title: "SDHI; SCC may stall if SDnH is not provided"
team: IO
key: d593a7de-95e8-11ea-8695-9f6a675410b7
status: Done
assignee: Wolfram

relationships:

bsp39x:
 - addfe9203290e3c274bca15e64e3f11638649e66 # mmc: renesas_sdhi: Fix hang up in HS400 timing mode selection

upstream:
 - torvalds: 24fdcb377892b44dfe9755a62e4b92e9c6a0b359 # mmc: core: when downgrading HS400, callback into drivers earlier
 - torvalds: 97a7d87e96b02fc5b3944d7735e0f6b8446d07da # mmc: core: add a 'doing_init_tune' flag and a 'mmc_doing_tune' helper
 - torvalds: 0f93db6542fa94262b611ff942b9b3ad7ecbea10 # mmc: renesas_sdhi: keep SCC clock active when tuning

comments:
 - "hang up is surprisingly hard to reproduce but we keep trying"
 - "updated manual calibration patchset makes the issue appear again"
 - "v1: https://patchwork.kernel.org/project/linux-renesas-soc/list/?series=298209&state=*"
 - "not all error paths covered; we need a new flag in the MMC core. Wolfram will work on it"
 - "delayed because of generic Gen3 board boot problems"
 - "patches for MMC core updates exists. However, currently not possible to trigger the issue again"
 - "issue could be triggered again once"
 - "patch to inject the fault for stalled SCC was created"
 - "MMC core updates could be tested and worked"
 - "v2: https://patchwork.kernel.org/project/linux-renesas-soc/list/?series=341697&state=*"
 - "v2 tested and reviewed by Shimoda-san"
 - "v2 merged"