e484879247963eb97d96622a980232e22a35e487dfe53b13708b686a EULA sha256 0f34f6175247762e2e1c38319aadf657a53f00ce124e569dfc61b30451549e7a COPYINGpackage/freescale-imx/firmware-imx/firmware-imx.mk +1
-1@@ -4,7 +4,7 @@ # ################################################################################ -FIRMWARE_IMX_VERSION = 7.4 +FIRMWARE_IMX_VERSION = 7.5 FIRMWARE_IMX_SITE = $(FREESCALE_IMX_SITE) FIRMWARE_IMX_SOURCE = firmware-imx-$(FIRMWARE_IMX_VERSION).bin
Signature · 6 lines
Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
PATCH 1/8 firmware-imx: bump to version 7.5
Hello,
Quoted 6 lines · Arnout Vandecappelle
Quoted 3 lines · Gary Bisson
This new package includes new binaries for i.MX8QXP.
Signed-off-by Gary Bisson <[email protected]>I've applied this patch to master, the rest is changes requested.
I don't know if we have been clear enough with Gary about what changes we want. Do we want to stay with a virtual package ? If so, what should be the name of the virtual package vs. the name of the package for the "old" i.MX VPUs ?
On my side, while I agree that imx-vpu-cnm violates the rule of "we name packages like their upstream name", I believe this is a violation that is acceptable because the naming chosen by Gary is the one that is the easiest to understand and the most obvious: imx-vpu is the virtual package, imx-vpu-hantro and imx-vpu-cnm are the providers.
Best regards,
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering https://bootlin.com
PATCH 1/8 firmware-imx: bump to version 7.5
Hi Thomas, Arnout,
Quoted 4 lines · Thomas Petazzoni
Hello,
Quoted 1 line · Arnout Vandecappelle
On 25-07-18 17:01, Gary Bisson wrote:
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
This new package includes new binaries for i.MX8QXP.
Signed-off-by Gary Bisson <[email protected]>I've applied this patch to master, the rest is changes requested.
I don't know if we have been clear enough with Gary about what changes we want. Do we want to stay with a virtual package ?
That's a good question. But I plan on offering the alternative series without the virtual package, maybe that will help making the decision. It shouldn't be too hard to do.
The way I see it using a virtual package is cleaner and is more future-proof (I wouldn't be surprised another imx-vpu-xxx package shows up in a near future).
However it also makes it "harder" to select imx-vpuwrap/libimxvpuapi since it depends on imx-vpu, the provider package isn't automatically selected as it used to.
Quoted 9 lines
If so, what should
be the name of the virtual package vs. the name of the package for the "old" i.MX VPUs ?On my side, while I agree that imx-vpu-cnm violates the rule of "we name packages like their upstream name", I believe this is a violation that is acceptable because the naming chosen by Gary is the one that is the easiest to understand and the most obvious: imx-vpu is the virtual package, imx-vpu-hantro and imx-vpu-cnm are the providers.
Glad to hear someone agrees with that naming which really makes more sense to me than what is done by NXP.
Regards,
Gary
PATCH 2/8 imx-vpu: rename package to imx-vpu-cnm
In reply to Gary BissonSo far the i.MX processors have been using a Chips&Media CODA VPU. But the latest generation of processors (starting with i.MX8MQ) uses VPU cores from Hantro which is provided as imx-vpu-hantro package.
NXP chose in Yocto to keep the imx-vpu naming for Chips&Media package and created a virtual "imxvpu" package that can either be provided by imx-vpu or imx-vpu-hantro.
Since this naming can be confusing (IMO), renaming the imx-vpu package to imx-vpu-cnm and then create a virtual imx-vpu one should be clearer.
---
package/freescale-imx/Config.in | 2 +-
package/freescale-imx/imx-codec/imx-codec.mk | 2 +-
.../{imx-vpu => imx-vpu-cnm}/Config.in | 12 +++---
.../imx-vpu-cnm.hash} | 0
.../freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk | 39 +++++++++++++++++++
package/freescale-imx/imx-vpu/imx-vpu.mk | 39 -------------------
package/freescale-imx/imx-vpuwrap/Config.in | 2 +-
.../freescale-imx/imx-vpuwrap/imx-vpuwrap.mk | 2 +-
package/libimxvpuapi/Config.in | 2 +-
package/libimxvpuapi/libimxvpuapi.mk | 2 +-
10 files changed, 51 insertions(+), 51 deletions(-)
rename package/freescale-imx/{imx-vpu => imx-vpu-cnm}/Config.in (54%)
rename package/freescale-imx/{imx-vpu/imx-vpu.hash => imx-vpu-cnm/imx-vpu-cnm.hash} (100%)
create mode 100644 package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk
delete mode 100644 package/freescale-imx/imx-vpu/imx-vpu.mkpackage/freescale-imx/Config.in +1 -1
@@ -79,7 +79,7 @@ source "package/freescale-imx/imx-lib/Config.in" source "package/freescale-imx/imx-m4fwloader/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-vpu-cnm/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)
package/freescale-imx/imx-codec/imx-codec.mk +1 -1
@@ -21,7 +21,7 @@ ifeq ($(BR2_ARM_EABIHF),y) IMX_CODEC_CONF_OPTS += --enable-fhw endif -ifeq ($(BR2_PACKAGE_IMX_VPU),y) +ifeq ($(BR2_PACKAGE_IMX_VPU_CNM),y) IMX_CODEC_CONF_OPTS += --enable-vpu endif
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +6 -6
@@ -1,17 +1,17 @@ -comment "imx-vpu needs an i.MX platform with VPU support" +comment "imx-vpu-cnm needs an i.MX platform with VPU support" depends on BR2_arm depends on !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU -config BR2_PACKAGE_IMX_VPU - bool "imx-vpu" +config BR2_PACKAGE_IMX_VPU_CNM + bool "imx-vpu-cnm" depends on BR2_arm # Only relevant for i.MX depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU select BR2_PACKAGE_FIRMWARE_IMX help Library of userspace helpers specific for the Freescale i.MX - platform. It wraps the kernel interfaces for the i.MX platform - Video Processing Unit (VPU) driver. It requires a kernel that - includes the i.MX specific headers to be built. + platform integrating a Chips&Media CODA Video Processing Unit + (VPU). It requires a kernel that includes the i.MX specific + headers to be built. This library is provided by Freescale as-is and doesn't have an upstream.
