| Message ID | 20221220031223.3890143-1-shaozhengchao@huawei.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | 9p/rdma: unmap receive dma buffer in rdma_request() | expand |
On Tue, Dec 20, 2022 at 11:12:23AM +0800, Zhengchao Shao wrote: > When down_interruptible() failed in rdma_request(), receive dma buffer > is not unmapped. Add unmap action to error path. > > Fixes: fc79d4b104f0 ("9p: rdma: RDMA Transport Support for 9P") > Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> > --- > net/9p/trans_rdma.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c > index 83f9100d46bf..da83023fecbf 100644 > --- a/net/9p/trans_rdma.c > +++ b/net/9p/trans_rdma.c > @@ -499,6 +499,8 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) > > if (down_interruptible(&rdma->sq_sem)) { > err = -EINTR; > + ib_dma_unmap_single(rdma->cm_id->device, c->busa, > + c->req->tc.size, DMA_TO_DEVICE); > goto send_error; > } It is not the only place where ib_dma_unmap_single() wasn't called. Even at the same function if ib_post_send() fails, the unmap is not called. Also post_recv() is missing call to ib_dma_unmap_single() too. Thanks > > -- > 2.34.1 >
On 2022/12/28 17:35, Leon Romanovsky wrote: > On Tue, Dec 20, 2022 at 11:12:23AM +0800, Zhengchao Shao wrote: >> When down_interruptible() failed in rdma_request(), receive dma buffer >> is not unmapped. Add unmap action to error path. >> >> Fixes: fc79d4b104f0 ("9p: rdma: RDMA Transport Support for 9P") >> Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> >> --- >> net/9p/trans_rdma.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c >> index 83f9100d46bf..da83023fecbf 100644 >> --- a/net/9p/trans_rdma.c >> +++ b/net/9p/trans_rdma.c >> @@ -499,6 +499,8 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) >> >> if (down_interruptible(&rdma->sq_sem)) { >> err = -EINTR; >> + ib_dma_unmap_single(rdma->cm_id->device, c->busa, >> + c->req->tc.size, DMA_TO_DEVICE); >> goto send_error; >> } > > It is not the only place where ib_dma_unmap_single() wasn't called. > Even at the same function if ib_post_send() fails, the unmap is not > called. Also post_recv() is missing call to ib_dma_unmap_single() too. > > Thanks > Hi Leon： Thank you for your review. I'm sorry I haven't answered your message for so long, I've got Coronavirus and it's a terrible feeling. I will send v2 soon. Zhengchao Shao >> >> -- >> 2.34.1 >>
diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c index 83f9100d46bf..da83023fecbf 100644 --- a/net/9p/trans_rdma.c +++ b/net/9p/trans_rdma.c @@ -499,6 +499,8 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) if (down_interruptible(&rdma->sq_sem)) { err = -EINTR; + ib_dma_unmap_single(rdma->cm_id->device, c->busa, + c->req->tc.size, DMA_TO_DEVICE); goto send_error; }
When down_interruptible() failed in rdma_request(), receive dma buffer is not unmapped. Add unmap action to error path. Fixes: fc79d4b104f0 ("9p: rdma: RDMA Transport Support for 9P") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> --- net/9p/trans_rdma.c | 2 ++ 1 file changed, 2 insertions(+)