![]() ![]() Warning: megasync-4.6.6.0-1 is up to date - reinstalling :: megasync-4.6.6.0-1 is up to date - skipping build Submodule path 'src/MEGASync/mega': checked out '4b40fcbf58cd7b0ccb4a9604f08693fc5ed62bf2' Submodule 'src/MEGASync/mega' () registered for path 'src/MEGASync/mega'Ĭloning into '/home/avangel/.cache/paru/clone/megasync/src/MEGAsync/src/MEGASync/mega'. > Creating working copy of meganz-sdk git repo. > Creating working copy of MEGAsync git repo. => Validating source files with sha256sums. ![]() Resolving deltas: 100% (637/637), completed with 104 local objects. Remote: Total 778 (delta 637), reused 744 (delta 610), pack-reused 0 The mystical problem with cmst was: => Making package: megasync 4.6.6.0-1 (Mon 04:49:15 PM CEST) So I manually upgraded cmst (went fine), then reinstalled megasync just to make sure. When upgrading megasync (with paru), for some reason there was some problem with cmst. ![]() but at least that would also hold true of the statfs.f_type value they're being compared against, which makes it more consistent, and it resolves the narrowing error. what concerns me is that these 2 values would set the sign bit of a 32-bit int. If anyone sees a problem with this, please let me know. +#define SMB2_MAGIC_NUMBER (_fsword_t)0xfe534d42 +#define CIFS_MAGIC_NUMBER (_fsword_t)0xFF534D42 tracking back through the include files, statfs.f_type is a _fsword_t which is #def'd to different types depending on the platform (32-bit vs 64-bit). Please let me know if this does or does not pass the smell test.Įdit: i've changed the cast to _fsword_t so that it matches the type of statfs.f_type because the only way these two constants are used are in a switch statement where they're compared to a statfs.f_type variable. +#define SMB2_MAGIC_NUMBER (int)0xfe534d42 +#define CIFS_MAGIC_NUMBER (int)0xFF534D42 ![]() the patch, which i'll post, allows the build not to fail, but i'd feel more comfortable if someone more familiar with the code base could give it a thumbs-up or thumbs-down, or even code in a fix for 32-bit platforms.ĭiff -git a/src/posix/fs.cpp b/src/posix/fs.cpp the error is one of narrowing in posix/fs.cpp - a 32-bit hex value is being assigned to an int via a #define. i was able to create a patch to allow the build to complete, but due to the context of the "fix", i'm not 100% comfortable that there wouldn't be negative ripple effects. ( git-vfs, git-git, git-run-command-patch-git) (make)Īm getting an error on the i686 build. ( libeudev-git, libeudev, lib32-eudev, libudev-zero-git, systemd-libs-homed-containers, systemd-pr23511-libs, systemd-libs-selinux, lib32-systemd-git, systemd-libs-git, lib32-systemd, systemd-libs) ( libpdfium-nojs, pdfium-binaries, pdfium-binaries-v8) ( ffmpeg-nox-git, ffmpeg-svt-vp9-git, ffmpeg-intel-full-git, ffmpeg-yt-dlp-bin, ffmpeg-shinobi, ffmpeg-git, ffmpeg-ndi, ffmpeg-mmal, ffmpeg-amd-full-git, ffmpeg-amd-full, ffmpeg-libfdk_aac, ffmpeg-full-git, ffmpeg-nocuda, ffmpeg-obs, ffmpeg-cuda, ffmpeg-decklink, ffmpeg-full, ffmpeg-headless, ffmpeg-v4l2-request-git) ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |