bpf: improve handling for short packets #25159
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
julianwiedmann
added
kind/bug
maintainer-s-little-helper
bot
added
dont-merge/needs-release-note-label
julianwiedmann
added
the
release-note/bug
maintainer-s-little-helper
bot
removed
the
dont-merge/needs-release-note-label
|
/test Job 'Cilium-PR-K8s-1.26-kernel-net-next' hit: #24697 (89.50% similarity) |
|
As mlh noted, |
|
/test-1.26-net-next Job 'Cilium-PR-K8s-1.26-kernel-net-next' failed: Click to show.Test NameFailure OutputJenkins URL: https://jenkins.cilium.io/job/Cilium-PR-K8s-1.26-kernel-net-next/1990/ If it is a flake and a GitHub issue doesn't already exist to track it, comment Then please upload the Jenkins artifacts to that issue. |
ldelossa
approved these changes
May 8, 2023
bpf/bpf_host.c
Outdated
| @@ -828,7 +821,16 @@ do_netdev(struct __ctx_buff *ctx, __u16 proto, const bool from_host) | |||
| #endif | |||
| #ifdef ENABLE_IPV4 | |||
| case bpf_htons(ETH_P_IP): | |||
| identity = resolve_srcid_ipv4(ctx, identity, &ipcache_srcid, | |||
| /* This is the first time revalidate_data() is going to be called in | |||
There was a problem hiding this comment.
Could you potentially drop this chunk all together? Looks like you nullify it in the next commit anyway
There was a problem hiding this comment.
You mean like shuffling the second and third patch around? Yeah let's try 🤔.
There was a problem hiding this comment.
yes, looks like you add this chunk, but then delete it in the third commit anyway.
Anyway, its a nit, is just to make the history a bit cleaner.
af_packet can craft packets without linear data. So make sure that we pull the IPv4 header, as resolve_srcid_ipv4() won't do it for us (we call it with from_host = true). Also update the comment to match current behaviour. This aligns the IPv4 path with handle_to_netdev_ipv6(). Signed-off-by: Julian Wiedmann <jwi@isovalent.com>
Don't trust that host traffic arrives with the IP header in the skb's linear space. So have resolve_srcid_ipv*() always pull the necessary data, without differentiating between from-netdev and from-host traffic. This means that for to-netdev traffic we now have two places that can handle the traffic. This is just temporary, and will go away with the next patch. Signed-off-by: Julian Wiedmann <jwi@isovalent.com>
resolve_srcid_ipv*() is meant to return an __u32, that represents the srcid. But it currently also validates the IP header, and returns DROP_INVALID from that. We use this value without further checks. Fix things up by performing the IP header validation & error handling outside the helper. Note that the to-netdev path already contains a revalidate_data_pull() call, so it's safe to lose this additional one inside resolve_srcid_ipv*(). Signed-off-by: Julian Wiedmann <jwi@isovalent.com>
julianwiedmann
force-pushed
the
1.14-bpf-resolve_srcid
branch
from
09b6526
to
e0b9fc2
Compare
|
/test |
maintainer-s-little-helper
bot
added
the
ready-to-merge
3 tasks
julianwiedmann
added
the
backport/author
1 task
julianwiedmann
added
the
backport-pending/1.13
julianwiedmann
added
backport-done/1.13
gentoo-root
moved this from Backport pending to v1.13
to Backport done to v1.13
in 1.13.5
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Align the different packet paths in how they handle traffic where the IP header is not in the skb's linear data.