Dear testers,
3.7.0-rc2 was just tagged. Please test, build binaries, upload to the
sftp, and report results to this thread.
A lot of fixes have been applied since rc1, both to the build script
and the code in general, so hopefully it will be less bumpy this time.
From this point, I will no longer accept merge requests to finish
existing features; it's now bug-fixes only.
Thanks again for helping with the release!
Cheers,
Hans
LNT is looking good on Ubuntu 14.04 x64, uploaded:
clang+llvm-3.7.0-rc2-x86_64-linux-gnu-ubuntu-14.04.tar.xz
The errors reported during build are:
Failing Tests (17):
AddressSanitizer-x86_64-linux :: TestCases/Posix/readv.cc
MemorySanitizer :: Linux/tcgetattr.cc
MemorySanitizer-Unit :: Msan-x86_64-Test/MemorySanitizer.gethostent
MemorySanitizer-Unit :: Msan-x86_64-Test/MemorySanitizer.gethostent_r
MemorySanitizer-Unit :: Msan-x86_64-Test/MemorySanitizer.getmntent
MemorySanitizer-Unit :: Msan-x86_64-Test/MemorySanitizer.getmntent_r
MemorySanitizer-Unit :: Msan-x86_64-with-call-Test/MemorySanitizer.gethostent
MemorySanitizer-Unit :: Msan-x86_64-with-call-Test/MemorySanitizer.gethostent_r
MemorySanitizer-Unit :: Msan-x86_64-with-call-Test/MemorySanitizer.getmntent
MemorySanitizer-Unit :: Msan-x86_64-with-call-Test/MemorySanitizer.getmntent_r
SanitizerCommon-asan :: Linux/getpass.cc
SanitizerCommon-asan :: Posix/decorate_proc_maps.cc
SanitizerCommon-lsan :: Linux/getpass.cc
SanitizerCommon-msan :: Linux/getpass.cc
SanitizerCommon-msan :: Posix/decorate_proc_maps.cc
SanitizerCommon-tsan :: Linux/getpass.cc
SanitizerCommon-tsan :: Posix/decorate_proc_maps.cc
Expected Passes : 24028
Expected Failures : 131
Unsupported Tests : 377
Unexpected Failures: 17
Ben
64bit Fedora
Failing Tests (8):
LeakSanitizer-AddressSanitizer :: TestCases/cleanup_in_tsd_destructor.cc
LeakSanitizer-AddressSanitizer :: TestCases/disabler.cc
LeakSanitizer-AddressSanitizer :: TestCases/disabler_in_tsd_destructor.cc
LeakSanitizer-AddressSanitizer :: TestCases/ignore_object.cc
LeakSanitizer-Standalone :: TestCases/cleanup_in_tsd_destructor.cc
LeakSanitizer-Standalone :: TestCases/disabler.cc
LeakSanitizer-Standalone :: TestCases/disabler_in_tsd_destructor.cc
LeakSanitizer-Standalone :: TestCases/ignore_object.cc
OS X 64Bit.
All tests passed.
Expected Passes : 28451
Expected Failures : 163
Unsupported Tests : 478
Built fine, and no test failures on FreeBSD (though I used autoconf again, like the previous rc).
Uploaded the following:
SHA256 (clang+llvm-3.7.0-rc2-amd64-unknown-freebsd10.tar.xz) = 23b5e3e79e6ce8375480a6cf0882831bfc88fb2cab476a5965a6510a69487680
SHA256 (clang+llvm-3.7.0-rc2-i386-unknown-freebsd10.tar.xz) = 09662a05a9ba46d22811b77502a1bbb74496bf778a8e048819858b4f324944d1
-Dimitry
As I mentioned to Hans on another thread I'm testing with two last-minute patches applied:
r243686 - Fixes an asan failure but is also suspected of breaking an x86 sanitizer buildbot. It hasn't been merged as a result. If it doesn't get merged it just adds a non-regression failure to mips-linux-gnu's check-all and possibly mipsel-linux-gnu's too.
r243745 - Fixes a rare register allocator bug potentially affecting all targets but currently affecting almabench on some mips configs. This patch just missed rc2 and was merged shortly after the tag was created.
I'm sure we'll need an rc3 due to the second of the two patches and I'll build that with a clean tag.
clang+llvm-3.7.0-rc2-mips-linux-gnu.tar.xz
check-all:
UBSan-ASan-mips :: TestCases/Float/cast-overflow.cpp
UBSan-Standalone-mips :: TestCases/Float/cast-overflow.cpp
Patch to fix this is committed to trunk but no response from code-owner (Richard Smith) for the merge request. I've asked Hans if this is still a merge candidate and pinged Richard again.
test-suite: Started. Will report back when it's done
clang+llvm-3.7.0-rc2-mipsel-linux-gnu.tar.xz
check-all: All ok (just the prune.m failure caused by NFS and noatime)
test-suite: Started. Will report back when it's done
clang+llvm-3.7.0-rc2-x86_64-linux-gnu-ubuntu-14.04.tar.xz (cross compiling to mips)
test-suite: Started. Will report back when it's done. I'm expecting almabench to fail due to the lack of r243745 so once the tag is tested I'll re-test using a local build that has it.
Hi,
I uploaded
e3c21f8567b86922e64d322c3413a31a8678affe clang+llvm-3.7.0-rc2-x86_64-apple-darwin.tar.xz
to SFTP. Testing itself was OK, except for differing binaries, e.g.
sed: RE error: illegal byte sequence
file YAMLBench.o differs between phase 2 and phase 3
Could be, that this comes from sed having some subtle differences on OS X in comparison to GNU sed. I’ll have a look into the script.
Cheers,
Sebastian
ARM and AArch64 binaries up and working. I got some XPASS on ASAN/ARM,
but that's hardly a problem. 
Test-suites green, all cool and dandy.
cheers,
--renato
Uploaded Windows binaries to the sftp:
42a402d852c2c11d83705b7683c853b3953c7665 LLVM-3.7.0-rc2-win32.exe
106434e853aad1aadbfcf89507793d45b37fc92a LLVM-3.7.0-rc2-win64.exe
I had one lit test failure on the 32-bit build:
MCJITCAPITest.addGlobalMapping. Apparently I had forgotten to merge
r243639. Besides that, everything looked good.
Attaching the script I used for building.
- Hans
build_llvm_370.bat|attachment (2.53 KB)
clang+llvm-3.7.0-rc2-mips-linux-gnu.tar.xz
check-all:
UBSan-ASan-mips :: TestCases/Float/cast-overflow.cpp
UBSan-Standalone-mips :: TestCases/Float/cast-overflow.cpp
Patch to fix this is committed to trunk but no response from code-owner (Richard Smith) for the merge request. I've asked Hans if this is still a merge candidate (it is) and pinged Richard again.
test-suite: All ok.
clang+llvm-3.7.0-rc2-mipsel-linux-gnu.tar.xz
check-all: All ok (just the prune.m failure caused by NFS and noatime)
test-suite: All ok
clang+llvm-3.7.0-rc2-x86_64-linux-gnu-ubuntu-14.04.tar.xz (cross compiling to mips)
test-suite: As expected almabench failed (but on fewer configs than last time, not sure why). Will re-test using a local build that has r243745 applied. Unexpectedly, kimwitu++/kc failed the reference check and tramp3d-v4 segfaulted on three configs (big-endian N32, mips64r[126]). Hopefully, it's the same issue as almabench but I'll investigate.