Created attachment 577660 [details] build.log /bin/sh: line 64: 816 Bus error srcdir="." ./${tname} >> ${log} 2>&1 ============================ dtypes Test Log ============================ Testing non-aligned conversions (ALIGNMENT=1).... Testing H5Tget_class() PASSED Testing H5Tcopy() PASSED Testing H5Tdetect_class() PASSED Testing compound datatypes PASSED Testing query functions of compound and enumeration types PASSED Testing transient datatypes PASSED Testing named datatypes PASSED Testing functions of encoding and decoding datatypes PASSED Testing encoding datatypes with the 'use the latest format' flag PASSED Testing exceptions for int <-> float conversions PASSED Testing indirectly reopening committed datatypes PASSED Testing deleting objects that use named datatypes PASSED Testing deleting objects that use named datatypes PASSED Testing H5Tset/get_order for compound type PASSED Testing string type creation using H5Tcreate PASSED Testing deprected API routines for datatypes PASSED Testing string conversions PASSED Testing random string conversion speed PASSED Testing some type functions for string PASSED Testing compound element reordering PASSED Testing compound subset conversions PASSED Testing compound element shrinking & reordering PASSED Testing optimized struct converter PASSED Testing compound element growing PASSED Testing compound element insertion PASSED Testing packing compound datatypes PASSED Testing compound datatype with VL string PASSED Testing array datatype of compound type with VL string PASSED Testing registering type conversion routine with compound conversions PASSED Testing adjust size of compound datatypes PASSED Testing compound datatypes of boundary size with latest format PASSED Testing unaligned VL strings in compound make[4]: *** [Makefile:3525: dtypes.chkexe_] Error 1
This still applies on sci-libs/hdf5-1.12.1-r1, but applying -fno-tree-ccp fixes it.
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=4a3fe05ed8aba8c04914b8dedd831ab1856a6e7f commit 4a3fe05ed8aba8c04914b8dedd831ab1856a6e7f Author: matoro <matoro@users.noreply.github.com> AuthorDate: 2022-05-03 01:33:00 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2022-05-03 02:05:52 +0000 sci-libs/hdf5: Stabilize 1.10.5-r1 sparc, #733578 Closes: https://bugs.gentoo.org/686620 Signed-off-by: matoro <matoro@users.noreply.github.com> Closes: https://github.com/gentoo/gentoo/pull/25291 Signed-off-by: Sam James <sam@gentoo.org> sci-libs/hdf5/hdf5-1.10.5-r1.ebuild | 7 ++++--- sci-libs/hdf5/hdf5-1.10.5.ebuild | 5 +++-- sci-libs/hdf5/hdf5-1.12.1-r1.ebuild | 3 ++- sci-libs/hdf5/hdf5-1.12.1.ebuild | 5 +++-- 4 files changed, 12 insertions(+), 8 deletions(-) Additionally, it has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=9acdbbd4c8debd771ac4a6e39c1073551f7ce43f commit 9acdbbd4c8debd771ac4a6e39c1073551f7ce43f Author: Sam James <sam@gentoo.org> AuthorDate: 2022-05-03 02:27:07 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2022-05-03 02:27:28 +0000 sci-libs/hdf5: add 1.12.2 Lots of correctness and sanitizer fixes in .2 so possibly helps with the sparc bug (although we've applied a workaround now anyway). Bug: https://bugs.gentoo.org/686620 Signed-off-by: Sam James <sam@gentoo.org> sci-libs/hdf5/Manifest | 1 + .../hdf5/files/hdf5-1.12.2-cmake_installdirs.patch | 163 +++++++++++++++++++++ sci-libs/hdf5/hdf5-1.12.2.ebuild | 119 +++++++++++++++ 3 files changed, 283 insertions(+)
Discussed a bit in #gentoo-sparc but I'm optimistic about the situation improving here as upstream seem very interested in correctness fixes and are looking at sanitizers, etc. Bit concerned about the option we're having to pass to avoid the SIGBUS as it's likely just avoiding UB but that's not an us problem, just that it's unlikely to be a compiler bug -- just avoiding an issue with upstream code. But it may well be fixed in later versions. Possibly in master if not in .2 which I just pushed.
Issue is NOT fixed in ~sci-libs/hdf5-1.12.2.
(In reply to matoro from comment #4) > Issue is NOT fixed in ~sci-libs/hdf5-1.12.2. Thanks! Could you report upstream? Did UBSAN catch it btw?
Does not get that far, dies during the build when it invokes one of its own binaries. [1579/2275] cd /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src && /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/H5detect H5Tinit.c && /usr/bin/cmake -E touch gen_SRCS.stamp1 FAILED: src/gen_SRCS.stamp1 src/H5Tinit.c /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src/gen_SRCS.stamp1 /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src/H5Tinit.c cd /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src && /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/H5detect H5Tinit.c && /usr/bin/cmake -E touch gen_SRCS.stamp1 /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2/src/H5detect.c:1275:5: runtime error: store to misaligned address 0x0100001c5001 for type 'short int', which requires 2 byte alignment 0x0100001c5001: note: pointer points here 00 00 21 00 00 00 00 10 00 01 c5 00 00 00 00 00 00 00 00 ff 00 ff ff c7 c0 01 04 00 00 00 00 00 ^ #0 0x10000044b04 in detect_C89_integers (/var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/H5detect+0x44b04)
(In reply to matoro from comment #6) > Does not get that far, dies during the build when it invokes one of its own > binaries. > > [1579/2275] cd > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src && > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/H5detect > H5Tinit.c && /usr/bin/cmake -E touch gen_SRCS.stamp1 > FAILED: src/gen_SRCS.stamp1 src/H5Tinit.c > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src/gen_SRCS. > stamp1 > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src/H5Tinit. > c > cd /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/src && > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/H5detect > H5Tinit.c && /usr/bin/cmake -E touch gen_SRCS.stamp1 > /var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2/src/H5detect.c: > 1275:5: runtime error: store to misaligned address 0x0100001c5001 for type > 'short int', which requires 2 byte alignment > 0x0100001c5001: note: pointer points here > 00 00 21 00 00 00 00 10 00 01 c5 00 00 00 00 00 00 00 00 ff 00 ff ff c7 > c0 01 04 00 00 00 00 00 > ^ > #0 0x10000044b04 in detect_C89_integers > (/var/tmp/portage/sci-libs/hdf5-1.12.2-r1/work/hdf5-1.12.2_build/bin/ > H5detect+0x44b04) OK, please file a bug.
(upstream)
The entire offending code was ripped out in https://github.com/HDFGroup/hdf5/pull/1400 two weeks ago, but it doesn't cleanly apply backwards. I'll keep an eye out for the next release keyword and check what state it's in then.
(In reply to matoro from comment #9) > The entire offending code was ripped out in > https://github.com/HDFGroup/hdf5/pull/1400 two weeks ago, but it doesn't > cleanly apply backwards. I'll keep an eye out for the next release keyword > and check what state it's in then. Thanks. Trying master would be useful too if you get time but not critical.
