PSOC6: initial integration of Cypress HAL #10692

vmedcy · 2019-05-28T17:56:22Z

Description

This PR brings the following major updates to the Cypress PSoC 6 targets for Mbed OS 5.13:

Initial implementation of the Cypress PSoC 6 hardware abstraction layer (targets/TARGET_Cypress/TARGET_PSOC6/psoc6csp/hal)
Ported Mbed OS HAL APIs implementation on top of the Cypress HAL (targets/TARGET_Cypress/TARGET_PSOC6/cy_*_api.c)
Initial implementation of the Cypress Wi-Fi Host Driver (targets/TARGET_Cypress/TARGET_PSOC6/TARGET_WHD)
WHD implementation of EMAC driver and NetworkInterfaceAPI (features/netsocket/emac-drivers/TARGET_Cypress/TARGET_PSOC6/TARGET_WHD)
USB device HAL implementation
Simplification of the CM0+ prebuilt application merging flow (replaced Python post-binary hex_filename merging hooks with prebuilt C arrays managed by the standard toolchain linker script).

Pull request type

[ ] Fix
[ ] Refactor
[X] Target update
[ ] Functionality change
[ ] Docs update
[ ] Test update
[ ] Breaking change

Reviewers

@ARMmbed/team-cypress

Release Notes

ciarmcom · 2019-05-28T19:00:48Z

@vmedcy, thank you for your changes.
@ARMmbed/mbed-os-maintainers @ARMmbed/mbed-os-ipcore please review.

maclobdell · 2019-05-29T15:43:56Z

I just pulled this and tried to build the PSA target version with the following command (using Arm Compiler 6)

mbed compile -m CY8CKIT_062_WIFI_BT_PSA -t ARM

I get the following error.

Compile [ 85.9%]: spm_init_api.c
[Fatal Error] spm_init_api.c@27,10: 'psoc6_utils.h' file not found
[ERROR] .\mbed-os\targets\TARGET_Cypress\TARGET_PSOC6\TARGET_CY8CKIT_062_WIFI_BT_PSA\spm_init_api.c:27:10: fatal error: 'psoc6_utils.h' file not found
#include "psoc6_utils.h"

It looks like that file was previously located at C:\Users\maclob01\Documents\mbed-os-5\pe\mbed-os\targets\TARGET_Cypress\TARGET_PSOC6\psoc6_utils.h, but now it is not present anywhere. I'm not sure if it is actually needed or not.

...es/netsocket/emac-drivers/TARGET_Cypress/TARGET_PSOC6/TARGET_WHD/interface/WhdSTAInterface.h

features/netsocket/emac-drivers/TARGET_Cypress/TARGET_PSOC6/TARGET_WHD/interface/whd_emac.cpp

features/netsocket/emac-drivers/TARGET_Cypress/TARGET_PSOC6/TARGET_WHD/network/whd_network.h

targets/TARGET_Cypress/TARGET_PSOC6/cy_crc_api.c

targets/TARGET_Cypress/TARGET_PSOC6/cy_serial_api.c

targets/TARGET_Cypress/TARGET_PSOC6/mbed_overrides.c

rtos/TARGET_CORTEX/mbed_rtx_conf.h

cypress-neil · 2019-06-27T21:10:33Z

features/lwipstack/lwipopts.h

-#ifndef LWIP_RAW
-#define LWIP_RAW                    0
-#endif
+#define LWIP_RAW                    1


These changes were added for PSOC6 testing but it is not clear if they should remain here as they affect all targets.
@vmedcy @morser499

It should not, this PR can be dependent on it (will be treated like that if required). I already mentioned few other issues like this, please create separate PR. They will get much faster in and easier to review.

