On 12/12/21 3:02 AM, Geert Uytterhoeven wrote:
> Hi Jens,
>
> On Fri, Dec 10, 2021 at 7:52 PM Jens Axboe <axboe@xxxxxxxxx> wrote:
>> On Mon, Dec 6, 2021 at 12:04 AM Christoph Hellwig <hch@xxxxxx> wrote:
>>> mtdblock / mtdblock_ro set part_bits to 0 and thus nevever scanned
>>> partitions. Restore that behavior by setting the GENHD_FL_NO_PART flag.
>>>
>>> Fixes: 1ebe2e5f9d68e94c ("block: remove GENHD_FL_EXT_DEVT")
>>> Reported-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
>>> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
>>> Tested-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
>
>>> @@ -355,9 +355,11 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
>>> "%s%c%c", tr->name,
>>> 'a' - 1 + new->devnum / 26,
>>> 'a' + new->devnum % 26);
>>> - else
>>> + } else {
>>> snprintf(gd->disk_name, sizeof(gd->disk_name),
>>> "%s%d", tr->name, new->devnum);
>>> + gd->flags |= GENHD_FL_NO_PART;
>>> + }
>>
>> Not sure why I didn't spot this until now, but:
>>
>> drivers/mtd/mtd_blkdevs.c: In function ‘add_mtd_blktrans_dev’:
>> drivers/mtd/mtd_blkdevs.c:362:30: error: ‘GENHD_FL_NO_PART’ undeclared (first use in this function); did you mean ‘GENHD_FL_NO_PART_SCAN’?
>> 362 | gd->flags |= GENHD_FL_NO_PART;
>> | ^~~~~~~~~~~~~~~~
>> | GENHD_FL_NO_PART_SCAN
>> drivers/mtd/mtd_blkdevs.c:362:30: note: each undeclared identifier is reported only once for each function it appears in
>>
>> Hmm?
>>
>> I'm going to revert this one for now, not sure how it could've been
>> tested in this form.
>
> Because next-20211130 and later have commit 46e7eac647b34ed4 ("block:
> rename GENHD_FL_NO_PART_SCAN to GENHD_FL_NO_PART").
I guess that explains it, it ended up in the wrong branch...
--
Jens Axboe