package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.hash renamed from package/freescale-imx/imx-vpu/imx-vpu.hash +0 -0
package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk new file +39 -0
@@ -0,0 +1,39 @@ +################################################################################ +# +# imx-vpu-cnm +# +################################################################################ + +IMX_VPU_CNM_VERSION = 5.4.37 +IMX_VPU_CNM_SITE = $(FREESCALE_IMX_SITE) +IMX_VPU_CNM_SOURCE = imx-vpu-$(IMX_VPU_CNM_VERSION).bin + +IMX_VPU_CNM_INSTALL_STAGING = YES + +IMX_VPU_CNM_MAKE_ENV = \ + $(TARGET_MAKE_ENV) \ + $(TARGET_CONFIGURE_OPTS) \ + CROSS_COMPILE="$(TARGET_CROSS)" \ + PLATFORM=$(BR2_PACKAGE_FREESCALE_IMX_PLATFORM) + +IMX_VPU_CNM_LICENSE = NXP Semiconductor Software License Agreement +IMX_VPU_CNM_LICENSE_FILES = EULA COPYING +IMX_VPU_CNM_REDISTRIBUTE = NO + +define IMX_VPU_CNM_EXTRACT_CMDS + $(call FREESCALE_IMX_EXTRACT_HELPER,$(IMX_VPU_CNM_DL_DIR)/$(IMX_VPU_CNM_SOURCE)) +endef + +define IMX_VPU_CNM_BUILD_CMDS + $(IMX_VPU_CNM_MAKE_ENV) $(MAKE1) -C $(@D) +endef + +define IMX_VPU_CNM_INSTALL_STAGING_CMDS + $(IMX_VPU_CNM_MAKE_ENV) $(MAKE1) -C $(@D) DEST_DIR=$(STAGING_DIR) install +endef + +define IMX_VPU_CNM_INSTALL_TARGET_CMDS + $(IMX_VPU_CNM_MAKE_ENV) $(MAKE1) -C $(@D) DEST_DIR=$(TARGET_DIR) install +endef + +$(eval $(generic-package))
package/freescale-imx/imx-vpu/imx-vpu.mk deleted +0 -39
@@ -1,39 +0,0 @@ -################################################################################ -# -# imx-vpu -# -################################################################################ - -IMX_VPU_VERSION = 5.4.37 -IMX_VPU_SITE = $(FREESCALE_IMX_SITE) -IMX_VPU_SOURCE = imx-vpu-$(IMX_VPU_VERSION).bin - -IMX_VPU_INSTALL_STAGING = YES - -IMX_VPU_MAKE_ENV = \ - $(TARGET_MAKE_ENV) \ - $(TARGET_CONFIGURE_OPTS) \ - CROSS_COMPILE="$(TARGET_CROSS)" \ - PLATFORM=$(BR2_PACKAGE_FREESCALE_IMX_PLATFORM) - -IMX_VPU_LICENSE = NXP Semiconductor Software License Agreement -IMX_VPU_LICENSE_FILES = EULA COPYING -IMX_VPU_REDISTRIBUTE = NO - -define IMX_VPU_EXTRACT_CMDS - $(call FREESCALE_IMX_EXTRACT_HELPER,$(IMX_VPU_DL_DIR)/$(IMX_VPU_SOURCE)) -endef - -define IMX_VPU_BUILD_CMDS - $(IMX_VPU_MAKE_ENV) $(MAKE1) -C $(@D) -endef - -define IMX_VPU_INSTALL_STAGING_CMDS - $(IMX_VPU_MAKE_ENV) $(MAKE1) -C $(@D) DEST_DIR=$(STAGING_DIR) install -endef - -define IMX_VPU_INSTALL_TARGET_CMDS - $(IMX_VPU_MAKE_ENV) $(MAKE1) -C $(@D) DEST_DIR=$(TARGET_DIR) install -endef - -$(eval $(generic-package))
package/freescale-imx/imx-vpuwrap/Config.in +1 -1
@@ -11,7 +11,7 @@ config BR2_PACKAGE_IMX_VPUWRAP depends on BR2_LINUX_KERNEL depends on BR2_arm # Only relevant for i.MX depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU - select BR2_PACKAGE_IMX_VPU + select BR2_PACKAGE_IMX_VPU_CNM help Wrapper library for the vpu library, giving it a different API.
package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk +1 -1
@@ -7,7 +7,7 @@ IMX_VPUWRAP_VERSION = 1.0.68 IMX_VPUWRAP_SITE = $(FREESCALE_IMX_SITE) IMX_VPUWRAP_SOURCE = imx-vpuwrap-$(IMX_VPUWRAP_VERSION).bin -IMX_VPUWRAP_DEPENDENCIES = imx-vpu +IMX_VPUWRAP_DEPENDENCIES = imx-vpu-cnm IMX_VPUWRAP_INSTALL_STAGING = YES IMX_VPUWRAP_LICENSE = NXP Semiconductor Software License Agreement
package/libimxvpuapi/Config.in +1 -1
@@ -6,7 +6,7 @@ config BR2_PACKAGE_LIBIMXVPUAPI bool "libimxvpuapi" depends on BR2_arm # Only relevant for i.MX depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU - select BR2_PACKAGE_IMX_VPU + select BR2_PACKAGE_IMX_VPU_CNM help This library provides an API for using the iMX6 VPU video engine. It is an alternative to Freescale's VPU
package/libimxvpuapi/libimxvpuapi.mk +1 -1
@@ -8,7 +8,7 @@ LIBIMXVPUAPI_VERSION = 0.10.3 LIBIMXVPUAPI_SITE = $(call github,Freescale,libimxvpuapi,$(LIBIMXVPUAPI_VERSION)) LIBIMXVPUAPI_LICENSE = LGPL-2.1+ LIBIMXVPUAPI_LICENSE_FILES = LICENSE -LIBIMXVPUAPI_DEPENDENCIES = host-pkgconf host-python imx-vpu +LIBIMXVPUAPI_DEPENDENCIES = host-pkgconf host-python imx-vpu-cnm LIBIMXVPUAPI_INSTALL_STAGING = YES $(eval $(waf-package))
Signature · 1 line
2.18.0
PATCH 2/8 imx-vpu: rename package to imx-vpu-cnm
Hi Gary,
Quoted 10 lines · Gary Bisson
So far the i.MX processors have been using a Chips&Media CODA VPU. But the latest generation of processors (starting with i.MX8MQ) uses VPU cores from Hantro which is provided as imx-vpu-hantro package.
NXP chose in Yocto to keep the imx-vpu naming for Chips&Media package and created a virtual "imxvpu" package that can either be provided by imx-vpu or imx-vpu-hantro.
Since this naming can be confusing (IMO), renaming the imx-vpu package to imx-vpu-cnm and then create a virtual imx-vpu one should be clearer.
What about Config.in.legacy?
[..]
Quoted 5 lines
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +2
-2-config BR2_PACKAGE_IMX_VPU - bool "imx-vpu" +config BR2_PACKAGE_IMX_VPU_CNM + bool "imx-vpu-cnm" depends on BR2_arm # Only relevant for i.MX
package/freescale-imx/imx-vpu-cnm/Config.in line 14
Quoted 10 lines
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +3
-3depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU select BR2_PACKAGE_FIRMWARE_IMX help Library of userspace helpers specific for the Freescale i.MX - platform. It wraps the kernel interfaces for the i.MX platform - Video Processing Unit (VPU) driver. It requires a kernel that - includes the i.MX specific headers to be built. + platform integrating a Chips&Media CODA Video Processing Unit + (VPU). It requires a kernel that includes the i.MX specific + headers to be built.
I think it is worth mentioning here which i.MX variants have this VPU. Otherwise the user has no easy way to know which package supports a specific SoC.
package/freescale-imx/imx-vpu-cnm/Config.in line 17
Quoted 2 lines
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +0
-0This library is provided by Freescale as-is and doesn't have an upstream.
baruch
Signature · 3 lines
http://baruch.siach.name/blog/ ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -PATCH 2/8 imx-vpu: rename package to imx-vpu-cnm
Hi Baruch,
Quoted 7 lines · Baruch Siach
Hi Gary,
Quoted 4 lines · Gary Bisson
So far the i.MX processors have been using a Chips&Media CODA VPU. But the latest generation of processors (starting with i.MX8MQ) uses VPU cores from Hantro which is provided as imx-vpu-hantro package.
Quoted 8 lines
Quoted 6 lines
NXP chose in Yocto to keep the imx-vpu naming for Chips&Media package and created a virtual "imxvpu" package that can either be provided by imx-vpu or imx-vpu-hantro.
Since this naming can be confusing (IMO), renaming the imx-vpu package to imx-vpu-cnm and then create a virtual imx-vpu one should be clearer.
What about Config.in.legacy?
Yep, good catch, I always forget about this.
Quoted 13 lines
[..]
