3 years agoi2c: PDK-6672: fix i2c eeprom bios application bug REL.CORESDK.07.00.00.29 REL.CORESDK.07.00.00.30 REL.CORESDK.07.00.00.31 REL.CORESDK.07.00.00.32
i2c: PDK-6672: fix i2c eeprom bios application bug
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
SBL: Update fix for MCU OCM RAM region for all SoC.
Adds the max possible MCU OCM RAM memory in this R5 MPU
region, to cover all possible KS3 SoCs with MCU RAM.
Doesn't need to be SoC-specific here.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Adds the max possible MCU OCM RAM memory in this R5 MPU
region, to cover all possible KS3 SoCs with MCU RAM.
Doesn't need to be SoC-specific here.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
IPC: echo test: Fix stack overflow in c7x echo test
The ipc_echo_test will sometimes throw stack overflow
error from the app task. This patch increases the size
of the task stack to avoid the error.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
The ipc_echo_test will sometimes throw stack overflow
error from the app task. This patch increases the size
of the task stack to avoid the error.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
PDK-6623:MCASP: Fix for device loopback example on c66xdsp_2
3 years agoapiguide:timeSync: Fix API guide build warnings REL.CORESDK.06.02.04.03 REL.CORESDK.06.02.04.04
apiguide:timeSync: Fix API guide build warnings
- Fix TimeSync API guide doxygen build warnings
Signed-off-by: Santhana Bharathi N <santa@ti.com>
- Fix TimeSync API guide doxygen build warnings
Signed-off-by: Santhana Bharathi N <santa@ti.com>
OSAL/test: add DebugP Log test for TPR12
- Enable the test for TPR12 only because BIOS-based trace is not required
for other SoCs.
- Although this test is applicable for all SoCs, the local BIOS config file
is required to enable the XDC/BIOS log operation.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- Enable the test for TPR12 only because BIOS-based trace is not required
for other SoCs.
- Although this test is applicable for all SoCs, the local BIOS config file
is required to enable the XDC/BIOS log operation.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
SBL: Fix J7 MCU OCM RAM region for SBL CUST builds
SBL CUST build on J7 family devices must support MCU-only
mode where only MCU domain resources are used for initial
bringup, even for signed applications. Full MCU OCM RAM
space must be utilized for app image verification in this
case.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
SBL CUST build on J7 family devices must support MCU-only
mode where only MCU domain resources are used for initial
bringup, even for signed applications. Full MCU OCM RAM
space must be utilized for app image verification in this
case.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Fix package issue for uniflash due to HS changes
- Uniflash target now gets build for HS and non-HS
from different dir
- So don't use relative patch for SRC and INC DIR
- Also fixed package variables and skip for HS build
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Uniflash target now gets build for HS and non-HS
from different dir
- So don't use relative patch for SRC and INC DIR
- Also fixed package variables and skip for HS build
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PRSDK-7753: HS build support
- The APPNAME or app -C directory (_PATH variable) should be
unique as the complier creates lto_$APPNAME optimization
file in the directory in which -C is called
- Because of this multiple lto_ files with same name can be
created if app name and app path are same
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- The APPNAME or app -C directory (_PATH variable) should be
unique as the complier creates lto_$APPNAME optimization
file in the directory in which -C is called
- Because of this multiple lto_ files with same name can be
created if app name and app path are same
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PRSDK-7753: HS build support
- Added SCICLIENT API to return the default board cfg
address and size
- This way the caller need not include the private header file
which has the size macros based on BUILD_HS flag
