Summary: | www-client/firefox: specify cbindgen header path | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | David Michael <fedora.dm0> |
Component: | Current packages | Assignee: | Mozilla Gentoo Team <mozilla> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | herrtimson, jstein |
Priority: | Normal | Keywords: | PATCH |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | screenshot from runtime test |
Description
David Michael
2020-12-22 21:09:12 UTC
And just so it's noted, the --target=${CHOST} in the diff does not seem to be redundant. When it's omitted, the search path does not include the GCC header directories, so it is missing standard C++ stuff. 5:59.36 #include <...> search starts here: 5:59.36 /usr/armv7a-gentoo-linux-gnueabihf/usr/include/nspr 5:59.36 /usr/armv7a-gentoo-linux-gnueabihf/usr/include/nss 5:59.36 /var/tmp/portage/www-client/firefox-84.0.1/work/firefox_build/dist/include/nss 5:59.36 /usr/armv7a-gentoo-linux-gnueabihf/usr/include/pixman-1 5:59.36 /var/tmp/portage/www-client/firefox-84.0.1/work/firefox_build/dist/include/cairo 5:59.36 /usr/lib/llvm/11/bin/../../../../lib/clang/11.0.0/include 5:59.36 /usr/armv7a-gentoo-linux-gnueabihf/usr/include 5:59.36 End of search list. 5:59.37 /var/tmp/portage/www-client/firefox-84.0.1/work/firefox_build/dist/include/nss/mozpkix/pkixtypes.h:28:10: fatal error: 'memory' file not found 5:59.37 /var/tmp/portage/www-client/firefox-84.0.1/work/firefox_build/dist/include/nss/mozpkix/pkixtypes.h:28:10: fatal error: 'memory' file not found, err: true 5:59.37 thread 'main' panicked at 'unable to generate bindings: ()', /var/tmp/portage/www-client/firefox-84.0.1/work/firefox-84.0.1/third_party/rust/neqo-crypto/build.rs:281:39 I've hit this ever since I first tried to cross compile firefox from amd64 to aarch64, and even though I invested half an evening into debugging this more than once I was unable to come up with a solution for it. The fix sounds very promissing, I'm going to test it. Ping. Test finished? Created attachment 685536 [details]
screenshot from runtime test
I did have some success with the ebuild patch, look at configure options to see
If you're asking because you want to add the patch to the tree, how about we do that for firefox-86.0 ebuild? This can give a bit more time to runtime test this with armv7a.
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=72ff7e264ae4b3f4a5eb5cc4d7c2b825c04c651d commit 72ff7e264ae4b3f4a5eb5cc4d7c2b825c04c651d Author: Thomas Deutschmann <whissi@gentoo.org> AuthorDate: 2021-02-05 19:04:18 +0000 Commit: Thomas Deutschmann <whissi@gentoo.org> CommitDate: 2021-02-05 20:02:12 +0000 www-client/firefox: bump to v85.0.1 Bug: https://bugs.gentoo.org/768825 Closes: https://bugs.gentoo.org/761277 Package-Manager: Portage-3.0.14, Repoman-3.0.2 Signed-off-by: Thomas Deutschmann <whissi@gentoo.org> www-client/firefox/Manifest | 97 +++ www-client/firefox/firefox-85.0.1.ebuild | 1137 ++++++++++++++++++++++++++++++ 2 files changed, 1234 insertions(+) |