[Pkg-rust-maintainers] rustc_1.14.0+dfsg1-1_amd64.changes ACCEPTED into unstable, unstable

Ximin Luo infinity0 at debian.org
Sat Dec 24 22:00:00 UTC 2016

Merry christmas everyone! Here is your present of 7 failed bootstrapping attempts :D


mips, mipsel, mips64el

e.g. configure: error: unknown CPU type: mips64

e.g. https://buildd.debian.org/status/fetch.php?pkg=rustc&arch=mips64el&ver=1.14.0%2Bdfsg1-1&stamp=1482615255

Probably easily fixable, but I'll have to try this in 2-3 weeks due to the holidays. Also these machines are sloooow which makes this a pain. If anyone has fast mips machines, please give us access.


Illegal instruction


rust upstream messed up disabling neon for v1.13 (what we're bootstrapping from), see https://github.com/rust-lang/rust/issues/36913

and you can check this yourself by extracting our orig-dl tarball and running:

rustc-1.13.0-armv7-unknown-linux-gnueabihf/rustc/lib$ objdump -d libstd-*.so | grep '<.*lang_start.*>:' -A30
0006cfec <_ZN3std2rt10lang_start17hcb77440d8735d978E>:
   6d030:	f2c00050 	vmov.i32	q8, #0	; 0x00000000
   6d03c:	f4410acd 	vst1.64	{d16-d17}, [r1]!
   6d040:	f4410acf 	vst1.64	{d16-d17}, [r1]

as the problematic instructions that cuviper mentioned in the above ticket.

ppc64el, s390x

Segmentation fault


No idea about this, but you people with these machines can now run "apt-get source rustc" to try to debug this for yourself. Also DDs with access to plummer.debian.org can try it (it's a very fast machine with 16x 3GHz cores).


Not yet built, but it's not an "official" Debian arch so is less important. Probably will segfault like the two above.


GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE

More information about the Pkg-rust-maintainers mailing list