- This way the caller need not be built special way using the
BUILD_HS macro
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Added SCICLIENT API to return the default board cfg
address and size
- This way the caller need not include the private header file
which has the size macros based on BUILD_HS flag
- This way the caller need not be built special way using the
BUILD_HS macro
Signed-off-by: Sivaraj R <sivaraj@ti.com>
spi: qspi-flash-test: use CSL address translation function for TPR12
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
sciclient: am64: Fix errors in gSciclientMap
The host mapping array, gSciclientMap had incorrect
entries defined. This patch removes all the incorrect
entries and adds a TBD M4 secure entry to match
sciclient_fmwMsgParams.h
Fixes: PDK-6724
Signed-off-by: Justin Sobota <jsobota@ti.com>
The host mapping array, gSciclientMap had incorrect
entries defined. This patch removes all the incorrect
entries and adds a TBD M4 secure entry to match
sciclient_fmwMsgParams.h
Fixes: PDK-6724
Signed-off-by: Justin Sobota <jsobota@ti.com>
Revert "PDK-6624: WORKAROUND for SBL OSPI Boot instability on J721E-HS"
This reverts commit 2c2cca942ff1c3f29b40dd23797fee80a7f3f83e.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
This reverts commit 2c2cca942ff1c3f29b40dd23797fee80a7f3f83e.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
PRSDK-7492: Board: Fix for Uniflash host failure on Windows
- Uniflash host utility on Windows fails when executed after using
the COM port with TeraTerm. Serial cable disconnect and reconnect
is needed to avoid this failure.
Failure is caused by COM port timeout settings done by TeraTerm.
Uniflash is not using the timeouts and failures are observed with the
timeout values configures by TeraTerm. This update fixes the issue by
resetting the COM port timeout settings to default.
- Uniflash host utility on Windows fails when executed after using
the COM port with TeraTerm. Serial cable disconnect and reconnect
is needed to avoid this failure.
Failure is caused by COM port timeout settings done by TeraTerm.
Uniflash is not using the timeouts and failures are observed with the
timeout values configures by TeraTerm. This update fixes the issue by
resetting the COM port timeout settings to default.
Fix Host EMU remark
- Remove -v: Disable verbose
- Add -c: do not warn if the library had to be created
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Remove -v: Disable verbose
- Add -c: do not warn if the library had to be created
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PDK-6685 - bug fix - Timer osal - Uninitialized variable
- gTimerStructs is uninitialized, initialized to zero
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
- gTimerStructs is uninitialized, initialized to zero
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
drv/usrt/test: use CSL address translation function for TPR12
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
OSAL/test: reduce TaskP test period for QT build
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
sciclient: Verify last hop IR output against boardcfg
The intermediate interrupt router hops in an
interrupt route have their outputs validating
against the board configuration assignments.
This patch updates the code to also validate
the last hop IR output against the board
configuration resource assignments.
Signed-off-by: Justin Sobota <jsobota@ti.com>
The intermediate interrupt router hops in an
interrupt route have their outputs validating
against the board configuration assignments.
This patch updates the code to also validate
the last hop IR output against the board
configuration resource assignments.
Signed-off-by: Justin Sobota <jsobota@ti.com>
sciclient: j721e: Some of the IR instances are not initialized correctly
Some of the J721e IR instances do not have
their inp0_mapping field initialized properly.
This can cause a failure if attempting to route
to the 0th output for these IRs.
Fixes: PDK-6679
Signed-off-by: Justin Sobota <jsobota@ti.com>
Some of the J721e IR instances do not have
their inp0_mapping field initialized properly.
This can cause a failure if attempting to route
to the 0th output for these IRs.
