General fix for MSPI drivers and Apollo3p mspi feature update #81762

swift-tk · 2024-11-22T11:47:46Z

Updates for the common files

Standarlized transfer priority and add a medium level.
Improve the mspi_async example.
Moved ambiq specific macro to mspi_ambiq header from shared drivers.
Add the CONFIG_MSPI_* macro for optional features in shared drivers
Add atxp032 driver to jesd216 example.

Ambiq specific update:
Introduce the am_hal_mspi_cq_scatter_xfer api with three modes(STREAM, NORMAL, LOOP) that can dynamically set transfer related settings in command queue.

Tests:
\zephyr\tests\drivers\mspi\api
\zephyr\tests\drivers\mspi\flash
\zephyr\samples\drivers\jesd216
\zephyr\samples\drivers\memc
\zephyr\samples\drivers\mspi\mspi_async
\zephyr\samples\drivers\mspi\mspi_flash

zephyrbot · 2024-11-22T11:48:47Z

The following west manifest projects have changed revision in this Pull Request:

Name	Old Revision	New Revision	Diff
hal_ambiq	zephyrproject-rtos/hal_ambiq@`87a188b` (`main`)	❌ Impostor SHA: zephyrproject-rtos/hal_ambiq@`1f30324`	zephyrproject-rtos/[email protected]

⛔ DNM label due to: 1 impostor SHA

Note: This message is automatically posted and updated by the Manifest GitHub Action.

drivers/mspi/mspi_ambiq_ap3.c

de-nordic · 2024-11-22T12:22:43Z

drivers/flash/flash_mspi_atxp032.c

@@ -745,7 +754,7 @@ static int flash_mspi_atxp032_read_jedec_id(const struct device *flash, uint8_t
 {
 	struct flash_mspi_atxp032_data *data = flash->data;

-	id = &data->jedec_id;
+	id = (uint8_t *)&data->jedec_id;


I do not think this is correct, as the flash_read_jedec_id expects ID to be copied from what is read on device to buffer pointed by the id, but the buffer is specified by function as 3 byte (which is wrong anyway), and here it will overflow if ID is longer and also there is a question whether byte order is not changed on assignment?

Yeah, your are right.

atxp032@0: SFDP v 1.7 AP fc with 4 PH
PH0: ff00 rev 1.7: 20 DW @ 28
Summary of BFP content:
DTR Clocking supported
Addressing: 4-Byte only
4-KiBy erase: uniform
Support 1-1-1
Flash density: 16777216 bytes
ET1: instr 20h for 4096 By; typ 144 ms, max 288 ms
ET2: instr 52h for 32768 By; typ 1152 ms, max 2304 ms
ET3: instr D8h for 65536 By; typ 2304 ms, max 4608 ms
Chip erase: typ 16384 ms, max 65536 ms
Byte program: type 12 + 12 * B us, max 48 + 48 * B us
Page program: typ 2048 us, max 8192 us
Page program size: 256 By
Suspend: B0h ; Resume: D0h
DPD: Enter B9h, exit ABh ; delay 8000 ns ; poll 0x3d
HOLD or RESET Disable: unsupported
QER: 0
0-4-4 Mode: not supported4-4-4 Mode sequences: enable 0x00 ; disable 0x0
4-byte addressing support: enter 0x40, exit 0x000
Soft Reset and Rescue Sequence support: 0x10
Status Register 1 support: 0x02
size = <134217728> bits;
sfdp-bfp = [
fd 20 8c ff ff ff ff 07 00 00 00 00 00 00 00 00
ee ff ff ff ff ff 00 00 ff ff 00 00 0c 20 0f 52
10 d8 00 00 80 42 46 01 81 ff 5a ff 20 61 06 44
d0 b0 d0 b0 f7 a7 d5 5c 00 00 00 ff 82 10 00 40
00 00 00 00 00 00 fc 01 21 00 10 00 00 00 42 62
];
PH1: ff05 rev 1.0: 5 DW @ 78
sfdp-ff05 = [
0c 0b 00 80 71 71 65 65 00 e4 94 dc 00 00 00 00
0c 57 24 00
];
PH2: ff87 rev 1.0: 28 DW @ 8c
sfdp-ff87 = [
00 00 00 00 80 00 00 00 08 d1 ff c7 08 d1 ff c7
00 05 c4 80 00 05 c4 81 00 05 c4 a5 00 05 c4 a5
71 65 03 d0 00 00 00 00 9c 00 00 00 88 23 59 9a
00 00 00 86 00 00 00 00 00 00 00 00 71 65 02 93
00 00 00 00 71 65 02 97 00 00 06 01 00 00 00 00
88 02 71 04 00 00 00 06 00 00 00 00 71 65 01 d6
71 65 01 d4 00 00 00 00 71 65 01 d0 00 00 38 05
];
PH3: ff0a rev 1.0: 4 DW @ d4
sfdp-ff0a = [
00 00 06 01 00 00 00 00 88 02 71 04 00 00 00 06
];
jedec-id = [1f a9 00];

drivers/mspi/mspi_ambiq_ap3.c

de-nordic · 2024-11-22T18:35:46Z

drivers/flash/flash_mspi_atxp032.c

@@ -31,7 +31,7 @@ LOG_MODULE_REGISTER(flash_mspi_atxp032, CONFIG_FLASH_LOG_LEVEL);
 #define NOR_WRITE_SIZE                      1
 #define NOR_ERASE_VALUE                     0xff

-#define ATXP032_VENDOR_ID                   0x43
+#define ATXP032_VENDOR_ID                   0x1F


Something is not right here, because adesto vendor number is 0x43, but their spec shows 9 bits in ID

ah, the one on my board is atxp128. Its datasheet says 0x1F as manufacturer ID LOL.

I will try to find a atxp032 and verify the ID again.

Standarlized transfer priority and add a medium level. Signed-off-by: Swift Tian <[email protected]>

The example now takes necessary information from the device tree and print better transfer status. Signed-off-by: Swift Tian <[email protected]>

1. Moved ambiq specific macro to mspi_ambiq header. 2. Always fill rx&tx dummy settings regardless of transfer direction. 3. Add the CONFIG_MSPI_* macro for optional features. 4. Fixed the ID read process. Signed-off-by: Swift Tian <[email protected]>

1. Use the newly introduced am_hal_mspi_cq_scatter_xfer api 2. Remove async PIO support 3. Use separate AM_HAL_MSPI_REQ_SCRAMB_CONFIG instead of am_hal_mspi_device_configure 4. Add MSPI_DEVICE_CONFIG_RX_DUMMY & MSPI_DEVICE_CONFIG_TX_DUMMY support and bug fix. 5. Unblock sync dma transcieve. Signed-off-by: Swift Tian <[email protected]>

Add the flash_mspi_atxp032 driver to the example and verified that flash_mspi_atxp032_read_sfdp is working as expected. Signed-off-by: Swift Tian <[email protected]>

swift-tk requested review from RichardSWheatley and aaronyegx November 22, 2024 11:48

zephyrbot added manifest manifest-hal_ambiq DNM This PR should not be merged (Do Not Merge) labels Nov 22, 2024

swift-tk requested review from FRASTM, de-nordic and ioannis-karachalios November 22, 2024 11:49

swift-tk force-pushed the apollo3p-mspi-feature-update branch from 970a5e9 to b0e390a Compare November 22, 2024 12:10

swift-tk changed the title ~~Apollo3p mspi feature update~~ General fix for MSPI drivers and Apollo3p mspi feature update Nov 22, 2024

swift-tk commented Nov 22, 2024

View reviewed changes

drivers/mspi/mspi_ambiq_ap3.c Outdated Show resolved Hide resolved

swift-tk marked this pull request as ready for review November 22, 2024 12:21

swift-tk self-assigned this Nov 22, 2024

zephyrbot added platform: Ambiq Ambiq area: MEMC area: Flash area: MSPI area: Samples Samples labels Nov 22, 2024

zephyrbot requested review from AlessandroLuo, kartben, nashif and rghaddab November 22, 2024 12:22

de-nordic reviewed Nov 22, 2024

View reviewed changes

swift-tk force-pushed the apollo3p-mspi-feature-update branch 3 times, most recently from 5435c3c to 6ca2b9d Compare November 22, 2024 16:27

de-nordic reviewed Nov 22, 2024

View reviewed changes

drivers/mspi/mspi_ambiq_ap3.c Outdated Show resolved Hide resolved

de-nordic reviewed Nov 22, 2024

View reviewed changes

drivers/mspi/mspi_ambiq_ap3.c Outdated Show resolved Hide resolved

de-nordic reviewed Nov 22, 2024

View reviewed changes

swift-tk force-pushed the apollo3p-mspi-feature-update branch from 6ca2b9d to 505d24f Compare November 23, 2024 08:59

swift-tk added 3 commits November 23, 2024 17:01

drivers: mspi: standarlized transfer priority

1764249

Standarlized transfer priority and add a medium level. Signed-off-by: Swift Tian <[email protected]>

samples: mspi: improve the mspi_async example

754222b

The example now takes necessary information from the device tree and print better transfer status. Signed-off-by: Swift Tian <[email protected]>

swift-tk force-pushed the apollo3p-mspi-feature-update branch from 505d24f to 5338b19 Compare November 23, 2024 09:02

swift-tk added 2 commits November 23, 2024 17:09

samples: jesd216: add atxp032 flash to example

9d14596

Add the flash_mspi_atxp032 driver to the example and verified that flash_mspi_atxp032_read_sfdp is working as expected. Signed-off-by: Swift Tian <[email protected]>

swift-tk force-pushed the apollo3p-mspi-feature-update branch from 5338b19 to 9d14596 Compare November 23, 2024 09:10

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

General fix for MSPI drivers and Apollo3p mspi feature update #81762

General fix for MSPI drivers and Apollo3p mspi feature update #81762

swift-tk commented Nov 22, 2024 •

edited

Loading

zephyrbot commented Nov 22, 2024 •

edited

Loading

de-nordic Nov 22, 2024

swift-tk Nov 22, 2024

swift-tk Nov 22, 2024

de-nordic Nov 22, 2024

swift-tk Nov 23, 2024 •

edited

Loading

General fix for MSPI drivers and Apollo3p mspi feature update #81762

Are you sure you want to change the base?

General fix for MSPI drivers and Apollo3p mspi feature update #81762

Conversation

swift-tk commented Nov 22, 2024 • edited Loading

zephyrbot commented Nov 22, 2024 • edited Loading

de-nordic Nov 22, 2024

Choose a reason for hiding this comment

swift-tk Nov 22, 2024

Choose a reason for hiding this comment

swift-tk Nov 22, 2024

Choose a reason for hiding this comment

de-nordic Nov 22, 2024

Choose a reason for hiding this comment

swift-tk Nov 23, 2024 • edited Loading

Choose a reason for hiding this comment

swift-tk commented Nov 22, 2024 •

edited

Loading

zephyrbot commented Nov 22, 2024 •

edited

Loading

swift-tk Nov 23, 2024 •

edited

Loading