*include the following changes:
bda2c9 - .gitmodules: use codeaurora url 7db258d - Merge pull request #520 in GITAM/odp from ~NXA07732/odp:fsl_odp_v16.07_qoriq to fsl_odp_v16.07_qoriq 6a2a62f - Updating the sub-module path to bitbucket from stash. 2923db9 - dpaa2: Fix odp restart issue in scheduler mode 946d32e - odp: Update qbman commit to 75ff61a7ca6acdbdbb780161b053cbcbc990f1be e354952 - Removing atomic/ordered context while enqueuing the buffer bef6398 - TM: Invalid flow crash fix 8e7a136 - Removed device based jiffies df492ef - Fixing legal issues 3e9c5da - Scripts: binaries path and odp_tm test cases updated c91dac8 - Werror flag enabled for ODP library ced3aaf - Updated MC flib to version 10.3.2 a68cd87 - odp_pktio_stats API updated 03e0e05 - LD_LIBRARY_PATH removed from loopback_script 7a64406 - CUNIT: Cunit installation enabled 048c6a6 - script: ODP library path fixed 16a0e99 - Removing local compilation patch from QBMAN repo b792e84 - Merge pull request #505 in GITAM/odp from ~B38457/odp:fsl_odp_v16.07_qoriq to fsl_odp_v16.07_qoriq 281a676 - Merge pull request #487 in GITAM/odp from ~B46187/odp:fsl_odp_v16.07_qoriq to fsl_odp_v16.07_qoriq 8e2888b - odp_dpaa2: Patch to replace build type flag from compile time to runtime. 26e4bdc - Fixed odp_pktout_send API fails to transmit > 8 packets c46b0bb - Patch to odp_l2fwd issue in queue mode.
*fix building error with GCC8
update CFLAGS
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*include the following changes: 2815572 - dpdmai: add --num-queues create parameter a5f268b - scripts: suppress unit_address_vs_reg warnings when using newer dtc 0e2846a - scripts: Fix MC version check d9cbf56 - dprc: use the corect specifier for uint16_t 1b49539 - dpni: print both rx and tx tcs number 10af575 - dpni: add support for 16 TX tcs ad33877 - dpni: add DPNI_OPT_SINGLE_SENDER option 887deae - scripts: Infer switch ports number from specified endpoints c8111df - scripts: Avoid word splitting 7a15d3e - ls-addsw: show netdev names when creating a dpsw 81fb178 - ls-addmux: fix endpoint discovery 254aef8 - update copyright years
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Otavio Salvador [Thu, 5 Jul 2018 20:20:29 +0000 (17:20 -0300)]
imx-test: Rework DEPENDS and VPU support
All variants depends on `virtual/kernel`, `imx-lib` and `alsa-lib`, so
this can be added by default.
The VPU support has been moved to a `PACKAGECONFIG` so it adds the
respective `make` parameter, and dependency, depending on the
`MACHINEOVERRIDES` value.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Tom Hochstein [Fri, 22 Jun 2018 20:17:19 +0000 (15:17 -0500)]
weston: Set package arch for all i.MX
Weston is built from a fork for all i.MX, not just
those with GPU, so set the package arch accordingly.
Change-Id: Ib8b2e639a8393dd41832c972d21812ee990a6c47 Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Tom Hochstein [Thu, 5 Jul 2018 18:18:39 +0000 (13:18 -0500)]
mesa-demos: Fix condition for GLES support
The condition for GLES support was incorrect and would
remove GLES for parts with both 2D and 3D. Fix the condition
so GLES is removed only for parts with 2D and no 3D.
Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*Fix the below error:
|libteec/src/teec_trace.c:110:24: error: '%s' directive output may be truncated
|writing up to 255 bytes into a region of size 246 [-Werror=format-truncation=]
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Following changes are included: 6a7f2ec - make: account for changes in the iproute2 4.15 sources 9231d65 - ceetm: dpaa2: Only print shaping info at root class level 85fd3f0 - ceetm: dpaa2: Update help text f98be8a - ceetm: choose between dpaa1/dpaa2 APIs at runtime ca49b40 - ceetm: add a DPAA version detector call 8d77ae9 - ceetm: dpaa2: rename the config type enum to reflect the arch aa9ef30 - ceetm: dpaa2: add the ceetm qdisc implementation a743094 - ceetm: dpaa1: rename the config type enum to reflect the arch c22d3d0 - ceetm: dpaa1: call the ceetm API from generic qdisc ops 4e24e6a - ceetm: dpaa1: move the ceetm API to a separate file d5b318f - ceetm: dpaa1: rename the ceetm ops to dpaa1_ceetm ops
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*Following changes are included: fe3b3da - Changed the target name from ls1012frwy to ls1012afrwy. 408ab3a - Merge branch 'master' of https://bitbucket.sw.nxp.com/scm/dash/optee_os c925812 - Merge pull request #6 in DASH/optee_os from ~NXA19713/optee_os:master to master f0451fd - Fix for PLATSEC-522 6d1e7cf - Merge pull request #5 in DASH/optee_os from ~NXA18717/dash_optee_os:master to master e0f2725 - plat-ls: Add support for armv8 platform flavour d812df8 - plat-ls: Add support for armv8 platform flavour
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*Include the following changes: 24d3133 - SOBJ_ENG_APP: Initialising uninitialised variables f979dc4 - SOBJ_APP: Adding option for creating fake pem file while importing object 2adc118 - APP: Miscellaneous changes 10f5c67 - APP: Resolving coverity issues 18d4422 - SEC_OBJ_ENG: Adding check if multiple objects are created with same index 9e21913 - Commit Update: SOBJ_APP:Adding option to save fake pem file
*Update COMPATIBLE_MACHINE for ls1012a
*Add ARCH for install path
*Fix install path issue.
*Add INSANE_SKIP for dev-elf,ldflags and dev-deps to fix QA Issue.
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Following changes are included: 24eec41 - Merge pull request #11 in DQNS/ovs-dpdk from ~NXA20386/ovs-dpdk-nxa20386:branch-2.9-qoriq to branch-2.9-qoriq 821dd5d - Merge remote-tracking branch 'upstream/branch-2.9' into branch-2.9-qoriq a34504b - Revert "netdev-dpdk: toggle descriptor size for ports" 1909a65 - Revert "netdev-dpdk: fix incorrect exclude mbuf configuration" 5cb8d0f - Prepare for 2.9.3. 7e61467 - Set release date for 2.9.2. c16e265 - ovn pacemaker: Fix promotion issue when the master node is reset 418be3d - dpif-netdev: Free packets on TUNNEL_PUSH if may_steal. adeda41 - netdev-dpdk: fix check for "net_nfp" driver 3e417c8 - netdev-dpdk: Don't use PMD driver if not configured successfully 9fc83bf - netdev-dpdk: Remove use of rte_mempool_ops_get_count. 2a3ee45 - datapath: compat: Fix ndo_size in RHEL 7.5 backport f26b724 - Prepare for 2.9.2. f8b6477 - Set release date for 2.9.1. bd4b503 - conntrack-tcp: Handle tcp session reuse. d5176ed - ovn: Set proper Neighbour Adv flag when replying for NS request for router IP 9260243 - datapath: compat: Fix build on RHEL 7.5 72a73e4 - sparse: Support newer GCC/glibc versions. c00f0c2 - tunnel: make tun_key_to_attr aware of tunnel type. 00e56c5 - netdev-dpdk: Fixed netdev_dpdk structure alignment db136da - Configurable Link State Change (LSC) detection mode
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Otavio Salvador [Thu, 26 Apr 2018 16:52:55 +0000 (13:52 -0300)]
wic: Fix image generation for i.MX MXS SoC family
The image needs to use the U-Boot generated, for the uSD card, so a
dependency on that must be added to ensure the file is available prior
'wic' uses it.
Besides that, the fstype needs to be changed so it sets the partition
mark allowing the bootrom to find it.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
We need to keep a reference to the original MACHINEOVERRIDES value so
it can be reprocessed as need. This allow the compatibility with
existing BSP's while fixing the complex multilib interactions.
I'd like to record that Richard Purdie was key to solve this issue.