Fixes: PDK-6679
Signed-off-by: Justin Sobota <jsobota@ti.com>
SBL Build Fix
- Also fixes the J7 SBL build for HS
- All other boot media except OSPI should use DMA mode
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Also fixes the J7 SBL build for HS
- All other boot media except OSPI should use DMA mode
Signed-off-by: Sivaraj R <sivaraj@ti.com>
AM64x SBL Fix
- AM64x SBL only non-DMA OSPI is supported
- use that library instead of dma ospi library
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- AM64x SBL only non-DMA OSPI is supported
- use that library instead of dma ospi library
Signed-off-by: Sivaraj R <sivaraj@ti.com>
3 years agoPRSDK-7753: HS build support - address review comments REL.CORESDK.06.02.03.10 REL.CORESDK.06.02.03.11 REL.CORESDK.06.02.04.01 REL.CORESDK.07.00.00.20
PRSDK-7753: HS build support - address review comments
- Now we don't have to pass BUILD_HS flag
- All targets are enabled by default
- The HS SBL will get generated in $(BOARD)_hs folder with
the same name
- Added variable SBL_CERT_KEY in make infra which can be
passed from individual SBL targets based on HS or non-HS SBL
- App image signed image is always generated for both HS and GP
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Now we don't have to pass BUILD_HS flag
- All targets are enabled by default
- The HS SBL will get generated in $(BOARD)_hs folder with
the same name
- Added variable SBL_CERT_KEY in make infra which can be
passed from individual SBL targets based on HS or non-HS SBL
- App image signed image is always generated for both HS and GP
Signed-off-by: Sivaraj R <sivaraj@ti.com>
i2c: ADASVISION-2476: fix check bus busy bug
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
OSAL: add debugP support
Note: only tirtos version is supported for now,
DebugP_log functions are implemented as stub
functions tentatively for nonos.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Note: only tirtos version is supported for now,
DebugP_log functions are implemented as stub
functions tentatively for nonos.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
sciclient: AM64: Fix mcu hosts assigned to IA VINT resources
The DMSS IA VINT ranges were mistakenly assigned
to the Secure MCU core hosts. This patch assigns
the VINT ranges to the non-secure hosts.
Signed-off-by: Justin Sobota <jsobota@ti.com>
The DMSS IA VINT ranges were mistakenly assigned
to the Secure MCU core hosts. This patch assigns
the VINT ranges to the non-secure hosts.
Signed-off-by: Justin Sobota <jsobota@ti.com>
SBL: Fix for packaging SBL when only OSPI is enabled
Also includes fix for building SBL from within ti/boot/sbl/build for
Am64x
Also includes fix for building SBL from within ti/boot/sbl/build for
Am64x
SBL: Disabling SBL targets not ready (yet) for AM64x
SBL:Deriving OSPI base/data addr from socGetConfig()
pdk: am64x: fix sbl build issue from the package
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-6642: Removed SDR_ECC_selfTest from J721E build.
Made SDR_ECC_selfTest() at start of Periodic tests only in AM65
build, since this test can fail on J721E.
Made SDR_ECC_selfTest() at start of Periodic tests only in AM65
build, since this test can fail on J721E.
[PDK-6644] UDMA AM65X Unit Test Issues Fix
-Update DMSC Events offset for AM65X with ref from https://bitbucket.itg.ti.com/projects/SYSFW/repos/system-firmware/browse/docs/public/5_soc_doc/am65x_sr2/interrupt_cfg.rst?at=refs%2Fheads%2Fdevel%2Fam64#6007
Signed-off-by: Don Dominic <a0486429@ti.com>
-Update DMSC Events offset for AM65X with ref from https://bitbucket.itg.ti.com/projects/SYSFW/repos/system-firmware/browse/docs/public/5_soc_doc/am65x_sr2/interrupt_cfg.rst?at=refs%2Fheads%2Fdevel%2Fam64#6007
Signed-off-by: Don Dominic <a0486429@ti.com>
Fix UDMA UT for C7x secure mode change
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
drv/i2c/test/master_slave: fix package build issue
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
[UDMA] Updated UDMA RM cfg for j721e to allow more block copy and fewer rx+tx channels for c6x-1, c6x-2
Bug Fix: PDK-6612: UDMA DRU Indirect Blockcpy DDR to DDR in polling mode fails for mcu2_1
- There are no DRU channels allocated for MCU2_1
- So disabled the DRU testcase for MCU2_1 and enabled it for MCU2_0
- Also there is no Main HC channel allocated for MCU2_2
- Removed that as well
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- There are no DRU channels allocated for MCU2_1
- So disabled the DRU testcase for MCU2_1 and enabled it for MCU2_0
- Also there is no Main HC channel allocated for MCU2_2
- Removed that as well
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Sciclient: Updating sysfw-vlab.bin with the latest
PDK-6624: WORKAROUND for SBL OSPI Boot instability on J721E-HS
WORKAROUND: On J721E-HS, the signed application is copied into DDR
where it is verified to be signed correctly. However, with OSPI
boot on the J721E-HS EVM, the verification is not proceeding in DDR.
If we change SBL_SCRATCH_MEM_START to use MSCM RAM, instead, then
it consistently boots fine.
The DDR stability should be fixed in CSL.
WORKAROUND: On J721E-HS, the signed application is copied into DDR
where it is verified to be signed correctly. However, with OSPI
boot on the J721E-HS EVM, the verification is not proceeding in DDR.
If we change SBL_SCRATCH_MEM_START to use MSCM RAM, instead, then
it consistently boots fine.
The DDR stability should be fixed in CSL.
pdk/build: remove utilsCopyVecs2ATcm.asm from TPR12 examples/tests
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
build/tpr12: cleanup linker command files
- Update L3 size to 0x390000
- Combine two L2 Banks into one
- Update HWA_RAM base
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- Update L3 size to 0x390000
- Combine two L2 Banks into one
- Update HWA_RAM base
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
drv/i2c/test: enable master_slave test for TPR12
- enable makefile build for TPR12
- add polling mode test
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- enable makefile build for TPR12
- add polling mode test
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
uart: am64x: fix unit test out of memory build issue for m4 core
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-6627: emac: Updates for Emac_Icssg_WithoutDDR_TestApp
fix sizing of ICSSG port queues
linker command and .cfg file updates to sync with build
infrastucture versions
Signed-off-by: Tinku Mannan <tmannan@ti.com>
fix sizing of ICSSG port queues
linker command and .cfg file updates to sync with build
infrastucture versions
Signed-off-by: Tinku Mannan <tmannan@ti.com>
SBL:AM64x: Update SMP example with updated SMP core IDs
SBL:AM64x: Using SBL_LAST_CORE_ID instead of specific core name
SBL:AM64x:Disable SBL log, fix M4F boot issue,cleanup OSAL test
-Disable SBL Logs
-Fix M4 boot issue
-Cleanup OSAL test for M4
-Disable SBL Logs
-Fix M4 boot issue
-Cleanup OSAL test for M4
SBL on AM64x: Addressing review comments
SBL: Disabling DDR Init for AM64x, bug fix for sbl_boot_test app on AM64x
PDK-5378: SBL: Add AM64x support
Signed-off-by: Mahesh Radhakrishnan <uda0875154local@UDA0875154>
Signed-off-by: Mahesh Radhakrishnan <uda0875154local@UDA0875154>
pdk: fix example build issue with GCC tools 9-2019-q4 for am3/4/5 and K2 devices
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
3 years agospi: qspi with local edma: move global address translation to test application REL.CORESDK.07.00.00.18 REL.CORESDK.07.00.00.19
spi: qspi with local edma: move global address translation to test application
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
spi: add build support for tpr12_qt
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
spi: address review comments for TPR12 qspi with edma support
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
spi: add edma support for TPR12 qspi
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
Bug Fix: PRSDK-8528: Patch 4
- This will break Yocto build as Yocto is XDC base build
- This flag can be overridden in Jenkin build scripts
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- This will break Yocto build as Yocto is XDC base build
- This flag can be overridden in Jenkin build scripts
Signed-off-by: Sivaraj R <sivaraj@ti.com>
v2020.04a migration
Migrate to v2020.04a release
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
Migrate to v2020.04a release
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
tools: firmwareHeaderGen.sh Update to generate separate header for HS
Create a separate header for HS
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
Create a separate header for HS
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
PDK-6556 : When built for mcu11, adding a dummy app for mcu10
This is required to ensure mcu11 is not in a higher
power states w.r.t to mcu10 when running in split mode
This same approach is used for mcu21, mcu31, etc...
