3 years ago[PDK-4913][PDK-7432] SBL QSPI mode and bug fix for large SBL img size REL.CORESDK.07.00.03.05 REL.CORESDK.07.00.03.06 REL.CORESDK.07.00.03.07 REL.CORESDK.07.00.03.08
[PDK-4913][PDK-7432] SBL QSPI mode and bug fix for large SBL img size
Migrate SBL QSPI to BoardFlash APIs and fix bug of large SBL img
size due to use of multiple discontinuous memory segments
Signed-off-by: Badri S <badri@ti.com>
Migrate SBL QSPI to BoardFlash APIs and fix bug of large SBL img
size due to use of multiple discontinuous memory segments
Signed-off-by: Badri S <badri@ti.com>
PDK-5085: Board: Updated j7200 evm diagnostic stress tests
- Updated memory allocations for some of the tests to use
DDR for data buffers
- Enabled stress tests for UART, Hyperflash & norflash
- Updated memory allocations for some of the tests to use
DDR for data buffers
- Enabled stress tests for UART, Hyperflash & norflash
emac: add example function to configure rx rate limiter
Signed-off-by: Vitaly Andrianov <vitalya@ti.com>
Signed-off-by: Vitaly Andrianov <vitalya@ti.com>
PDK-7402: Fix typos in PRU ICSS enum names
Signed-off-by: Dhaval Khandla <dhavaljk@ti.com>
Signed-off-by: Dhaval Khandla <dhavaljk@ti.com>
[PDK-7986] IPC: QNX: Update to support 64-bit virtual addresses for vring and mailbox
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
Signed-off-by: Praveen Rao <prao@ti.com>
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
Signed-off-by: Praveen Rao <prao@ti.com>
osal: Fix osal_ClearInterrupt
- Only the event instead of the interrupt should be cleared in the case
of the Event combiner.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- Only the event instead of the interrupt should be cleared in the case
of the Event combiner.
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
PDK-7586: ECC SDR for Main and MCU ESM, MSMC and CBASS AGGR
Added reception of ECC events from both MCU ESM and Main ESM.
For MSMC ECC Aggregator base address, the SDR needs to use RAT to
map the address.
Added more details in ECC callback to report bitErrorOffset, group,
and RAM ID.
Added RAT mapping of ECC Aggregators for MSMC instance.
Confirmed through ECC example that we can use these SDR updates to
receive MSMC Interconnect ECC events properly, and MCU R5F wrapper
and interconnect ECC events.
Added basic CBASS ECC aggegator support.
Added Group information for a single Interconnect RAM ID on MCU R5F
ECC aggregator.
Tied in updated J721E CSLR values for ECC aggregators and RAM ID's.
Ensured that AM65xx SDR still builds correctly.
Modified how SDR memEntries tables are searched (since there is no
guarantee that Wrapper types start at index 0).
Added reception of ECC events from both MCU ESM and Main ESM.
For MSMC ECC Aggregator base address, the SDR needs to use RAT to
map the address.
Added more details in ECC callback to report bitErrorOffset, group,
and RAM ID.
Added RAT mapping of ECC Aggregators for MSMC instance.
Confirmed through ECC example that we can use these SDR updates to
receive MSMC Interconnect ECC events properly, and MCU R5F wrapper
and interconnect ECC events.
Added basic CBASS ECC aggegator support.
Added Group information for a single Interconnect RAM ID on MCU R5F
ECC aggregator.
Tied in updated J721E CSLR values for ECC aggregators and RAM ID's.
Ensured that AM65xx SDR still builds correctly.
Modified how SDR memEntries tables are searched (since there is no
guarantee that Wrapper types start at index 0).
[PDK-7605] UDMA UT : Temp workaround for C7x Stack Size Error
- Temp workaround for:
[C71X_0] ti.sysbios.family.c7x.TaskSupport: line 111: assertion failure: A_stackSizeTooSmall: Task stack size must be >= 16KB.
xdc.runtime.Error.raise: terminating execution
- unitl PDK-7605 is resolved.
Signed-off-by: Don Dominic <a0486429@ti.com>
- Temp workaround for:
[C71X_0] ti.sysbios.family.c7x.TaskSupport: line 111: assertion failure: A_stackSizeTooSmall: Task stack size must be >= 16KB.
xdc.runtime.Error.raise: terminating execution
- unitl PDK-7605 is resolved.
