On 07/08/24 19:21, Andrew Davis wrote:
On 8/7/24 1:22 AM, Beleswar Padhi wrote:
Acquire the mailbox handle during device probe and do not release handle
in stop/detach routine or error paths. This removes the redundant
requests for mbox handle later during rproc start/attach. This also
allows to defer remoteproc driver's probe if mailbox is not probed yet.
Fixes: b8431920391d ("remoteproc: k3-dsp: Add support for IPC-only
mode for all K3 DSPs")
Not sure this patch counts as a "fix". There was a bug yes, and this
certainly
is an improvment that solves the issue, but I like to reserve "Fixes"
tags
for more serious issues. Otherwise this patch will be backported to
"stable"
versions where it might cause things to be *less stable* given the
size of
Understood. Will remove Fixes tag in revision.
the "fix". Also, the commit you selected isn't the source of the issue,
it only adds another instance of it, getting the mailbox after probe has
been the case since the first version of this driver.
Correct. My idea was, since we are also making k3_attach()/detach()
functions NOP, select the later commit because that was where the
k3_attach()/detach() functions were introduced as well as the "mailbox
after probe" issue was present.
Rest of the patch LGTM,
Acked-by: Andrew Davis <afd@xxxxxx>
BTW, I've folded this change (getting mbox in probe) into the new
K3 M4 driver[0] I just posted, so we should be aligned here accross
all K3 rproc drivers.
Thanks for this!
Andrew
[0]
https://lore.kernel.org/linux-arm-kernel/20240802152109.137243-4-afd@xxxxxx/
[...]
