On 28 August 2017 at 07:07, Oliver Kowalke via Boost wrote:

could you you add following commits:

boost.context: - https://github.com/boostorg/context/commit/420554fe9ac014829 90e8ae314f4f7ba81010c9a

boost.fiber: - https://github.com/boostorg/fiber/commit/63f1041f0e484e64dec 437a0fa780ea2463abda2 - https://github.com/boostorg/fiber/commit/63f1041f0e484e64dec437a0fa780ea2463... https://github.com/boostorg/fiber/commit/440c7f799ae2e5a7718e5126d5b982bbd65...

The fiber commit is fine, but the context commit has the same problem as thread - it will pull in the previous changes as well. So I need a branch with the appropriate commits cherry-picked. I'll do it if you want.

Can we also do https://github.com/boostorg/thread/commit/39c23ddf5a3082d76ebda107dce02fb86e... It fixes an outstanding deadlock issue. On Mon, Aug 28, 2017 at 9:01 AM, Oliver Kowalke via Boost wrote:

2017-08-28 13:22 GMT+02:00 Daniel James via Boost :

...

On 28 August 2017 at 07:07, Oliver Kowalke via Boost wrote:

...

could you you add following commits:

boost.context: - https://github.com/boostorg/context/commit/420554fe9ac014829 90e8ae314f4f7ba81010c9a

boost.fiber: - https://github.com/boostorg/fiber/commit/63f1041f0e484e64dec 437a0fa780ea2463abda2 - <https://github.com/boostorg/fiber/commit/63f1041f0e484e64dec437a0fa780e

a2463abda2>

...

https://github.com/boostorg/fiber/commit/440c7f799ae2e5a7718e5126d5b982 bbd656cb72

The fiber commit is fine, but the context commit has the same problem as thread - it will pull in the previous changes as well. So I need a branch with the appropriate commits cherry-picked. I'll do it if you want.

OK, that would be fine. ty

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

On Mon, Aug 28, 2017 at 11:29 AM, Daniel James via Boost wrote:

On 28 August 2017 at 16:09, Gary Furnish via Boost wrote:

...

Can we also do https://github.com/boostorg/thread/commit/39c23ddf5a3082d76ebda107dce02fb86e... It fixes an outstanding deadlock issue.

Sorry, but support for the new Visual C++ release seems like a fairly large job that is best handled in the next full release. I don't need it personally need it put in, but it likely effects all versions using the windows API.

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

On 28 August 2017 at 10:19, Raffi Enficiaud via Boost wrote:

For the release notes, I suggest taking the one of boost 1.65 and add a section "Changes since 1.65.0" for the point release.

We have separate release notes for point releases, so that download links and things like that work, and also because that's how other releases have been handled. But I copied in the 1.65.0 release notes, you can see them here: http://www.boost.org/users/history/in_progress.html Ignore the 1.66.0 skeleton release notes at the bottom, that page displays all the 'in progress' releases and I never put any effort into formatting it.

On 28 August 2017 at 17:51, John Maddock via Boost wrote:

I've had a request for this PR: https://github.com/boostorg/config/pull/175 to be included as without it the next CUDA release won't compile Boost as the meaning of __CUDACC_VER__ has changed in a non-backwards compatible way :(

It would need to be cherry picked to a branch I'm afraid.

Go ahead. I'm considerably more lax about this change as it only affects CUDA users. Didn't someone say something about the Visual C++ version check? That seems like something that's simple enough to fix quickly in this release.

On 28 August 2017 at 20:45, Antony Polukhin via Boost wrote:

2017-08-28 2:39 GMT+03:00 Daniel James via Boost :

...

Hello all,

As no one objected, I'm starting work on a point release. <...> If I can cleanly incorporate changes from your master branch, let me know the commit hash.

Please merge all the Stacktrace changes from master branch up to https://github.com/boostorg/stacktrace/commit/af78623e8a525f28830289da7e7d5e... Master branch contains multiple fixes for builds and installation.

OK. IMO the '.pp' problem should be fixed in the build system, but it's too late for that.

On 30 August 2017 at 19:33, Antony Polukhin via Boost wrote:

I've just merged the FreeBSD fixes into master.

Please incorporate all the Stacktrace master branch changes, up to https://github.com/boostorg/stacktrace/commit/8364ad1f5e93e1db9ea0cf5306ba8a...

Done.

On 28 August 2017 at 22:49, Peter Dimov via Boost wrote:

I have just received a request to incorporate

https://github.com/boostorg/smart_ptr/pull/40

into 1.65.1, which seems reasonable. That's commit

https://github.com/boostorg/smart_ptr/commit/10c6233029b1766bd9c58499ca42c85...

which, in addition to the PR, also contains a few trivial fixes for the new MSVC version, but this should be OK.

OK, but please just say which changes you want to include.

On Tue, Aug 29, 2017 at 2:21 AM, Daniel James via Boost wrote:

OK, but please just say which changes you want to include.

Can we please do something about "Unknown compiler version - please run configure tests and report the result" when using the latest Visual Studio update? Thanks!

On Tue, Aug 29, 2017 at 10:07 AM, John Maddock via Boost wrote:

That should have been fixed in the 1.65.0 release anyway?

When I do git fetch upstream git checkout master git submodule update after updating my Visual Studio, I get the visualc.hpp warnings "Unknown compiler version" Thanks

On Tue, Aug 29, 2017 at 10:27 AM, Daniela Engert via Boost wrote:

The fix is in 'master' - one commit after tag 'boost-1.65.0'

From the commit titled "Kick off 1.65.1" on master, I only see updates to fiber, context, stacktrace, smart_ptr, and a bug fix for thread hash. I don't see any update to the config submodule. Am I reading this correctly?

https://github.com/boostorg/boost/commits/master Thanks

On 29/08/2017 19:22, Daniel James via Boost wrote:

On 29 August 2017 at 19:16, John Maddock via Boost wrote:

...

There's now a fixes/1.65 branch for Boost.Config containing the msvc and CUDA changes. Thanks, I've updated the super project.

Great! Thanks Daniel, John. --- This email has been checked for viruses by AVG. http://www.avg.com