Its implications were beyond my understanding until he guided me and
provided a prototype fix.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*include the following changes: e34ec59 - apps: update elfs before tag f24816b - firmware no.: 8.2.1 6e9f4b5 - aiopsl : Fix compile warnings and cosmetics c19e5be - aiopsl: fix only one core processing packets issue 95ef568 - aiopsl: add support for dpni label
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
This includes the following change: e9abf79 - LS1088ARDB: Change DPL address SD in input file 74d1ebe - LS1088ARDB: Change DPL address in input file 1e600c9 - Merge pull request #87 in SDK/cst from ~NXA27423/cst-ls1012:master to master 11e6437 - Merge pull request #90 in SDK/cst from ~NXA08484/cst:master to master 7abaf13 - create_header: Add command line options f581f73 - Changes based on utilities supported in OpenSSL 1.1.0. 4daeb19 - LS1012AFRWY: Change kernel address in inputfile 20837ba - CH3: Fix issue in SRK offset Calculation 42f33a8 - LS1012AFRWY: Fix error in generating srk_hash.txt with flexbuild be43538 - LS1012AFRWY: Add input files for the platform
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*include the following changes: 1f43bef - Fix PBI command length for both ls1088ard and ls1088aqds boards. 32f8b4f - ls2088ardb_rev1.1: Add rcw_2100.rcw for serdes protocol 2a_41 84078ad - Include auto swapped binaries for qspiboot during install target 8646cf1 - ls2088ardb: Add rcw_2100.rcw for serdes protocol 2a_41 591b59d - ls1012a: Add ls1012afrwy RCW configuration files fd6b9ef - Add support to auto swap rcw for qspiboot if required. 6f514c1 - Ls1043aqds: Fix building error for rcw_1600.rcw
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
*include the following changes: dcbe27b - Adding policy.h flag for spectre v4 mitigation 4fbad6a - Fix suppression flag on sec_init 24d6891 - Adding SUPPRESS_TZC and SUPPRESS_SEC to policy cfg file; Adding platforms qds & rdb to device LX2160; Adding platform emu t 006a0e5 - Correcting disable load-pass-store in CPUACTLR from bit [53] to bit [55] 8e557f9 - Adding spectre v4 mitigation for a57, a72 cores. b0a07cf - Updating reset function for LX2160. 74814cf - Increasing L3 cache flush timeout from 100ms to 333ms. 9d5ecd2 - Enabled OP-TEE support on LS1012FRWY baord. 9950f2b - Removing wfe from core 0 sequence when waiting for soc init tasks to complete. 508ed94 - Enabling initialization of the CCN-508 interconnect. fcede9a - Adding support for ls1043 variant ls1023, and ls1046 variant ls1026. 7d5dcc1 - Fixing PSCI_SYS_OFF so that this function never returns. 8e8fc09 - Removing \r from line endings of shell script gen_its. 1c4fc78 - its: Correct config node as per upstream u-boot guideline f00e7ef - Adding smc function SIP_HW_UNQ_KEY_32. c87863a - Fixing usage of _check_sec_disabled(). d0ec59c - Fix for PLATSEC-437 & PLATSEC-457. 926fb97 - Adding smc function (SIP_ALLOW_L2_CLR_32) to allow clearing of L2 memory errs. 781d7b5 - Fixing broken build of LS1043, LS1012. 7289e86 - Adding smc function SIP_ALLOW_L1L2_ERR_32; Added test for SPD interface. 8804017 - Providing very fast-path for SMCC_ARCH_WORKAROUND_1 e9dfa28 - Moved _soc_check_sec_enabled() to _check_sec_disabled(); Fixed conflicting defs in LX2160 soc.h. 474b797 - Changing LX2160 cluster 3 handling from RCW frequency-based to RCW[850] selectable. f56bd49 - fuse: Correct OTPMK value mapping to OTPMK registers
Signed-off-by: Chunrong Guo <chunrong.guo@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
mx7dsabresd: Remove the mx7dsabresd_secure_defconfig target
mx7dsabresd_secure_defconfig was introduced to allow booting NXP kernel
that has CAAM support and needs to boot in secure mode.
Instead of keeping two different config targets for the same board,
remove mx7dsabresd_secure_defconfig and select
CONFIG_ARMV7_BOOT_SEC_DEFAULT inside mx7dsabresd_defconfig so that
this target could be used to boot both mainline and the vendor kernel.
This makes maintenance task easier and avoid potentially confusion
for the end user.
