| Message ID | 20230808075233.3337922-1-idosch@nvidia.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | nexthop: Nexthop dump fixes | expand |
Hello: This series was applied to netdev/net.git (main) by Jakub Kicinski <kuba@kernel.org>: On Tue, 8 Aug 2023 10:52:30 +0300 you wrote: > Patches #1 and #3 fix two problems related to nexthops and nexthop > buckets dump, respectively. Patch #2 is a preparation for the third > patch. > > The pattern described in these patches of splitting the NLMSG_DONE to a > separate response is prevalent in other rtnetlink dump callbacks. I > don't know if it's because I'm missing something or if this was done > intentionally to ensure the message is delivered to user space. After > commit 0642840b8bb0 ("af_netlink: ensure that NLMSG_DONE never fails in > dumps") this is no longer necessary and I can improve these dump > callbacks assuming this analysis is correct. > > [...] Here is the summary with links: - [net,1/3] nexthop: Fix infinite nexthop dump when using maximum nexthop ID https://git.kernel.org/netdev/net/c/913f60cacda7 - [net,2/3] nexthop: Make nexthop bucket dump more efficient https://git.kernel.org/netdev/net/c/f10d3d9df49d - [net,3/3] nexthop: Fix infinite nexthop bucket dump when using maximum nexthop ID https://git.kernel.org/netdev/net/c/8743aeff5bc4 You are awesome, thank you!
Patches #1 and #3 fix two problems related to nexthops and nexthop buckets dump, respectively. Patch #2 is a preparation for the third patch. The pattern described in these patches of splitting the NLMSG_DONE to a separate response is prevalent in other rtnetlink dump callbacks. I don't know if it's because I'm missing something or if this was done intentionally to ensure the message is delivered to user space. After commit 0642840b8bb0 ("af_netlink: ensure that NLMSG_DONE never fails in dumps") this is no longer necessary and I can improve these dump callbacks assuming this analysis is correct. No regressions in existing tests: # ./fib_nexthops.sh [...] Tests passed: 230 Tests failed: 0 Ido Schimmel (3): nexthop: Fix infinite nexthop dump when using maximum nexthop ID nexthop: Make nexthop bucket dump more efficient nexthop: Fix infinite nexthop bucket dump when using maximum nexthop ID net/ipv4/nexthop.c | 28 ++++++--------------- tools/testing/selftests/net/fib_nexthops.sh | 10 ++++++++ 2 files changed, 17 insertions(+), 21 deletions(-)