ci: remove LTO job
diff mbox series

Message ID 1573241531-16484-1-git-send-email-david.marchand@redhat.com
State Accepted, archived
Delegated to: David Marchand
Headers show
Series
  • ci: remove LTO job
Related show

Checks

Context Check Description
ci/Intel-compilation fail apply issues
ci/travis-robot success Travis build: passed
ci/checkpatch success coding style OK

Commit Message

David Marchand Nov. 8, 2019, 7:32 p.m. UTC
The LTO job using gcc-7 has two issues at the moment:
- warnings about implicit fallthroughs trigger build errors:

In file included from ...common/include/rte_memory.h:22:0,
                 from ...linux/eal/eal_hugepage_info.c:24:
...common/include/rte_common.h: In function ‘rte_str_to_size’:
...common/include/rte_common.h:744:27: error: this statement may
  fall through [-Werror=implicit-fallthrough=]
  case 'G': case 'g': size *= 1024; /* fall-through */
                      ~~~~~^~~~~~~

- if we disable this warning, linking the binaries takes too much time
  and the job is terminated by Travis because it reaches the maximum
  time limit

Fixes: 098cc0fea3be ("build: add option to enable LTO")

Signed-off-by: David Marchand <david.marchand@redhat.com>
---
 .travis.yml | 9 ---------
 1 file changed, 9 deletions(-)

Comments

Thomas Monjalon Nov. 8, 2019, 7:55 p.m. UTC | #1
08/11/2019 20:32, David Marchand:
> The LTO job using gcc-7 has two issues at the moment:
> - warnings about implicit fallthroughs trigger build errors:
> 
> In file included from ...common/include/rte_memory.h:22:0,
>                  from ...linux/eal/eal_hugepage_info.c:24:
> ...common/include/rte_common.h: In function ‘rte_str_to_size’:
> ...common/include/rte_common.h:744:27: error: this statement may
>   fall through [-Werror=implicit-fallthrough=]
>   case 'G': case 'g': size *= 1024; /* fall-through */
>                       ~~~~~^~~~~~~
> 
> - if we disable this warning, linking the binaries takes too much time
>   and the job is terminated by Travis because it reaches the maximum
>   time limit
> 
> Fixes: 098cc0fea3be ("build: add option to enable LTO")
> 
> Signed-off-by: David Marchand <david.marchand@redhat.com>

Acked-by: Thomas Monjalon <thomas@monjalon.net>

Thank you, I should have tested it in Travis before merging.
David Marchand Nov. 8, 2019, 8:07 p.m. UTC | #2
On Fri, Nov 8, 2019 at 8:56 PM Thomas Monjalon <thomas@monjalon.net> wrote:
>
> 08/11/2019 20:32, David Marchand:
> > The LTO job using gcc-7 has two issues at the moment:
> > - warnings about implicit fallthroughs trigger build errors:
> >
> > In file included from ...common/include/rte_memory.h:22:0,
> >                  from ...linux/eal/eal_hugepage_info.c:24:
> > ...common/include/rte_common.h: In function ‘rte_str_to_size’:
> > ...common/include/rte_common.h:744:27: error: this statement may
> >   fall through [-Werror=implicit-fallthrough=]
> >   case 'G': case 'g': size *= 1024; /* fall-through */
> >                       ~~~~~^~~~~~~
> >
> > - if we disable this warning, linking the binaries takes too much time
> >   and the job is terminated by Travis because it reaches the maximum
> >   time limit
> >
> > Fixes: 098cc0fea3be ("build: add option to enable LTO")
> >
> > Signed-off-by: David Marchand <david.marchand@redhat.com>
>
> Acked-by: Thomas Monjalon <thomas@monjalon.net>

Thanks, applied.

> Thank you, I should have tested it in Travis before merging.

Well, the robot could have helped, but I can't find the job for the
7128 series, and the job for the 7333 series is still pending.
Bruce Richardson Nov. 11, 2019, 11:31 a.m. UTC | #3
On Fri, Nov 08, 2019 at 08:55:52PM +0100, Thomas Monjalon wrote:
> 08/11/2019 20:32, David Marchand:
> > The LTO job using gcc-7 has two issues at the moment:
> > - warnings about implicit fallthroughs trigger build errors:
> > 
> > In file included from ...common/include/rte_memory.h:22:0,
> >                  from ...linux/eal/eal_hugepage_info.c:24:
> > ...common/include/rte_common.h: In function ‘rte_str_to_size’:
> > ...common/include/rte_common.h:744:27: error: this statement may
> >   fall through [-Werror=implicit-fallthrough=]
> >   case 'G': case 'g': size *= 1024; /* fall-through */
> >                       ~~~~~^~~~~~~
> > 
> > - if we disable this warning, linking the binaries takes too much time
> >   and the job is terminated by Travis because it reaches the maximum
> >   time limit
> > 
> > Fixes: 098cc0fea3be ("build: add option to enable LTO")
> > 
> > Signed-off-by: David Marchand <david.marchand@redhat.com>
> 
> Acked-by: Thomas Monjalon <thomas@monjalon.net>
> 
> Thank you, I should have tested it in Travis before merging.
> 
I'm also interested in whether someone can give steps to reproduce this
issue outside of the Travis environment. I've tested with gcc-7 on both
Ubuntu 16.04 (as used by Travis) and Ubuntu 18.04, and the builds I tested
all worked fine, without any warnings being seen. I am therefore wondering
what is special about the travis setup that causes it to fail.

