| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Signed-off-by: Javier Viguera <javier.viguera@digi.com>
|
|
|
|
|
|
|
| |
The encoding library was not previously being installed.
Signed-off-by: Chris Dimich <Chris.Dimich@boundarydevices.com>
(cherry picked from commit 0a435d64e588843f6943d6e5723e5ecc12990aa2)
|
|
|
|
|
|
|
|
| |
These recipes are really not meant for non imx architectures, therefore
pin them as such
Signed-off-by: Khem Raj <raj.khem@gmail.com>
(cherry picked from commit 346713517e26729f91fd35423c8934cf59e7d64d)
|
|
|
|
|
| |
Signed-off-by: Chris Dimich <Chris.Dimich@boundarydevices.com>
(cherry picked from commit 7fecdcede764dc2081f6105f03bcc385d3219aba)
|
|
|
|
|
| |
Signed-off-by: Chris Dimich <Chris.Dimich@boundarydevices.com>
(cherry picked from commit 01d65a00275c2e6016dc9003cc38cd63f6b0fe71)
|
|
|
|
|
| |
Signed-off-by: Chris Dimich <Chris.Dimich@boundarydevices.com>
(cherry picked from commit ab82bf33d4f95dacfdfbc56a5b34c6fa51d67e9c)
|
|
|
|
|
| |
Signed-off-by: Chris Dimich <Chris.Dimich@boundarydevices.com>
(cherry picked from commit 3b91fa80566510e6f787a4212ef93fcc89c54955)
|
|
|
|
|
|
|
|
|
|
| |
When building a eSDK, sdma-imx6q/7d files are installed by
both firmware-imx and linux-firmware causing a abort.
Prevent firmware-imx to install sdma-imx6q/7d allows
linux-firmware to install them exclusively.
Signed-off-by: Vinicius Aquino <voa.aquino@gmail.com>
(cherry picked from commit 9f5ecc8396a9b247109236cb5d94ab560ff62be9)
|
|
|
|
|
|
| |
This makes u-boot-fslc and linux-fslc support the nxp mfgtool.
Signed-off-by: Maxim De Clercq <maximdeclercq00@gmail.com>
|
|
|
|
|
|
| |
using append and += together is undefined behavior in bitbake
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
| |
Install additionally the vpu fw needed for the mx8qm/mx8qxp SoC famiiles.
Use a wildcard as only vpu_fw_imx8_dec.bin and vpu_fw_imx8_enc.bin files
are not yet installed. No other recipe is providing them.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
|
|
|
|
| |
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
|
|
|
|
|
|
| |
This version uses EULA v15 August 2020
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
| |
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
|
| |
This version uses EULA v15 August 2020.
imx-seco-libs: Expand the compatibility list for better tracking.
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
| |
This version uses EULA v15 August 2020.
