Skip to content

{bp-19502} drivers/mmcsd: fix eMMC bus width switch sequencing - #19638

Merged
xiaoxiang781216 merged 1 commit into
apache:releases/13.0from
jerpelea:bp-19502
Aug 3, 2026
Merged

{bp-19502} drivers/mmcsd: fix eMMC bus width switch sequencing#19638
xiaoxiang781216 merged 1 commit into
apache:releases/13.0from
jerpelea:bp-19502

Conversation

@jerpelea

@jerpelea jerpelea commented Aug 3, 2026

Copy link
Copy Markdown
Contributor

Summary

Two sequencing problems in the MMC wide bus path break eMMC 4-bit operation on hosts that program the bus width in the SDIO widebus / clock callbacks (e.g. STM32H7):

  1. The SWITCH command (CMD6) is issued before the host has switched to wide bus operation. When the card completes the switch while the host is still in 1-bit mode the switch never takes effect and every following data transfer times out. Switch the host to wide bus operation before issuing CMD6.

  2. The transfer clock is selected only at the end of mmcsd_widebus(), so the whole switch sequence runs at ID-mode clock and, on the affected hosts, the final clock update does not take effect either, leaving the bus at ~400 kHz. Select the MMC transfer clock before calling mmcsd_widebus(), and pick CLOCK_MMC_TRANSFER_4BIT when wide bus operation is active (mirroring the SD card path) so a later clock selection cannot revert the host to 1-bit.

No behavior change for SD cards, and no change on hosts whose widebus callback only records the requested state.

Tested on a custom STM32H743 board with eMMC: sd_bench sequential write ~4.1 MB/s, sequential read ~6.3 MB/s (previously all data transfers timed out).

Impact

RELEASE

Testing

CI

Two sequencing problems in the MMC wide bus path break eMMC 4-bit
operation on hosts that program the bus width in the SDIO widebus /
clock callbacks (e.g. STM32H7):

1. The SWITCH command (CMD6) is issued before the host has switched
   to wide bus operation.  When the card completes the switch while
   the host is still in 1-bit mode the switch never takes effect and
   every following data transfer times out.  Switch the host to wide
   bus operation before issuing CMD6.

2. The transfer clock is selected only at the end of mmcsd_widebus(),
   so the whole switch sequence runs at ID-mode clock and, on the
   affected hosts, the final clock update does not take effect either,
   leaving the bus at ~400 kHz.  Select the MMC transfer clock before
   calling mmcsd_widebus(), and pick CLOCK_MMC_TRANSFER_4BIT when wide
   bus operation is active (mirroring the SD card path) so a later
   clock selection cannot revert the host to 1-bit.

No behavior change for SD cards, and no change on hosts whose widebus
callback only records the requested state.

Tested on a custom STM32H743 board with eMMC: sd_bench sequential
write ~4.1 MB/s, sequential read ~6.3 MB/s (previously all data
transfers timed out).

Signed-off-by: DuoYuWang <thirteenking.wang@gmail.com>
@github-actions github-actions Bot added Area: Drivers Drivers issues Size: S The size of the change in this PR is small labels Aug 3, 2026
@xiaoxiang781216
xiaoxiang781216 merged commit ec9891c into apache:releases/13.0 Aug 3, 2026
15 of 41 checks passed
@jerpelea
jerpelea deleted the bp-19502 branch August 3, 2026 14:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Area: Drivers Drivers issues Size: S The size of the change in this PR is small

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants