| Message ID | 1697009600-22367-1-git-send-email-alibuda@linux.alibaba.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | net/smc: bugfixs for smc-r | expand |
The subject of the thread says 'smc-r', but some of the changes affect smc-d alike, don't they? On 11.10.23 09:33, D. Wythe wrote: > From: "D. Wythe" <alibuda@linux.alibaba.com> > > This patches contains bugfix following: > > 1. hung state > 2. sock leak > 3. potential panic > I may be helpful for the reviewers, when you point out, which patch fixes which problem. Were they all found by code reviews? Or did some occur in real life? If so, then what were the symptoms? A description of the symptoms is helpful for somebody who is debugging and wants to check whether the issue was already fixed upstream.
On 10/12/23 9:43 PM, Alexandra Winter wrote: > The subject of the thread says 'smc-r', but some of the changes affect smc-d alike, > don't they? Yes, sorry for this mistake, it should be bugfix for smc. > > > On 11.10.23 09:33, D. Wythe wrote: >> From: "D. Wythe" <alibuda@linux.alibaba.com> >> >> This patches contains bugfix following: >> >> 1. hung state >> 2. sock leak >> 3. potential panic >> > I may be helpful for the reviewers, when you point out, which patch fixes which problem. > > Were they all found by code reviews? > Or did some occur in real life? If so, then what were the symptoms? > A description of the symptoms is helpful for somebody who is debugging and wants to check > whether the issue was already fixed upstream. Hi Alexandra, Except for the issue with the barrier, which was feedback from the review, all other issues have actually occurred in our environment and have been verified through internal testing. However, most of these issues are caused by reference leakage rather than panic, so it is difficult to provide a representative phenomenon. But what you said is do necessary, so I will post some phenomena in the next version, such as lsmod | grep smc or smcss - a In that case, we can found the issues of reference residue or the connection residue. Hope it can be helpful to you. Thanks, D. Wythe
From: "D. Wythe" <alibuda@linux.alibaba.com> This patches contains bugfix following: 1. hung state 2. sock leak 3. potential panic We have been testing these patches for some time, but if you have any questions, please let us know. Thanks, D. Wythe D. Wythe (5): net/smc: fix dangling sock under state SMC_APPFINCLOSEWAIT net/smc: fix incorrect barrier usage net/smc: allow cdc msg send rather than drop it with NULL sndbuf_desc net/smc: protect connection state transitions in listen work net/smc: put sk reference if close work was canceled net/smc/af_smc.c | 9 +++++++-- net/smc/smc.h | 5 +++++ net/smc/smc_cdc.c | 11 +++++------ net/smc/smc_close.c | 5 +++-- net/smc/smc_core.c | 21 +++++++++++++-------- 5 files changed, 33 insertions(+), 18 deletions(-)