Tom Hochstein [Fri, 22 Jun 2018 20:17:18 +0000 (15:17 -0500)]
imx-base.inc: Add imxgpu to MACHINEOVERRIDES_EXTENDER
Add imxgpu to simplify the conditional for imxgpu2d OR
imxgpu3d. Note that i.MX 8MQ has no 2D, so the previous
idiom of using imxgpu2d to mean any GPU is no longer correct.
Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Otavio Salvador [Fri, 29 Jun 2018 16:56:17 +0000 (13:56 -0300)]
imx-base: Use i.MX's fork of libdrm for i.MX6 and i.MX7
There few components which require the new libdrm fork, so we must use
it when using NXP's BSP components. Besides adding the preferred
version we need also to add it to the SoC architecture filter.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Tom Hochstein [Fri, 22 Jun 2018 20:17:08 +0000 (15:17 -0500)]
imx-dpu-g2d: Add new G2D implementation based on DPU
Change-Id: Ib3114b9962a2bb89e95da8757870050c1d816abd Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Fabio Berton [Mon, 18 Jun 2018 19:07:18 +0000 (16:07 -0300)]
gstreamer1.0-plugins-imx: Bump to revision 963aea6
- Rename recipe to use git package name
- Set PV to use latest version plus git revision
- Add O.S. Systems License
This commit includes the following changes:
- 963aea6 Merge pull request #180 from clopez/bug-179
- 81d3233 Merge pull request #188 from dougnazar/fixes
- 68715a6 g2dpango: Add missing header and cleanup a few warnings
- f4aa2db audio: Save codec_data
- 2b2edd7 Endless loop at gst_imx_vpu_decoder_finish() when in no_wait mode (#179)
- ecd884a v4l2: Add phys mem meta only if the driver/card is known to support this
- 9772200 ipu: Reintroduce interlace methods
- 3404b56 v4l2video: Define V4L2_COLORSPACE_DEFAULT if it doesn't exist
Fabio Berton [Mon, 18 Jun 2018 17:57:02 +0000 (14:57 -0300)]
xf86-video-imxfb-vivante: Update to version 6.2.4.p1.2
- Remove .inc file to use only one .bb file
- Update O.S. Systems License
- Remove unused patch
- Set SDKTARGETSYSROOT to use STAGING_DIR_HOST
- Remove unnecessary CFLAGS
Fabio Berton [Thu, 14 Jun 2018 18:56:09 +0000 (15:56 -0300)]
kernel-module-imx-gpu-viv: Update to version 6.2.4.p1.2-based fork
This includes the following changes:
- 8635400 Merge tag 'upstream/6.2.4.p1.2'
- 9bbacfe Import 6.2.4.p1.2 original version
- 26c273f Kbuild: fix AQROOT reassignment and remove CONFIG_MXC_GPU_VIV
- 09b4ae4 Merge tag 'upstream/6.2.4.p1.0'
- 7957949 Import 6.2.4.p1.0 original version
- 3b9e057 Merge tag 'upstream/6.2.2.p0'
- b3604bf Merge pull request #8 from gibsson/upstream
- 0d83ec5 Import 6.2.2.p0 original version
Fabio Berton [Mon, 11 Jun 2018 18:54:36 +0000 (15:54 -0300)]
imx-test: Update to version 7.0
- Remove .inc file to use only one .bb file
- Add alsa-lib to DEPENDS
- Add patch to use include path when compile sources from test directory
- Drop unused patch
- Update O.S. Systems Software Copyright
Fabio Berton [Mon, 11 Jun 2018 14:23:26 +0000 (11:23 -0300)]
imx-gpu-viv: Update to 6.2.4.p1.2
Key Features of the Vivante 6.x Driver Series
- Full support for Khronos OpenGL ES 3.2 API. The new OpenGL ES 3.2 and OpenGL ES Shading Language 3.20 specifications bring AEP,
plus additional functionality, into core OpenGL ES. Vivante OpenGL ES 3.2 implementation has passed Khronos OpenGL ES 3.2 Conformance Test.
And Vivante GC7000XS* GPU core has been certified by Khronos as an OpenGL ES 3.2 compliant product.
- Backward Compatible with 5.x and 4.x Drivers. Existing OpenGL ES 3.x/2.0 applications can run on these drivers without any changes.