Quoted 11 lines
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +2
-4-config BR2_PACKAGE_IMX_VPU - bool "imx-vpu" +config BR2_PACKAGE_IMX_VPU_CNM + bool "imx-vpu-cnm" depends on BR2_arm # Only relevant for i.MX depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU select BR2_PACKAGE_FIRMWARE_IMX help Library of userspace helpers specific for the Freescale i.MX - platform. It wraps the kernel interfaces for the i.MX platform - Video Processing Unit (VPU) driver. It requires a kernel that
Quoted 8 lines
package/freescale-imx/imx-vpu-cnm/Config.in line 14
Quoted 4 lines
package/freescale-imx/imx-vpu-cnm/Config.in renamed from package/freescale-imx/imx-vpu/Config.in +3
-1- includes the i.MX specific headers to be built. + platform integrating a Chips&Media CODA Video Processing Unit + (VPU). It requires a kernel that includes the i.MX specific + headers to be built.I think it is worth mentioning here which i.MX variants have this VPU. Otherwise the user has no easy way to know which package supports a specific SoC.
That makes sense, will be fixed in V2.
Thanks,
Gary
PATCH 3/8 imx-vpu-cnm: bump version to 5.4.38
In reply to Gary BissonNo changelog provided by NXP.
Added COPYING hash value to track license changes.
--- package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.hash | 3 ++- package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.hash +2 -1
@@ -1,2 +1,3 @@ # Locally computed -sha256 ee265e88d17c7369bd9cb917e7cce035b8c7ee2ba4491645fdab9f382f54beb0 imx-vpu-5.4.37.bin +sha256 34bed0ddf6c797f444bddfa5d5495adc751ff268b6431d8beb48129c89c9e47f imx-vpu-5.4.38.bin +sha256 78696664f7afe7f2b081c99af60082d6403fab935b1b10056ed41aec123ff7fa COPYING
package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -IMX_VPU_CNM_VERSION = 5.4.37 +IMX_VPU_CNM_VERSION = 5.4.38 IMX_VPU_CNM_SITE = $(FREESCALE_IMX_SITE) IMX_VPU_CNM_SOURCE = imx-vpu-$(IMX_VPU_CNM_VERSION).bin
Signature · 1 line
2.18.0
PATCH 4/8 imx-vpu-hantro: new package
In reply to Gary BissonThis package provides the user-space libraries needed to use the Hantro VPU present in processors such as the i.MX8MQ.
--- package/freescale-imx/Config.in | 4 +- ...on.h-header-inclusion-to-be-standard.patch | 44 +++++++++++++++++++ .../freescale-imx/imx-vpu-hantro/Config.in | 15 +++++++ .../imx-vpu-hantro/imx-vpu-hantro.hash | 2 + .../imx-vpu-hantro/imx-vpu-hantro.mk | 42 ++++++++++++++++++ 5 files changed, 106 insertions(+), 1 deletion(-) create mode 100644 package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch create mode 100644 package/freescale-imx/imx-vpu-hantro/Config.in create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk
package/freescale-imx/Config.in +3 -1
@@ -63,7 +63,8 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU default y if BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX27ADS || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX51 || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \ - BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q || \ + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8M config BR2_PACKAGE_FREESCALE_IMX_HAS_VIV_GPU bool
@@ -80,6 +81,7 @@ source "package/freescale-imx/imx-m4fwloader/Config.in" source "package/freescale-imx/imx-parser/Config.in" source "package/freescale-imx/imx-uuc/Config.in" source "package/freescale-imx/imx-vpu-cnm/Config.in" +source "package/freescale-imx/imx-vpu-hantro/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)
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +44 -0
44 lines not on this page · raw message
package/freescale-imx/imx-vpu-hantro/Config.in new file +15 -0
@@ -0,0 +1,15 @@ +comment "imx-vpu-hantro needs an i.MX platform with VPU support" + depends on BR2_aarch64 + depends on !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + +config BR2_PACKAGE_IMX_VPU_HANTRO + bool "imx-vpu-hantro" + depends on BR2_aarch64 # Only relevant for i.MX8 + depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + help + Library of userspace helpers specific for the NXP i.MX CPUs + integrating a Hantro Video Processing Unit (VPU). It requires + a kernel that includes the i.MX specific headers to be built. + + This library is provided by Freescale as-is and doesn't have + an upstream.
package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash new file +2 -0
@@ -0,0 +1,2 @@ +sha256 cbc648e41f005aad209f74c9e5dd346138dca12efeb7b27e471de7474c4da302 imx-vpu-hantro-1.6.0.bin +sha256 0f34f6175247762e2e1c38319aadf657a53f00ce124e569dfc61b30451549e7a COPYING
package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk new file +42 -0
42 lines not on this page · raw message
Signature · 1 line
2.18.0
PATCH 4/8 imx-vpu-hantro: new package
Hi Gary,
package/freescale-imx/imx-vpu-hantro/Config.in line 12
Quoted 8 lines · Gary Bisson
package/freescale-imx/imx-vpu-hantro/Config.in new file +8
-0+config BR2_PACKAGE_IMX_VPU_HANTRO + bool "imx-vpu-hantro" + depends on BR2_aarch64 # Only relevant for i.MX8 + depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + help + Library of userspace helpers specific for the NXP i.MX CPUs + integrating a Hantro Video Processing Unit (VPU). It requires + a kernel that includes the i.MX specific headers to be built.
Here also, a mention of i.MX8 would be nice, IMO.
baruch
Signature · 3 lines
http://baruch.siach.name/blog/ ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -PATCH 4/8 imx-vpu-hantro: new package
Hi Baruch,
Quoted 13 lines · Baruch Siach
Hi Gary,
package/freescale-imx/imx-vpu-hantro/Config.in line 12
Quoted 8 lines · Gary Bisson
package/freescale-imx/imx-vpu-hantro/Config.in new file +8
-0+config BR2_PACKAGE_IMX_VPU_HANTRO + bool "imx-vpu-hantro" + depends on BR2_aarch64 # Only relevant for i.MX8 + depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + help + Library of userspace helpers specific for the NXP i.MX CPUs + integrating a Hantro Video Processing Unit (VPU). It requires + a kernel that includes the i.MX specific headers to be built.Here also, a mention of i.MX8 would be nice, IMO.
Yes that would be clearer indeed, will be fixed in V2.
Thanks,
Gary
PATCH 4/8 imx-vpu-hantro: new package
In reply to Gary BissonQuoted 21 lines · Gary Bisson
This package provides the user-space libraries needed to use the Hantro VPU present in processors such as the i.MX8MQ.
Signed-off-by Gary Bisson <[email protected]>--- package/freescale-imx/Config.in | 4 +- ...on.h-header-inclusion-to-be-standard.patch | 44 +++++++++++++++++++ .../freescale-imx/imx-vpu-hantro/Config.in | 15 +++++++ .../imx-vpu-hantro/imx-vpu-hantro.hash | 2 + .../imx-vpu-hantro/imx-vpu-hantro.mk | 42 ++++++++++++++++++ 5 files changed, 106 insertions(+), 1 deletion(-) create mode 100644 package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch create mode 100644 package/freescale-imx/imx-vpu-hantro/Config.in create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mkpackage/freescale-imx/Config.in +0
-0
package/freescale-imx/Config.in config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU · line 67
Quoted 6 lines
package/freescale-imx/Config.in +2
-1@@ -63,7 +63,8 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU default y if BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX27ADS || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX51 || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \ - BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q || \ + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8M
Wouldn't it make more sense to make a new _HAS_VPU_HANTRO option? You're later
changing some 'depends' options to depending on the provider instead of depending on the VPU, so that makes sense.
