Hi Philipp,
(Adding firmware loader maintainers to Cc).
On Thu, Jan 19, 2017 at 10:44:54AM +0100, Philipp Zabel wrote:
> On Wed, 2017-01-18 at 21:33 +0200, Baruch Siach wrote:
> > > - if (dev->firmware == 1) {
> > > + if (dev->firmware > 0) {
> >
> > Why would vpu/vpu_fw_*.bin and v4l-coda960-*.bin be considered fallback
> > firmware?
>
> That was meant in the sense of a firmware loaded from fallback location.
>
> See the comment below, I needed a string to tell the user that the
> preceding firmware not found error messages can be safely ignored. If
> you have an idea for better wording, feel free submit a change.
>
> > > /*
> > > * Since we can't suppress warnings for failed asynchronous
> > > * firmware requests, report that the fallback firmware was
> > > * found.
> > > */
> > > dev_info(&pdev->dev, "Using fallback firmware %s\n",
> > > dev->devtype->firmware[dev->firmware]);
> > > }
For the context, we are talking about these boot time messages:
[ 2.152822] coda 2040000.vpu: Direct firmware load for vpu_fw_imx6q.bin failed with error -2
[ 2.162669] coda 2040000.vpu: Using fallback firmware vpu/vpu_fw_imx6q.bin
Looks bad. There must be a better way to load a firmware file from a few
optional locations. Actually, this use of the 'cont' parameter of
request_firmware_nowait() to retry firmware load does not look like the
intended use for the callback.
How about extending the firmware loader to accept multiple possible firmware
locations?
baruch
--
http://baruch.siach.name/blog/ ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- baruch@xxxxxxxxxx - tel: +972.52.368.4656, http://www.tkos.co.il -
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