Signed-off-by: sujith <sujith.s@ti.com>
This is required to ensure mcu11 is not in a higher
power states w.r.t to mcu10 when running in split mode
This same approach is used for mcu21, mcu31, etc...
Signed-off-by: sujith <sujith.s@ti.com>
PDK-6626: Unable to rebuild board LLD since top level makefile is not packaged
- This is because we use "jacinto" as the directory name when we package PDK now
- Due to this the PDK_SOC variable is set to jacinto when the steenv script is called
- Instead set a default SOC so that the compilation goes through by default
- User can always override the PDK_SOC variable if need to build for another SOC
in the same family
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- This is because we use "jacinto" as the directory name when we package PDK now
- Due to this the PDK_SOC variable is set to jacinto when the steenv script is called
- Instead set a default SOC so that the compilation goes through by default
- User can always override the PDK_SOC variable if need to build for another SOC
in the same family
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Added DRU trigger and event testcases
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
diag: sdr/sdtf_test: Fix typo error
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
PDK-6547: Board: Fix for AM65xx CPSW2G Ethernet diag test failure
- emac_poll function is not able to detect the PHY link up/down
which is making the diagnostic test to hang while running the
cable disconnect/reconnet test.
Updated the diagnostic test to use PHY registers to detect the
link up/down as a workaround till emac driver API is fixed.
- emac_poll function is not able to detect the PHY link up/down
which is making the diagnostic test to hang while running the
cable disconnect/reconnet test.
Updated the diagnostic test to use PHY registers to detect the
link up/down as a workaround till emac driver API is fixed.
PDK-6546: Board: Fix for AM65xx SPI NOR flash diag test failure
- SPI NOR diagnostic test is based on CSL for AM65xx. Need to pick
the source files specific to AM65xx in diag make file.
- SPI NOR diagnostic test is based on CSL for AM65xx. Need to pick
the source files specific to AM65xx in diag make file.
PDK-6614: Bug Fix: Ring attach failure
- The drv handle was not initialized due to
the recent changes in ring API for AM64x support
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- The drv handle was not initialized due to
the recent changes in ring API for AM64x support
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Bug Fix: PRSDK-8528: Patch 4
- Top level scripts files were missing
- There are part of https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/processor-pdk-packages/browse/release/full
repo.
- These were packaged as part of Yocto and was missed
- These are now maintained here and will be removed from Yocto flow
and from above repo as well
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Top level scripts files were missing
- There are part of https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/processor-pdk-packages/browse/release/full
repo.
- These were packaged as part of Yocto and was missed
- These are now maintained here and will be removed from Yocto flow
and from above repo as well
Signed-off-by: Sivaraj R <sivaraj@ti.com>
sciclient: am64: Update VINT resource allocation
Update the DMASS IA VINT resource allocation in
the RM board configuration so that the proper
hosts are assigned to the processors connected
to the VINT ranges.
Signed-off-by: Justin Sobota <jsobota@ti.com>
Update the DMASS IA VINT resource allocation in
the RM board configuration so that the proper
hosts are assigned to the processors connected
to the VINT ranges.
Signed-off-by: Justin Sobota <jsobota@ti.com>
sdr: sdtf-test: Add unity support for automation
Also update passing tests with default image for J721E
- CCM tests not covered part of run_all for J721E due to HW issue
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
Also update passing tests with default image for J721E
- CCM tests not covered part of run_all for J721E due to HW issue
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
PDK/Build: undo PDK-4378 changes for TPR12
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
pdk: update GCC tools to 9-2019-q4 for am3/4/5 and K2 devices
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-6625: Fix cust boot issue on J7HS
J7HS currently copies the appimage into MSMC to validate the HS signature. "CUST" boot only initializes the MCU domain, and thus does not have access to MSMC. As such, "CUST" boot must now use R5 OCMC as scratch memory to authenticate the image.