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
| |
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
| |
This version uses EULA v15 August 2020
Following commits are included:
* 6b2c608 MA-17652-2 Rebuild vc8000e library
* 0d477ad M865SW-342-[VC8000E] Support VUI info in encoder
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This version uses EULA v15 August 2020
Following commits are included:
* cd8a7a6 M865SW-326-2: VP9 decoder return error stream for a special stream
* f607b7c M865SW-342-[H1] support VUI info in H1 encoder
* 1419c97 M865SW-326-[G2] Fix vp9 error of 'shou_existing_frame'
* 48ae7d6 MA-17522 modify android.bp to pass build both in android 10 and 11
* 2b8639a M865SW-217 - [G1/G2] Avoid use virtual framebuffer address to support secure mode
* ac91e5c M865SW-231: [G1/G2]: Fix building error in type conversion
* a09dba2 MA-17424-3 - [MPEG2] Implement DWLFlushCache
* 048a93d M865SW-217 - [MPEG2] Call flush cache after CPU touch framebuffer
* 9f11483 M865SW-231: [G1/G2]: Avoid use virtual YUV buffer address as handler to support security mode
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
| |
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
New commits are included:
---
c949a888e MLK-24599 Revert "spd: opteed: enable NS_TIMER_SWITCH"
84fc81fb6 MLK-24721 plat: imx8m: Fix the out of bound access to rank setting array
77c519ec2 MLK-24513 plat: imx8mp: disable the memrepair clock when do domain power down
9c8d998cf MLK-24502 plat: imx8m: Add mem repair done check for domain that need repair
7ad9e7ab4 MLK-24474: Add SIP call to enable FIPS mode
fac431ed6 MLK-24474: Add SECO API sc_seco_set_fips_mode
e39a0c329 MLK-24480 plat: imx8dxl: Enable CPU/FP/L2 retention counter
fbf6df269 MLK-24457 plat: imx8m: Fix the current fsp init
a97de4c80 MLK-24414 plat: imx8mp: Add the dsp low power audio basic support
aa3fee73c MLK-24360 plat: imx8mp: fix the vpu noc nttp handshake hang issue
2eb979f14 MLK-23930-2 plat: imx8m: Add csu and rdc test
5e705b7aa MLK-23930-1 plat: imx8mq: cleanup csu and rdc implementation
9b8cf9247 MLK-23969 imx: Update SCFW API message type according to latest SCFW
8cd4fa6df MLK-23821-04 plat: imx8m: Fix the rank to rank issue
e9dfde639 MLK-23821-03 plat: imx8m: Fix the dfiphymaster setting after dvfs
57bf8a00a MLK-23821-02 plat: imx8m: update the ddr4 dvfs flow to include ddr3l support
bc9bc03f3 MLK-23821-01 plat: imx8m: Correct the rank number get from mstr
---
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
| |
- Drop the mxc_v4l2_test patch - the compilation error fix is included in the source code
- Add pxp_test patch to fix format-security error
Signed-off-by: Cristinel Panfir <cristinel.panfir@nxp.com>
|
|
|
|
|
|
|
|
|
|
| |
The i.MX8 QuadMax and QuadXPlus SoCs can use the ION allocator.
Also, if there is a DPU, then G2D will be emulated via the DPU. However,
the emulation's allocation functions are broken, so disable the G2D
allocator then.
Signed-off-by: Carlos Rafael Giani <crg7475@mailbox.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This includes following commit:
88c58453be8 Revert "pico-imx7d: Allow bringing up the two cores"
We are going to fix this on Linux kernel side; the new commit:
b31249c40147 ARM: imx: add smp support for imx7d
Is being included on our 5.4.x+fslc branch to solve this for i.MX7D
users.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Change-Id: Ieaa700a864169be7c2b75bffe9be4111a59551df
|
|
|
|
|
|
|
|
|
|
| |
The following changes are included:
c28472955be pico-imx7d: Allow bringing up the two cores
e2cb327cff2 mx6ul_14x14_evk: Change environment offset to overlap with U-Boot
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Change-Id: I0ebc39d19556f1a7e65138ac5af1a6753d9fe4f0
|
|
|
|
|
|
|
|
|
|
|
| |
I got under mistake here and moved the SDMA firmware files when the only
ones which should have been moved were the VPU ones. This commit
partially reverts 979cac97 "firmware-imx: Move VPU and SDMA firmware
files to the correct path"
Fixes: 979cac97 "firmware-imx: Move VPU and SDMA firmware files to the correct path"
Change-Id: Ic380889dbbba673311fce93263ea402f81c5c871
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
|
|
|
|
|
|
|
|
|
|
| |
During update the location of SDMA and VPU firmware files were moved to
the 'imx' sub-directory, inside /lib/firmware, but it is not where the
Linux kernel looks for it. We moved the those to /lib/firmware/sdma and
/lib/firmware/vpu, respectively, so the probe now works as expected.
