Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12619

rogeryou · 2020-03-12T05:30:56Z

Summary of changes

Add OSPI HAL driver and OSPIF block device driver for using Macronix octaflash MX25LM51245G.

ST Octo-SPI was ported on QSPI MBED API, but we can only use in SPI mode and the Octo mode is not supported. So, we add OSPI driver to enable Octo mode to improve read and write performance.

Impact of changes

Enable the octa mode of MX25LM51245G.

Migration actions required

N/A

Documentation

N/A

Pull request type

[] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[x] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

Test results

[] No Tests required for this change (E.g docs only update)
[x] Covered by existing mbed-os tests (Greentea or Unittest)
[] Tests / results supplied as part of this PR

OSPI HAL test was executed.

target	platform_name	test suite	test case	passed	result	elapsed_time (sec)
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi frequency setting test	1	OK	0.12
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi init/free test	1	OK	0.07
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi memory id test	1	OK	0.11
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(1-1-1)/x1 read(1-1-1)/x1 repeat/x1 test	1	OK	0.1
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(1-1-1)/x1 read(1-1-1)/x1 repeat/x4 test	1	OK	0.19
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(1-1-1)/x1 read(1-1-1)/x4 repeat/x1 test	1	OK	0.1
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(1-1-1)/x4 read(1-1-1)/x1 repeat/x1 test	1	OK	0.13
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8D_8D_8D)/x1 read(8D_8D_8D)/x1 repeat/x1 test	1	OK	0.12
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8D_8D_8D)/x1 read(8D_8D_8D)/x1 repeat/x4 test	1	OK	0.18
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8D_8D_8D)/x1 read(8D_8D_8D)/x4 repeat/x1 test	1	OK	0.12
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8D_8D_8D)/x4 read(8D_8D_8D)/x1 repeat/x1 test	1	OK	0.12
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8_8_8)/x1 read(8_8_8)/x1 repeat/x1 test	1	OK	0.11
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8_8_8)/x1 read(8_8_8)/x1 repeat/x4 test	1	OK	0.18
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8_8_8)/x1 read(8_8_8)/x4 repeat/x1 test	1	OK	0.11
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-ospi	ospi write(8_8_8)/x4 read(8_8_8)/x1 repeat/x1 test	1	OK	0.1

OSPIF driver test was executed.

target	platform_name	test suite	test case	passed	result	elapsed_time (sec)
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing BlockDevice erase functionality	1	OK	0.16
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing Deinit block device	1	OK	0.13
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing Init block device	1	OK	0.12
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing contiguous erase, write and read	1	OK	0.16
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing multi threads erase program read	1	OK	0.16
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing program read small data sizes	1	OK	0.17
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing read write random blocks	1	OK	0.15
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing unaligned erase blocks	1	OK	0.13
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OSPIF Testing write -> deinit -> init -> read	1	OK	0.01

Reviewers

sync with source repo

Sync to latest mbed-os

ciarmcom · 2020-03-12T06:00:33Z

@rogeryou, thank you for your changes.
@ARMmbed/mbed-os-core @ARMmbed/mbed-os-wan @ARMmbed/mbed-os-pan @ARMmbed/mbed-os-storage @ARMmbed/mbed-os-hal @ARMmbed/mbed-os-ipcore @ARMmbed/mbed-os-connectivity @ARMmbed/mbed-os-maintainers @ARMmbed/mbed-os-test @ARMmbed/mbed-os-crypto @ARMmbed/mbed-os-tls please review.

jeromecoutant

Great work!
I am going to execute tests asap!

@ARMmbed/team-st-mcd

jeromecoutant · 2020-03-12T10:47:39Z

targets/targets.json

@@ -12525,7 +12525,7 @@
            }
        },
        "components_add": [
-            "FLASHIAP"
+            "OSPIF"


Why FLASHIAP is removed ?

Why FLASHIAP is removed ?

@jeromecoutant

Before I add the OSPI driver, I have executed the test "features-storage-tests-blockdevice-general_block_device" of FLASHIAP, but it failed.