J7HS currently copies the appimage into MSMC to validate the HS signature. "CUST" boot only initializes the MCU domain, and thus does not have access to MSMC. As such, "CUST" boot must now use R5 OCMC as scratch memory to authenticate the image.
Updated the dependency for Doxygen rule
Signed-off-by: Aravind Batni <aravindbr@ti.com>
Signed-off-by: Aravind Batni <aravindbr@ti.com>
[PDK-4949][UDMA][Docs] Addressed additional review comments in Migration Guide
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
i2c: ADASVISION-2476: fix polling mode timeout issue
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PRSDK-7753: HS build support - address review comments
- Made changes so that make targets remain same for both HS and GP
- Only the library and generated SBL image path gets changed
- This is minimal required so that both HS/GP libraries and SBL
images can co-esit without a clean build
Signed-off-by: Sivaraj R <sivaraj@ti.com>
- Made changes so that make targets remain same for both HS and GP
- Only the library and generated SBL image path gets changed
- This is minimal required so that both HS/GP libraries and SBL
images can co-esit without a clean build
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PRSDK-8493: Fix the Am64x XDC packaging for remaining modules
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PDK-6604: SBL: Fix Boot perf test to fit in MCU SRAM
Builds the correct (smaller) Boot performance test for
all SBL CUST builds, by checking for the proper flag set
in the CUST set of build flags. If the SBL_ENABLE_DEV_GRP_MCU
flag is set for the CUST build set of flags, then the small
Boot performance test is used (which fits into MCU SRAM).
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Builds the correct (smaller) Boot performance test for
all SBL CUST builds, by checking for the proper flag set
in the CUST set of build flags. If the SBL_ENABLE_DEV_GRP_MCU
flag is set for the CUST build set of flags, then the small
Boot performance test is used (which fits into MCU SRAM).
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
3 years ago [PDK-4949][UDMA][Docs] Replaced LCDMA with DMSS REL.CORESDK.07.00.00.15 REL.CORESDK.07.00.00.16
[PDK-4949][UDMA][Docs] Replaced LCDMA with DMSS
-Also replaced UDMA(old) with NAVSS UDMA
Signed-off-by: Don Dominic <a0486429@ti.com>
-Also replaced UDMA(old) with NAVSS UDMA
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-4949][UDMA][Docs] Updates for PKTDMA
Highlights:
- 1.2 :- New
- 2.2 :- Updated + New Sup Sections
- 3.1.6 :- New
Signed-off-by: Don Dominic <a0486429@ti.com>
Highlights:
- 1.2 :- New
- 2.2 :- Updated + New Sup Sections
- 3.1.6 :- New
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-4949][UDMA][Docs] Addressed Review Comments
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-4949][UDMA][Docs] Application Note: Migrating_Applications_from_UDMA_to_LCDMA_using_TI-RTOS.doc
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
Minor Build issues resolution
Signed-off-by: sujith <sujith.s@ti.com>
Signed-off-by: sujith <sujith.s@ti.com>
PDK-5543 : Moving to a GA version of sysbios
Signed-off-by: sujith <sujith.s@ti.com>
Signed-off-by: sujith <sujith.s@ti.com>
board: Add am64x_svb board support
NOTE: AM64X SVB will use LPDDR4
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
NOTE: AM64X SVB will use LPDDR4
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
board: am64x_evm: Add board support for am64x_evm
NOTE: am64x_evm will use DDR4
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
NOTE: am64x_evm will use DDR4
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
board: j721e, j7200 evm: Update to cleanup hard coding and use CSL defines
The naming of BOARD_DDR_SS_BASE was wrong. It is actually the CTRL_CFG
register base
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
The naming of BOARD_DDR_SS_BASE was wrong. It is actually the CTRL_CFG
register base
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
sdr: sdtf-test: Move core specific define to be platform & core specific
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
PRSDK-8529: Update ESM Event numbers for J721E
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
PRSDK-8529: Update alignment of DED vector
With thumb mode this alignment to 32 byte is required to avoid exception
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
With thumb mode this alignment to 32 byte is required to avoid exception
Signed-off-by: Sam Nelson <sam.nelson@ti.com>
sciclient: sciclient_component.mk Build Fixes for Sciclient
Build fixes for RTOS build for M4F.