Fixes: b3eff3e4 "firmware-imx: upgrade to version 8.8"
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Change-Id: I42396f9bb606e8c4761c368150dba5090c66bf4c
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This fixes following build error:
,----
| pxp_lib_test/pxp_test.c: In function 'main':
| pxp_lib_test/pxp_test.c:541:2: error: format not a string literal and no format arguments [-Werror=format-security]
| 541 | printf(usage);
| | ^~~~~~
`----
Fixes: #506.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Change-Id: Ie6df1a64ba7126967861cb8fc889d4ca6f22b67a
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
With the current call to imx-mkimage and SOC_TARGET = "iMX8M" the u-boot
device tree file name is set to $(PLAT)-evk.dtb in iMX8M/soc.mak which
happens to match ${UBOOT_DTB_NAME} if building for an NXP machine.
Setting the dtbs environment variable to ${UBOOT_DTB_NAME} makes the build
succeed also for other machines.
In the hardcoded version $(dtbs) is an intermediate file and gets delete
as the last step of make. Thus for several make targets we have to redo
the copy step before each make invocation or the second one will fail as
the device tree file is then missing.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
|
|
|
|
|
|
|
|
|
| |
Introduce recipe which produces i.MX VC8000E Encoder
library package.
VC8000E Video Encoder is integrated in i.MX8M Plus SoC.
Signed-off-by: Andrey Zhizhikin <andrey.z@gmail.com>
|
|
|
|
|
|
|
| |
Introduce new recipe, which produces Sound Open Firmware package
required for imx8mp machine.
Signed-off-by: Andrey Zhizhikin <andrey.z@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Update all recipe PN to match the one from upstream.
Latest delivery from NXP has VPU firmware for imx27 in the package, so
enable it to be included in the image.
It also also lacks the VPU firmware for imx8qxp, so disable the package
for now until it is provided from NXP.
firmware-imx has been re-worked to provide a more unified mechanism for
installing firmware files into ${D} and package split is expalined in
the comments section.
NOTE: This update uses new EULA v11 from NXP.
Signed-off-by: Andrey Zhizhikin <andrey.z@gmail.com>
|
|
|
|
|
|
|
|
|
| |
NXP release imx_5.4.24_2.1.0 includes support for wm8960 codec, which
could be found on various EVKs of i.MX8 derivatives.
Update file to include support for the codec.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
|
|
|
| |
Upstream component version has been updated, align recipe to pick up the
new version.
This version is licensed under new LA_OPT_NXP_Software_License v10
dated December 2019.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
|
| |
This upgrade contains of following modifications:
- introduce new i.MX8 derivatives;
- upgrade imx-mkimage recipe to match NXP release;
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
|
|
|
| |
Upstream version has been bumped up, update the recipe to match it.
SECO FW name is picked up from the machine definition, as it is now
defined in a separate bbclass which allows FW names to be set based on
machine's SoC.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
| |
New version also uses new version of EULA from NXP and excludes i.MX8M
series in COMPATIBLE_MACHINE.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
| |
This adds support for additional i.MX8-based machines
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Drop patch appplied upstream, remove additional install step which is
handled by imx-boot recipe.
Update u-boot dtb names in machine description files, upstream U-Boot
from NXP has DTB files renamed.
Upgrade u-boot-imx-tools and u-boot-imx-mfgtool to 2020.04 as well to
match u-boot-imx recipe.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
| |
Backport patch required to fix build with -fno-common.
Signed-off-by: Ricardo Salveti <ricardo@foundries.io>
|
|
|
|
|
|
|
|
|
|
|
| |
GCC 10.2 is recently configured to include -fcommon per default, which
results in compilation error with multiple definitions of enum names.
Intoduce a patch which corrects double-defined enum names and resolves
compilation of the component.
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
Cc: Tom Hochstein <tom.hochstein@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This version uses updated EULA and limits compatibility to mx8mq, mx8mm
and and mx8mp machines (mx8mn does not have video accelerator IP block).