Have you tested FLASHIAP on DISCO_L4R9I before?

You can see the error information below.

mbedgt: processing target 'DISCO_L4R9I' toolchain 'GCC_ARM' compatible platforms... (note: switch set to --parallel 1)
mbedgt: test case filter (specified with -n option)
features-storage-tests-blockdevice-general_block_device
test filtered in 'features-storage-tests-blockdevice-general_block_device'
mbedgt: running 1 test for platform 'DISCO_L4R9I' and toolchain 'GCC_ARM'
mbedgt: mbed-host-test-runner: started
mbedgt: :741::FAIL: Expected -1 Was 0
mbedgt: :741::FAIL: Expected -1 Was 0
mbedgt: :747::FAIL: Expected 0 Was 1
mbedgt: :741::FAIL: Expected -1 Was 0
mbedgt: :747::FAIL: Expected 0 Was 2
mbedgt: :258::FAIL: Expected 0 Was -1
mbedgt: :274::FAIL: Expected 92 Was 23
mbedgt: :274::FAIL: Expected 163 Was 0
mbedgt: :274::FAIL: Expected 105 Was 0
mbedgt: :274::FAIL: Expected 177 Was 0
mbedgt: :274::FAIL: Expected 133 Was 0
mbedgt: :274::FAIL: Expected 106 Was 56

Working in mbed-os master branch:

target platform_name test suite result elapsed_time (sec) copy_method

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-buffered_block_device OK 15.64 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-flashsim_block_device OK 15.32 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-general_block_device OK 26.57 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-heap_block_device OK 17.56 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-mbr_block_device OK 17.6 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I features-storage-tests-blockdevice-util_block_device OK 16.34 default

DISCO_L4R9I-ARMC6 DISCO_L4R9I tests-mbed_hal-qspi OK 38.15 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-buffered_block_device OK 15.58 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-flashsim_block_device OK 15.26 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-general_block_device OK 26.53 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-heap_block_device OK 17.63 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-mbr_block_device OK 16.44 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I features-storage-tests-blockdevice-util_block_device OK 16.07 default

DISCO_L4R9I-GCC_ARM DISCO_L4R9I tests-mbed_hal-qspi OK 37.88 default

Hi,
I execute the command below, and I got failed result.
sudo mbed test -m DISCO_L4R9I -t GCC_ARM -n features-storage-tests-blockdevice-general_block_device

Could you tell me the command you executed to test many test suite in one time?
Thanks a lot!

adbridge · 2020-03-12T15:41:01Z

@rogeryou thanks for the PR but it looks like you have modified the PR header template to remove some fields (or are using a very old template). Could you please update to the correct template and fill in the missing information? Thanks.

jeromecoutant · 2020-03-13T15:09:24Z

Hi

I have try this branch:

Could you squash some commits and rebase on top of mbed-os master ?
I got compilation error:

[Error] @0,0: L6218E: Undefined symbol ospi_init_direct (referred from BUILD/tests/DISCO_L4R9I/ARM/drivers/source/OSPI.o).

in drivers/source/SFDP.cpp
#if (DEVICE_SPI || DEVICE_QSPI)
You should add DEVICE_OSPI ?

rogeryou · 2020-03-16T06:21:51Z

@rogeryou thanks for the PR but it looks like you have modified the PR header template to remove some fields (or are using a very old template). Could you please update to the correct template and fill in the missing information? Thanks.

Hi,
I have filled in some missing information.Please review.@adbridge
Thanks.

jeromecoutant · 2020-03-16T09:21:15Z

I got compilation error:

[Error] @0,0: L6218E: Undefined symbol ospi_init_direct (referred from BUILD/tests/DISCO_L4R9I/ARM/drivers/source/OSPI.o).

OSPI code is missing in static_pinmap.cpp/h files

rogeryou · 2020-03-16T09:46:28Z

Hi

I have try this branch:

* Could you squash some commits and rebase on top of mbed-os master ?

