Fixing flow control events in watermark buffer#1315
Merged
mattklein123 merged 1 commit intoenvoyproxy:masterfrom Jul 24, 2017
Merged
Fixing flow control events in watermark buffer#1315mattklein123 merged 1 commit intoenvoyproxy:masterfrom
mattklein123 merged 1 commit intoenvoyproxy:masterfrom
Conversation
dnoe
reviewed
Jul 24, 2017
Contributor
dnoe
left a comment
There was a problem hiding this comment.
This doesn't look too unclean to me, I think it's probably the best way. The static cast you do to call postProcess() was already done a few lines earlier so this doesn't seem to be introducing a new assumption.
Looks OK to me if it is OK with @mattklein123
htuch
approved these changes
Jul 24, 2017
mattklein123
approved these changes
Jul 24, 2017
rshriram
pushed a commit
to rshriram/envoy
that referenced
this pull request
Oct 30, 2018
…yproxy#1318) Automatic merge from submit-queue. Use StopIterationAndWatermark instead of StopIterationAndBuffer **What this PR does / why we need it**:Do not buffer request body during Mixer Check() call. This is to avoid buffer overflow when POST request has large body and Mixer filter is waiting for Check() response. **Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes envoyproxy#1315 **Special notes for your reviewer**: **Release note**: ```release-note NONE ```
jpsim
pushed a commit
that referenced
this pull request
Nov 28, 2022
Description: Resolves a security issue flagged by GitHub. Risk Level: Low Testing: CI Signed-off-by: Mike Schore <mike.schore@gmail.com> Signed-off-by: JP Simard <jp@jpsim.com>
jpsim
pushed a commit
that referenced
this pull request
Nov 29, 2022
Description: Resolves a security issue flagged by GitHub. Risk Level: Low Testing: CI Signed-off-by: Mike Schore <mike.schore@gmail.com> Signed-off-by: JP Simard <jp@jpsim.com>
mathetake
pushed a commit
that referenced
this pull request
Mar 3, 2026
**Description** Many AI providers have recently supported prompt caching on the provider side. Cached token prices are significantly cheaper than normal token processing. For example, in Open AI, cached token price is 10x cheaper than normal token [1]. Thus, Envoy AI Gateway would like to take cached token count into account in calculating `llmRequestCosts` in `AIGatewayRequestCosts`. Moreover, for self-hosted LLMs, cached tokens can drastically reduce GPU usage. Thus, in such a case, users would like to care about cached token usage in `llmRequestCosts`. 1: https://openai.com/api/pricing/ --------- Signed-off-by: Shingo Omura <everpeace@gmail.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Previously, low watermark events didn't fire when data was moved out of a watermark buffer. This wasn't caught by existing tests as this code path appears to not be used in the pure TCP case, but is in the H2 path.
I'm not thrilled at buffer_impl having to be mildly aware of watermark buffer's existence but this is the cleanest way I could support move semantics and also watermarks working as expected.