The patchset mainly fixes the sge_num bug relates to sq_inline.
Changes since v2:
1. Move the refactor rq_inline and new CQE inline patches to go
into for-next after this bugfix apply.
2. Change the commit message to better explain the bug.
V2 Link: https://patchwork.kernel.org/project/linux-rdma/cover/[email protected]/
Luoyouming (2):
RDMA/hns: Fix ext_sge num error when post send
RDMA/hns: Fix the problem of sge nums
drivers/infiniband/hw/hns/hns_roce_device.h | 3 +
drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 +-
drivers/infiniband/hw/hns/hns_roce_main.c | 18 +++-
drivers/infiniband/hw/hns/hns_roce_qp.c | 107 ++++++++++++++++----
include/uapi/rdma/hns-abi.h | 15 +++
5 files changed, 125 insertions(+), 21 deletions(-)
--
2.30.0
On Tue, Nov 08, 2022 at 09:38:45PM +0800, Haoyue Xu wrote:
> The patchset mainly fixes the sge_num bug relates to sq_inline.
>
> Changes since v2:
> 1. Move the refactor rq_inline and new CQE inline patches to go
> into for-next after this bugfix apply.
> 2. Change the commit message to better explain the bug.
> V2 Link: https://patchwork.kernel.org/project/linux-rdma/cover/[email protected]/
>
> Luoyouming (2):
> RDMA/hns: Fix ext_sge num error when post send
> RDMA/hns: Fix the problem of sge nums
>
> drivers/infiniband/hw/hns/hns_roce_device.h | 3 +
> drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 +-
> drivers/infiniband/hw/hns/hns_roce_main.c | 18 +++-
> drivers/infiniband/hw/hns/hns_roce_qp.c | 107 ++++++++++++++++----
> include/uapi/rdma/hns-abi.h | 15 +++
Jason,
I think that it was you who asked to send it to -rc. Given the amount of
changes together with UAPI in second patch, are you sure that it is -rc4
material?
Thanks
> 5 files changed, 125 insertions(+), 21 deletions(-)
>
> --
> 2.30.0
>
Hi Jason and Leon,
Actually, I have another bugfix patch set needed to send, and I am waiting for this patch to apply. That one is not related to this patch set.
Could I send it out, since it is a little late for Linux 6.1?
Also, I wonder is that ok to send more than one patch set in the review list.
Sincerely,
haoyue(Aaron)
On 2022/11/10 2:07:35, Leon Romanovsky wrote:
> On Tue, Nov 08, 2022 at 09:38:45PM +0800, Haoyue Xu wrote:
>> The patchset mainly fixes the sge_num bug relates to sq_inline.
>>
>> Changes since v2:
>> 1. Move the refactor rq_inline and new CQE inline patches to go
>> into for-next after this bugfix apply.
>> 2. Change the commit message to better explain the bug.
>> V2 Link: https://patchwork.kernel.org/project/linux-rdma/cover/[email protected]/
>>
>> Luoyouming (2):
>> RDMA/hns: Fix ext_sge num error when post send
>> RDMA/hns: Fix the problem of sge nums
>>
>> drivers/infiniband/hw/hns/hns_roce_device.h | 3 +
>> drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 +-
>> drivers/infiniband/hw/hns/hns_roce_main.c | 18 +++-
>> drivers/infiniband/hw/hns/hns_roce_qp.c | 107 ++++++++++++++++----
>> include/uapi/rdma/hns-abi.h | 15 +++
>
> Jason,
>
> I think that it was you who asked to send it to -rc. Given the amount of
> changes together with UAPI in second patch, are you sure that it is -rc4
> material?
>
> Thanks
>
>> 5 files changed, 125 insertions(+), 21 deletions(-)
>>
>> --
>> 2.30.0
>>
> .
>
On Fri, Nov 11, 2022 at 07:41:43PM +0800, xuhaoyue (A) wrote:
> Hi Jason and Leon,
>
> Actually, I have another bugfix patch set needed to send, and I am waiting for this patch to apply. That one is not related to this patch set.
> Could I send it out, since it is a little late for Linux 6.1?
Up to you and the change you want to send. At this stage, your commit
message must be very descriptive and patch should fix kernel panic and/or
UAPI issue introduced in previous cycle.
> Also, I wonder is that ok to send more than one patch set in the review list.
Yes, you can send, but with two small caveats:
1. You need to try and avoid merge conflicts if series are applied not in
the order of submission.
2. Series should be small.
Bottom line, take a look on RXE patches and DON'T do like them.
Thanks
On Tue, Nov 08, 2022 at 09:38:45PM +0800, Haoyue Xu wrote:
> The patchset mainly fixes the sge_num bug relates to sq_inline.
>
> Changes since v2:
> 1. Move the refactor rq_inline and new CQE inline patches to go
> into for-next after this bugfix apply.
> 2. Change the commit message to better explain the bug.
> V2 Link: https://patchwork.kernel.org/project/linux-rdma/cover/[email protected]/
>
> Luoyouming (2):
> RDMA/hns: Fix ext_sge num error when post send
> RDMA/hns: Fix the problem of sge nums
It is still not clear what the bugs are from the commit messages, and
the second is a bit big, so I put this in for-next
Thanks,
Jason
Sorry about that. I will pay more attention to the commit messages in the future.
Could you give me some advice about it? Which patch makes it confusing?
Thanks,
haoyue
On 2022/11/19 8:23:21, Jason Gunthorpe wrote:
> On Tue, Nov 08, 2022 at 09:38:45PM +0800, Haoyue Xu wrote:
>> The patchset mainly fixes the sge_num bug relates to sq_inline.
>>
>> Changes since v2:
>> 1. Move the refactor rq_inline and new CQE inline patches to go
>> into for-next after this bugfix apply.
>> 2. Change the commit message to better explain the bug.
>> V2 Link: https://patchwork.kernel.org/project/linux-rdma/cover/[email protected]/
>>
>> Luoyouming (2):
>> RDMA/hns: Fix ext_sge num error when post send
>> RDMA/hns: Fix the problem of sge nums
>
> It is still not clear what the bugs are from the commit messages, and
> the second is a bit big, so I put this in for-next
>
> Thanks,
> Jason
> .
>