mirrors/arti.git/crates/tor-proto/src, branch arti-v1.3.0 mirror of https://gitlab.torproject.org/tpo/core/arti http://git.dilluti0n.com/mirrors/arti.git/atom?h=arti-v1.3.0 2024-10-22T13:06:32Z Merge branch 'mq-fix' into 'main' 2024-10-22T13:06:32Z Ian Jackson iwj@torproject.org 2024-10-22T13:06:32Z urn:sha1:2ceea6972e29fdf30187bb41207c0a3d77b0131d memquota: Fix account lifetime bugs, and arrange to test mq in shadow See merge request tpo/core/arti!2560 memquota: Use _ rather than allow(dead_code) (fmt) 2024-10-22T12:19:22Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-22T12:19:15Z urn:sha1:905dad6dd31a4927c8d35b081105d69e1063cf77 memquota: Use _ rather than allow(dead_code) 2024-10-22T12:19:22Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-22T12:16:26Z urn:sha1:eed28245887b87a7bf1e884b568f8062148cdd76 Promote the associated comments. As suggested here: https://gitlab.torproject.org/tpo/core/arti/-/merge_requests/2560#note_3097188 Merge branch 'test_create_hop_count_reinstate' into 'main' 2024-10-21T16:21:27Z gabi-250 gabi@torproject.org 2024-10-21T16:21:27Z urn:sha1:6b737142ccd0f3c94e1ca22ac17f9c3107e9c47d tor-proto: Reinstate circuit hop check in test_create() See merge request tpo/core/arti!2546 memquota: fix data stream account lifetime 2024-10-21T10:11:20Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-16T15:57:44Z urn:sha1:6092b87c6a2351dfcf90e7f81505b06820e95eaf The DataStream is sometimes disassembled, eg by split. When that happens, the StreamAccount would be dropped - and that was the only strong reference. Put a StreamAccount in each of the pieces, instead of just in the combined DataStream struct. memquota: Fix resolve stream account lifetime 2024-10-21T10:11:20Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-16T15:25:28Z urn:sha1:a67433aa9a0a079d1e070d40000fdb74fb6bbc06 We need the mq account for the stream not to collapse. The ResolveStream object needs to contain a strong reference to it. Have begin_stream_impl return the StreamAccount, rather than taking it as a parameter. That makes this bug a little more obvious. It also centralises the StreamAccount creation. memquota: Fix circuit account lifetime (fmt) 2024-10-21T10:11:20Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-21T10:09:34Z urn:sha1:c017ccaf28fe1089ddd313fe0b874e6bc71546cd memquota: Fix circuit account lifetime (clippy churn) 2024-10-21T10:11:20Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-16T17:01:36Z urn:sha1:0be7b09361f1820179f9ec1f6d995c773b726433 Now that it doesn't call CircuitAccount::new() it has no error paths, and clippy demands we remove the Result, so it must once again become infallible. memquota: Fix circuit account lifetime 2024-10-21T10:11:20Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-16T15:08:49Z urn:sha1:09b80b32e316634324b0f99258df66773642da88 We foolishly made *two* CircuitAccounts, one of which gets immediately dropped. But we need to hold onto the account somewhere, because an mq_queue doesn't keep the account alive. Otherwise everything breaks when mq tracking is enabled. Disable a lot of dead code warnings (fmt) 2024-10-17T10:25:33Z Ian Jackson ijackson@chiark.greenend.org.uk 2024-10-17T10:25:20Z urn:sha1:731adee5ad984dd1226eb917b20bb1380065aa09