/Bruce
David Marchand Nov. 11, 2019, 12:27 p.m. UTC | #4
On Mon, Nov 11, 2019 at 12:32 PM Bruce Richardson
<bruce.richardson@intel.com> wrote:
>
> On Fri, Nov 08, 2019 at 08:55:52PM +0100, Thomas Monjalon wrote:
> > 08/11/2019 20:32, David Marchand:
> > > The LTO job using gcc-7 has two issues at the moment:
> > > - warnings about implicit fallthroughs trigger build errors:
> > >
> > > In file included from ...common/include/rte_memory.h:22:0,
> > >                  from ...linux/eal/eal_hugepage_info.c:24:
> > > ...common/include/rte_common.h: In function ‘rte_str_to_size’:
> > > ...common/include/rte_common.h:744:27: error: this statement may
> > >   fall through [-Werror=implicit-fallthrough=]
> > >   case 'G': case 'g': size *= 1024; /* fall-through */
> > >                       ~~~~~^~~~~~~
> > >
> > > - if we disable this warning, linking the binaries takes too much time
> > >   and the job is terminated by Travis because it reaches the maximum
> > >   time limit
> > >
> > > Fixes: 098cc0fea3be ("build: add option to enable LTO")
> > >
> > > Signed-off-by: David Marchand <david.marchand@redhat.com>
> >
> > Acked-by: Thomas Monjalon <thomas@monjalon.net>
> >
> > Thank you, I should have tested it in Travis before merging.
> >
> I'm also interested in whether someone can give steps to reproduce this
> issue outside of the Travis environment. I've tested with gcc-7 on both
> Ubuntu 16.04 (as used by Travis) and Ubuntu 18.04, and the builds I tested
> all worked fine, without any warnings being seen. I am therefore wondering
> what is special about the travis setup that causes it to fail.

Did you try with the custom apt repository that the LTO job was using ?
-      apt:
-        sources:
-            - ubuntu-toolchain-r-test
Bruce Richardson Nov. 11, 2019, 1:57 p.m. UTC | #5
On Mon, Nov 11, 2019 at 01:27:23PM +0100, David Marchand wrote:
> On Mon, Nov 11, 2019 at 12:32 PM Bruce Richardson
> <bruce.richardson@intel.com> wrote:
> >
> > On Fri, Nov 08, 2019 at 08:55:52PM +0100, Thomas Monjalon wrote:
> > > 08/11/2019 20:32, David Marchand:
> > > > The LTO job using gcc-7 has two issues at the moment:
> > > > - warnings about implicit fallthroughs trigger build errors:
> > > >
> > > > In file included from ...common/include/rte_memory.h:22:0,
> > > >                  from ...linux/eal/eal_hugepage_info.c:24:
> > > > ...common/include/rte_common.h: In function ‘rte_str_to_size’:
> > > > ...common/include/rte_common.h:744:27: error: this statement may
> > > >   fall through [-Werror=implicit-fallthrough=]
> > > >   case 'G': case 'g': size *= 1024; /* fall-through */
> > > >                       ~~~~~^~~~~~~
> > > >
> > > > - if we disable this warning, linking the binaries takes too much time
> > > >   and the job is terminated by Travis because it reaches the maximum
> > > >   time limit
> > > >
> > > > Fixes: 098cc0fea3be ("build: add option to enable LTO")
> > > >
> > > > Signed-off-by: David Marchand <david.marchand@redhat.com>
> > >
> > > Acked-by: Thomas Monjalon <thomas@monjalon.net>
> > >
> > > Thank you, I should have tested it in Travis before merging.
> > >
> > I'm also interested in whether someone can give steps to reproduce this
> > issue outside of the Travis environment. I've tested with gcc-7 on both
> > Ubuntu 16.04 (as used by Travis) and Ubuntu 18.04, and the builds I tested
> > all worked fine, without any warnings being seen. I am therefore wondering
> > what is special about the travis setup that causes it to fail.
> 
> Did you try with the custom apt repository that the LTO job was using ?
> -      apt:
> -        sources:
> -            - ubuntu-toolchain-r-test
> 
Yes, that repo was needed to get gcc-7 on Ubuntu 16.04, IIRC
Andrzej Ostruszka Nov. 12, 2019, 8:55 a.m. UTC | #6
On 11/11/19 2:57 PM, Bruce Richardson wrote:
> On Mon, Nov 11, 2019 at 01:27:23PM +0100, David Marchand wrote:
[...]
>> Did you try with the custom apt repository that the LTO job was using ?
>> -      apt:
>> -        sources:
>> -            - ubuntu-toolchain-r-test
>>
> Yes, that repo was needed to get gcc-7 on Ubuntu 16.04, IIRC

That is correct - Travis build env was said to be Ubuntu 16.04 and in
order to install gcc 7 package from this PPA is used.  I tested this on
a fresh VM with that version and it worked, so I'm a bit curious what is
the difference too.

However I'm fine with removing of travis build entry for LTO - the build
time problems are enough of a reason for me.

Regards
Andrzej

Patch
diff mbox series

diff --git a/.travis.yml b/.travis.yml
index a1c31d4..3d6ef29 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -105,15 +105,6 @@  matrix:
       apt:
         packages:
           - *extra_packages
-  - env: DEF_LIB="static" OPTS="-Db_lto=true" EXTRA_PACKAGES=1 CC=gcc-7
-    compiler: gcc
-    addons:
-      apt:
-        sources:
-            - ubuntu-toolchain-r-test
-        packages:
-          - *extra_packages
-          - gcc-7
 
 
 script: ./.ci/${TRAVIS_OS_NAME}-build.sh