Quoted 17 lines
package/freescale-imx/Config.in +1
-0config BR2_PACKAGE_FREESCALE_IMX_HAS_VIV_GPU bool@@ -80,6 +81,7 @@ source "package/freescale-imx/imx-m4fwloader/Config.in" source "package/freescale-imx/imx-parser/Config.in" source "package/freescale-imx/imx-uuc/Config.in" source "package/freescale-imx/imx-vpu-cnm/Config.in" +source "package/freescale-imx/imx-vpu-hantro/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)package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +0
-0
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch line 10
Quoted 10 lines
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +10
-0@@ -0,0 +1,44 @@ +From 872c82e7cbb9a0a0e761e8ac70fc28e19a55b4c3 Mon Sep 17 00:00:00 2001 +From: Gary Bisson <[email protected]> +Date: Thu, 12 Jul 2018 11:38:28 +0200 +Subject: [PATCH] Fix ion.h header inclusion to be standard + +NXP "solution" was to manually copy the header to include/linux. +Let's point the Makefile to the proper (mainline) location instead: +https://elixir.bootlin.com/linux/v4.17/source/drivers/staging/android/uapi/ion.h + +Signed-off-by: Gary Bisson <[email protected]>
Is it now possible to upstream things to codeaurora? Did you do that?
Quoted 40 lines
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +34
-0+--- + Makefile | 2 ++ + decoder_sw/software/linux/dwl/dwl_linux.c | 2 +- + 2 files changed, 3 insertions(+), 1 deletion(-) + +diff --git a/Makefile b/Makefile +index b74e23a..a5ce22b 100755 +--- a/Makefile ++++ b/Makefile +@@ -11,6 +11,8 @@ INCLUDE_HEADERS = -I./decoder_sw -I$(SOURCE_ROOT)/source/inc -I$(SOURCE_ROOT)/so + INCLUDE_HEADERS += -I$(SOURCE_ROOT)/linux/memalloc + #INCLUDE_HEADERS += -I$(SOURCE_ROOT)/linux/ldriver + INCLUDE_HEADERS += -I$(LINUX_KERNEL_ROOT)/include/uapi -I$(LINUX_KERNEL_ROOT)/include ++# ION header location ++INCLUDE_HEADERS += -I$(LINUX_KERNEL_ROOT)/drivers/staging/android/uapi + + CFLAGS += -DDEC_MODULE_PATH=\"/dev/mxc_hantro\" -DUSE_FAKE_RFC_TABLE -DFIFO_DATATYPE=void* -DNDEBUG -DDOWN_SCALER \ + -DUSE_EXTERNAL_BUFFER -DUSE_FAST_EC -DUSE_VP9_EC -DGET_FREE_BUFFER_NON_BLOCK \ +diff --git a/decoder_sw/software/linux/dwl/dwl_linux.c b/decoder_sw/software/linux/dwl/dwl_linux.c +index 8183660..ed37d86 100644 +--- a/decoder_sw/software/linux/dwl/dwl_linux.c ++++ b/decoder_sw/software/linux/dwl/dwl_linux.c +@@ -41,7 +41,7 @@ + #include "dwl.h" + #include <linux/hantrodec.h> + #ifdef USE_ION +-#include <linux/ion.h> ++#include <ion.h> + #ifdef ANDROID + #include <linux/mxc_ion.h> + #endif +-- +2.18.0 +package/freescale-imx/imx-vpu-hantro/Config.in new file +0
-0
package/freescale-imx/imx-vpu-hantro/Config.in line 7
Quoted 7 lines
package/freescale-imx/imx-vpu-hantro/Config.in new file +7
-0@@ -0,0 +1,15 @@ +comment "imx-vpu-hantro needs an i.MX platform with VPU support" + depends on BR2_aarch64 + depends on !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + +config BR2_PACKAGE_IMX_VPU_HANTRO + bool "imx-vpu-hantro" + depends on BR2_aarch64 # Only relevant for i.MX8
If you make a new BR2_PACKAGE_FREESCALE_IMX_HAS_VPU_HANTRO, this becomes
unnecessary. Actually the dependency on BR2_arm in imx-vpu is also redundant now IMO.
package/freescale-imx/imx-vpu-hantro/Config.in line 12
Quoted 5 lines
package/freescale-imx/imx-vpu-hantro/Config.in new file +5
-0+ depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + help + Library of userspace helpers specific for the NXP i.MX CPUs + integrating a Hantro Video Processing Unit (VPU). It requires + a kernel that includes the i.MX specific headers to be built.
You copied this from imx-vpu, but it is wrong there... Look at commit0ac0982fb7e936c48c0925b272eb18e0195e1c47 which removed the Linux dependency but failed to update the help text...
So, you need to add a dependency on BR2_LINUX_KERNEL and the corresponding comment.
Regards,
Arnout
Quoted 60 lines
package/freescale-imx/imx-vpu-hantro/Config.in new file +3
-0+ + This library is provided by Freescale as-is and doesn't have + an upstream.package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash new file +2
-0@@ -0,0 +1,2 @@ +sha256 cbc648e41f005aad209f74c9e5dd346138dca12efeb7b27e471de7474c4da302 imx-vpu-hantro-1.6.0.bin +sha256 0f34f6175247762e2e1c38319aadf657a53f00ce124e569dfc61b30451549e7a COPYINGpackage/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk new file +42
-0@@ -0,0 +1,42 @@ +################################################################################ +# +# imx-vpu-hantro +# +################################################################################ + +IMX_VPU_HANTRO_VERSION = 1.6.0 +IMX_VPU_HANTRO_SITE = $(FREESCALE_IMX_SITE) +IMX_VPU_HANTRO_SOURCE = imx-vpu-hantro-$(IMX_VPU_HANTRO_VERSION).bin + +IMX_VPU_HANTRO_INSTALL_STAGING = YES + +IMX_VPU_HANTRO_MAKE_ENV = \ + $(TARGET_MAKE_ENV) \ + $(TARGET_CONFIGURE_OPTS) \ + CROSS_COMPILE="$(TARGET_CROSS)" \ + SDKTARGETSYSROOT=$(STAGING_DIR) \ + LINUX_KERNEL_ROOT=$(LINUX_DIR) + +IMX_VPU_HANTRO_LICENSE = NXP Semiconductor Software License Agreement +IMX_VPU_HANTRO_LICENSE_FILES = EULA COPYING +IMX_VPU_HANTRO_REDISTRIBUTE = NO + +define IMX_VPU_HANTRO_EXTRACT_CMDS + $(call FREESCALE_IMX_EXTRACT_HELPER,$(IMX_VPU_HANTRO_DL_DIR)/$(IMX_VPU_HANTRO_SOURCE)) +endef + +define IMX_VPU_HANTRO_BUILD_CMDS + $(IMX_VPU_HANTRO_MAKE_ENV) $(MAKE1) -C $(@D) +endef + +define IMX_VPU_HANTRO_INSTALL_STAGING_CMDS + $(IMX_VPU_HANTRO_MAKE_ENV) $(MAKE1) -C $(@D) \ + DEST_DIR=$(STAGING_DIR) libdir=/usr/lib install +endef + +define IMX_VPU_HANTRO_INSTALL_TARGET_CMDS + $(IMX_VPU_HANTRO_MAKE_ENV) $(MAKE1) -C $(@D) \ + DEST_DIR=$(TARGET_DIR) libdir=/usr/lib install +endef + +$(eval $(generic-package))
Signature · 6 lines
Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
PATCH 4/8 imx-vpu-hantro: new package
Hi Arnout,
Quoted 19 lines · Arnout Vandecappelle
Quoted 16 lines · Gary Bisson
This package provides the user-space libraries needed to use the Hantro VPU present in processors such as the i.MX8MQ.