* I got compilation error:

[Error] @0,0: L6218E: Undefined symbol ospi_init_direct (referred from BUILD/tests/DISCO_L4R9I/ARM/drivers/source/OSPI.o).

* in drivers/source/SFDP.cpp
  #if (DEVICE_SPI || DEVICE_QSPI)
  You should add DEVICE_OSPI ?

Hi,
I have squashed some commits and add DEVICE_OSPI in SFDP.cpp. @jeromecoutant
Thanks!

rogeryou · 2020-03-16T10:23:10Z

I got compilation error:
[Error] @0,0: L6218E: Undefined symbol ospi_init_direct (referred from BUILD/tests/DISCO_L4R9I/ARM/drivers/source/OSPI.o).
OSPI code is missing in static_pinmap.cpp/h files

Hi,

There is no compilation error here without OSPI code in static_pinmap.cpp/h.
But, I have added OSPI code in static_pinmap.cpp/h. Please review.
Thanks a lot!@jeromecoutant

0xc0170 · 2020-03-16T14:42:51Z

@rogeryou Hi, can you please rebase your branch to remove merge commits ? To have clean linear git history, just adding new files (or any fixes to these).

jeromecoutant · 2020-03-16T14:50:33Z

There is no compilation error here without OSPI code in static_pinmap.cpp/h.
But, I have added OSPI code in static_pinmap.cpp/h. Please review.

Still compilation error with ARM.
And static_pinmap files...

mergify · 2020-03-17T07:40:56Z

This PR cannot be merged due to conflicts. Please rebase to resolve them.

jeromecoutant · 2020-03-17T08:19:46Z

@rogeryou Hi, can you please rebase your branch to remove merge commits ? To have clean linear git history, just adding new files (or any fixes to these).

I agree... I currently can't work with that branch...
Starts from mbed-os master and just add your files (no merged branches)

Modify the astyle Add DEVICE_OSPI in SFDP.cpp

rogeryou · 2020-03-18T08:35:47Z

@rogeryou Hi, can you please rebase your branch to remove merge commits ? To have clean linear git history, just adding new files (or any fixes to these).

I agree... I currently can't work with that branch...
Starts from mbed-os master and just add your files (no merged branches)

@jeromecoutant
I have created a new macronix_ospi branch from mbed-os master and a new pull request PR #12644 .
Please review.

mergify · 2020-03-20T15:26:51Z

This PR cannot be merged due to conflicts. Please rebase to resolve them.

VeijoPesonen · 2020-03-23T08:23:25Z

@adbridge the bot doesn't seem to realize that this PR has been closed.

adbridge · 2020-03-23T11:10:29Z

@VeijoPesonen yeah it's a known issue, it is on the list of investigations.

juliensu added 2 commits February 7, 2020 15:58

Merge pull request #1 from ARMmbed/master

90c1f89

sync with source repo

Merge pull request #2 from ARMmbed/master

a1d9695

Sync to latest mbed-os

mergify bot added the needs: work label Mar 12, 2020

ciarmcom requested review from a team March 12, 2020 06:00

ciarmcom added the needs: review label Mar 12, 2020

rogeryou changed the title ~~Macronix ospi~~ Add OSPI driver to support the Octa mode of Macronix Octa flash MX25LM51245G Mar 12, 2020

rogeryou marked this pull request as ready for review March 12, 2020 06:19

rogeryou changed the title ~~Add OSPI driver to support the Octa mode of Macronix Octa flash MX25LM51245G~~ Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G Mar 12, 2020

mergify bot removed the needs: review label Mar 12, 2020

jeromecoutant reviewed Mar 12, 2020

View reviewed changes

0xc0170 removed request for a team March 16, 2020 14:41

rogeryou force-pushed the macronix_ospi branch from 1910ff4 to 8dc1acf Compare March 17, 2020 07:40

rogeryou force-pushed the macronix_ospi branch from 8dc1acf to 6e63129 Compare March 17, 2020 08:09

add OSPI HAL and OSPIF driver

14c0d25

Modify the astyle Add DEVICE_OSPI in SFDP.cpp

rogeryou force-pushed the macronix_ospi branch from 6e63129 to 14c0d25 Compare March 17, 2020 09:18

