💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#issuecomment-2652428792)
Windows code signature:
[signature-win.tar.gz](https://github.com/user-attachments/files/18760608/signature-win.tar.gz)

It looks like I have a mismatch.

```
$ find guix-build-46e44a35b858/output/ -type f -print0 | env LC_ALL=C sort -z | xargs -r0 sha256sum
c92e7d0b4d0d1f929bfe8d61e15a953738725d530fc64eb936634571c17936b7 guix-build-46e44a35b858/output/aarch64-linux-gnu/SHA256SUMS.part
ad3992a66f9de8039cabbc9f222f7369f8002fc5350a01b03a3194d574100770 guix-build-46e44a35b858/output/aarch
...

📝 luke-jr opened a pull request: "Add -pruneduringinit option to temporarily use another prune target during IBD"
(https://github.com/bitcoin/bitcoin/pull/31845)
Low prune sizes can force flushing long before dbcache is full, making a significant impact on real-world IBD time.

This new option allows a higher (or lower) prune size during IBD to mitigate this. When assumeutxo is active, the entire additional budget is allocated to the background chainstate.

💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#discussion_r1951862122)
Done

💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#discussion_r1951862158)
Done

💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#issuecomment-2652459686)
I believe I identified the non-determinism, pushed a fix for that.

💬 andrewtoth commented on pull request "Add -pruneduringinit option to temporarily use another prune target during IBD":
(https://github.com/bitcoin/bitcoin/pull/31845#issuecomment-2652484368)
After #28280 how significant is the impact of this?

💬 luke-jr commented on pull request "Add -pruneduringinit option to temporarily use another prune target during IBD":
(https://github.com/bitcoin/bitcoin/pull/31845#issuecomment-2652488514)
>After #28280 and #30039 how significant is the impact of this?

I don't know, but logically I would expect this to still improve things significantly.

💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#issuecomment-2652507129)
Windows code signature for 096525e92cc2f5a4318bae13cedd2cf36b928d5f:
[signature-win.tar.gz](https://github.com/user-attachments/files/18761067/signature-win.tar.gz)

# Non-Codesigned guix hashes
```
$ find guix-build-096525e92cc2/output -type f -print0 | env LC_ALL=C sort -z | xargs -r0 sha256sum
bff1786e0fe30ac2c70eb8abd39af1764eff33aef98c92cccd74b70db0cf63d8 guix-build-096525e92cc2/output/aarch64-linux-gnu/SHA256SUMS.part
d7308491e32e076f40e58aaa9092ceae8a16b39e66937af72bdc8b879164f30
...

💬 pinheadmz commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#issuecomment-2652521109)
New sigs for 096525e92cc2f5a4318bae13cedd2cf36b928d5f:

[signature-osx-arm64.tar.gz](https://github.com/user-attachments/files/18761206/signature-osx-arm64.tar.gz)

[signature-osx-x86_64.tar.gz](https://github.com/user-attachments/files/18761208/signature-osx-x86_64.tar.gz)

# All osx

```

111016205f0a2ac732feb934acb3e8a36d5251f119d8fa9215790310ba46c31d bitcoin-096525e92cc2-arm64-apple-darwin.tar.gz
a1228ddcb27fc5802ff5cab8f99026286070b4938aba08982f8e49d8e01b49cf bitcoin-096525e92
...

💬 achow101 commented on pull request "guix: Notarize MacOS app bundle and codesign all MacOS and Windows binaries":
(https://github.com/bitcoin/bitcoin/pull/31407#issuecomment-2652543539)
Looks like everything matches:

```
$ find guix-build-096525e92cc2/output -wholename "*codesigned*" -type f -print0 | env LC_ALL=C sort -z | xargs -r0 sha256sum
b329c253e4c791d5622adc16ed0c06759ce29309e7d72221c76de4becb58b409 guix-build-096525e92cc2/output/arm64-apple-darwin-codesigned/SHA256SUMS.part
111016205f0a2ac732feb934acb3e8a36d5251f119d8fa9215790310ba46c31d guix-build-096525e92cc2/output/arm64-apple-darwin-codesigned/bitcoin-096525e92cc2-arm64-apple-darwin.tar.gz
a1228ddcb27fc580
...

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951957283)
added

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951957432)
fixed

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951957531)
Moved into `NeedsTrim`

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951962043)
not entirely sure... could be cleaned up

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951964695)
made consistent + made both logs %u

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951964799)
fixed

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951965002)
(thanks, I took your test with just a few tweaks)

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951966069)
Added a couple pushes ago, but gone again. After a bit of offline discussion with @mzumsande and @sipa, it seemed better to just synchronously remove wtxids from worksets when they are removed as announcements. This means that the work set is always a subset of announcement set (added this to `SanityCheck`). Also, potentially failing to add things to work set seemed to make this less useful.

💬 glozow commented on pull request "p2p: improve TxOrphanage denial of service bounds and increase -maxorphantxs":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r1951966734)
Yes, they're both meant to be ratios :+1:

💬 davidgumberg commented on pull request "Benchmark Chainstate::ConnectBlock duration":
(https://github.com/bitcoin/bitcoin/pull/31689#discussion_r1952062784)
not blocking, just a question: here and below, why `reserve()` for vectors that are outside of the benchmark's path?