LWIP_RAW change moved to #10978

rtos/TARGET_CORTEX/mbed_rtx_conf.h

hennadiykytsun · 2019-07-02T06:35:25Z

Issue #10515 is not applicable anymore. It was fixed in mbed-os 5.13.0 per the PR#10747 according to the following issues: #10613 and #10731 that were introduced by PR#10078.

maclobdell · 2019-07-02T12:17:24Z

@hennadiykytsun - perhaps this PR needs to be rebased to the latest master then, to avoid possibility that PR 10747 gets reverted? I will let the maintainers determine if that is necessary.

hennadiykytsun · 2019-07-02T14:41:15Z

@maclobdell - I have verified each Cypress target in this pull request - each of them has fix from the latest master. All asm files are correct.

morser499 · 2019-07-03T19:21:45Z

Test reports for latest PR:
GTEA_KIT_062_4343W_IAR_WIN_NETWORK.txt
GTEA_KIT_062_BLE__ARMC6_UBU.txt
GTEA_KIT_062_BLE__GCC_ARM_MAC.txt
GTEA_KIT_062_WIFI_BT__IAR_WIN.txt
GTEA_KIT_062_WIFI_BT_ARMC6_WIN_NETSOCKET.txt
GTEA_KIT_062_WIFI_BT_ARMC6_WIN_NETWORK.txt
GTEA_KIT_062_WIFI_BT_GCC_ARM_MAC_NETSOCKET.txt
GTEA_KIT_062_WIFI_BT_GCC_ARM_MAC_NETWORK.txt
GTEA_KIT_062_WIFI_BT_IAR_WIN_NETSOCKET.txt
GTEA_KIT_062_WIFI_BT_IAR_WIN_NETWORK.txt
GTEA_KIT_EVB_43012_ARMC6_WIN_NETSOCKET.txt
GTEA_KIT_EVB_43012_ARMC6_WIN_NETWORK.txt
GTEA_KIT_EVB_43012_GCC_ARM_MAC_NETSOCKET.txt
GTEA_KIT_EVB_43012_GCC_ARM_MAC_NETWORK.txt
GTEA_KIT_EVB_43012_IAR_WIN_NETSOCKET.txt
GTEA_KIT_EVB_43012_IAR_WIN_NETWORK.txt
GTEA_PROTO_062_4343W__ARMC6_WIN.txt
GTEA_PROTO_062_4343W__IAR_WIN.txt
GTEA_USBDEV_KIT_062_4343W__ARMC6_WIN .txt
GTEA_USBDEV_KIT_062_4343W__GCC_ARM_WIN.txt
GTEA_USBDEV_KIT_062_4343W__IAR_WIN.txt
GTEA_USBDEV_KIT_062_WIFI_BT__ARMC6_WIN.txt
GTEA_USBDEV_KIT_062_WIFI_BT__GCC_ARM_WIN.txt
GTEA_USBDEV_KIT_062_WIFI_BT__IAR_WIN.txt
GTEA_KIT_062_4343W_ARMC6_WIN_NETSOCKET.txt
GTEA_KIT_062_4343W_ARMC6_WIN_NETWORK.txt
GTEA_KIT_062_4343W_GCC_ARM_MAC_NETSOCKET.txt
GTEA_KIT_062_4343W_GCC_ARM_MAC_NETWORK.txt
GTEA_KIT_062_4343W_IAR_WIN_NETSOCKET.txt

cypress-neil · 2019-07-03T23:41:07Z

This should be the complete set of changes for the release.

0xc0170 · 2019-07-04T11:25:28Z

Are all commits ready for release (no need to clean - squash/rebase) ? I can see some merge commits in the history, they cause issues in patch releases. Is this targeting the next minor release?

vmedcy · 2019-07-04T12:25:18Z

@0xc0170: yes, this PR is ready for release, targeting 5.13.1.

Can you clarify your request to remove merge commits: is rebase required?

0xc0170 · 2019-07-04T12:41:19Z

Can you clarify your request to remove merge commits: is rebase required?

If this is targeting the patch release (the next one will be 5.13.2), then yes. Cherry-pick will be difficult with merge commits

vmedcy · 2019-07-04T14:36:30Z

Rebased against latest ARMmbed/master (this removed merge commits)

0xc0170 · 2019-07-04T14:37:37Z

I triggered CI run, will review tomorrow

mbed-ci · 2019-07-11T12:16:00Z

Test run: FAILED

Summary: 1 of 4 test jobs failed
Build number : 4
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_build-ARM

vmedcy · 2019-07-11T12:28:41Z

I reviewed the fail log: mbed-os-greentea_FUTURE_SEQUANA_ARM.log.