rogeryou closed this Mar 18, 2020

rogeryou deleted the macronix_ospi branch March 18, 2020 06:49

rogeryou restored the macronix_ospi branch March 18, 2020 06:52

rogeryou reopened this Mar 18, 2020

rogeryou closed this Mar 18, 2020

rogeryou deleted the macronix_ospi branch March 18, 2020 07:18

This was referenced Mar 18, 2020

dele #12643

Closed

Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12644

Merged

adbridge removed the needs: work label Mar 20, 2020

mergify bot added the needs: work label Mar 20, 2020

adbridge added mergify skip and removed needs: work labels Mar 23, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12619

Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12619

rogeryou commented Mar 12, 2020 •

edited

ciarmcom commented Mar 12, 2020

jeromecoutant left a comment

jeromecoutant Mar 12, 2020

rogeryou Mar 13, 2020 •

edited

jeromecoutant Mar 17, 2020

rogeryou Mar 18, 2020

adbridge commented Mar 12, 2020

jeromecoutant commented Mar 13, 2020

rogeryou commented Mar 16, 2020 •

edited

jeromecoutant commented Mar 16, 2020

rogeryou commented Mar 16, 2020

rogeryou commented Mar 16, 2020

0xc0170 commented Mar 16, 2020

jeromecoutant commented Mar 16, 2020

mergify bot commented Mar 17, 2020

jeromecoutant commented Mar 17, 2020

rogeryou commented Mar 18, 2020

mergify bot commented Mar 20, 2020

VeijoPesonen commented Mar 23, 2020

adbridge commented Mar 23, 2020

target	platform_name	test suite	result	elapsed_time (sec)	copy_method
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-buffered_block_device	OK	15.64	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-flashsim_block_device	OK	15.32	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OK	26.57	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-heap_block_device	OK	17.56	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-mbr_block_device	OK	17.6	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	features-storage-tests-blockdevice-util_block_device	OK	16.34	default
DISCO_L4R9I-ARMC6	DISCO_L4R9I	tests-mbed_hal-qspi	OK	38.15	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-buffered_block_device	OK	15.58	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-flashsim_block_device	OK	15.26	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-general_block_device	OK	26.53	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-heap_block_device	OK	17.63	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-mbr_block_device	OK	16.44	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	features-storage-tests-blockdevice-util_block_device	OK	16.07	default
DISCO_L4R9I-GCC_ARM	DISCO_L4R9I	tests-mbed_hal-qspi	OK	37.88	default

Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12619

Add OSPI driver to support the Octa mode of Macronix octaflash MX25LM51245G #12619

Conversation

rogeryou commented Mar 12, 2020 • edited

Summary of changes

Impact of changes

Migration actions required

Documentation

N/A

Pull request type

Test results

Reviewers

ciarmcom commented Mar 12, 2020

jeromecoutant left a comment

Choose a reason for hiding this comment

jeromecoutant Mar 12, 2020

Choose a reason for hiding this comment

rogeryou Mar 13, 2020 • edited

Choose a reason for hiding this comment

jeromecoutant Mar 17, 2020

Choose a reason for hiding this comment

rogeryou Mar 18, 2020

Choose a reason for hiding this comment

adbridge commented Mar 12, 2020

jeromecoutant commented Mar 13, 2020

rogeryou commented Mar 16, 2020 • edited

jeromecoutant commented Mar 16, 2020

rogeryou commented Mar 16, 2020

rogeryou commented Mar 16, 2020

0xc0170 commented Mar 16, 2020

jeromecoutant commented Mar 16, 2020

mergify bot commented Mar 17, 2020

jeromecoutant commented Mar 17, 2020

rogeryou commented Mar 18, 2020

mergify bot commented Mar 20, 2020

VeijoPesonen commented Mar 23, 2020

adbridge commented Mar 23, 2020

rogeryou commented Mar 12, 2020 •

edited

rogeryou Mar 13, 2020 •

edited

rogeryou commented Mar 16, 2020 •

edited