Skip to content

Conversation

@lenary
Copy link
Contributor

@lenary lenary commented Jun 25, 2020

I don't quite know what's going on here, meson should raise an error with a good explanation if it tries to set the gnu linker to anything that isn't the exact string bfd, gold or lld and yet it just… doesn't? Instead you get "unable to find dynamic linker" with no useful error message, and a useless log.

Anyway, turns out I got #26 wrong. Selecting a linker is required for clang, but not for the gnu toolchain (because the gnu toolchain is built as one piece, not as a cross-compiler). So this change removes these lines in the gcc meson configuration.

I tested this by trying the HEAD build from 15 minutes ago, and then discovering that meson was totally hosed. Applying this fix sorts it out, once I debugged how meson was trying to find the linker version.

Signed-off-by: Sam Elliott <selliott@lowrisc.org>
@lenary lenary requested a review from luismarques June 25, 2020 23:41
@lenary
Copy link
Contributor Author

lenary commented Jun 26, 2020

Performed both a clang build and a gcc build using the resulting combined toolchain, both succeeded on morty, which was the reason for doing these changes in #26. I'm going to merge now so we can tag a new release with this fix.

@lenary lenary merged commit 2dc0b1f into lowRISC:master Jun 26, 2020
@luismarques
Copy link
Collaborator

I guess this is good for merging? If it works that way...

@luismarques
Copy link
Collaborator

Oh well, too late by 2 minutes :-)

@lenary
Copy link
Contributor Author

lenary commented Jun 26, 2020

Sorry :(

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants