llvm dev - Jan 2015 - [LLVMdev] [3.6 Release] RC1 has been tagged, Testing Phase I begins

Hi testers!

3.6.0-rc1 was just tagged, so testing can begin.

I'm not super familiar with the testing script yet, but I believe the
idea is to run it as follows:

  $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
x86_64-unknown-linux-gnu

With the -triple value substituted for your platform. Bill's 3.5rc1
email also said 'test-suite' needs to be run by hand.

Please upload binaries to the sftp and report your results to this thread.

Thanks for helping with the release!

 - Hans

Hi,

I uploaded the binary for OS X. Nothing suspicious in testing...

Cheers,
Sebastian

2015-01-15 5:44 GMT+01:00 Hans Wennborg <hans at
chromium.org>:
> Hi testers!
>
> 3.6.0-rc1 was just tagged, so testing can begin.
>
> I'm not super familiar with the testing script yet, but I believe the
> idea is to run it as follows:
>
>   $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
> x86_64-unknown-linux-gnu
>
> With the -triple value substituted for your platform. Bill's 3.5rc1
> email also said 'test-suite' needs to be run by hand.
>
> Please upload binaries to the sftp and report your results to this thread.
>
> Thanks for helping with the release!
>
>  - Hans

On 15 January 2015 at 04:44, Hans Wennborg <hans at chromium.org>
wrote:
> Please upload binaries to the sftp and report your results to this thread.
AArch64 tested and uploaded. ARMv7 in progress.

cheers,
--renato

On 15 Jan 2015, at 05:44, Hans Wennborg <hans at chromium.org>
wrote:
> 
> 3.6.0-rc1 was just tagged, so testing can begin.
> 
> I'm not super familiar with the testing script yet, but I believe the
> idea is to run it as follows:
> 
>  $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
> x86_64-unknown-linux-gnu
> 
> With the -triple value substituted for your platform. Bill's 3.5rc1
> email also said 'test-suite' needs to be run by hand.
> 
> Please upload binaries to the sftp and report your results to this thread.
> 
> Thanks for helping with the release!
> 
> - Hans
Here are binaries for FreeBSD 10; I did not see any failures during testing.

As usual, I don't have access to sftp (and don't even know which server
;), but the binaries can be found here:

http://www.andric.com/freebsd/clang/clang+llvm-3.6.0-rc1-i386-unknown-freebsd10.tar.xz
http://www.andric.com/freebsd/clang/clang+llvm-3.6.0-rc1-amd64-unknown-freebsd10.tar.xz

SHA256 (clang+llvm-3.6.0-rc1-i386-unknown-freebsd10.tar.xz) =
304c444cd005819e93f3b580fb1f821a0ab2ec69bb56e61902412cc7abd59883
SHA256 (clang+llvm-3.6.0-rc1-amd64-unknown-freebsd10.tar.xz) =
117f763ba7afd1d6f3f4ca4f534464d68be215dcdf52b4edd219930badee0d46

-Dimitry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: Message signed with OpenPGP using GPGMail
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150115/f5b54421/attachment.sig>

On Wed, Jan 14, 2015 at 8:44 PM, Hans Wennborg <hans at chromium.org>
wrote:
> Hi testers!
>
> 3.6.0-rc1 was just tagged, so testing can begin.
>
> I'm not super familiar with the testing script yet, but I believe the
> idea is to run it as follows:
>
>   $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
> x86_64-unknown-linux-gnu
>
> With the -triple value substituted for your platform. Bill's 3.5rc1
> email also said 'test-suite' needs to be run by hand.
>
> Please upload binaries to the sftp and report your results to this thread.
>
> Thanks for helping with the release!
I've uploaded the Windows binary. Everything looked good except
Modules/compiler_builtins.m (PR20995). I'll take a look at that
tomorrow.

 - Hans

I've uploaded the mips binaries and started the test-suite. We unexpectedly*
fail two 'make check-all' tests at the moment. One of them is a faulty
test and the fix was committed to trunk yesterday. The other is a recently-added
test (test/DebugInfo/multiline.ll) that we're looking into.

* There's one expected fail on the little endian build caused by running
test/Modules/prune.m on an NFS mount that doesn't track access times
correctly. It's still there but it's not clangs fault.
> Dimitry:
> As usual, I don't have access to sftp (and don't even know which
server ;)
Anton Korobeynikov <anton at korobeynikov.info> should be able to sort
this out for you. You'll need to provide an ssh key.
________________________________________
From: Dimitry Andric [dimitry at andric.com]
Sent: 15 January 2015 18:59
To: Hans Wennborg
Cc: llvmdev; cfe-dev; Ben Pope; Renato Golin; Sylvestre Ledru; Sebastian
Dreßler; Daniel Sanders; Nikola Smiljanić
Subject: Re: [3.6 Release] RC1 has been tagged, Testing Phase I begins

On 15 Jan 2015, at 05:44, Hans Wennborg <hans at chromium.org>
wrote:
>
> 3.6.0-rc1 was just tagged, so testing can begin.
>
> I'm not super familiar with the testing script yet, but I believe the
> idea is to run it as follows:
>
>  $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
> x86_64-unknown-linux-gnu
>
> With the -triple value substituted for your platform. Bill's 3.5rc1
> email also said 'test-suite' needs to be run by hand.
>
> Please upload binaries to the sftp and report your results to this thread.
>
> Thanks for helping with the release!
>
> - Hans
Here are binaries for FreeBSD 10; I did not see any failures during testing.

As usual, I don't have access to sftp (and don't even know which server
;), but the binaries can be found here:

http://www.andric.com/freebsd/clang/clang+llvm-3.6.0-rc1-i386-unknown-freebsd10.tar.xz
http://www.andric.com/freebsd/clang/clang+llvm-3.6.0-rc1-amd64-unknown-freebsd10.tar.xz

SHA256 (clang+llvm-3.6.0-rc1-i386-unknown-freebsd10.tar.xz) =
304c444cd005819e93f3b580fb1f821a0ab2ec69bb56e61902412cc7abd59883
SHA256 (clang+llvm-3.6.0-rc1-amd64-unknown-freebsd10.tar.xz) =
117f763ba7afd1d6f3f4ca4f534464d68be215dcdf52b4edd219930badee0d46

-Dimitry

On 15 January 2015 at 13:35, Renato Golin <renato.golin at linaro.org>
wrote:
> AArch64 tested and uploaded. ARMv7 in progress.
Hi Hans,

ARMv7 is giving me a bit of a headache... I've tried building on two
different machines and I get the same error... Phase 2 completes
successfully, but while building Phase 3, I get the error on *every*
file:

1.     
/usr/lib/gcc/arm-linux-gnueabihf/4.9/../../../../include/c++/4.9/bits/ptr_traits.h:77:25:
at annotation token
2.     
/usr/lib/gcc/arm-linux-gnueabihf/4.9/../../../../include/c++/4.9/bits/ptr_traits.h:37:1:
parsing namespace 'std'
3.     
/usr/lib/gcc/arm-linux-gnueabihf/4.9/../../../../include/c++/4.9/bits/ptr_traits.h:73:5:
parsing struct/union/class body '__ptrtr_rebind_helper'
clang: error: unable to execute command: Bus error
clang: error: clang frontend command failed due to signal (use -v to
see invocation)

Phase 1 passes check-all with flying colours, while Phase 2 fails *a
lot* if C++ Clang tests with seg-faults in the parser. But this is
something we're not seeing in the self-hosting buildbot for the same
commit as 3.6 branched off of:

http://lab.llvm.org:8011/builders/clang-cmake-armv7-a15-selfhost/builds/2343

Could it be some of the merges later?

cheers,
--renato

Hello,

On 15/01/2015 05:44, Hans Wennborg wrote:
> Hi testers!
>
> 3.6.0-rc1 was just tagged, so testing can begin.
>
> I'm not super familiar with the testing script yet, but I believe the
> idea is to run it as follows:
>
>   $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
> x86_64-unknown-linux-gnu
>
> With the -triple value substituted for your platform. Bill's 3.5rc1
> email also said 'test-suite' needs to be run by hand.
>
> Please upload binaries to the sftp and report your results to this thread.
>
I uploaded in Debian 3.6rc1. Most of the archs are fine beside kfreebsd.
I had to write some patches to make it build (I will see to forward them).
David and I did a rebuild of the whole Debian archive using clang 3.6rc1:
http://clang.debian.net/status.php?version=3.6.0rc1

It didn't show any regressions compared to 3.5. There are a few more
failures but nothing critical: 5.7 % of failure with 3.5, 5.9% with 3.6

I found two new kind of failures:
http://clang.debian.net/status.php?version=3.6.0rc1&key=KEYWORD_PRECEDE
http://clang.debian.net/status.php?version=3.6.0rc1&key=INCORRECT_CONVERSION

I will do another rebuild once 3.6 stable is released.

Cheers,
Sylvestre

On Tue, Jan 27, 2015 at 1:29 PM, Sylvestre Ledru <sylvestre at debian.org>
wrote:
> Hello,
>
> On 15/01/2015 05:44, Hans Wennborg wrote:
>> Hi testers!
>>
>> 3.6.0-rc1 was just tagged, so testing can begin.
>>
>> I'm not super familiar with the testing script yet, but I believe
the
>> idea is to run it as follows:
>>
>>   $ utils/release/test-release.sh -release 3.6.0 -rc 1 -triple
>> x86_64-unknown-linux-gnu
>>
>> With the -triple value substituted for your platform. Bill's 3.5rc1
>> email also said 'test-suite' needs to be run by hand.
>>
>> Please upload binaries to the sftp and report your results to this
thread.
>>
> I uploaded in Debian 3.6rc1.
I couldn't find anything on the sftp. Did you upload somewhere else?
> Most of the archs are fine beside kfreebsd.
> I had to write some patches to make it build (I will see to forward them).
Are the patches for kfreebsd, or for building in general? It would be
nice if we could get them in before tagging rc2.
> David and I did a rebuild of the whole Debian archive using clang 3.6rc1:
> http://clang.debian.net/status.php?version=3.6.0rc1
>
> It didn't show any regressions compared to 3.5. There are a few more
> failures but nothing critical: 5.7 % of failure with 3.5, 5.9% with 3.6
>
> I found two new kind of failures:
> http://clang.debian.net/status.php?version=3.6.0rc1&key=KEYWORD_PRECEDE
This wording was added in
http://llvm.org/viewvc/llvm-project?rev=221630&view=rev (which is new
in 3.6). But IIUC, we rejected the code before too, but probably with
a less nice message.
>
http://clang.debian.net/status.php?version=3.6.0rc1&key=INCORRECT_CONVERSION
Looks like good finds :-)

Thank you very much for testing!

Cheers,
Hans