[PATCH 2/9] block: add scatterlist-less DMA mapping helpers
Daniel Gomez
da.gomez at kernel.org
Mon Jun 16 05:52:15 PDT 2025
On 16/06/2025 14.42, Christoph Hellwig wrote:
> On Mon, Jun 16, 2025 at 02:37:43PM +0200, Daniel Gomez wrote:
>>
>> 2. In the if/switch case default case (line 181-183):
>> It sets the status to invalid and *iter_start() returns false:
>>
>> default:
>> iter->status = BLK_STS_INVAL;
>> return false;
>>
>>
>> Removing the invalid assignment "makes it" valid (because of the
>> initialization)
>
> Huhhh? How do you make a field access valid by dropping an assignment?
>
It should be valid already. But if the initialized value at line 160 can be
changed through the previous code then, this should suffice:
diff --git a/block/blk-mq-dma.c b/block/blk-mq-dma.c
index 37f8fba077e6..7d5ebe1a3f49 100644
--- a/block/blk-mq-dma.c
+++ b/block/blk-mq-dma.c
@@ -179,7 +179,7 @@ bool blk_rq_dma_map_iter_start(struct request *req, struct device *dma_dev,
req->cmd_flags &= ~REQ_P2PDMA;
break;
default:
- iter->status = BLK_STS_INVAL;
+ iter->status = BLK_STS_OK;
return false;
}
}
More information about the Linux-nvme
mailing list