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7493][PDK-7055] MIBSPI migrate to use CSL and DebugP
Modify MIBSPI driver to avoid app callbacks and
instead use CSL and DebugP APIs.Also removed
pinmux config from app as it is supported in
BoardLib now
Signed-off-by: Badri S <badri@ti.com>
Modify MIBSPI driver to avoid app callbacks and
instead use CSL and DebugP APIs.Also removed
pinmux config from app as it is supported in
BoardLib now
Signed-off-by: Badri S <badri@ti.com>
[PDK-6611] UDMA UT - J721E : Temp disable MCU NAVSS Multiple Tasks TC on mpu1_0
- temp disbale UDMA UT Multiple tasks testcases (3495 and 3497) on J721E mpu1_0
- until Bug PDK-6611 is resolved
Signed-off-by: Don Dominic <a0486429@ti.com>
- temp disbale UDMA UT Multiple tasks testcases (3495 and 3497) on J721E mpu1_0
- until Bug PDK-6611 is resolved
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7957] UART Test: Remove overriding of default Udma_rmInitiPrms
- Removed overriding of default number of VINTR in Udma_rmInitPrms
Background:
- Sciclient BoardCfg will be alligned with linux, with host specific entries and no more HOST_ID_ALL
- Overriding default Udma_RmInitPrms(which is alligned with the BoardCfg) won't work.
- Sysfw will report error when resources are out of range for the particular host.
Signed-off-by: Don Dominic <a0486429@ti.com>
- Removed overriding of default number of VINTR in Udma_rmInitPrms
Background:
- Sciclient BoardCfg will be alligned with linux, with host specific entries and no more HOST_ID_ALL
- Overriding default Udma_RmInitPrms(which is alligned with the BoardCfg) won't work.
- Sysfw will report error when resources are out of range for the particular host.
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7957] UDMA UT: Remove overriding of default Udma_rmInitiPrms
- Since Sciclient BoardCfg will be alligned with linux with host specific entries and no more HOST_ID_ALL
overriding default Udma_RmInitPrms(which is alligned with the BoardCfg) won't work.
Sysfw will report error when resources are out of range for the particular host.
- Removed overring of UHC/HC channels range in Unit Test
- HC Block Copy tests can't be run on various cores since HC blok copy channel allocations are 0 for all cores in default BoardCfg
Signed-off-by: Don Dominic <a0486429@ti.com>
- Since Sciclient BoardCfg will be alligned with linux with host specific entries and no more HOST_ID_ALL
overriding default Udma_RmInitPrms(which is alligned with the BoardCfg) won't work.
Sysfw will report error when resources are out of range for the particular host.
- Removed overring of UHC/HC channels range in Unit Test
- HC Block Copy tests can't be run on various cores since HC blok copy channel allocations are 0 for all cores in default BoardCfg
Signed-off-by: Don Dominic <a0486429@ti.com>
3 years ago[PRSDK-7872] UDMA: Leave TX CH FIFO Depth Reg with default reset value REL.CORESDK.07.01.00.11 REL.CORESDK.07.01.00.12 REL.CORESDK.07.01.00.14
[PRSDK-7872] UDMA: Leave TX CH FIFO Depth Reg with default reset value
- Temporary workaround until SYSFW-4176 is fixed
- When fdepth is passed to Sciclient_rmUdmapTxChCfg API, It corrupts TCHAN_TFIFO_DEPTH register
- Therefore, Not supporting configuration of TX FIFO Depth.
- fdepth valid param flag - TISCI_MSG_VALUE_RM_UDMAP_CH_TX_FDEPTH_VALID is not passed to Sciclient_rmUdmapTxChCfg
- As a result SYSFW won't touch this register and it will have its default reset value
- So,FIFO Depth for various channels will be as given below:
* Normal Capacity Channel - (128 bytes)
* High Capacity Channel - (1024 bytes)
* Ultra High Capacity Channel - (4096 bytes)
- PDK-8033 to track enabling this back once SYSFW-4176 is resolved.
Signed-off-by: Don Dominic <a0486429@ti.com>
- Temporary workaround until SYSFW-4176 is fixed
- When fdepth is passed to Sciclient_rmUdmapTxChCfg API, It corrupts TCHAN_TFIFO_DEPTH register
- Therefore, Not supporting configuration of TX FIFO Depth.
- fdepth valid param flag - TISCI_MSG_VALUE_RM_UDMAP_CH_TX_FDEPTH_VALID is not passed to Sciclient_rmUdmapTxChCfg
- As a result SYSFW won't touch this register and it will have its default reset value
- So,FIFO Depth for various channels will be as given below:
* Normal Capacity Channel - (128 bytes)
* High Capacity Channel - (1024 bytes)
* Ultra High Capacity Channel - (4096 bytes)
- PDK-8033 to track enabling this back once SYSFW-4176 is resolved.
