| Message ID | 20200318150257.198402-3-maxg@mellanox.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
| Series | None | expand |
On Wed, Mar 18, 2020 at 05:02:54PM +0200, Max Gurtovoy wrote: > This is a preparetion patch for the SRQ per completion vector feature. > > Signed-off-by: Max Gurtovoy <maxg@mellanox.com> > --- > drivers/nvme/target/rdma.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) Max, how are you sending these emails, and why don't they thread properly on patchworks: https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 This patch is missing from the series v1 had the same issue Very strange. Can you fix it? Jason
On 2020-03-18 08:02, Max Gurtovoy wrote:
> This is a preparetion patch for the SRQ per completion vector feature.
^^^^^^^^^^^
preparation?
Thanks,
Bart.
On 3/19/2020 1:32 AM, Jason Gunthorpe wrote: > On Wed, Mar 18, 2020 at 05:02:54PM +0200, Max Gurtovoy wrote: >> This is a preparetion patch for the SRQ per completion vector feature. >> >> Signed-off-by: Max Gurtovoy <maxg@mellanox.com> >> --- >> drivers/nvme/target/rdma.c | 6 ++++-- >> 1 file changed, 4 insertions(+), 2 deletions(-) > Max, how are you sending these emails, and why don't they thread > properly on patchworks: > > https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 > > This patch is missing from the series > > v1 had the same issue > > Very strange. Can you fix it? I'm using "git send-email". Should I use some special flags or CC another email for it ? How do you suggest sending patches so we'll see it on patchworks ? > > Jason
On Thu, Mar 19, 2020 at 10:48:22AM +0200, Max Gurtovoy wrote: > > On 3/19/2020 1:32 AM, Jason Gunthorpe wrote: > > On Wed, Mar 18, 2020 at 05:02:54PM +0200, Max Gurtovoy wrote: > > > This is a preparetion patch for the SRQ per completion vector feature. > > > > > > Signed-off-by: Max Gurtovoy <maxg@mellanox.com> > > > --- > > > drivers/nvme/target/rdma.c | 6 ++++-- > > > 1 file changed, 4 insertions(+), 2 deletions(-) > > Max, how are you sending these emails, and why don't they thread > > properly on patchworks: > > > > https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 > > > > This patch is missing from the series > > > > v1 had the same issue > > > > Very strange. Can you fix it? > > I'm using "git send-email". > > Should I use some special flags or CC another email for it ? > > How do you suggest sending patches so we'll see it on patchworks ? It can be related to your SMTP relay, I see that you are sending patches through labmailer, and it is known as unreliable. Use the same SMTP server as you are using to send emails. Thanks > > > > > Jason
On 3/19/2020 11:14 AM, Leon Romanovsky wrote: > On Thu, Mar 19, 2020 at 10:48:22AM +0200, Max Gurtovoy wrote: >> On 3/19/2020 1:32 AM, Jason Gunthorpe wrote: >>> On Wed, Mar 18, 2020 at 05:02:54PM +0200, Max Gurtovoy wrote: >>>> This is a preparetion patch for the SRQ per completion vector feature. >>>> >>>> Signed-off-by: Max Gurtovoy <maxg@mellanox.com> >>>> --- >>>> drivers/nvme/target/rdma.c | 6 ++++-- >>>> 1 file changed, 4 insertions(+), 2 deletions(-) >>> Max, how are you sending these emails, and why don't they thread >>> properly on patchworks: >>> >>> https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 >>> >>> This patch is missing from the series >>> >>> v1 had the same issue >>> >>> Very strange. Can you fix it? >> I'm using "git send-email". >> >> Should I use some special flags or CC another email for it ? >> >> How do you suggest sending patches so we'll see it on patchworks ? > It can be related to your SMTP relay, I see that you are sending > patches through labmailer, and it is known as unreliable. Use the > same SMTP server as you are using to send emails. I'll use it in V3. Thanks. > > Thanks > >>> Jason