Key Features
============
- OpenCL improvements
- Wayland-weston DRM Support
- Wayland-weston with 2D accelerated renderer
- Wayland-Weston EGL DRM Modifier support
- Wayland-Weston EGL Framebuffer compression
- Wayland-Weston DMABUF support
- OpenCV tests with OpenCL
- Add GPU DRM support
- GLES CTS version 3.2.4
6.2.4P1 Improvements
==================
OCL: MGS-3255-34 add opencl memory optimzation
OCL: MGS-3255-33 fix opencl test buffer copy failure
OCL: MGS-3640 fix opencl corruption for arm_compute_library
OCL: MGS-3710 SoftISP meet CL_INVALID_WORK_GROUP_SIZE (-54) on mscale and qxp
OCL: cache invalidate for uncached buffer get accesing user space memory outsied uaccess.h error
OCL: ovx and ocl running together
EGL MGS-3711 met display blurred when run weston-nested
EGL: Enable extension EGL_EXT_image_dma_buf_import_modifiers
EGL: weston-nested met process aborted and core dump
EGL: MGS-3399 fix wayland egl resize cts failure
EGL: support direct rendering on external drm buffer
EGL: MGS-3840-2 fix 6qp display flicker with fbdev test
EGL: MGS-3791 imx6q shared egl context rendering issue
EGL: MGS-3744 fix GC355 OpenVG flicker on wayland
EGL: MA-11528 Support enable framebuffer direct rendering
GLES3.0: 6QP FB have CTS failures with ESCTS 3.2.2.0
GLES: MGS-3570 glGenerateMipmap takes too much time on 5.x/6.x driver
Weston: Disable surface compress if g2d-renderer
GBM: Enable DRM compression for mscale board
GBM: MGS-3625 6.2.4.p1: Fix xserver build break with GBM change
GBM :MGS-3694 enable fb fd and tile status fd in GPU and DCSS on wayland
Android:MA-11232 Enable HDR10 support on imx8mq platform.
Android: MA-10844 GPU driver pass build after refine some NXP git path
Android : MA-11101 Add HAL_PIXEL_FORMAT_NV12_TILED format to support VPU tile
QNX:iMX8QM MGS-2906 VDK app fail if it's started as the first app after screen
QNX:iMX8QM MGS-3289 [iMX8] Rotation using blitter seems broken
QNX:iMX8QM MGS-3338 - Deadlock seen with Screen API test
QNX:iMX8QM MGS-3323 QNX -[iMX8] Fix NV12 handling
iMX8QM Disable depth compression for LPDDR4 parts
OCL: MGS-3848-5 OCL1.2: test_image_streams failures
OCL:MGS-3832 cl_convolution of ACL cause MMU exception
OVX: MGS-3579 Fix the invalid scope of VX graph for 8QM
OVG: MGS-3826 met core dump and system prompt out of memory when do OpenVG memory leak test.
OVG: MGS-3750 fix gc355 openvg stuck issue on 6q and 6qp
OVG: MGS-3791 VGCTS failed to run on i.MX6SL wayland
OGL: MGS-3570 glGenerateMipmap takes too much time on 5.x/6.x driver
OGL: MGS-3741 glmark2 test met error message
General: MGS-3848 refinements for pm and isr
General: MGS-3840 Display resume back very slow and has snow screen after suspend and resume.
General: MGS-3751 met GPU core dump when run multiple es20 sdk apps
Known Issues
============
MGS-3770 [6q/6qp] can not generate correct reference file after do api retrace and sometimes met stuck
- Need wayland support in waffle. will be addressed in next release
MGS-3747 [6dl/6q] X11: copytex can't work after "Qt5_NMapper" app test.
YOCIMX-2354 Some qt3-3d demos not work. Issue related to QTwayland in 5.8 later
MGS-3739 can not pass webgl 1.0.2 CTS test. Chromium 64 known issue. Same reproduced on desktop
Signed-off-by: Fabio Berton <fabio.berton@ossystems.com.br> Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com> Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Fabio Berton [Fri, 8 Jun 2018 20:56:14 +0000 (17:56 -0300)]
linux-imx: Update to version 4.9.88
- Update to latest revision of 4.9.88
- Sync defconfig file with imx_v7_defconfig
- Remove do_configure_prepend from linux-imx-mfgtool, once
now the defconfig are the same imx_v7_defconfig
- Add and update O.S. Systems Software Copyright