Signed-off-by Gary Bisson <[email protected]>--- package/freescale-imx/Config.in | 4 +- ...on.h-header-inclusion-to-be-standard.patch | 44 +++++++++++++++++++ .../freescale-imx/imx-vpu-hantro/Config.in | 15 +++++++ .../imx-vpu-hantro/imx-vpu-hantro.hash | 2 + .../imx-vpu-hantro/imx-vpu-hantro.mk | 42 ++++++++++++++++++ 5 files changed, 106 insertions(+), 1 deletion(-) create mode 100644 package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch create mode 100644 package/freescale-imx/imx-vpu-hantro/Config.in create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk
Quoted 15 lines
package/freescale-imx/Config.in config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU · line 67
Quoted 11 lines
package/freescale-imx/Config.in +2
-1@@ -63,7 +63,8 @@ config BR2_PACKAGE_FREESCALE_IMX_HAS_VPU default y if BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX27ADS || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX51 || \ BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX53 || \ - BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX6Q || \ + BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8MWouldn't it make more sense to make a new
_HAS_VPU_HANTROoption? You're later
changing some 'depends' options to depending on the provider instead of depending on the VPU, so that makes sense.
Yes we could, I just didn't do it as first as I recall someone not being fond of the FREESCALE_IMX_HAS_XXX options.
But if we remove the virtual package as Thomas suggested that would be best indeed.
Quoted 20 lines
Quoted 20 lines
package/freescale-imx/Config.in +1
-0config BR2_PACKAGE_FREESCALE_IMX_HAS_VIV_GPU bool@@ -80,6 +81,7 @@ source "package/freescale-imx/imx-m4fwloader/Config.in" source "package/freescale-imx/imx-parser/Config.in" source "package/freescale-imx/imx-uuc/Config.in" source "package/freescale-imx/imx-vpu-cnm/Config.in" +source "package/freescale-imx/imx-vpu-hantro/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)package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +4
-0@@ -0,0 +1,44 @@ +From 872c82e7cbb9a0a0e761e8ac70fc28e19a55b4c3 Mon Sep 17 00:00:00 2001 +From: Gary Bisson <[email protected]> +Date: Thu, 12 Jul 2018 11:38:28 +0200 +Subject: [PATCH] Fix ion.h header inclusion to be standard
Quoted 8 lines
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch line 10
Quoted 6 lines
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +6
-0+ +NXP "solution" was to manually copy the header to include/linux. +Let's point the Makefile to the proper (mainline) location instead: +https://elixir.bootlin.com/linux/v4.17/source/drivers/staging/android/uapi/ion.h + +Signed-off-by: Gary Bisson <[email protected]>Is it now possible to upstream things to codeaurora? Did you do that?
Not that I know of. I've tried sending emails to the owner of the repo + main contributor but never received an answer.
The best approach is to have that patch in the meta-freescale BSP, then NXP will look at it (it is the case here).
Quoted 43 lines
Quoted 43 lines
package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch new file +34
-0+--- + Makefile | 2 ++ + decoder_sw/software/linux/dwl/dwl_linux.c | 2 +- + 2 files changed, 3 insertions(+), 1 deletion(-) + +diff --git a/Makefile b/Makefile +index b74e23a..a5ce22b 100755 +--- a/Makefile ++++ b/Makefile +@@ -11,6 +11,8 @@ INCLUDE_HEADERS = -I./decoder_sw -I$(SOURCE_ROOT)/source/inc -I$(SOURCE_ROOT)/so + INCLUDE_HEADERS += -I$(SOURCE_ROOT)/linux/memalloc + #INCLUDE_HEADERS += -I$(SOURCE_ROOT)/linux/ldriver + INCLUDE_HEADERS += -I$(LINUX_KERNEL_ROOT)/include/uapi -I$(LINUX_KERNEL_ROOT)/include ++# ION header location ++INCLUDE_HEADERS += -I$(LINUX_KERNEL_ROOT)/drivers/staging/android/uapi + + CFLAGS += -DDEC_MODULE_PATH=\"/dev/mxc_hantro\" -DUSE_FAKE_RFC_TABLE -DFIFO_DATATYPE=void* -DNDEBUG -DDOWN_SCALER \ + -DUSE_EXTERNAL_BUFFER -DUSE_FAST_EC -DUSE_VP9_EC -DGET_FREE_BUFFER_NON_BLOCK \ +diff --git a/decoder_sw/software/linux/dwl/dwl_linux.c b/decoder_sw/software/linux/dwl/dwl_linux.c +index 8183660..ed37d86 100644 +--- a/decoder_sw/software/linux/dwl/dwl_linux.c ++++ b/decoder_sw/software/linux/dwl/dwl_linux.c +@@ -41,7 +41,7 @@ + #include "dwl.h" + #include <linux/hantrodec.h> + #ifdef USE_ION +-#include <linux/ion.h> ++#include <ion.h> + #ifdef ANDROID + #include <linux/mxc_ion.h> + #endif +-- +2.18.0 +package/freescale-imx/imx-vpu-hantro/Config.in new file +3
-0@@ -0,0 +1,15 @@ +comment "imx-vpu-hantro needs an i.MX platform with VPU support" + depends on BR2_aarch64 + depends on !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU
Quoted 8 lines
package/freescale-imx/imx-vpu-hantro/Config.in line 7
Quoted 4 lines
package/freescale-imx/imx-vpu-hantro/Config.in new file +4
-0+ +config BR2_PACKAGE_IMX_VPU_HANTRO + bool "imx-vpu-hantro" + depends on BR2_aarch64 # Only relevant for i.MX8If you make a new
BR2_PACKAGE_FREESCALE_IMX_HAS_VPU_HANTRO, this becomes
unnecessary. Actually the dependency onBR2_armin imx-vpu is also redundant now IMO.
True.
Quoted 9 lines
package/freescale-imx/imx-vpu-hantro/Config.in line 12
Quoted 5 lines
package/freescale-imx/imx-vpu-hantro/Config.in new file +5
-0+ depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + help + Library of userspace helpers specific for the NXP i.MX CPUs + integrating a Hantro Video Processing Unit (VPU). It requires + a kernel that includes the i.MX specific headers to be built.You copied this from imx-vpu, but it is wrong there... Look at commit
0ac0982fb7e936c48c0925b272eb18e0195e1c47which removed the Linux dependency but failed to update the help text...
Yep, will update help text for imx-vpu. Here it is required.
Quoted 1 line
So, you need to add a dependency on
BR2_LINUX_KERNELand the corresponding comment.
Yes.
Thanks,
Gary
PATCH 5/8 imx-vpu: new virtual package
In reply to Gary BissonSince it can be provided by either imx-vpu-cnm or imx-vpu-hantro.