Signed-off-by: Don Dominic <a0486429@ti.com>
PDK-5608: SBL: optimized Linux boot DTB update
Include pre-built DTBs for j721e and j7200 Linux boot appimage
creation that sets up SD card filesystem as the default FS for
Linux to search for when it boots.
This is not set up in the standard DTB images that are put in
LCPD nightly build server.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Include pre-built DTBs for j721e and j7200 Linux boot appimage
creation that sets up SD card filesystem as the default FS for
Linux to search for when it boots.
This is not set up in the standard DTB images that are put in
LCPD nightly build server.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
[PDK-8027] UDMA RM : Updates in J721E udma_rmcfg to allign with latest BoardCfg
- Allocates 2 DRU Channels to MCU2_0 and 2 for MCU2_1
- SYSFW Reserves 8 more MCU NAVSS VINTR and Global Events, Start range for MCU Domain cores alligned with this change
- updates in allignment with k3-image-gen/master and https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/pdk/pull-requests/1160/overview
Signed-off-by: Don Dominic <a0486429@ti.com>
- Allocates 2 DRU Channels to MCU2_0 and 2 for MCU2_1
- SYSFW Reserves 8 more MCU NAVSS VINTR and Global Events, Start range for MCU Domain cores alligned with this change
- updates in allignment with k3-image-gen/master and https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/pdk/pull-requests/1160/overview
Signed-off-by: Don Dominic <a0486429@ti.com>
J721E: sciclient: Update boardCfg_RM for Linux+RTOS boot
Updated sciclient boardCfg_RM files for "linux" to match
latest LCPD updates from SYSCONFIG tool.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Updated sciclient boardCfg_RM files for "linux" to match
latest LCPD updates from SYSCONFIG tool.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Test Automation : Corrected the bin names
for j721e, am65x, j7200 and am64x
Signed-off-by: sujith <sujith.s@ti.com>
for j721e, am65x, j7200 and am64x
Signed-off-by: sujith <sujith.s@ti.com>
[UDMA] Changing resource allocation to move DMPAC channels to MCU2-1
fixed c++ build issue
Signed-off-by: Prasad Konnur <prasadkonnur@ti.com>
Signed-off-by: Prasad Konnur <prasadkonnur@ti.com>
UDMA: Copy eventPrms [OUT] fields to eventHandle->eventPrms
- Copy following allocated resource info to eventHandle->eventPrms
*intrStatusReg
*intrClearReg
*intrMask
*coreIntrNum
Signed-off-by: Don Dominic <a0486429@ti.com>
- Copy following allocated resource info to eventHandle->eventPrms
*intrStatusReg
*intrClearReg
*intrMask
*coreIntrNum
Signed-off-by: Don Dominic <a0486429@ti.com>
PDK-5104: Board: Updated j7200 diag memory allocation for framework execution on R5
[PDK-7488] Migrate sysbios/xdc/cgtools
Migrate to sysbios 6.83 version and
associated xdc, cgtools arm,dsp to be
in sync with K3 processor SDK
The ARM CGT version is 20.2.2.LTS as
it contains fix for CODEGEN-7346
Signed-off-by: Badri S <badri@ti.com>
Migrate to sysbios 6.83 version and
associated xdc, cgtools arm,dsp to be
in sync with K3 processor SDK
The ARM CGT version is 20.2.2.LTS as
it contains fix for CODEGEN-7346
Signed-off-by: Badri S <badri@ti.com>
PDK-7762: Mailbox open returning error
- Fixed multi channel.
- Updated example for multi channel test
Signed-off-by: Prasad Konnur <prasadkonnur@ti.com>
- Fixed multi channel.
- Updated example for multi channel test
Signed-off-by: Prasad Konnur <prasadkonnur@ti.com>
[PDK-7088] UDMA : Support UDMA Virtualization
- Expose ‘virt_id’ param in Udma_RingPrms and pass it to Sciclient_rmRingCfg
- All the checks are done in System firmware, UDMA don’t do any validation for atype and virt_id.
- Application(User or someone else) should make sure PVU / SMMU is programmed correctly, to route the UDMA traffic via them.
- TX/RX Channel 'aType' is already exposed as 'addrType' in Udma_ChTxPrms/Udma_ChRxPrms
Signed-off-by: Don Dominic <a0486429@ti.com>
- Expose ‘virt_id’ param in Udma_RingPrms and pass it to Sciclient_rmRingCfg
- All the checks are done in System firmware, UDMA don’t do any validation for atype and virt_id.
