On 26/04/18 14:28, Shawn Lin wrote:
> On 2018/4/26 18:09, Adrian Hunter wrote:
>> On 24/04/18 12:13, Shawn Lin wrote:
>>>
>>> On 2018/4/13 15:55, Adrian Hunter wrote:
>>>> On 13/04/18 10:21, Shawn Lin wrote:
>>>
>>>>
>>>> No we always remove children (card) before the parent (host controller).
>>>>
>>>
>>> Ok, I agree with this. One thing now come to my mind is could we allow
>>> mmc_remove_host() to help remove children in advance or is it okay to
>>> add a new helper for host drivers to remove children in their
>>> ->remove()?
>>
>> Lets' step back a bit. At what point does the remove end up waiting? Does
>> it get to mmc_blk_remove()?
>
> I guess the remove might get to mmc_blk_remove for sd/(e)MMC cards.
Yes, this works for me:
diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 03e3d48b083e..54457d8745cd 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -2968,9 +2968,11 @@ static void mmc_blk_remove(struct mmc_card *card)
mmc_blk_remove_debugfs(card, md);
mmc_blk_remove_parts(card, md);
pm_runtime_get_sync(&card->dev);
- mmc_claim_host(card->host);
- mmc_blk_part_switch(card, md->part_type);
- mmc_release_host(card->host);
+ if (md->part_curr != md->part_type) {
+ mmc_claim_host(card->host);
+ mmc_blk_part_switch(card, md->part_type);
+ mmc_release_host(card->host);
+ }
if (card->type != MMC_TYPE_SD_COMBO)
pm_runtime_disable(&card->dev);
pm_runtime_put_noidle(&card->dev);
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
