New Boost.Build regression/failure

John Maddock

3 Feb 2021 3 Feb '21

6:24 p.m.

This: https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r... Has just started failing. Any ideas anyone? Thanks, John. -- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus

Show replies by date

Peter Dimov

3 Feb 3 Feb

6:51 p.m.

John Maddock wrote:

...

This: https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it: commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, origin/HEAD) Author: Rene Rivera Date: Wed Feb 3 10:19:15 2021 -0600 Remove fixed list of configure relevant features. This change removes the "hard wired" set of features that are relevant for a configure check. And instead uses the dynamic base properties of the target property set. Hence always giving the minimal set of unique properties for the configure check. This also adds output to the configure items showing what that feature set is.

René Ferdinand Rivera Morell

7:08 p.m.

On Wed, Feb 3, 2021 at 12:51 PM Peter Dimov via Boost wrote:

...

John Maddock wrote:

...
This:

https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

...
Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it:

commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, Remove fixed list of configure relevant features.

Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that. -- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supone Nada -- Robot Dreams - http://robot-dreams.net

Peter Dimov

7:26 p.m.

René Ferdinand Rivera Morell wrote:

...

On Wed, Feb 3, 2021 at 12:51 PM Peter Dimov via Boost wrote:

...
John Maddock wrote:

...
This: https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it:

commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, Remove fixed list of configure relevant features.

Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that.

No idea about that. It only fails with Cygwin GCC though. Also, under msvc there's a change in the output: - symlinks supported : no (cached) : - junctions supported : yes (cached) : - hardlinks supported : yes (cached) : The second : was not there before. No idea where it comes from or what's the significance.

René Ferdinand Rivera Morell

7:38 p.m.

On Wed, Feb 3, 2021 at 1:34 PM Peter Dimov wrote:

...

René Ferdinand Rivera Morell wrote:

...
On Wed, Feb 3, 2021 at 12:51 PM Peter Dimov via Boost wrote:

...
John Maddock wrote:

...
This:

https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

...
Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it:

commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, Remove fixed list of configure relevant features.

Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that.

No idea about that. It only fails with Cygwin GCC though.

Interesting.. That's the one setup I don't test :-(

...

Also, under msvc there's a change in the output:

- symlinks supported : no (cached) : - junctions supported : yes (cached) : - hardlinks supported : yes (cached) :

The second : was not there before. No idea where it comes from or what's the significance.

That's part of the added info for the configure variation it's reporting the result of. Doesn't impact it in any way. And it would have also been blank in the past if we did show it. -- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supone Nada -- Robot Dreams - http://robot-dreams.net

Peter Dimov

7:55 p.m.

René Ferdinand Rivera Morell wrote:

...

On Wed, Feb 3, 2021 at 1:34 PM Peter Dimov wrote:

...
René Ferdinand Rivera Morell wrote: ...

...
Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that.

No idea about that. It only fails with Cygwin GCC though.

Interesting.. That's the one setup I don't test :-(

Even more interesting, it works with Cygwin Clang (toolset=clang). Only toolset=gcc fails. C:\boost-git\develop\libs\config\test>b2 print_config_info toolset=clang Performing configuration checks - default architecture : x86 - default address-model : 64-bit - symlinks supported : no (cached) : - junctions supported : yes (cached) : - hardlinks supported : yes (cached) : ...found 185 targets... ...updating 7 targets... clang-linux.compile.c++.without-pch ..\..\..\bin.v2\libs\config\test\clang-linux-8.0.1\debug\threading-multi\visibility-hidden\config_info.obj clang-linux.link ..\..\..\bin.v2\libs\config\test\clang-linux-8.0.1\debug\threading-multi\visibility-hidden\config_info_printer.exe clang-8: warning: argument unused during compilation: '-pthread' [-Wunused-command-line-argument] Jamfile.print-run print_config_info

Andrey Semashev

5 Feb 5 Feb

9:12 p.m.

On 2/3/21 10:08 PM, René Ferdinand Rivera Morell via Boost wrote:

...

On Wed, Feb 3, 2021 at 12:51 PM Peter Dimov via Boost wrote:

...
John Maddock wrote:

...
This:

https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

...
Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it:

commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, Remove fixed list of configure relevant features.

Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that.

I've created a ticket: https://github.com/boostorg/build/issues/709

René Ferdinand Rivera Morell

8 Feb 8 Feb

5:21 p.m.

On Fri, Feb 5, 2021 at 3:12 PM Andrey Semashev via Boost < boost@lists.boost.org> wrote:

...

On 2/3/21 10:08 PM, René Ferdinand Rivera Morell via Boost wrote:

...
On Wed, Feb 3, 2021 at 12:51 PM Peter Dimov via Boost < boost@lists.boost.org> wrote:

...
John Maddock wrote:

...
This:

https://ci.appveyor.com/project/jzmaddock/config/builds/37592647/job/l6ogy7r...

...
...
Has just started failing. Any ideas anyone?

This worked for me locally before I updated tools/build, then started failing with the above. I suppose this the change that caused it:

commit 95c875b1dd12a52f0d120480c8b498c9c3da36ab (HEAD, origin/develop, Remove fixed list of configure relevant features.

Could be. Although I find it weird that the boostcpp module has recursive configuration checks. Because if it is.. It should really not do that.

I've created a ticket:

https://github.com/boostorg/build/issues/709

You read the root of the problem and resolution on that ticket. But the end result is that this is now fixed / worked around. -- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supone Nada -- Robot Dreams - http://robot-dreams.net

John Maddock

6:01 p.m.

...

You read the root of the problem and resolution on that ticket. But the end result is that this is now fixed / worked around.

Thanks Rene! -- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus

1387

Age (days ago)

1392

Last active (days ago)

List overview

Download

8 comments

4 participants

participants (4)

Andrey Semashev
John Maddock
Peter Dimov
René Ferdinand Rivera Morell