- Application(User or someone else) should make sure PVU / SMMU is programmed correctly, to route the UDMA traffic via them.
- TX/RX Channel 'aType' is already exposed as 'addrType' in Udma_ChTxPrms/Udma_ChRxPrms
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-5992] UDMA : Added check for Ring Occupancy in Udma_eventUnRegister
- Add check for non-zero ring occupancy in Udma_eventUnRegister for Ring / DMA Completion Events
- API returns error when the ring occupancy is non-zero.
- This is to make sure that there is no resource leak, because unregistering these events will reset the ring.
- All the unprocessed descriptors in the ring / processed descriptors returned to the ring,
should be dequeued using #Udma_ringFlushRaw / #Udma_ringDequeueRaw before unregistering these events.
- Updated unit test TC:udmaTestRingEventTc (PDK-3509/PDK-3510) with Error Check,
which fails when Udma_eventUnRegister dosen't return error while calling with ringOcc != 0
- Updated memcpy and dru examples, to do ringFlush before unregistering DMA completion events
Signed-off-by: Don Dominic <a0486429@ti.com>
- Add check for non-zero ring occupancy in Udma_eventUnRegister for Ring / DMA Completion Events
- API returns error when the ring occupancy is non-zero.
- This is to make sure that there is no resource leak, because unregistering these events will reset the ring.
- All the unprocessed descriptors in the ring / processed descriptors returned to the ring,
should be dequeued using #Udma_ringFlushRaw / #Udma_ringDequeueRaw before unregistering these events.
- Updated unit test TC:udmaTestRingEventTc (PDK-3509/PDK-3510) with Error Check,
which fails when Udma_eventUnRegister dosen't return error while calling with ringOcc != 0
- Updated memcpy and dru examples, to do ringFlush before unregistering DMA completion events
Signed-off-by: Don Dominic <a0486429@ti.com>
PDK-7989: sciclient: Fix to support 64bit addressing
Signed-off-by: Praveen Rao <prao@ti.com>
Signed-off-by: Praveen Rao <prao@ti.com>
PDK-5038: Board: Updated board clock configurations for tpr12 evm
UDMA rmcfg - AM64x : Updates in SAUL resources and VINTR
- Update in allignment with the updates in Sciclient_defaultBoardCfg_rm.c
in PR: https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/pdk/pull-requests/1023/overview
Signed-off-by: Don Dominic <a0486429@ti.com>
- Update in allignment with the updates in Sciclient_defaultBoardCfg_rm.c
in PR: https://bitbucket.itg.ti.com/projects/PROCESSOR-SDK/repos/pdk/pull-requests/1023/overview
Signed-off-by: Don Dominic <a0486429@ti.com>
Add Linux support to AM64x RM Configuration
Update RM configuration based on changes needed for booting linux on A53. Effectively uses less HOST_ID_ALL and allocates resources to specific cores so A53 can have control over resources it needs for Linux
Update RM configuration based on changes needed for booting linux on A53. Effectively uses less HOST_ID_ALL and allocates resources to specific cores so A53 can have control over resources it needs for Linux
SBL: Fix for critical mcu1_0 sequenece when LOG_LEVEL=3
Removes extra prints added by the helper function
when LOG_LEVEL=3. These prints were getting in the way
of a critical section of the final boot sequence for
mcu1_0.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Removes extra prints added by the helper function
when LOG_LEVEL=3. These prints were getting in the way
of a critical section of the final boot sequence for
mcu1_0.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
osal: RegisterIntr: add configurable parameter enableIntr
- add configurable parameter enableIntr
- enableIntr is set to TRUE by default to keep backward-compatibility
- Pass the same setting to HwiP
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- add configurable parameter enableIntr
- enableIntr is set to TRUE by default to keep backward-compatibility
- Pass the same setting to HwiP
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
board/tpr12: add emac_mode definitions for cpsw lld
- add ENUM emac_mode at board_ethernet_config.h
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- add ENUM emac_mode at board_ethernet_config.h
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
PDK7938:add interrupt trigger sensitiviety
PDK7485:add missing files
PDK7485: syn with master
PDK7485:hwa driver update using CSL defines
[PDK-7899] UDMA: OSPI Example - Debug Build Linking Issue Fix
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
Update ndk/ns tools version
- Updates in NDK and NS tools for SDK7.1
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
- Updates in NDK and NS tools for SDK7.1
Signed-off-by: Prasad Jondhale <prasad.jondhale@ti.com>
gpio: PDK-5522: remove m4f bios build for GPIO_LedBlink_TestApp
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
Enable Reset Isolation
In SBL, do not touch the M4 if the MMR Magic Register has been set
In SBL, do not touch the M4 if the MMR Magic Register has been set
SBL Pinmux : do not configure pinmux on reset
SBL checks the MMR reset register to determine if a reset was previously issued. If so, do not configure pinmux, as the settings from prior boot are stil valid.
