meson 0.52.1 breaking shared linking with D #6359
Labels
Comments
|
Apparently this only happens with ldc, gonna start bisecting... |
|
I think this is due to 49d20a5. |
|
So |
|
Actually nevermind, it's due to 0efab59 |
|
I was able to compile your project with #6359 applied, can you confirm? |
|
Ah sure, I'll give it a go, thanks. Also, can I maybe help with tests for the D lang? |
|
The project does compile (it also compiled before that change), the problem is the linkage. If I run a binary compiled with ldc2 with (or without this change) on meson >0.51.2 the binary blows up with: The shared library shouldn't be
Here's a verbose log of the compilation |
|
You're right, that still isn't fixed. |
dcbaker
added a commit
to dcbaker/meson
that referenced
this issue
Mar 10, 2020
Some compilers that act as linker drivers (dmd and ldc) need to split arguments that GCC combines with , (ie, -Wl,-foo,bar -> -L=-foo -L=bar). As such we need to detect that the previous argument contained -soname, and not wrap that in a --start-group/--end-group Fixes mesonbuild#6359 ../../../tmplckitmb1.o
|
Okay, I think I have a fix. I've updated the PR. it turns out some things changed and resulted in the soname argument not getting applied correctly. |
dcbaker
added a commit
to dcbaker/meson
that referenced
this issue
Mar 10, 2020
Some compilers that act as linker drivers (dmd and ldc) need to split arguments that GCC combines with , (ie, -Wl,-foo,bar -> -L=-foo -L=bar). As such we need to detect that the previous argument contained -soname, and not wrap that in a --start-group/--end-group This modifies the shared library test to demonstrate the problem, with a test case. Fixes mesonbuild#6359
|
alright great, I'll test in a bit. |
|
Seems like that fixed it, thank you! :) |
dcbaker
added a commit
to dcbaker/meson
that referenced
this issue
Mar 10, 2020
Some compilers that act as linker drivers (dmd and ldc) need to split arguments that GCC combines with , (ie, -Wl,-foo,bar -> -L=-foo -L=bar). As such we need to detect that the previous argument contained -soname, and not wrap that in a --start-group/--end-group This modifies the shared library test to demonstrate the problem, with a test case. Fixes mesonbuild#6359
Cogitri
added a commit
to Cogitri/corecollector
that referenced
this issue
Mar 11, 2020
…eson versions Meson >=0.52 and <0.54 contain a bug that causes bad linkage with ldc and dmd. See mesonbuild/meson#6359 for more info. fixes #24
dcbaker
added a commit
to dcbaker/meson
that referenced
this issue
Mar 11, 2020
Some compilers that act as linker drivers (dmd and ldc) need to split arguments that GCC combines with , (ie, -Wl,-foo,bar -> -L=-foo -L=bar). As such we need to detect that the previous argument contained -soname, and not wrap that in a --start-group/--end-group This modifies the shared library test to demonstrate the problem, with a test case. Fixes mesonbuild#6359
Cogitri
added a commit
to Cogitri/corecollector
that referenced
this issue
Mar 12, 2020
…ersions Meson >=0.52 and <0.54 contain a bug that causes bad linkage with ldc and dmd. See mesonbuild/meson#6359 for more info. fixes #24
Cogitri
added a commit
to Cogitri/corecollector
that referenced
this issue
Mar 12, 2020
…ersions Meson >=0.52 and <0.54 contain a bug that causes bad linkage with ldc and dmd. See mesonbuild/meson#6359 for more info. fixes #24
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
Since meson 0.52.1 building my project does succeed, but running the resulting binary doesn't work. See the following ldd output:
With meson 0.51.2 this works:
To Reproduce
Build https://github.com/Cogitri/corecollector
system parameters
Native build on Alpine Linux Edge with meson 0.52.1. Tested with latest scm (being 585ccfc) too.
The text was updated successfully, but these errors were encountered: