It's broken on arm again. This is a known issue and can be fixed by importing appropriate patched from other ports (which I had previously written to fix the same issue). Will take care of it soonish. ../../../../../3rdparty/lzma/C/7zCrc.c:158:1: warning: unknown architecture 'armv8-a+crc' in the 'target' attribute string; 'target' attribute ignored [-Wignored-attributes] ATTRIB_CRC ^ ../../../../../3rdparty/lzma/C/7zCrc.c:89:58: note: expanded from macro 'ATTRIB_CRC' #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc"))) ^ ../../../../../3rdparty/lzma/C/7zCrc.c:160:1: warning: unknown architecture 'armv8-a+crc' in the 'target' attribute string; 'target' attribute ignored [-Wignored-attributes] ATTRIB_CRC ^ ../../../../../3rdparty/lzma/C/7zCrc.c:89:58: note: expanded from macro 'ATTRIB_CRC' #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc"))) ^ ../../../../../3rdparty/lzma/C/7zCrc.c:190:1: warning: unknown architecture 'armv8-a+crc' in the 'target' attribute string; 'target' attribute ignored [-Wignored-attributes] ATTRIB_CRC ^ ../../../../../3rdparty/lzma/C/7zCrc.c:89:58: note: expanded from macro 'ATTRIB_CRC' #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc"))) ^ ../../../../../3rdparty/lzma/C/7zCrc.c:192:1: warning: unknown architecture 'armv8-a+crc' in the 'target' attribute string; 'target' attribute ignored [-Wignored-attributes] ATTRIB_CRC ^ ../../../../../3rdparty/lzma/C/7zCrc.c:89:58: note: expanded from macro 'ATTRIB_CRC' #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc"))) ^ fatal error: error in backend: Cannot select: intrinsic %llvm.arm.crc32b
Hey Robert, Thanks for opening report. The emulators/qmc2 build is also failing with the same fatel error: https://pkg-status.freebsd.org/ampere1/data/124arm64-quarterly/adc92393850d/logs/qmc2-0.243_1.log This stuff is quite beyond me, thanks for looking into it.
Got patches for both, stolen from archivers/7-zip. Will commit with my next batch. BTW, why didn't you unbundle lua? There's WITH_SYSTEM_LIB_LUA or something like that to make that work.
(In reply to Robert Clausecker from comment #2) > Got patches for both, stolen from archivers/7-zip. Will commit with my next batch. What a relief, thank you so much. > BTW, why didn't you unbundle lua? There's WITH_SYSTEM_LIB_LUA or something like that to make that work. MAME now requires Lua-5.4 built with C++ and C++ unwinding support; I had a quick look, Arch User Repository has a package for Lua-5.4 with C++ unwind support, not on my TODO at the moment, tho, it will probably need to be sorted eventually. From MAME-0.253 https://www.mamedev.org/releases/whatsnew_0253.txt: -Various updates, mostly around Lua: [Vas Crabb] * Compile Lua as C++. When Lua is compiled as C, it uses setjmp/longjmp for error handling, resulting in failure to unwind intermediate stack frames. Trying to ensure no objects with non-trivial destructors are in scope when raising a Lua error is error-prone. In particular, converting an exception to a Lua error becomes convoluted, and raising a Lua error from a constructor is effectively impossible. * Updated sol2 to version 3.3.0 - this adds support for Lua 5.4 and fixes a number of issues, including not correctly handling errors when Lua is built as C++.
A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=cb40b0a43949a621f407fa3b65dff7349b024ba2 commit cb40b0a43949a621f407fa3b65dff7349b024ba2 Author: Robert Clausecker <fuz@FreeBSD.org> AuthorDate: 2023-06-25 16:58:55 +0000 Commit: Robert Clausecker <fuz@FreeBSD.org> CommitDate: 2023-06-26 23:30:28 +0000 emulators/mame: fix build on armv7 and arm64 Port now bundles LZMA with known build issues. Import patch set from archivers/7-zip to fix the build on armv7 and arm64. PR: 271987 Approved by: portmgr (build fix blanket) Obtained from: archivers/7-zip emulators/mame/Makefile | 3 ++ ...atch-3rdparty_lzma_CPP_7zip_7zip__gcc.mak (new) | 27 ++++++++++++++ ...patch-3rdparty_lzma_CPP_7zip_var__gcc.mak (new) | 12 +++++++ ...-3rdparty_lzma_CPP_Windows_SystemInfo.cpp (new) | 27 ++++++++++++++ .../mame/files/patch-3rdparty_lzma_C_7zCrc.c (new) | 11 ++++++ .../mame/files/patch-3rdparty_lzma_C_Aes.c (new) | 11 ++++++ .../files/patch-3rdparty_lzma_C_AesOpt.c (new) | 38 ++++++++++++++++++++ .../files/patch-3rdparty_lzma_C_CpuArch.c (new) | 42 ++++++++++++++++++++++ .../files/patch-3rdparty_lzma_C_Sha256Opt.c (new) | 11 ++++++ 9 files changed, 182 insertions(+)
A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=1ebb93fd8095f4cc77bd8f0681580e9ef747b965 commit 1ebb93fd8095f4cc77bd8f0681580e9ef747b965 Author: Robert Clausecker <fuz@FreeBSD.org> AuthorDate: 2023-06-25 16:55:58 +0000 Commit: Robert Clausecker <fuz@FreeBSD.org> CommitDate: 2023-06-26 23:30:28 +0000 emulators/qmc2: fix build on armv7 and arm64 Port now bundles LZMA with known build issues. Import patch set from archivers/7-zip to fix the build on armv7 and arm64. PR: 271987 Approved by: portmgr (build fix blanket) Obtained from: archivers/7-zip emulators/qmc2/Makefile | 3 +- emulators/qmc2/files/patch-src_lzma_7zCrc.c (new) | 11 ++++++ emulators/qmc2/files/patch-src_lzma_Aes.c (new) | 11 ++++++ emulators/qmc2/files/patch-src_lzma_AesOpt.c (new) | 38 ++++++++++++++++++++ .../qmc2/files/patch-src_lzma_CpuArch.c (new) | 42 ++++++++++++++++++++++ .../qmc2/files/patch-src_lzma_Sha256Opt.c (new) | 11 ++++++ 6 files changed, 115 insertions(+), 1 deletion(-)
That should do it!
(In reply to Robert Clausecker from comment #6) Great work, thanks.