[PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[Jack Wang]

ib_dma_map_sg return 0 on error.

Cc: Keith Busch <kbusch@kernel.org>
Cc: Jens Axboe <axboe@fb.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: linux-nvme@lists.infradead.org
Cc: linux-kernel@vger.kernel.org

Signed-off-by: Jack Wang <jinpu.wang@ionos.com>
---
 drivers/nvme/host/rdma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 3100643be299..d70bccbcba3e 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -1551,7 +1551,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq,
 
 	*count = ib_dma_map_sg(ibdev, req->data_sgl.sg_table.sgl,
 			       req->data_sgl.nents, rq_dma_dir(rq));
-	if (unlikely(*count <= 0)) {
+	if (unlikely(!*count)) {
 		ret = -EIO;
 		goto out_free_table;
 	}
@@ -1574,7 +1574,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq,
 					  req->metadata_sgl->sg_table.sgl,
 					  req->metadata_sgl->nents,
 					  rq_dma_dir(rq));
-		if (unlikely(*pi_count <= 0)) {
+		if (unlikely(!*pi_count)) {
 			ret = -EIO;
 			goto out_free_pi_table;
 		}
-- 
2.34.1

[PATCH v1 15/19] nvme-fc: Fix the error check for dma_map_sg

[Jack Wang]

dma_map_sg return 0 on error, add missing check.

Cc: James Smart <james.smart@broadcom.com>
Cc: Keith Busch <kbusch@kernel.org>
Cc: Jens Axboe <axboe@fb.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: linux-nvme@lists.infradead.org
Cc: linux-kernel@vger.kernel.org

Signed-off-by: Jack Wang <jinpu.wang@ionos.com>
---
 drivers/nvme/host/fc.c   | 2 +-
 drivers/nvme/target/fc.c | 5 +++++
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index 127abaf9ba5d..95050f2e0a06 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -2616,7 +2616,7 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 	WARN_ON(op->nents > blk_rq_nr_phys_segments(rq));
 	freq->sg_cnt = fc_dma_map_sg(ctrl->lport->dev, freq->sg_table.sgl,
 				op->nents, rq_dma_dir(rq));
-	if (unlikely(freq->sg_cnt <= 0)) {
+	if (unlikely(!freq->sg_cnt)) {
 		sg_free_table_chained(&freq->sg_table, NVME_INLINE_SG_CNT);
 		freq->sg_cnt = 0;
 		return -EFAULT;
diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c
index ab2627e17bb9..3749ca28860b 100644
--- a/drivers/nvme/target/fc.c
+++ b/drivers/nvme/target/fc.c
@@ -2089,6 +2089,11 @@ nvmet_fc_alloc_tgt_pgs(struct nvmet_fc_fcp_iod *fod)
 				((fod->io_dir == NVMET_FCP_WRITE) ?
 					DMA_FROM_DEVICE : DMA_TO_DEVICE));
 				/* note: write from initiator perspective */
+	if (!fod->data_sg_cnt) {
+		sgl_free(fod->data_sg);
+		fod->data_sg = NULL;
+		goto out;
+	}
 	fod->next_sg = fod->data_sg;
 
 	return 0;
-- 
2.34.1

Re: [PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[Christoph Hellwig]

Please don't send me just two random patches out of a series, as I
have no way to review them.  If the patches are independent, send them
independently and if they depend on common prep work send the entire
series to everyone.

Re: [PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[Jinpu Wang]

On Sun, Aug 21, 2022 at 7:57 AM Christoph Hellwig <hch@lst.de> wrote:
>
> Please don't send me just two random patches out of a series, as I
> have no way to review them.  If the patches are independent, send them
> independently and if they depend on common prep work send the entire
> series to everyone.
Hi Christoph,

As mentioned in the cover letter, the bugfixes are all independent,
for each of the drivers.

I will wait a bit to see if there is other comments,  resend as you
suggested later.

Thx!

Re: [PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[Christoph Hellwig]

On Mon, Aug 22, 2022 at 07:12:23AM +0200, Jinpu Wang wrote:
> On Sun, Aug 21, 2022 at 7:57 AM Christoph Hellwig <hch@lst.de> wrote:
> >
> > Please don't send me just two random patches out of a series, as I
> > have no way to review them.  If the patches are independent, send them
> > independently and if they depend on common prep work send the entire
> > series to everyone.
> Hi Christoph,
> 
> As mentioned in the cover letter, the bugfixes are all independent,
> for each of the drivers.

I still haven't seen the cover letter as it was never sent to me.
But if patches are independent please send a series per subsystem
as that makes that very clear.

Re: [PATCH v1 14/19] nvme-rdma: Fix error check for ib_dma_map_sg

[Jinpu Wang]

On Mon, Aug 22, 2022 at 8:25 AM Christoph Hellwig <hch@lst.de> wrote:
>
> On Mon, Aug 22, 2022 at 07:12:23AM +0200, Jinpu Wang wrote:
> > On Sun, Aug 21, 2022 at 7:57 AM Christoph Hellwig <hch@lst.de> wrote:
> > >
> > > Please don't send me just two random patches out of a series, as I
> > > have no way to review them.  If the patches are independent, send them
> > > independently and if they depend on common prep work send the entire
> > > series to everyone.
> > Hi Christoph,
> >
> > As mentioned in the cover letter, the bugfixes are all independent,
> > for each of the drivers.
>
> I still haven't seen the cover letter as it was never sent to me.
> But if patches are independent please send a series per subsystem
> as that makes that very clear.
Thanks for the hint, this definitely helps the subsystem maintainer to
get a clear background.
I'll do it in the future.