SBL checks the MMR reset register to determine if a reset was previously issued. If so, do not configure pinmux, as the settings from prior boot are stil valid.
i2c: PROC_SOC-185: remove bios build for AM64x m4f
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-7507: OSAL: remove SoC-independent libraries - patch 2
Throw error incase SOC type is not set
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Throw error incase SOC type is not set
Signed-off-by: Sivaraj R <sivaraj@ti.com>
[PDK-7899] UDMA: OSPI Example - Cleanup
- Removed BOARD_INIT_PLL, not required
- Added print after configuring RCLK
Signed-off-by: Don Dominic <a0486429@ti.com>
- Removed BOARD_INIT_PLL, not required
- Added print after configuring RCLK
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7899] UDMA: OSPI Example - uart boot mode appimage test issue fix
- Fixed OCMRAM memory overlap with SBL on J721E/J7200/AM65xx
Signed-off-by: Don Dominic <a0486429@ti.com>
- Fixed OCMRAM memory overlap with SBL on J721E/J7200/AM65xx
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7899] UDMA UT - J7200 : Temp disable MCU NAVSS Multiple Tasks TC on mpu1_0
- temp disbale UDMA UT Multiple tasks testcases (3495 and 3497) on J7200 mpu1_0
- until Bug PDK-6611 is resolved
Signed-off-by: Don Dominic <a0486429@ti.com>
- temp disbale UDMA UT Multiple tasks testcases (3495 and 3497) on J7200 mpu1_0
- until Bug PDK-6611 is resolved
Signed-off-by: Don Dominic <a0486429@ti.com>
PDK-7894 : Fixed the OSAL unit test failures on MCU 20
Root Cause
The sysbios uses timers 13 as default time when
timer-any is opted for timers.
In J7200 Timers 8 to 19 are controlled by
seperate LPSC bit (w.r.t) J721E
Fix : Updated board to power up timers 8 and others
as part of main domain peripherals
Signed-off-by: sujith <sujith.s@ti.com>
Root Cause
The sysbios uses timers 13 as default time when
timer-any is opted for timers.
In J7200 Timers 8 to 19 are controlled by
seperate LPSC bit (w.r.t) J721E
Fix : Updated board to power up timers 8 and others
as part of main domain peripherals
Signed-off-by: sujith <sujith.s@ti.com>
WIP - UDMA : J7200 Unit Test Issues Fix
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7671]Sciclient: Fixes for hostemulation build
Signed-off-by: Don Dominic <a0486429@ti.com>
Signed-off-by: Don Dominic <a0486429@ti.com>
icss_dualemac: Include cmp_cap event in resource table
Include cap and compare event(7) in PRU0 resource table. system_event 7 is
being mapped to channel 7.
Increase the firmware version build number to determine this change.
Fixes: LCPD-18771
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Include cap and compare event(7) in PRU0 resource table. system_event 7 is
being mapped to channel 7.
Increase the firmware version build number to determine this change.
Fixes: LCPD-18771
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
PDK-7828: IPC: QNX: Fix locking around recv_buffer
For QNX case, the interrupt handling is done in thread context, and
so the thread lock needs to be taken to protect the recv_buffer, which
is accessed from both user API as well as IPC isr handler.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
For QNX case, the interrupt handling is done in thread context, and
so the thread lock needs to be taken to protect the recv_buffer, which
is accessed from both user API as well as IPC isr handler.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
PDK-7757: PSDKQA-198: QNX: Fix the IPC error
New warning is enabled in QNX build that throws an
error when strncpy is used with length less than the
size of the length of the source string. Although in
this case, the next line was already taking care to
make sure the string is NULL-terminated.
Fix the compilation error by using length that is
the same size as the source name entry since the source
and destination name entries are the same size.
Signed-off-by: Praveen Rao <prao@ti.com>
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
New warning is enabled in QNX build that throws an
error when strncpy is used with length less than the
size of the length of the source string. Although in
this case, the next line was already taking care to
make sure the string is NULL-terminated.
