Re: [PATCH 2/2] nvme: blk_rq_dma_map_iter_next is no longer using iova state

From: Nitesh Shetty

Date: Mon Jan 12 2026 - 09:32:41 EST

On 12/01/26 03:08PM, Damien Le Moal wrote:
On 1/12/26 14:57, Nitesh Shetty wrote:
DMA IOVA state is not used inside blk_rq_dma_map_iter_next

Signed-off-by: Nitesh Shetty <nj.shetty@xxxxxxxxxxx>
---
drivers/nvme/host/pci.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 3b528369f5454..065555576d2f9 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -823,7 +823,7 @@ static bool nvme_pci_prp_iter_next(struct request *req, struct device *dma_dev,

if (iter->len)
return true;
- if (!blk_rq_dma_map_iter_next(req, dma_dev, &iod->dma_state, iter))
+ if (!blk_rq_dma_map_iter_next(req, dma_dev, iter))

Hu... Why is this not squashed with the previous patch ? If only patch 1 is
applied, this will not compile, right ?

I couldn’t decide whether to use the layering convention or a unified patch,
so I chose one patch per layer.
Agreed, independently this doesn't compile, merging make sense.
I will resend.

Thanks,
Nitesh