Uh oh!
There was an error while loading. Please reload this page.
Add rustc SHA to released DWARF debuginfo - #53829
Conversation
rust-highfive
commented
Aug 30, 2018
(rust_highfive has picked a reviewer for you, use r? to override) |
alexcrichton
commented
Aug 30, 2018
cc @luser, this will likely affect Gecko if we decide to merge! cc @yurydelendik, I think this should do what you were requesting on IRC @bors: try I'd like to poke around these artifacts before landing to double check it's all as expected. |
bors
commented
Aug 30, 2018
⌛ Trying commit 8bdc6f79ac792303051bb23a053f7c45f4616267 with merge 5eaeeb3b3e3e2540410740229dcfc8b883091b10... |
luser
commented
Aug 30, 2018
Oh, excellent! We currently just fetch the sha from |
luser
commented
Aug 30, 2018
One other thing to consider here would be trying to apply this to vendored C sources that you compile as well. |
bors
commented
Aug 30, 2018
☀️ Test successful - status-travis |
8bdc6f7 to
4b790adComparealexcrichton
commented
Aug 30, 2018
Sure enough C code was still mentioning @bors: try |
bors
commented
Aug 30, 2018
⌛ Trying commit 4b790ad1f8e78f61e2847058071b42a264e712f3 with merge 37655c8000de5017c6d28454888e55af3d6e5040... |
4b790ad to
cc4c606Comparealexcrichton
commented
Aug 30, 2018
@bors: try |
bors
commented
Aug 30, 2018
Add rustc SHA to released DWARF debuginfo This commit updates the debuginfo that is encoded in all of our released artifacts by default. Currently it has paths like `/checkout/src/...` but these are a little inconsistent and have changed over time. This commit instead attempts to actually define the file paths in our debuginfo to be consistent between releases. All debuginfo paths are now intended to be `/rustc/$sha` where `$sha` is the git sha of the released compiler. Sub-paths are all paths into the git repo at that `$sha`.
rust-highfive
commented
Aug 30, 2018
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
bors
commented
Aug 31, 2018
💔 Test failed - status-travis |
rust-highfive
commented
Aug 31, 2018
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
bors
commented
Aug 31, 2018
☀️ Test successful - status-travis |
cc4c606 to
3286a69Comparealexcrichton
commented
Aug 31, 2018
Alright should be good to go! |
There was a problem hiding this comment.
Why only Clang? Gcc should accept the same flag.
There was a problem hiding this comment.
Oh sure! I was unaware of that
(eventually we'll need to figure out how to pass this to clang-cl as well)
There was a problem hiding this comment.
clang-cl seems to accept it if you stick -Xclang before it, from cursory testing:
c:\build>c:\Users\ted\.mozbuild\clang\bin\clang-cl.exe -Fohello.obj -c c:\build\hello.c -Ic:\build -Z7 -Xclang -fdebug-prefix-map=c:\build=\src
$ stringsext hello.obj | rg src
\src\src\hello.c
michaelwoerister
commented
Aug 31, 2018
Looks great, thanks Alex! @bors r+ |
bors
commented
Aug 31, 2018
📌 Commit 3286a6957193e34c1f20a2d0f0c15bcdcbab9462 has been approved by |
bors
commented
Sep 12, 2018
⌛ Testing commit 5595aeb with merge 0ca767d460157af164fd4a7adb1cc7f61b8dfce4... |
bors
commented
Sep 12, 2018
💔 Test failed - status-travis |
rust-highfive
commented
Sep 12, 2018
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
alexcrichton
commented
Sep 12, 2018
@bors: retry sigh |
bors
commented
Sep 12, 2018
⌛ Testing commit 5595aeb with merge e12dcff5e7e8f632039fe9b88e2c5cc7b91f171f... |
bors
commented
Sep 13, 2018
💔 Test failed - status-travis |
rust-highfive
commented
Sep 13, 2018
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
alexcrichton
commented
Sep 13, 2018
@bors: retry |
mati865
commented
Sep 13, 2018
3rd failure because of 50-minute timeout. Travis broke? |
kennytm
commented
Sep 13, 2018
@mati865 Yes, and Travis is aware of the bug already. |
bors
commented
Sep 13, 2018
⌛ Testing commit 5595aeb with merge 48e6312571a6498927ddf63f91e3649e05e36ab9... |
bors
commented
Sep 13, 2018
💔 Test failed - status-travis |
rust-highfive
commented
Sep 13, 2018
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
kennytm
commented
Sep 13, 2018
@bors retry treeclosed=1 50 minute timeout |
| // except according to those terms. | ||
| // ignore-musl | ||
| // ignore-x86 |
There was a problem hiding this comment.
I don't see this discussed explicitly. What is causing the problem these ignores are, well, ignoring? Was an issue ever opened to track it.
This showed up again in #58140 (comment), and while I'm fine with copying the two directives from here, this seems like a bug that should be tracked, if not investigated right away.
There was a problem hiding this comment.
The failure happened here and I wasn't really interested in debugging this, but I agree I should have opened up an issue to track it.
This commit updates the debuginfo that is encoded in all of our released
artifacts by default. Currently it has paths like
/checkout/src/...but theseare a little inconsistent and have changed over time. This commit instead
attempts to actually define the file paths in our debuginfo to be consistent
between releases.
All debuginfo paths are now intended to be
/rustc/$shawhere$shais the gitsha of the released compiler. Sub-paths are all paths into the git repo at that
$sha.