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
Build fixes for RTOS build for M4F.
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
PDK-5148: Board: Enabled auto config for J721E board pinmux
PRSDK-7908: Board: Fix for R5 diag execution hang through SD boot framework
- R5 diagnostic tests on K3 platform are hanging while running from diagnostic
framework through SD boot due to cache configurations and osal interrupts.
This was fixed earlier but diag entry code execution flow changed due to thumb
mode updates. Added the R5 issue fix for new execution sequence of diag entry code.
- R5 diagnostic tests on K3 platform are hanging while running from diagnostic
framework through SD boot due to cache configurations and osal interrupts.
This was fixed earlier but diag entry code execution flow changed due to thumb
mode updates. Added the R5 issue fix for new execution sequence of diag entry code.
udma rm cfg - allocated RX channel to mcu2_1
- cpsw examples failed on mcu2_1 due to not having RX channels allcocated and CPSW needs
one RX channels
- Allocated single RX channel to mcu2_1
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
- cpsw examples failed on mcu2_1 due to not having RX channels allcocated and CPSW needs
one RX channels
- Allocated single RX channel to mcu2_1
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
PDK-4724: Fix for J7 CPSW9G Ethernet test
PDK-5239: Board: Removed unused diagnostic test source files
PDK-6391: Board: Enabled MPU build for AM65xx diagnostic tests
PDK-5239: Board: Update board diag binary and build names
- Board diagnostic tests are using non-standatd names build and binary names
to match with existing platforms. Updated these names for K3 platforms
to allow system test automation
- Board diagnostic tests are using non-standatd names build and binary names
to match with existing platforms. Updated these names for K3 platforms
to allow system test automation
PDK-6462: Removes SBL XIP testcases's dependency on MSMC
As the SBL now supports MCU only boot mode, the XIP testcase has been
updated to use OCMC memory instead of MSMC.
This testcase now uses the memory where the SBL normally loads up the
sysfw temporarily and the memory where the RoM loads hte SBL
Signed-off-by: Madan Srinivas <madans@ti.com>
As the SBL now supports MCU only boot mode, the XIP testcase has been
updated to use OCMC memory instead of MSMC.
This testcase now uses the memory where the SBL normally loads up the
sysfw temporarily and the memory where the RoM loads hte SBL
Signed-off-by: Madan Srinivas <madans@ti.com>
PRSDK-8535: use only mcpu and avoid march and mtune
reference: https://community.arm.com/developer/tools-software/tools/b/tools-software-ides-blog/posts/compiler-flags-across-architectures-march-mtune-and-mcpu
Signed-off-by: Aravind Batni <aravindbr@ti.com>
reference: https://community.arm.com/developer/tools-software/tools/b/tools-software-ides-blog/posts/compiler-flags-across-architectures-march-mtune-and-mcpu
Signed-off-by: Aravind Batni <aravindbr@ti.com>
AM64x: Testing SCICLIENT Unit test on VLAB
SCiclient unit test based on RTOS enabled and tested on VLAB
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
SCiclient unit test based on RTOS enabled and tested on VLAB
Signed-off-by: Piyali Goswami <piyali_g@ti.com>
uart/test: remove vector copy section for tpr12
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>