--- package/freescale-imx/Config.in | 1 + package/freescale-imx/imx-codec/imx-codec.mk | 2 +- package/freescale-imx/imx-vpu-cnm/Config.in | 8 ++++++++ package/freescale-imx/imx-vpu-hantro/Config.in | 8 ++++++++ package/freescale-imx/imx-vpu/Config.in | 6 ++++++ package/freescale-imx/imx-vpu/imx-vpu.mk | 7 +++++++ package/freescale-imx/imx-vpuwrap/Config.in | 7 +++---- package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk | 2 +- package/gstreamer/gst-fsl-plugins/Config.in | 2 +- package/gstreamer/gst-fsl-plugins/gst-fsl-plugins.mk | 2 +- package/libimxvpuapi/Config.in | 7 +++---- package/libimxvpuapi/libimxvpuapi.mk | 2 +- 12 files changed, 41 insertions(+), 13 deletions(-) create mode 100644 package/freescale-imx/imx-vpu/Config.in create mode 100644 package/freescale-imx/imx-vpu/imx-vpu.mk
package/freescale-imx/Config.in +1 -0
@@ -80,6 +80,7 @@ source "package/freescale-imx/imx-lib/Config.in" source "package/freescale-imx/imx-m4fwloader/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-vpu-cnm/Config.in" source "package/freescale-imx/imx-vpu-hantro/Config.in" source "package/freescale-imx/imx-vpuwrap/Config.in"
package/freescale-imx/imx-codec/imx-codec.mk +1 -1
@@ -21,7 +21,7 @@ ifeq ($(BR2_ARM_EABIHF),y) IMX_CODEC_CONF_OPTS += --enable-fhw endif -ifeq ($(BR2_PACKAGE_IMX_VPU_CNM),y) +ifeq ($(BR2_PACKAGE_HAS_IMX_VPU),y) IMX_CODEC_CONF_OPTS += --enable-vpu endif
package/freescale-imx/imx-vpu-cnm/Config.in +8 -0
@@ -7,6 +7,7 @@ config BR2_PACKAGE_IMX_VPU_CNM depends on BR2_arm # Only relevant for i.MX depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU select BR2_PACKAGE_FIRMWARE_IMX + select BR2_PACKAGE_HAS_IMX_VPU help Library of userspace helpers specific for the Freescale i.MX platform integrating a Chips&Media CODA Video Processing Unit
@@ -15,3 +16,10 @@ config BR2_PACKAGE_IMX_VPU_CNM This library is provided by Freescale as-is and doesn't have an upstream. + +if BR2_PACKAGE_IMX_VPU_CNM + +config BR2_PACKAGE_PROVIDES_IMX_VPU + default "imx-vpu-cnm" + +endif
package/freescale-imx/imx-vpu-hantro/Config.in +8 -0
@@ -6,6 +6,7 @@ config BR2_PACKAGE_IMX_VPU_HANTRO bool "imx-vpu-hantro" depends on BR2_aarch64 # Only relevant for i.MX8 depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + select BR2_PACKAGE_HAS_IMX_VPU help Library of userspace helpers specific for the NXP i.MX CPUs integrating a Hantro Video Processing Unit (VPU). It requires
@@ -13,3 +14,10 @@ config BR2_PACKAGE_IMX_VPU_HANTRO This library is provided by Freescale as-is and doesn't have an upstream. + +if BR2_PACKAGE_IMX_VPU_HANTRO + +config BR2_PACKAGE_PROVIDES_IMX_VPU + default "imx-vpu-hantro" + +endif
package/freescale-imx/imx-vpu/Config.in new file +6 -0
@@ -0,0 +1,6 @@ +config BR2_PACKAGE_HAS_IMX_VPU + bool + +config BR2_PACKAGE_PROVIDES_IMX_VPU + depends on BR2_PACKAGE_HAS_IMX_VPU + string
package/freescale-imx/imx-vpu/imx-vpu.mk new file +7 -0
@@ -0,0 +1,7 @@ +################################################################################ +# +# imx-vpu +# +################################################################################ + +$(eval $(virtual-package))
package/freescale-imx/imx-vpuwrap/Config.in +3 -4
@@ -2,16 +2,15 @@ comment "imx-vpuwrap needs an imx-specific Linux kernel to be built" depends on BR2_arm depends on !BR2_LINUX_KERNEL -comment "imx-vpuwrap needs an i.MX platform with VPU support" +comment "imx-vpuwrap needs an i.MX VPU backend provider" depends on BR2_arm - depends on BR2_LINUX_KERNEL && !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + depends on BR2_LINUX_KERNEL && !BR2_PACKAGE_HAS_IMX_VPU 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 - select BR2_PACKAGE_IMX_VPU_CNM + depends on BR2_PACKAGE_HAS_IMX_VPU help Wrapper library for the vpu library, giving it a different API.
package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk +1 -1
@@ -7,7 +7,7 @@ IMX_VPUWRAP_VERSION = 1.0.68 IMX_VPUWRAP_SITE = $(FREESCALE_IMX_SITE) IMX_VPUWRAP_SOURCE = imx-vpuwrap-$(IMX_VPUWRAP_VERSION).bin -IMX_VPUWRAP_DEPENDENCIES = imx-vpu-cnm +IMX_VPUWRAP_DEPENDENCIES = imx-vpu IMX_VPUWRAP_INSTALL_STAGING = YES IMX_VPUWRAP_LICENSE = NXP Semiconductor Software License Agreement
package/gstreamer/gst-fsl-plugins/Config.in +1 -1
@@ -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_IMX_VPUWRAP if BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + select BR2_PACKAGE_IMX_VPUWRAP if BR2_PACKAGE_HAS_IMX_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
@@ -18,7 +18,7 @@ GST_FSL_PLUGINS_AUTORECONF = YES GST_FSL_PLUGINS_DEPENDENCIES += host-pkgconf gstreamer gst-plugins-base \ imx-lib imx-parser imx-codec -ifeq ($(BR2_PACKAGE_FREESCALE_IMX_HAS_VPU),y) +ifeq ($(BR2_PACKAGE_HAS_IMX_VPU),y) GST_FSL_PLUGINS_DEPENDENCIES += imx-vpuwrap endif
package/libimxvpuapi/Config.in +3 -4
@@ -1,12 +1,11 @@ -comment "libimxvpuapi needs an i.MX platform with VPU support" +comment "libimxvpuapi needs an i.MX VPU backend provider" depends on BR2_arm - depends on !BR2_PACKAGE_FREESCALE_IMX_HAS_VPU + depends on !BR2_PACKAGE_HAS_IMX_VPU config BR2_PACKAGE_LIBIMXVPUAPI bool "libimxvpuapi" depends on BR2_arm # Only relevant for i.MX - depends on BR2_PACKAGE_FREESCALE_IMX_HAS_VPU - select BR2_PACKAGE_IMX_VPU_CNM + depends on BR2_PACKAGE_HAS_IMX_VPU help This library provides an API for using the iMX6 VPU video engine. It is an alternative to Freescale's VPU
package/libimxvpuapi/libimxvpuapi.mk +1 -1
@@ -8,7 +8,7 @@ LIBIMXVPUAPI_VERSION = 0.10.3 LIBIMXVPUAPI_SITE = $(call github,Freescale,libimxvpuapi,$(LIBIMXVPUAPI_VERSION)) LIBIMXVPUAPI_LICENSE = LGPL-2.1+ LIBIMXVPUAPI_LICENSE_FILES = LICENSE -LIBIMXVPUAPI_DEPENDENCIES = host-pkgconf host-python imx-vpu-cnm +LIBIMXVPUAPI_DEPENDENCIES = host-pkgconf host-python imx-vpu LIBIMXVPUAPI_INSTALL_STAGING = YES $(eval $(waf-package))
Signature · 1 line
2.18.0
PATCH 6/8 imx-vpuwrap: bump version to 4.3.5
In reply to Gary BissonNo changelog provided for 4.3.5 but it can now builds against the following VPU backeds:
- Chips&Media
- Hantro
- Amphion (not supported in BR yet)
Changes in 4.3.2:
- Add vpu_wrapper_test
- Add gst_dump tools
- Add gl sink
- Enable RMVB/VC1/DIVX3
- Add webp support
Changes in 4.3.4:
- set hasColorDesc and fullRange in VPU_DecGetInitialInfo.
- Fix H264 failure due to too many MBI buffers.
--- package/freescale-imx/imx-vpuwrap/Config.in | 4 ++-- package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash | 3 ++- package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk | 2 +- 3 files changed, 5 insertions(+), 4 deletions(-)
package/freescale-imx/imx-vpuwrap/Config.in +2 -2
@@ -1,5 +1,5 @@ comment "imx-vpuwrap needs an imx-specific Linux kernel to be built" - depends on BR2_arm + depends on BR2_arm || BR2_aarch64 depends on !BR2_LINUX_KERNEL comment "imx-vpuwrap needs an i.MX VPU backend provider"
@@ -9,7 +9,7 @@ comment "imx-vpuwrap needs an i.MX VPU backend provider" 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_arm || BR2_aarch64 # Only relevant for i.MX depends on BR2_PACKAGE_HAS_IMX_VPU help Wrapper library for the vpu library, giving it a different
package/freescale-imx/imx-vpuwrap/imx-vpuwrap.hash +2 -1
@@ -1,2 +1,3 @@ # locally computed -sha256 282e7f8766ce385d8752bd29f04ddeff709ece0846be97547cf982183bbe241e imx-vpuwrap-1.0.68.bin +sha256 5c08b4b7c771404c998779f0e27a75564b57958d463e2df152c910d76cca9e44 imx-vpuwrap-4.3.5.bin +sha256 78696664f7afe7f2b081c99af60082d6403fab935b1b10056ed41aec123ff7fa COPYING
package/freescale-imx/imx-vpuwrap/imx-vpuwrap.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -IMX_VPUWRAP_VERSION = 1.0.68 +IMX_VPUWRAP_VERSION = 4.3.5 IMX_VPUWRAP_SITE = $(FREESCALE_IMX_SITE) IMX_VPUWRAP_SOURCE = imx-vpuwrap-$(IMX_VPUWRAP_VERSION).bin IMX_VPUWRAP_DEPENDENCIES = imx-vpu
Signature · 1 line
2.18.0
PATCH 7/8 imx-codec: bump version to 4.3.5
In reply to Gary BissonNo changelog provided by NXP for v4.3.5.
Added COPYING hash to track license change.
Changelog for v4.3.4
1. Add INSANE_SKIP for file-rdeps to fix QA Issue
2. Update mp3_enc version to 02.03.01
3. Fix crash in lib_mp3_enc_arm12_elinux.so This crash error is caused by division by zero in noOfElements_mp3e() function When encoding 0 size file by using unit test
so before performing a division operation, need to check whether divisor is zero or not to fix it.
Changelog for v4.3.2
1. Upgrade EULA to v20
2. Add i.MX8 support
3. Bugfix/Optimization
- Remove video codec in imx-codec package.
- Fix logically dead code issue for aac_dec.
- Fix wrong sizeof argument for flac_dec.
- Cancel unused code in nb_amr/wb_amr/flac_dec/bsac_dec.
--- package/freescale-imx/imx-codec/imx-codec.hash | 3 ++- package/freescale-imx/imx-codec/imx-codec.mk | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
package/freescale-imx/imx-codec/imx-codec.hash +2 -1
@@ -1,2 +1,3 @@ # locally computed -sha256 20d3f9b4187fcd9e7007c94558a00bab1191513eee74b6f0d8c7b43f874e06ed imx-codec-4.2.1.bin +sha256 6bb54f91c3ca18567e14d95d3858022dc9be00dc86e9edfdb544d1240a3b2b04 imx-codec-4.3.5.bin +sha256 78696664f7afe7f2b081c99af60082d6403fab935b1b10056ed41aec123ff7fa COPYING
package/freescale-imx/imx-codec/imx-codec.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -IMX_CODEC_VERSION = 4.2.1 +IMX_CODEC_VERSION = 4.3.5 IMX_CODEC_SITE = $(FREESCALE_IMX_SITE) IMX_CODEC_SOURCE = imx-codec-$(IMX_CODEC_VERSION).bin IMX_CODEC_INSTALL_STAGING = YES
Signature · 1 line
2.18.0
PATCH 8/8 imx-parser: bump version to 4.3.5
In reply to Gary BissonAdded COPYING hash to track license changes.
No changelog provided by NXP for v4.3.5.
Changelog for v4.3.4:
- Fix crash issue when parse index list.
- Add mett atom in mp4 parser.
- MP4 PARSER 06.13.00 add interface PARSER_API_GET_TEXT_TRACK_MIME to get mime type from mett atom.
- Add protection for contentDescriptionObject in case the size is not correct and larger than qwSize.
Changelog for v4.3.2:
- Fix creating parser error
When the variable value is negetive, if(bytesToRead) will return true, modify this condition to if(bytesToRead > 0).
- FLAC PARSER 03.00.07
Check read actual length in case of parser reach eos.
- MP4 PARSER 06.12.11
Fix crash issue when parse id3 metadata.
- MPG2 PARSER 04.05.16
Fix parser return wrong frame size after fastforward.
- ASFPARSER_01.00.55
Update for HEVC tag and audio AC3 support.
--- package/freescale-imx/imx-parser/imx-parser.hash | 3 ++- package/freescale-imx/imx-parser/imx-parser.mk | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
package/freescale-imx/imx-parser/imx-parser.hash +2 -1
@@ -1,2 +1,3 @@ # locally computed -sha256 36d3ae7285f3a83a87abf680b8a52b3c07df869d2443de844fb5f0ff528ca862 imx-parser-4.2.1.bin +sha256 9cd8c49a0bb1050afcfbcc7a677a348443fda4ecacb621cc6e38897619e27c67 imx-parser-4.3.5.bin +sha256 78696664f7afe7f2b081c99af60082d6403fab935b1b10056ed41aec123ff7fa COPYING
package/freescale-imx/imx-parser/imx-parser.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -IMX_PARSER_VERSION = 4.2.1 +IMX_PARSER_VERSION = 4.3.5 IMX_PARSER_SITE = $(FREESCALE_IMX_SITE) IMX_PARSER_SOURCE = imx-parser-$(IMX_PARSER_VERSION).bin IMX_PARSER_INSTALL_STAGING = YES
Signature · 1 line
2.18.0
cover imx: update multimedia packages to 4.9.88_2.0.0_ga
In reply to Gary BissonQuoted 11 lines · Gary Bisson
Hi,
This series updates the i.MX multimedia packages to be in line with the NXP 4.9.88-2.0.0_ga release [1].
The main benefit of this release is that it works for both 32-bit i.MX6/7 processors as well as 64-bit i.MX8MQ.
That means that new packages had to be added to support the i.MX8MQ VPU IP (from Hantro).
Quoted 8 lines
In Yocto, the imx-vpu package for the Chips&Media VPU present on i.MX6 wasn't renamed. But packages such as imx-vpuwrap now depend on virtual/imxvpu which can either be provided by imx-vpu (Chips&Media) or imx-vpu-hantro (Hantro) [2][3].
As I thought it'd be confusing to have a virtual imxvpu virtual package with an imx-vpu provider, I renamed imx-vpu to imx-vpu-cnm and added an imx-vpu virtual package.
I'm sorry for all the work that you did, but I don't really agree that this is
a good idea. The upstream package really is called imx-vpu, so we prefer to keep that name. We changed the name for openssl because there really was no other way, but I do prefer to avoid that.
So I think it's just a matter of finding a better name for the virtual package.
What about imx-vpu-provider?
Quoted 9 lines
Each patch for this imx-vpu change modifies several packages (those depending on it) so that it doesn't break the build in case someone bisects the tree.
Finally, I've verified those packages on both i.MX6Q and i.MX8MQ. But in the case of i.MX8MQ, the VPU testing was limited to the package unit tests since the open-source imx plugin doesn't support Hantro VPU (yet). Note that I don't plan on integrating NXP plugin since it implies changing all the standard plugins to be "NXP-compliant" [4].
Hm, I thought that NXP had understood open source, but apparently there is
still a ways to go...
Regards,
Arnout
Quoted 55 lines
Let me know if you are ok with this approach. As usual, comments are welcome.