Following commits are added in this release (including command):
$ git log --oneline --no-decorate 904a9ab..ee99dc1
ee99dc1 NVJM850-82 - [RV8] RV8 video can't be decoded
b2aa5a0 M865SW-147: [G1/G2]: enhancement for h264 error handing
cfffc5f MMIOT-418: Allow to configure CFG_SECURE_DATA_PATH on Linux
0a741cc NVJM850-81 - [G2]:hevc error handling
2f3f4fa M865SW-132: [G1G2]: HEVC decoder illegal memory access in
security mode
8182f16 MA-14108-1 Remove ION_FLAG_CACHED flag for hantro vpu buffer
65d5f3a MMFMWK-8595 Enable VC8000E encoder for 865
b434f22 M865SW-87:[G1/G2]: Integrate some code for VSI platform
4dc2d6f M865SW-82 - [G1/G2]: H264 conformance test failure caused by
change in 1.0.6 release
dd606d2 [G1/G2]: Integrate VSI 1.0.7_20191212 release
d62fd74 [G1/G2]: Coverity issues fix
6460a5f [G1/G2]: Refine Test Makefile
7805f7f [G1/G2]: Coverity issues fix
eca7409 [G1/G2]: Coverity issues fix
a210f04 [G1/G2]: Coverity issues fix
dd0e5f4 [G1/G2]: Coverity issues fix
c12e5a8 [G1/G2]: Coverity issues fix
7631926 [G1/G2]: Coverity issues fix
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
Cc: Tom Hochstein <tom.hochstein@nxp.com>
|
|
|
|
|
|
|
|
|
| |
Following commits are added in this release (including command):
$ git log --oneline --no-decorate 3e47ffe..d8f988d
d8f988d MA-14108 Remove ION_FLAG_CACHED flag for cnm vpu buffer
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
Cc: Tom Hochstein <tom.hochstein@nxp.com>
|
|
|
|
|
|
|
| |
This includes one commit from upstream:
d6afb27 Always use ipv6 address to mount rootfs
Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin@leica-geosystems.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
0001-fix-fiptool-build-error.patch fixes below issue:
| gcc fiptool.o tbbr_config.o -o fiptool -lcrypto
| /opt/ci/yocto/dunfell/bld-lx2160ardb/tmp/hosttools/ld: cannot find -lcrypto
| collect2: error: ld returned 1 exit status
| Makefile:39: recipe for target 'fiptool' failed
| make[1]: *** [fiptool] Error 1
The issue was fixed in atf_git.bb in below commit from meta-freescale:
commit: fa4994b2d8da73fd3ae30875b3af8d7a682cdd53
author: Daniel Klauer <daniel.klauer@gin.de>
atf: Fix builds of native tools during build process
0001-Makefile-add-CC-gcc.patch fixes below issue:
| byte_swap.c:9:10: fatal error: stdio.h: No such file or directory
| 9 | #include <stdio.h>
| | ^~~~~~~~~
| compilation terminated.
| create_pbl.c:10:10: fatal error: stdio.h: No such file or directory
| 10 | #include <stdio.h>
| | ^~~~~~~~~
| compilation terminated.
The issue was fixed in atf source code:
commit: 9aae9712ec09ddeae528a1896247e5c432a6a19f
author: Biwen Li <biwen.li@nxp.com>
plat/nxp/tools: fix a makefiles bug that will use default implicit rules
build and test have no problem without them.
