2016-10-19 19:37:54 +03:00
branches :
- master
- auto
- try
2016-12-07 00:29:39 +03:00
required : true
2017-07-18 12:59:04 +03:00
context : f26-primary
2016-12-07 00:29:39 +03:00
2016-10-19 19:37:54 +03:00
container :
2017-07-18 12:59:04 +03:00
image : registry.fedoraproject.org/fedora:26
2016-12-08 05:26:43 +03:00
packages :
2017-03-30 23:47:57 +03:00
- git
2016-12-08 05:26:43 +03:00
2016-11-09 05:37:43 +03:00
env :
2017-06-21 00:19:01 +03:00
# Enable all the sanitizers for this primary build.
# We only use -Werror=maybe-uninitialized here with a "fixed" toolchain
2017-06-28 18:59:36 +03:00
CFLAGS : '-fsanitize=undefined -fsanitize-undefined-trap-on-error -fsanitize=address -O2 -Wp,-D_FORTIFY_SOURCE=2'
2017-06-27 01:12:33 +03:00
# Only for CI with a known g-ir-scanner
GI_SCANNERFLAGS : '--warn-error'
2016-12-08 05:26:43 +03:00
ASAN_OPTIONS : 'detect_leaks=0' # Right now we're not fully clean, but this gets us use-after-free etc
# TODO when we're doing leak checks: G_SLICE: "always-malloc"
2016-11-09 05:37:43 +03:00
2016-10-19 19:37:54 +03:00
tests :
2017-06-08 20:47:20 +03:00
- ci/ci-commitmessage-submodules.sh
- ci/build-check.sh
2017-06-19 18:25:25 +03:00
- ci/ci-release-build.sh
2016-10-19 19:37:54 +03:00
timeout : 30m
artifacts :
2017-06-08 20:47:20 +03:00
- test-suite.log
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
---
2017-06-27 23:39:27 +03:00
context : c7-primary
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
inherit : true
2017-06-09 17:57:40 +03:00
required : true
2017-06-27 23:39:27 +03:00
packages :
host :
distro : centos/7/atomic
2017-06-09 17:57:40 +03:00
env :
CFLAGS : ''
tests :
2017-06-27 23:39:27 +03:00
- docker run --privileged -v $PWD:$PWD --workdir $PWD
registry.centos.org/centos/centos:7 sh -c
'yum install -y git && ci/build-check.sh'
2017-06-09 17:57:40 +03:00
---
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
2017-07-18 12:59:04 +03:00
context : f26-rust
2017-06-09 17:57:40 +03:00
inherit : true
container :
2017-07-18 12:59:04 +03:00
image : registry.fedoraproject.org/fedora:26
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
packages :
- cargo
env :
2017-06-08 20:47:20 +03:00
CONFIGOPTS : '--enable-rust'
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
tests :
2017-06-08 20:47:20 +03:00
- ci/build.sh
oxidation: Add implementation of bupsplit in Rust
This is an initial drop of "oxidation", or adding implementation
of components in Rust. The bupsplit code is a good target - no
dependencies, just computation.
Translation into Rust had a few twists -
- The C code relies a lot on overflowing unsigned ints, and
also on the C promotion rules for e.g. `uint8_t -> int32_t`
- There were some odd loops that I introduced bugs in while
translating...in particular, the function always returns `len`,
but I mistakenly translated to `len+1`, resulting in an OOB
read on the C side, which was hard to debug.
On the plus side, an off-by-one array indexing in the Rust code paniced nicely.
In practice, we'll need a lot more build infrastructure to make this work, such
as using `cargo vendor` when producing build artifacts for example. Also, Cargo
is yet another thing we need to cache.
Where do we go with this? Well, I think we should merge this, it's not a lot of
code. We can just have it be an alternative CI target. Should we do a lot more
right now? Probably not immediately, but I find the medium/long term prospects
pretty exciting!
Closes: #656
Approved by: jlebon
2017-01-25 05:43:53 +03:00
- make check TESTS=tests/test-rollsum
2017-05-19 17:39:48 +03:00
---
inherit : true
2017-07-18 12:59:04 +03:00
context : f26-experimental-api
2017-05-19 17:39:48 +03:00
env :
2017-06-08 20:47:20 +03:00
CONFIGOPTS : '--enable-experimental-api'
2017-05-19 17:39:48 +03:00
tests :
2017-06-08 20:47:20 +03:00
- ci/build-check.sh
2017-05-19 17:39:48 +03:00
2016-12-08 05:02:30 +03:00
---
inherit : true
required : true
2017-07-18 12:59:04 +03:00
context : f26-curl-openssl
2016-12-08 05:02:30 +03:00
packages :
- pkgconfig(libcurl)
2017-03-13 21:41:14 +03:00
- pkgconfig(openssl)
2016-12-08 05:02:30 +03:00
2017-06-08 20:47:20 +03:00
env :
CONFIGOPTS : "--with-curl --with-openssl"
2016-12-08 05:02:30 +03:00
tests :
2017-06-08 20:47:20 +03:00
- ci/build-check.sh
2017-04-24 21:40:06 +03:00
---
inherit : false
branches :
- master
- auto
- try
2017-07-18 12:59:04 +03:00
context : f26ah-insttest
2017-04-24 21:40:06 +03:00
required : false
cluster :
hosts :
- name : vmcheck
2017-07-18 12:59:04 +03:00
distro : fedora/26/atomic
2017-04-24 21:40:06 +03:00
container :
2017-07-18 12:59:04 +03:00
image : registry.fedoraproject.org/fedora:26
2017-04-24 21:40:06 +03:00
# Copy the build from the container to the host; ideally down the line
# this is installing an RPM via https://github.com/jlebon/redhat-ci/issues/10
tests :
2017-06-08 20:47:20 +03:00
- ci/build.sh
2017-04-24 21:40:06 +03:00
- make install DESTDIR=$(pwd)/insttree
2017-07-18 12:59:04 +03:00
- yum -y install rsync
2017-04-24 21:40:06 +03:00
- rsync -rl -e 'ssh -o User=root' . vmcheck:ostree/
2017-05-08 23:44:42 +03:00
- ssh root@vmcheck './ostree/tests/installed/fah-prep.sh && ostree admin unlock && rsync -rlv ./ostree/insttree/usr/ /usr/ && ./ostree/tests/installed/run.sh'
2017-04-24 21:40:06 +03:00
2017-04-28 16:32:35 +03:00
---
inherit : false
branches :
- master
- auto
- try
2017-07-18 12:59:04 +03:00
context : f26-flatpak
2017-04-28 16:32:35 +03:00
required : false
# This test case wants an "unprivileged container with bubblewrap",
# which we don't have right now; so just provision a VM and do a
# docker --privileged run.
host :
2017-07-18 12:59:04 +03:00
distro : fedora/26/atomic
2017-05-02 01:02:27 +03:00
specs :
ram : 4096 # build-bundle is a static delta, which needs RAM right now
2017-04-28 16:32:35 +03:00
tests :
2017-07-18 12:59:04 +03:00
- docker run --rm --privileged -v $(pwd):/srv/code registry.fedoraproject.org/fedora:26 /bin/sh -c "cd /srv/code && ./ci/flatpak.sh"
2017-04-28 16:32:35 +03:00
artifacts :
- test-suite.log