The error is related to FUTURE_SEQUANA targets inheriting TARGET_PSOC6 but not implementing DEVICE_I2CSLAVE and DEVICE_SPISLAVE. Fixing targets.json to add device_has_remove appropriately.

vmedcy · 2019-07-11T12:47:13Z

Removed DEVICE_I2CSLAVE for FUTURE_SEQUANA targets in b652407.
DEVICE_SPISLAVE is actually implemented in targets/TARGET_Cypress/TARGET_PSOC6_FUTURE/spi_api.c, no change needed.
Checked FUTURE_SEQUANA and FUTURE_SEQUANA_PSA targets compile successfully.

adbridge · 2019-07-11T13:38:13Z

ci restarted

AnttiKauppila

Went through the code and left couple of questions. Those won't be blocking this PR.

mbed-ci · 2019-07-11T16:26:10Z

Test run: SUCCESS

Summary: 11 of 11 test jobs passed
Build number : 5
Build artifacts

adbridge · 2019-07-11T17:16:31Z

@vmedcy We are having problems patching this across to the 5.13 release branch, getting these errors:
error: Failed to merge in the changes.
Patch failed at 0014 PSOC6: remove PSA targets
The copy of the patch that failed is found in: .git/rebase-apply/patch
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.

Are we sure there are no dependencies on any other PR on master that this is sitting on top of ?

adbridge · 2019-07-11T17:19:48Z

@adbridge It looks like your comment dropped the actual errors.

There now

adbridge · 2019-07-11T17:21:51Z

I've tried applying as a full patch and by cherry picking individual commits , both ways get stuck at that same commit

ifyall · 2019-07-11T17:25:21Z

@adbridge Thank you. I have asked someone in our US offices to look at this, as the day has ended for @vmedcy.

cypress-neil · 2019-07-11T18:02:34Z

The changes from this pull request conflict with 39f0997 and 6084069 which exist on mbed-os-5.13.1 but not master. I was able to merge the commits using cherry-pick and manual conflict resolution.

ifyall · 2019-07-11T18:15:48Z

@adbridge How should @cypress-neil show his work on the cherry-picking and manual resolution? A new PR?

adbridge · 2019-07-11T18:55:27Z

I suggest making an equivalent PR to the mbed-os-5.13 branch for these changes. That may well then need rebasing once we drop the rest of the patch release there

morser499 · 2019-07-11T18:55:30Z

To address the conflict with 'remove PSA targets' the PSA targets should be removed from the targets.json file and the 5 .hex files should be deleted.

adbridge · 2019-07-11T19:17:51Z

@morser499 can you guys please create a new PR directly against the mbed-os-5.13 branch, perhaps combining this PR with 11027 so we can bring both across easily ? NOTE we wouldn't normally do it this way but I think it will be the quickest way to try and ensure both get into the patch release.

ifyall · 2019-07-11T19:34:41Z

@adbridge Ryan and his team will be working on the new PR starting in about 30 minutes or so.

cypress-neil · 2019-07-11T20:16:02Z

Created #11030

ifyall · 2019-07-11T20:26:02Z

@adbridge Neil has created the requested PR and linked it above. Thanks.

0xc0170 · 2019-08-12T09:47:29Z

I came across this PR today while searching for answer why some PSA targets were removed - in this PR. There is a commit "PSOC6: remove PSA targets" - does not explain why it is being removed. Please in the future, add more content to the commit message. I don't understand how "initial integration of Cypress HAL" is related to removing targets.

Removing targets is not considered only "a target update" but rather functionality change - whoever used this target won't be able to use it after the release 5.13.1. We need to be more careful with changes like this and do not mix multiple PR types to avoid the surprises like this one.

