| Message ID | 168321371754.16695.4217960864733718685.stgit@oracle-102.nfsv4bat.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | Bug fixes for net/handshake | expand |
On Thu, 04 May 2023 11:24:12 -0400 Chuck Lever wrote:
> I plan to send these as part of a 6.4-rc PR.
Can you elaborate? You'll send us the same code as PR?
I'm about to send the first batch of fixes to Linus,
I was going to apply this series.
On Fri, May 05, 2023 at 01:39:18PM -0700, Jakub Kicinski wrote: > On Thu, 04 May 2023 11:24:12 -0400 Chuck Lever wrote: > > I plan to send these as part of a 6.4-rc PR. > > Can you elaborate? You'll send us the same code as PR? > I'm about to send the first batch of fixes to Linus, > I was going to apply this series. Since I am listed as a maintainer/supporter of net/handshake, I assumed I can and should be sending changes through nfsd or some other repo I can commit to. netdev@ is also listed in MAINTAINERS, so I Cc'd you all on this series. I did not intend for you to be responsible for merging the series. We'll need to agree on a workflow going forward. Since you had some review questions about one of the patches, maybe that patch should not be merged at this time by either of us.
On Fri, 5 May 2023 19:16:40 -0400 Chuck Lever wrote: > On Fri, May 05, 2023 at 01:39:18PM -0700, Jakub Kicinski wrote: > > On Thu, 04 May 2023 11:24:12 -0400 Chuck Lever wrote: > > > I plan to send these as part of a 6.4-rc PR. > > > > Can you elaborate? You'll send us the same code as PR? > > I'm about to send the first batch of fixes to Linus, > > I was going to apply this series. > > Since I am listed as a maintainer/supporter of net/handshake, I > assumed I can and should be sending changes through nfsd or some > other repo I can commit to. > > netdev@ is also listed in MAINTAINERS, so I Cc'd you all on this > series. I did not intend for you to be responsible for merging the > series. We'll need to agree on a workflow going forward. Let me talk to DaveM and Paolo -- with NFS being the main user taking it via your trees is likely fine. But if it's a generic TLS handshake and other users will appear - netdev trees may be a more natural central point :S DaveM and Paolo are more familiar with existing cases of similar nature (rxrpc?).. > Since you had some review questions about one of the patches, > maybe that patch should not be merged at this time by either of us. Right, I noticed the cover message first then started looking more in depth at the code :)
On Fri, May 05, 2023 at 04:47:15PM -0700, Jakub Kicinski wrote: > On Fri, 5 May 2023 19:16:40 -0400 Chuck Lever wrote: > > On Fri, May 05, 2023 at 01:39:18PM -0700, Jakub Kicinski wrote: > > > On Thu, 04 May 2023 11:24:12 -0400 Chuck Lever wrote: > > > > I plan to send these as part of a 6.4-rc PR. > > > > > > Can you elaborate? You'll send us the same code as PR? > > > I'm about to send the first batch of fixes to Linus, > > > I was going to apply this series. > > > > Since I am listed as a maintainer/supporter of net/handshake, I > > assumed I can and should be sending changes through nfsd or some > > other repo I can commit to. > > > > netdev@ is also listed in MAINTAINERS, so I Cc'd you all on this > > series. I did not intend for you to be responsible for merging the > > series. We'll need to agree on a workflow going forward. > > Let me talk to DaveM and Paolo -- with NFS being the main user > taking it via your trees is likely fine. But if it's a generic TLS > handshake and other users will appear - netdev trees may be a more > natural central point :S DaveM and Paolo are more familiar with > existing cases of similar nature (rxrpc?).. Makes sense. We expect NVMe to become a consumer in the near future, and have considered a putative in-kernel QUIC implementation to be another potential consumer down the road.
On Fri, 2023-05-05 at 16:47 -0700, Jakub Kicinski wrote: > On Fri, 5 May 2023 19:16:40 -0400 Chuck Lever wrote: > > On Fri, May 05, 2023 at 01:39:18PM -0700, Jakub Kicinski wrote: > > > On Thu, 04 May 2023 11:24:12 -0400 Chuck Lever wrote: > > > > I plan to send these as part of a 6.4-rc PR. > > > > > > Can you elaborate? You'll send us the same code as PR? > > > I'm about to send the first batch of fixes to Linus, > > > I was going to apply this series. > > > > Since I am listed as a maintainer/supporter of net/handshake, I > > assumed I can and should be sending changes through nfsd or some > > other repo I can commit to. > > > > netdev@ is also listed in MAINTAINERS, so I Cc'd you all on this > > series. I did not intend for you to be responsible for merging the > > series. We'll need to agree on a workflow going forward. > > Let me talk to DaveM and Paolo -- with NFS being the main user > taking it via your trees is likely fine. But if it's a generic TLS > handshake and other users will appear - netdev trees may be a more > natural central point :S DaveM and Paolo are more familiar with > existing cases of similar nature (rxrpc?).. Really, I' not ;) My guess is that net/handshake is going to be dependent more on core networking changes than anything else. If later developments will require/use/leverage a new core net helper, it would be quite straight- forward going trough the netdev trees. Otherwise such changes will require extra coordination and/or an additional RTT WRT kernel releases. All the above very much IMHO ;) /P
I plan to send these as part of a 6.4-rc PR. --- Chuck Lever (5): net/handshake: Remove unneeded check from handshake_dup() net/handshake: Fix handshake_dup() ref counting net/handshake: Fix uninitialized local variable net/handshake: handshake_genl_notify() shouldn't ignore @flags net/handshake: Enable the SNI extension to work properly Documentation/netlink/specs/handshake.yaml | 4 ++++ Documentation/networking/tls-handshake.rst | 5 +++++ include/net/handshake.h | 1 + include/uapi/linux/handshake.h | 1 + net/handshake/netlink.c | 17 +++++------------ net/handshake/tlshd.c | 8 ++++++++ 6 files changed, 24 insertions(+), 12 deletions(-) -- Chuck Lever