Fix the compilation error by using length that is
the same size as the source name entry since the source
and destination name entries are the same size.
Signed-off-by: Praveen Rao <prao@ti.com>
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
[PDK-6682][DSS APP]Display sample application is printing wrong number of frames sent/displayed despite of correct FPS calculations
- Reduced test time from 120 seconds to 30 seconds
- Issue:
- Sample application was printing wrong number of frames displayed
- Root-cause
- App has an additional layer on top of normal UART/CIO print statements
- Argument parsing done in this function was incorrect
- Resolution:
- Add 'vsnprintf()' along with 'va_start()' and 'va_end()' for argument parsing
Signed-off-by: Vivek Dhande <a0132295@ti.com>
- Reduced test time from 120 seconds to 30 seconds
- Issue:
- Sample application was printing wrong number of frames displayed
- Root-cause
- App has an additional layer on top of normal UART/CIO print statements
- Argument parsing done in this function was incorrect
- Resolution:
- Add 'vsnprintf()' along with 'va_start()' and 'va_end()' for argument parsing
Signed-off-by: Vivek Dhande <a0132295@ti.com>
[PDK-7671] Use IR Interrupt Idx instead of Core Interrupt Idx in UDMA RM
- Before configuring the interrupt, translate IR Interrupt to Core Interrupt Idx using Sciclient_rmIrqTranslateIrOutput
- In case of devices like AM64x, where there are no Interrupt Routers, irIntrNum refers to coreIntrNum number itself.
- Added new param 'startC66xCoreIntr' in 'Udma_RmInitPrms' since For C66x Sciclient translates NAVSS IR Idx to corresponding C66 IR Idx, Not the Core Interrupt Idx.
- Added new param 'devIdIr'- Interrupt Router TISCI Dev ID in UDMA driver object, to pass to Sciclient_rmIrqTranslateIrOutput API
- Added new param 'irIntrNum' in 'Udma_EventObj'
Signed-off-by: Don Dominic <a0486429@ti.com>
- Before configuring the interrupt, translate IR Interrupt to Core Interrupt Idx using Sciclient_rmIrqTranslateIrOutput
- In case of devices like AM64x, where there are no Interrupt Routers, irIntrNum refers to coreIntrNum number itself.
- Added new param 'startC66xCoreIntr' in 'Udma_RmInitPrms' since For C66x Sciclient translates NAVSS IR Idx to corresponding C66 IR Idx, Not the Core Interrupt Idx.
- Added new param 'devIdIr'- Interrupt Router TISCI Dev ID in UDMA driver object, to pass to Sciclient_rmIrqTranslateIrOutput API
- Added new param 'irIntrNum' in 'Udma_EventObj'
Signed-off-by: Don Dominic <a0486429@ti.com>
PDK-7719 : Addressed review comments on J7200 noos OSAL Timer
Signed-off-by: sujith <sujith.s@ti.com>
Signed-off-by: sujith <sujith.s@ti.com>
i2c: PROC_SOC-185: fix I2C eeprom test package build issue for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
[ADASVISION-4202] Adding common SBL API for setting QoS
i2c: PROC_SOC-185: fix I2C eeprom test package build issue for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
[dmautils] Add hostemu support for 16bit to 8bit conversion
Signed-off-by: Anshu Jain <anshu.jain@ti.com>
Signed-off-by: Anshu Jain <anshu.jain@ti.com>
ti/osal/soc: add missing definition OSAL_ARCH_TIMER_INST_FOR_TS for tpr12
- add OSAL_ARCH_TIMER_INST_FOR_TS for tpr12
- fix an editing typo at osal.h
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- add OSAL_ARCH_TIMER_INST_FOR_TS for tpr12
- fix an editing typo at osal.h
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
ti/osal: remove nested comments
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
ti/osal: doxygen enhancements
- add missing comments
- put related macro definitions into groups
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
- add missing comments
- put related macro definitions into groups
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
PDK-7758: IPC: Use custom MMU config to C7x
IPC tests need the Linux/C7x vring memory to be set as non-cached, and
this is not done in the default MMU configuration provided in PDK APIs.
Instead, use a custom function to configure the MMU settings for C7x.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
IPC tests need the Linux/C7x vring memory to be set as non-cached, and
this is not done in the default MMU configuration provided in PDK APIs.