JanneKiiskila · 2019-08-12T12:08:59Z

This breaks our CI now, as we are using CY8CKIT_062_WIFI_BT_PSA as one of our targets. I am not really sure was this intended?
@maclobdell

ciarmcom requested review from a team May 28, 2019 19:00

ciarmcom added the needs: review label May 28, 2019

maclobdell added the devices: cypress label May 29, 2019

0xc0170 requested changes Jun 4, 2019

View reviewed changes

0xc0170 added needs: work release-version: 5.13.1 and removed needs: review labels Jun 4, 2019

This was referenced Jun 19, 2019

Slow context switch after releasing a semaphore #10197

Closed

Reading SD Card returns wrong 0x00h data value - CY8CKIT_062_WIFI_BT #10245

Closed

morser499 reviewed Jun 27, 2019

View reviewed changes

rtos/TARGET_CORTEX/mbed_rtx_conf.h Outdated Show resolved Hide resolved

cypress-neil reviewed Jun 27, 2019

View reviewed changes

rtos/TARGET_CORTEX/mbed_rtx_conf.h Outdated Show resolved Hide resolved

maclobdell mentioned this pull request Jun 28, 2019

ARM Compiler 5 crash when running mbed tls on Cypress Platform (CY8CPROTO_062_4343W) #10515

Closed

0xc0170 self-requested a review July 3, 2019 10:03

0xc0170 added needs: review and removed needs: work release-version: 5.13.1 labels Jul 3, 2019

evedon removed the needs: review label Jul 11, 2019

PSOC6: TARGET_PSOC6_FUTURE doesn't implement DEVICE_I2CSLAVE

b652407

vmedcy mentioned this pull request Jul 11, 2019

PSOC6: add LwIP configuration for Cypress boards #11027

Merged

AnttiKauppila approved these changes Jul 11, 2019

View reviewed changes

evedon merged commit 22c82dc into ARMmbed:master Jul 11, 2019

evedon removed the needs: CI label Jul 11, 2019

cypress-neil mentioned this pull request Jul 11, 2019

PSOC6: initial integration of Cypress HAL (5.13.1) #11030

Closed

maclobdell mentioned this pull request Jul 16, 2019

Cypress QSPI and QSPIF removed #11061

Closed

cypress-neil mentioned this pull request Jul 18, 2019

PSOC6: USB device implementation #11071

Merged

vmedcy mentioned this pull request Sep 19, 2019

LwIP: add ICMPv4 Socket support #10978

Merged

michalpasztamobica mentioned this pull request Oct 31, 2019

Add API to get ipv6 link local address #11279

Merged

PSOC6: initial integration of Cypress HAL #10692

PSOC6: initial integration of Cypress HAL #10692

Conversation

vmedcy commented May 28, 2019 • edited Loading

Description

Pull request type

Reviewers

Release Notes

ciarmcom commented May 28, 2019

maclobdell commented May 29, 2019

cypress-neil Jun 27, 2019

Choose a reason for hiding this comment

0xc0170 Jul 5, 2019

Choose a reason for hiding this comment

vmedcy Jul 5, 2019 • edited Loading

Choose a reason for hiding this comment

hennadiykytsun commented Jul 2, 2019

maclobdell commented Jul 2, 2019

hennadiykytsun commented Jul 2, 2019 • edited Loading

morser499 commented Jul 3, 2019

cypress-neil commented Jul 3, 2019

0xc0170 commented Jul 4, 2019

vmedcy commented Jul 4, 2019

0xc0170 commented Jul 4, 2019

vmedcy commented Jul 4, 2019

0xc0170 commented Jul 4, 2019

mbed-ci commented Jul 11, 2019

Test run: FAILED

vmedcy commented Jul 11, 2019

vmedcy commented Jul 11, 2019

adbridge commented Jul 11, 2019

AnttiKauppila left a comment

Choose a reason for hiding this comment

mbed-ci commented Jul 11, 2019

Test run: SUCCESS

adbridge commented Jul 11, 2019 • edited Loading

adbridge commented Jul 11, 2019

adbridge commented Jul 11, 2019

ifyall commented Jul 11, 2019

cypress-neil commented Jul 11, 2019

ifyall commented Jul 11, 2019

adbridge commented Jul 11, 2019

morser499 commented Jul 11, 2019

adbridge commented Jul 11, 2019

ifyall commented Jul 11, 2019

cypress-neil commented Jul 11, 2019

ifyall commented Jul 11, 2019

0xc0170 commented Aug 12, 2019

JanneKiiskila commented Aug 12, 2019

vmedcy commented May 28, 2019 •

edited

Loading

vmedcy Jul 5, 2019 •

edited

Loading

hennadiykytsun commented Jul 2, 2019 •

edited

Loading

adbridge commented Jul 11, 2019 •

edited

Loading