Regards,
Gary[1] https://community.nxp.com/docs/DOC-340805
[2] https://github.com/Freescale/meta-freescale/blob/master/recipes-bsp/imx-vpu/imx-vpu_5.4.38.bb#L11
[3] https://github.com/Freescale/meta-freescale/blob/master/recipes-bsp/imx-vpu-hantro/imx-vpu-hantro_1.6.0.bb#L9
[4] https://github.com/Freescale/meta-freescale/tree/master/recipes-multimedia/gstreamerGary Bisson (8):
firmware-imx: bump to version 7.5
imx-vpu: rename package to imx-vpu-cnm
imx-vpu-cnm: bump version to 5.4.38
imx-vpu-hantro: new package
imx-vpu: new virtual package
imx-vpuwrap: bump version to 4.3.5
imx-codec: bump version to 4.3.5
imx-parser: bump version to 4.3.5package/freescale-imx/Config.in | 5 ++- .../firmware-imx/firmware-imx.hash | 2 +- .../firmware-imx/firmware-imx.mk | 2 +- .../freescale-imx/imx-codec/imx-codec.hash | 3 +- package/freescale-imx/imx-codec/imx-codec.mk | 4 +- .../freescale-imx/imx-parser/imx-parser.hash | 3 +- .../freescale-imx/imx-parser/imx-parser.mk | 2 +- package/freescale-imx/imx-vpu-cnm/Config.in | 25 +++++++++++ .../imx-vpu-cnm/imx-vpu-cnm.hash | 3 ++ .../freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk | 39 ++++++++++++++++ ...on.h-header-inclusion-to-be-standard.patch | 44 +++++++++++++++++++ .../freescale-imx/imx-vpu-hantro/Config.in | 23 ++++++++++ .../imx-vpu-hantro/imx-vpu-hantro.hash | 2 + .../imx-vpu-hantro/imx-vpu-hantro.mk | 42 ++++++++++++++++++ package/freescale-imx/imx-vpu/Config.in | 21 +++------ package/freescale-imx/imx-vpu/imx-vpu.hash | 2 - package/freescale-imx/imx-vpu/imx-vpu.mk | 34 +------------- package/freescale-imx/imx-vpuwrap/Config.in | 11 +++-- .../imx-vpuwrap/imx-vpuwrap.hash | 3 +- .../freescale-imx/imx-vpuwrap/imx-vpuwrap.mk | 2 +- package/gstreamer/gst-fsl-plugins/Config.in | 2 +- .../gst-fsl-plugins/gst-fsl-plugins.mk | 2 +- package/libimxvpuapi/Config.in | 7 ++- 23 files changed, 210 insertions(+), 73 deletions(-) create mode 100644 package/freescale-imx/imx-vpu-cnm/Config.in create mode 100644 package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.hash create mode 100644 package/freescale-imx/imx-vpu-cnm/imx-vpu-cnm.mk create mode 100644 package/freescale-imx/imx-vpu-hantro/0001-Fix-ion.h-header-inclusion-to-be-standard.patch create mode 100644 package/freescale-imx/imx-vpu-hantro/Config.in create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.hash create mode 100644 package/freescale-imx/imx-vpu-hantro/imx-vpu-hantro.mk delete mode 100644 package/freescale-imx/imx-vpu/imx-vpu.hash
Signature · 6 lines
Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
cover imx: update multimedia packages to 4.9.88_2.0.0_ga
Hello,
Quoted 7 lines · Arnout Vandecappelle
I'm sorry for all the work that you did, but I don't really agree that this is
a good idea. The upstream package really is called imx-vpu, so we prefer to keep that name. We changed the name for openssl because there really was no other way, but I do prefer to avoid that.So I think it's just a matter of finding a better name for the virtual package.
What about imx-vpu-provider?
I have not reviewed the patch series carefully enough yet, but a question is: do we need a virtual package at all?
I guess the imx-vpu API is not going to be used by gazillions of packages. If I read PATCH 5/8 correctly, it's in fact only used by two packages, right ?
Can't we simply have those two packages do:
ifeq ($(BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8M),y) use the new imx-vpu package
else
use the old imx-vpu package
endif
And ditto in their Config.in ?
Virtual packages are great when there is really an arbitrary number of providers and/or an arbitrary number of users.
OpenSSL for examples has only two providers, but it has a very large number of users. Ditto jpeg. MySQL does not have a lot of users, but it might potentially have.
Also, the i.MX VPU stuff is highly platform-specific, it provides a very specialized API, it's very unlikely that we will see gazillions of packages depending on the i.MX VPU API.
Best regards,
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering https://bootlin.com
cover imx: update multimedia packages to 4.9.88_2.0.0_ga
Hi,
Quoted 9 lines · Thomas Petazzoni
Hello,
Quoted 5 lines · Arnout Vandecappelle
I'm sorry for all the work that you did, but I don't really agree that this is
a good idea. The upstream package really is called imx-vpu, so we prefer to keep that name. We changed the name for openssl because there really was no other way, but I do prefer to avoid that.
Quoted 5 lines
Quoted 2 lines
So I think it's just a matter of finding a better name for the virtual package.
What about imx-vpu-provider?I have not reviewed the patch series carefully enough yet, but a question is: do we need a virtual package at all?
Up to you, I'm ok either way, just thought it'd be cleaner this way.
Quoted 3 lines
I guess the imx-vpu API is not going to be used by gazillions of packages. If I read PATCH 5/8 correctly, it's in fact only used by two packages, right ?
Yes, libimxvpuapi and imx-vpuwrap are the only 2 users of imx-vpu right now.
Quoted 9 lines
Can't we simply have those two packages do:
ifeq ($(BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8M),y) use the new imx-vpu package
else
use the old imx-vpu package
endifAnd ditto in their Config.in ?
Ok.
Quoted 10 lines
Virtual packages are great when there is really an arbitrary number of providers and/or an arbitrary number of users.
OpenSSL for examples has only two providers, but it has a very large number of users. Ditto jpeg. MySQL does not have a lot of users, but it might potentially have.
Also, the i.MX VPU stuff is highly platform-specific, it provides a very specialized API, it's very unlikely that we will see gazillions of packages depending on the i.MX VPU API.
Ok, I'll make a v3 like this then.
Regards,
Gary
cover imx: update multimedia packages to 4.9.88_2.0.0_ga
Quoted 12 lines · Gary Bisson
Hi,
Quoted 9 lines · Thomas Petazzoni
Hello,
Quoted 5 lines · Arnout Vandecappelle
I'm sorry for all the work that you did, but I don't really agree that this is
a good idea. The upstream package really is called imx-vpu, so we prefer to keep that name. We changed the name for openssl because there really was no other way, but I do prefer to avoid that.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
So I think it's just a matter of finding a better name for the virtual package.
What about imx-vpu-provider?I have not reviewed the patch series carefully enough yet, but a question is: do we need a virtual package at all?
Up to you, I'm ok either way, just thought it'd be cleaner this way.
If you don't use the virtual package infra, the code just gets a little more
difficult to understand, so I really don't see a reason NOT to use the virtual package infra. Yes, a virtual package is a little more verbose than "manually" handling it, but it's a pattern that is understandable. The only disadvantage that I see is that you have to find a name for it :-)
That said, since there are only two users, doing it manually isn't that bad
either. So if you've already done the work, please stick to it and don't keep running around in circles...
Regards,
Arnout
Quoted 36 lines
Quoted 4 lines
I guess the imx-vpu API is not going to be used by gazillions of packages. If I read PATCH 5/8 correctly, it's in fact only used by two packages, right ?
Yes, libimxvpuapi and imx-vpuwrap are the only 2 users of imx-vpu right now.
Quoted 9 lines
Can't we simply have those two packages do:
ifeq ($(BR2_PACKAGE_FREESCALE_IMX_PLATFORM_IMX8M),y) use the new imx-vpu package
else
use the old imx-vpu package
endifAnd ditto in their Config.in ?
Ok.
Quoted 10 lines
Virtual packages are great when there is really an arbitrary number of providers and/or an arbitrary number of users.
OpenSSL for examples has only two providers, but it has a very large number of users. Ditto jpeg. MySQL does not have a lot of users, but it might potentially have.
Also, the i.MX VPU stuff is highly platform-specific, it provides a very specialized API, it's very unlikely that we will see gazillions of packages depending on the i.MX VPU API.
Ok, I'll make a v3 like this then.
Regards,
Gary
Signature · 6 lines
Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF