On 2/4/21 1:54 PM, Yann GAUTIER wrote:
On 2/4/21 1:26 PM, Marek Vasut wrote:
On 2/4/21 1:05 PM, yann.gautier@xxxxxxxxxxx wrote:
From: Yann Gautier <yann.gautier@xxxxxxxxxxx>
To properly manage commands awaiting R1B responses, the capability
MMC_CAP_NEED_RSP_BUSY is enabled in mmci driver, for variants that
manage busy detection.
This R1B management needs both the flags MMC_CAP_NEED_RSP_BUSY and
MMC_CAP_WAIT_WHILE_BUSY to be enabled together.
Shouldn't this have Fixes: tag ?
Hi Marek,
There is no unique patch that brought the issue, but a combination of
several things:
- The series that brought the MMC_CAP_NEED_RSP_BUSY flag [1]
- The series that enabled MMC_ERASE for all hosts [2] (removal of
MMC_CAP_ERASE)
But you're right, this patch may go on v5.8.x kernel and newer versions.
I think there will be quite some interest in 5.10.y LTS on the MP1 from
the various industrial/embedded users, so it would be nice to have that
5.10.y well maintained with necessary backports / fixes :)