Instead, use a custom function to configure the MMU settings for C7x.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
i2c: PROC_SOC-185: fix I2C eeprom test package build issue for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-7028: pdksetupenv.bat does not work on Windows
Resolution:
We now have SOC specific package for AM65xx and j7200
So we can now use j721e for Jacinto
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Resolution:
We now have SOC specific package for AM65xx and j7200
So we can now use j721e for Jacinto
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Migrating to SYSFW version v2020.07-RC3
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
gpio: PDK-5522: fix pdk build error for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
i2c: PROC_SOC-185: add I2C EEPROM example support for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-7023: IPC: Update mailbox assignment
There were some incorrect user IDs in the am65xx assignment for TX
interrupt. This isn't actually used by IPC LLD so no issue was observed,
but it is fixed here for correctness.
Also, the am64x mailbox assignment is updated based on latest
alignment for usage between Linux and RTOS/baremetal.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
There were some incorrect user IDs in the am65xx assignment for TX
interrupt. This isn't actually used by IPC LLD so no issue was observed,
but it is fixed here for correctness.
Also, the am64x mailbox assignment is updated based on latest
alignment for usage between Linux and RTOS/baremetal.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
PDK-7060 : VCL OSAL updates
OSAL UT works for mcu 10, mcu20 and mpu 10
mcusw can_app works for mcu21
Signed-off-by: sujith <sujith.s@ti.com>
OSAL UT works for mcu 10, mcu20 and mpu 10
mcusw can_app works for mcu21
Signed-off-by: sujith <sujith.s@ti.com>
PDK-7060 : used instance 6 & 7 timer for MCU 20 and MCU 21
Signed-off-by: sujith <sujith.s@ti.com>
Signed-off-by: sujith <sujith.s@ti.com>
PDK-7648 : MMCSD
Signed-off-by: sujith <sujith.s@ti.com>
Signed-off-by: sujith <sujith.s@ti.com>
PDK-7685: Build fixes for AM65xx alone package
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
OSAL: remove SoC-independent libraries for all active SoCs
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
Signed-off-by: Eric Ruei <e-ruei1@ti.com>
PDK-7717: Board: Fix for j7200 evm spi nor flash test faiure running from sbl
- Enabled clock for MCSPI instance connected to nor flash in the board lib.
- Enabled clock for MCSPI instance connected to nor flash in the board lib.
PDK-4999: Board: Updated calibration values for j7200 evm current monitor test
PDK-4972: Board: Updated PMIC diagnostic test for j7200 evm
- Updated the voltage rails and values being tested to be under
safe limits.
- Updated the voltage rails and values being tested to be under
safe limits.
PDK-5093: Board: Updated j7200 evm boot switch diagnostic test as per HW design change
- One of the boot switch pins is used as oscillator input after boot which cannot
be used as gpio input by the test. Excluded the pin from the test.
- One of the boot switch pins is used as oscillator input after boot which cannot
be used as gpio input by the test. Excluded the pin from the test.
gpio: PDK-5522: fix led blink example build error for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
[PDK-7690] Sciclient : Support translation of IA VINT output
- reuse internal API 'Sciclient_rmTranslateIrOutput' by adding support for translating an interrupt aggregator output to the peripheral input it's connected to.
- Renamed 'Sciclient_rmTranslateIrOutput' to 'Sciclient_rmTranslateIntOutput'
- Add new external API 'Sciclient_rmIrqTranslateIaOutput'
- Translates an interrupt aggregator output to the peripheral input it's connected to. The primary use of the function is to retrieve the processor input IRQ(or IR input) an interrupt aggregator output is connected to.
Signed-off-by: Don Dominic <a0486429@ti.com>
- reuse internal API 'Sciclient_rmTranslateIrOutput' by adding support for translating an interrupt aggregator output to the peripheral input it's connected to.
- Renamed 'Sciclient_rmTranslateIrOutput' to 'Sciclient_rmTranslateIntOutput'
- Add new external API 'Sciclient_rmIrqTranslateIaOutput'
- Translates an interrupt aggregator output to the peripheral input it's connected to. The primary use of the function is to retrieve the processor input IRQ(or IR input) an interrupt aggregator output is connected to.
Signed-off-by: Don Dominic <a0486429@ti.com>
[PDK-7689] Sciclient : Added New API Sciclient_rmIrqTranslateIrqInput
- Translates a peripheral input to the connected interrupt router or aggregator output.
- The primary use of the function is to retrieve the interrupt router or aggregator output connected to a processor input IRQ.
Signed-off-by: Don Dominic <a0486429@ti.com>
- Translates a peripheral input to the connected interrupt router or aggregator output.
