improve error handling for einval with lpm trie. #1720
Draft
+6
−1
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
In case of invalid key (i.e. not following lpm rules) the default error message is quite misleading: `noPrealloc flag may be incompatible with map type` In fact, lpm-trie requires no prealloc to be enabled so this error is likely to waste a confused programmers time. Instead, we validate key params and return a specific error. Signed-off-by: Tom Hadlaw <[email protected]>
chenhengqi
reviewed
Mar 12, 2025
| @@ -556,7 +556,12 @@ func handleMapCreateError(attr sys.MapCreateAttr, spec *MapSpec, err error) erro | |||
| if errors.Is(err, unix.EINVAL) && spec.Type == UnspecifiedMap { | |||
| return fmt.Errorf("map create: cannot use type %s", UnspecifiedMap) | |||
| } | |||
| if errors.Is(err, unix.EINVAL) && spec.Flags&sys.BPF_F_NO_PREALLOC > 0 { | |||
| // LPM trie key size must be a multiple of 8 and not exceed 2048. | |||
There was a problem hiding this comment.
This seems incorrect according to my experience and the doc https://docs.ebpf.io/linux/map-type/BPF_MAP_TYPE_LPM_TRIE/.
ti-mo
requested changes
Mar 12, 2025
| if errors.Is(err, unix.EINVAL) && spec.Flags&sys.BPF_F_NO_PREALLOC > 0 { | ||
| // LPM trie key size must be a multiple of 8 and not exceed 2048. | ||
| // Ref: https://docs.kernel.org/bpf/map_lpm_trie.html#bpf-map-type-lpm-trie | ||
| if errors.Is(err, unix.EINVAL) && spec.Type == LPMTrie && (spec.KeySize%8 != 0 || spec.KeySize > 2048) { |
There was a problem hiding this comment.
This is becoming a bit too specific for me. The goal is not to be exhaustive, and the wording here has always been soft on purpose. Could you drop this change?
The actual problem is a false-positive on the error message. I suggest adding the the following to types.go:
func (mt MapType) mustHaveNoPrealloc() bool { }
I'd check the kernel sources for all currently-known cases and make this function return true for LPM, cgroup storage and others that require NO_PREALLOC to be set.
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.
In case of invalid key (i.e. not following lpm rules) the default error message is quite misleading:
noPrealloc flag may be incompatible with map typeIn fact, lpm-trie requires no prealloc to be enabled so this error is likely to waste a confused programmers time.
Instead, we validate key params and return a specific error.