ci: bump rustc to 1.60 for fuzz test#2109
Conversation
apoelstra
left a comment
There was a problem hiding this comment.
ACK 6aa1b297d0dd633935636e168ea5d645c4944e08
|
Huh. It looks like |
|
Yeah I was looking at it now Then we have also i will fix now |
|
Trying to reproduce it on my machine it works, looks like the CI gets the wrong cargo version Ah ofcs maybe the CI round Can you give another shot @apoelstra the ./contrib/test.sh is running on my system now so |
6aa1b29 to
cf313ae
Compare
Lol. Great catch! |
|
Ugh. Let's start trying random versions then. 1.64? 1.70? |
|
Mh this looks another kind of error, for example the fuzzer is a link error https://github.com/rust-bitcoin/rust-bitcoin/actions/runs/6436379883/job/17480345960?pr=2109 Then there is this, but this is not modified by me. So I will look more into this tomorrow |
cf313ae to
8b03531
Compare
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
Clippy warns: redundant guard As suggested, remove the redundant guard.
8b03531 to
05d3dc5
Compare
error[E0308]: mismatched types
--> bitcoin/src/psbt/raw.rs:87:24
|
87 | return Err(encode::Error::OversizedVectorAllocation {
| ________________________^
88 | | requested: key_byte_size as usize,
89 | | max: MAX_VEC_SIZE,
90 | | });
| |_____________^ expected enum `psbt::error::Error`, found enum `consensus::encode::Error`
|
help: try wrapping the expression in `psbt::error::Error::ConsensusEncoding`
|
87 ~ return Err(psbt::error::Error::ConsensusEncoding(encode::Error::OversizedVectorAllocation {
88 | requested: key_byte_size as usize,
89 | max: MAX_VEC_SIZE,
90 ~ }));
|
----
Compiling bitcoin v0.30.0 (/home/vincent/github/work/rust-btc/rust-bitcoin/bitcoin)
Checking bitcoin-fuzz v0.0.1 (/home/vincent/github/work/rust-btc/rust-bitcoin/fuzz)
error: redundant clone
--> bitcoin/examples/taproot-psbt.rs:453:77
|
453 | witness_utxo: { Some(TxOut { value, script_pubkey: script_pubkey.clone() }) },
| ^^^^^^^^ help: remove this
|
= note: `-D clippy::redundant-clone` implied by `-D warnings`
note: this value is dropped without further use
--> bitcoin/examples/taproot-psbt.rs:453:64
|
453 | witness_utxo: { Some(TxOut { value, script_pubkey: script_pubkey.clone() }) },
| ^^^^^^^^^^^^^
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#redundant_clone
error: could not compile `bitcoin` due to previous error
warning: build failed, waiting for other jobs to finish...
error: redundant clone
--> bitcoin/src/psbt/mod.rs:1095:13
|
1095 | .clone()
| ^^^^^^^^ help: remove this
|
= note: `-D clippy::redundant-clone` implied by `-D warnings`
note: this value is dropped without further use
--> bitcoin/src/psbt/mod.rs:1094:17
|
1094 | assert!(psbt
| _________________^
1095 | | .clone()
| |____________^
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#redundant_clone
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
|
OK, I see the light with the CI :) see my self-pr for the result vincenzopalazzo#2 I had also some more clippy fixes in 98513ef Now with the new rust version, there is a fuzz test that it is failing, but I have no experience with the fuzzing test, so there is someone who can give me some tips? The failure is here https://github.com/vincenzopalazzo/rust-bitcoin/actions/runs/6440404417/job/17489177949?pr=2 Then, I had to fix the version of the byteorder to 1.4.3 due to the recent update https://crates.io/crates/byteorder/versions |
|
Thanks @vincenzopalazzo! See #2111 -- the fuzzing issue was fixed in rust-bech32 but hasn't been released and propagated to rust-bitcoin yet. |
apoelstra
left a comment
There was a problem hiding this comment.
ACK cf313ae6a1ab08555147a0f5a04faf814145d7c8
|
Nice @apoelstra somedays I will get someone from rust-bitcoin in as room to teach me how do fuzz testing haha |
|
Hah. And thanks a ton for keeping on this! It looks like CI is passing so far. I will merge this once we get all the green checkmarks in. |
|
Ok, this looks like a pass to me, other than some things that appear to be Github bugs (complete jobs still marked as "in progress") and the deserialize_address thing which we can address sepaarately. |
apoelstra
left a comment
There was a problem hiding this comment.
ACK cf313ae6a1ab08555147a0f5a04faf814145d7c8
3e6021b ci: fuzz test YAML toolchain future-proof (Einherjar) Pull request description: To avoid YAML automatic parsing of inputs as floats, e.g. [`1.60` being parsed as `1.6`](#2109 (comment)), it is best to future-proof the `toolchain` input in CI fuzz testing as a string. ACKs for top commit: vincenzopalazzo: ACK 3e6021b apoelstra: ACK 3e6021b tcharding: ACK 3e6021b Tree-SHA512: 838afb401fcfbb00717d6d3efff013cda08dabe4962ee0e914c3cf438a5cc8e83fa336dbd920b299fe95f935c0c1d76cf15d3f6fc6f763ea15a1746e7254d94f
Ci looks like broken, so this should fix
it