- The primary use of the function is to retrieve the interrupt router or aggregator output connected to a processor input IRQ.
Signed-off-by: Don Dominic <a0486429@ti.com>
esm: clear group1 error status using the current CSL-FL API function
Signed-off-by: Hongmei Gou <h-gou@ti.com>
Signed-off-by: Hongmei Gou <h-gou@ti.com>
PDK-7685: Build fixes for J7200 alone package
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PDK-7687: IPC: Fix negative test timeout test
The timeout test was incorrectly returning fail when it should have
been returning pass. Fix it to return the proper status.
Also, init the timer interrupts for the negative test for
C66x and C7x because it is needed for testing the timeout functionality,
and also the test uses Task_sleep.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
The timeout test was incorrectly returning fail when it should have
been returning pass. Fix it to return the proper status.
Also, init the timer interrupts for the negative test for
C66x and C7x because it is needed for testing the timeout functionality,
and also the test uses Task_sleep.
Signed-off-by: Angela Stegmaier <angelabaker@ti.com>
gpio: PDK-5522: add gpio lib/example support for AM64x
Signed-off-by: Hao Zhang <hzhang@ti.com>
Signed-off-by: Hao Zhang <hzhang@ti.com>
PDK-5528: SBL: J7200: Fallback for xSPI PHY tuning failure
Allows to continue reading the App image from xSPI flash
in xSPI+SFDP boot mode when the PHY tuning has failed.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Allows to continue reading the App image from xSPI flash
in xSPI+SFDP boot mode when the PHY tuning has failed.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
PDK-7685: Build fixes for AM65xx alone package
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PDK-7616: SBL: Build fixes for HLOS boot variant
Fixes HLOS builds for non-J7 SoC variants.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Fixes HLOS builds for non-J7 SoC variants.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Fix SBL warning - repo redundant make target
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
PDK-5086: Board: Disabled OSPI PHY mode for J7200 Uniflash programmer
- DAC mode with PHY enabled cannot be used for write operations on J7200
EVM
- DAC mode with PHY enabled cannot be used for write operations on J7200
EVM
PDK-7682: spi: Fix for OSPI flash test failure at 166MHz on j7200 evm
- Disabled the OSPI DAC mode tests since write operations are not
allowed in DAC mode for flash used on j7200 platform.
- Disabled the OSPI DAC mode tests since write operations are not
allowed in DAC mode for flash used on j7200 platform.
PDK-7616: SBL: Add build targets for HLOS boot variants
Added SBL HLOS boot variants for mmcsd, ospi, & hyperflash,
since separate compile flags needed to be used for SBL builds
that support HLOS boot.
Supported, currently for j721e & j7200. But, the infrastructure
is there for all K3 SoCs now, and others can easily be added.
Also, fixed the CUST build flags for j7200 sbl_cust_img build.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Added SBL HLOS boot variants for mmcsd, ospi, & hyperflash,
since separate compile flags needed to be used for SBL builds
that support HLOS boot.
Supported, currently for j721e & j7200. But, the infrastructure
is there for all K3 SoCs now, and others can easily be added.
Also, fixed the CUST build flags for j7200 sbl_cust_img build.
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
esm: clear error status with esm CSL-FL functions
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
watchdog: configure the stall model using CSL-FL
Signed-off-by: Hongmei Gou <hgou@ti.com>
Signed-off-by: Hongmei Gou <hgou@ti.com>
Fix doxygen warnings for J7200 only build
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
3 years agoPDK-5079: Board: Fix for J7200 board library build failure on mpu core REL.CORESDK.07.01.00.04 REL.CORESDK.07.01.00.05 REL.CORESDK.07.01.00.06
PDK-5079: Board: Fix for J7200 board library build failure on mpu core
Combined boot flow image gen enhancements to support automation
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
Signed-off-by: Sivaraj R <sivaraj@ti.com>
Signed-off-by: Jonathan Bergsagel <jbergsagel@ti.com>
build: Enable Enet LLD into build system
Enable building of Enet LLD.
Signed-off-by: Misael Lopez Cruz <misael.lopez@ti.com>
Enable building of Enet LLD.
Signed-off-by: Misael Lopez Cruz <misael.lopez@ti.com>
PDK-7627: Fixed in SBL to allow loading multi-stage FW on HS device
Signed-off-by: Brijesh Jadav <brijesh.jadav@ti.com>
Signed-off-by: Brijesh Jadav <brijesh.jadav@ti.com>
PDK-5079: Board: Integrated ospi phy tuning algorithm with j7200 evm