On Thu, Mar 19, 2020 at 10:48:22AM +0200, Max Gurtovoy wrote: > > On 3/19/2020 1:32 AM, Jason Gunthorpe wrote: > > On Wed, Mar 18, 2020 at 05:02:54PM +0200, Max Gurtovoy wrote: > > > This is a preparetion patch for the SRQ per completion vector feature. > > > > > > Signed-off-by: Max Gurtovoy <maxg@mellanox.com> > > > drivers/nvme/target/rdma.c | 6 ++++-- > > > 1 file changed, 4 insertions(+), 2 deletions(-) > > Max, how are you sending these emails, and why don't they thread > > properly on patchworks: > > > > https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 > > > > This patch is missing from the series > > > > v1 had the same issue > > > > Very strange. Can you fix it? > > I'm using "git send-email". > > Should I use some special flags or CC another email for it ? > > How do you suggest sending patches so we'll see it on patchworks ? I looked at these mails and they seem properly threaded/etc. I've added Konstantin, perhaps he knows why patchworks is acting weird here? Jason
On Thu, Mar 19, 2020 at 08:54:31AM -0300, Jason Gunthorpe wrote: > > > Max, how are you sending these emails, and why don't they thread > > > properly on patchworks: > > > > > > https://patchwork.kernel.org/project/linux-rdma/list/?series=258271 > > > > > > This patch is missing from the series > > > > > > v1 had the same issue > > > > > > Very strange. Can you fix it? > > > > I'm using "git send-email". > > > > Should I use some special flags or CC another email for it ? > > > > How do you suggest sending patches so we'll see it on patchworks ? > > I looked at these mails and they seem properly threaded/etc. > > I've added Konstantin, perhaps he knows why patchworks is acting > weird here? Not sure. Everything appears to be properly threaded. I see 2/5 arriving at precisely the same time as the cover letter, so 2/5 probably got processed before 0/5. I have no idea if that actually matters to patchwork -- a whole bunch of series would break if arrival ordering was that important. At least I assume that would be the case. I'll check with upstream. -K
On Thu, Mar 19, 2020 at 08:54:31AM -0300, Jason Gunthorpe wrote: > > I'm using "git send-email". > > > > Should I use some special flags or CC another email for it ? > > > > How do you suggest sending patches so we'll see it on patchworks ? > > I looked at these mails and they seem properly threaded/etc. > > I've added Konstantin, perhaps he knows why patchworks is acting > weird here? Looks like it's hitting a race condition in patchwork that will be fixed in the next release. Until then, you can grab these patches from lore.kernel.org (e.g. with b4), though it won't help with git-patchwork-bot integration. -K
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 37d262a..04062af 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -38,6 +38,7 @@ struct nvmet_rdma_cmd { struct scatterlist inline_sg[NVMET_RDMA_MAX_INLINE_SGE]; struct nvme_command *nvme_cmd; struct nvmet_rdma_queue *queue; + struct ib_srq *srq; }; enum { @@ -461,8 +462,8 @@ static int nvmet_rdma_post_recv(struct nvmet_rdma_device *ndev, cmd->sge[0].addr, cmd->sge[0].length, DMA_FROM_DEVICE); - if (ndev->srq) - ret = ib_post_srq_recv(ndev->srq, &cmd->wr, NULL); + if (cmd->srq) + ret = ib_post_srq_recv(cmd->srq, &cmd->wr, NULL); else ret = ib_post_recv(cmd->queue->cm_id->qp, &cmd->wr, NULL); @@ -882,6 +883,7 @@ static int nvmet_rdma_init_srq(struct nvmet_rdma_device *ndev) ndev->srq_size = srq_size; for (i = 0; i < srq_size; i++) { + ndev->srq_cmds[i].srq = srq; ret = nvmet_rdma_post_recv(ndev, &ndev->srq_cmds[i]); if (ret) goto out_free_cmds;
This is a preparetion patch for the SRQ per completion vector feature. Signed-off-by: Max Gurtovoy <maxg@mellanox.com> --- drivers/nvme/target/rdma.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)