Signed-off-by: Ting Liu <ting.liu@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Newer buildhost distributions as Fedora 32 come with GCC10+, which causes:
gcc -isystemtmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/usr/include -O2 -pipe -Ltmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/usr/lib -Ltmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/lib -Wl,--enable-new-dtags -Wl,-rpath-link,tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/usr/lib -Wl,-rpath-link,tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/lib -Wl,-rpath,tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/usr/lib -Wl,-rpath,tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/recipe-sysroot-native/lib -Wl,-O1 -Wl,--allow-shlib-undefined -Wl,--dynamic-linker=tmp/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2 -o scripts/dtc/dtc scripts/dtc/dtc.o scripts/dtc/flattree.o scripts/dtc/fstree.o scripts/dtc/data.o scripts/dtc/livetree.o scripts/dtc/treesource.o scripts/dtc/srcpos.o scripts/dtc/checks.o scripts/dtc/util.o scripts/dtc/dtc-lexer.lex.o scripts/dtc/dtc-parser.tab.o
tmp/hosttools/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x10): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here
collect2: error: ld returned 1 exit status
make[3]: *** [scripts/Makefile.host:106: scripts/dtc/dtc] Error 1
make[2]: *** [tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/git/scripts/Makefile.build:432: scripts/dtc] Error 2
make[1]: *** [tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/git/Makefile:528: scripts] Error 2
make[1]: Leaving directory 'tmp/work/lx2160acex7-poky-linux/u-boot-qoriq/2019.10+fslgit-r0/build/mdr_gt_tfa_defconfig
'Backporting' fix from mainline to this failure.
Signed-off-by: Jens Rehsack <sno@netbsd.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Generally, rcw are deployed to ${DEPLOY_DIR_IMAGE}/rcw/${MACHINE}/.
But there are some machines that don't follow this rule. For example,
lx2160ardb deployed RCW in folder lx2160ardb_rev2 for boards with
Rev2 processor, but folder lx2160ardb for boards with Rev1 processor.
As there are only difference on RCW for them, use variable RCW_FOLDER
to allow selecting rcw folder for building atf targets. For exmaple,
to build atf which supports lx2160ardb with Rev2 processor:
RCW_FOLDER_lx2160ardb = "lx2160ardb_rev2"
Signed-off-by: Ting Liu <ting.liu@nxp.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Forward to commit 1e55b2f9e7 (LSDK-20.04-update-290520), which contains
since LSDK-20.04 (3cd9bc3993):
1e55b2f9e7 lx2160a : Update eMMC boot environment variable
98247c2b65 mmc: fsl_esdhc: workaround for hardware 3.3v IO reliability issue
9cd843a51c mmc: fsl_esdhc: read register once for card inserted status
8ded21cab5 configs: ls1012afrwy: drop env qspi_bootcmd
8911beb6ca armv8: ls1028a: clean up the environment variables
93d1089cbb configs: ls1012afrwy: fix board name
aab85e08a4 configs: ls1046aqds: support distro boot
d78a7f85ec configs: ls2080aqds: support distro boot
8c65af759f configs: ls1088aqds: support distro boot
eae5fa98d3 include/configs: ls1012aqds: add default environment variable
9a9f2f8f34 armv8: ls1028aqds: add some environments
be032ac3b4 configs: ls1028aqds: add lpuart config
bd2c372fdc armv8: ls1028aqds: add lpuart dts support
e8c89a28e7 arm: dts: ls1028a: add lpuart nodes
c4cfcc2a34 board: freescale: ls1028a: mux changes for lpuart
Signed-off-by: Jens Rehsack <sno@netbsd.org>
|
|
|
|
|
|
|
|
| |
Forward rcw to commit e0fab6d (LSDK-20.04-update-290520), which contains:
e0fab6d lx2160a_rev2 : Add proper dividers for flexspi for top platform freq
05975f0 lx2160aqds_rev2: add RCW for 2000_800_2900_13_5_2
Signed-off-by: Jens Rehsack <sno@netbsd.org>
|
|
|
|
|
|
|
|
|
| |
Bumping SRCREV to fbc036b for tag LSDK-20.04-update-290520, which contains
following changes since 14d03e6 (LSDK-20.04):
fbc036b lx2: ddr: phy: add fip file for 2019.04 firmware
f56e808 lx2: ddr: phy: Add 2019.04 ddr phy binaries
Signed-off-by: Jens Rehsack <sno@netbsd.org>
|