💬 instagibbs commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372314425)
huh, huge reading comprehension issues. I distinctly recall checking, sad!

💬 sipa commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372336256)
Done.

💬 sipa commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372336983)
Done something like this.

💬 sipa commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372342553)
Added a comment that `BlockBuilder`s are excluded from the memory usage.

Agreed w.r.t `ordered_set`, but will leave for a follow-up.

💬 vasild commented on pull request "net: support overriding the proxy selection in ConnectNode()":
(https://github.com/bitcoin/bitcoin/pull/33454#issuecomment-3324028544)
`48c5939cf4...ec7c215345`: do https://github.com/bitcoin/bitcoin/pull/33454#discussion_r2372128145

💬 sipa commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372343531)
Added an `Assume(graph_idx != GraphIndex(-1));` here for extra safety.

💬 vasild commented on pull request "net: support overriding the proxy selection in ConnectNode()":
(https://github.com/bitcoin/bitcoin/pull/33454#discussion_r2372343874)
Right, done!

💬 sipa commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#discussion_r2372343983)
Done.

✅ fanquake closed an issue: "oss-fuzz: build failing"
(https://github.com/bitcoin/bitcoin/issues/33366)

💬 Kruwed commented on pull request "docs: Undeprecate datacarrier and datacarriersize configuration options":
(https://github.com/bitcoin/bitcoin/pull/33453#issuecomment-3324053860)
>There are more people who want to set it to non-default value today than ever before.

This is a justification for why the deprecated options should be removed as quickly as possible, not a justification for making a footgun available to these people.

🚀 fanquake merged a pull request: "rpc: Always return per-wtxid entries in submitpackage tx-results"
(https://github.com/bitcoin/bitcoin/pull/33427)

💬 moonsettler commented on pull request "docs: Undeprecate datacarrier and datacarriersize configuration options":
(https://github.com/bitcoin/bitcoin/pull/33453#issuecomment-3324066344)
> This is a justification for why the deprecated options should be removed as quickly as possible, not a justification for making a footgun available to these people.

It's not any more a "footgun" than `blocksonly`. Which is not deprecated. That word should be reserved for situations that can result in loss of funds.

💬 waketraindev commented on pull request "docs: Undeprecate datacarrier and datacarriersize configuration options":
(https://github.com/bitcoin/bitcoin/pull/33453#issuecomment-3324075316)
> > This is a justification for why the deprecated options should be removed as quickly as possible, not a justification for making a footgun available to these people.
>
> It's not any more a "footgun" than `blocksonly`. Which is not deprecated. That word should be reserved for situations that can result in loss of funds.

Blocks only doesn't request transactions that after are denied only to be downloaded again when they are confirmed or have a child

💬 fanquake commented on pull request "ci: Skip Centos job on GHA runners":
(https://github.com/bitcoin/bitcoin/pull/33365#issuecomment-3324082582)
@maflcko any opinion here?

👍 vasild approved a pull request: "ci: Update Clang in "tidy" job"
(https://github.com/bitcoin/bitcoin/pull/33445#pullrequestreview-3258115531)
ACK 1bb0f7f8b6d927046fefa2eb29d0132713415cca

💬 maflcko commented on pull request "ci: Skip Centos job on GHA runners":
(https://github.com/bitcoin/bitcoin/pull/33365#issuecomment-3324110754)
It seems a bit odd to have this task here, but not on forks. Somehow defeats the point of even supporting GHA runners, if they don't run the same matrix.

So I think I'd go with the clear-files action, albeit it is ugly to have to maintain it here.

Although, in the meantime, it seems to be passing: https://github.com/fanquake/bitcoin/actions/runs/17928384544/job/50981349378?

👋 fanquake's pull request is ready for review: "ci: run native_fuzz_with_msan"
(https://github.com/bitcoin/bitcoin/pull/33425)

💬 fanquake commented on pull request "ci: always use tag for LLVM checkout":
(https://github.com/bitcoin/bitcoin/pull/33364#issuecomment-3324130842)
This broke msan fuzz, but could be resolved in #33425.

💬 instagibbs commented on pull request "cluster mempool: control/optimize TxGraph memory usage":
(https://github.com/bitcoin/bitcoin/pull/33157#issuecomment-3324131315)
reACK 99e144778eeffc1074d5a1c193896bea335db8af

via `git range-diff master 66401148ddcba04c27751e39e53fc41e6deb5d14 99e144778eeffc1074d5a1c193896bea335db8af`

💬 fanquake commented on pull request "ci: remove Clang build from msan fuzz job":
(https://github.com/bitcoin/bitcoin/pull/33425#issuecomment-3324137303)
Dropped adding the job to the CI (just dropping the Clang build for now).