octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #65573] Bytecode-interpreter branch won´t get


From: Philip Nienhuis
Subject: [Octave-bug-tracker] [bug #65573] Bytecode-interpreter branch won´t get crossbuilt
Date: Tue, 9 Apr 2024 10:39:30 -0400 (EDT)

URL:
  <https://savannah.gnu.org/bugs/?65573>

                 Summary: Bytecode-interpreter branch won´t get crossbuilt
                   Group: GNU Octave
               Submitter: philipnienhuis
               Submitted: Tue 09 Apr 2024 04:39:29 PM CEST
                Category: Configuration and Build System
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Build Failure
                  Status: None
             Assigned to: None
         Originator Name: Philip Nienhuis
        Originator Email: 
             Open/Closed: Open
                 Release: bytecode-evaluator
         Discussion Lock: Any
        Operating System: Microsoft Windows
           Fixed Release: 10.1.0 (current default)
         Planned Release: None


    _______________________________________________________

Follow-up Comments:


-------------------------------------------------------
Date: Tue 09 Apr 2024 04:39:29 PM CEST By: Philip Nienhuis <philipnienhuis>
(see also Discourse thread
https://octave.discourse.group/t/do-we-have-buildbots-for-dev-10-0-0-octave/5480)

Since a few days the bytecode-interpreter branch of default Octave cannot get
crossbuilt anymore. AFAICS it *may* have started with this cset:
abdb846bafe8 "Instantiate octave_base_matrix and octave_base_scalar template
classes".
That is, merging the cset before (714c6d2e253e "maint: Merge stable to
default.") with the bytecode-interpreter branch (cset af9c0617f9f9 "maint:
Merge default to bytecode-interpreter.") from a day earlier leads to a
successful crossbuild, merging the b.-i. branch with later csets yields
crossbuild failures.
Why it builds fine for Linux but not for Windows is beyond me. Apparently
mxe-octave is a bit more fragile?

I just tried with a clean checkout updated to the bytecode-interpreter branch.
It builds and AFAICS runs fine on Linux, although with a few warnings, a.o.,
these:

  CXX      libinterp/octave-value/liboctave_value_la-ov-float.lo
../dev/libinterp/octave-value/ov-base-mat-inst.cc:34:30: warning: type
attributes ignored after type is already defined [-Wattributes]
   34 | template class OCTINTERP_API octave_base_matrix<boolNDArray>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../dev/libinterp/octave-value/ov-base-mat-inst.cc:44:30: warning: type
attributes ignored after type is already defined [-Wattributes]
   44 | template class OCTINTERP_API octave_base_matrix<ComplexNDArray>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../dev/libinterp/octave-value/ov-base-mat-inst.cc:45:30: warning: type
attributes ignored after type is already defined [-Wattributes]
   45 | template class OCTINTERP_API octave_base_matrix<FloatComplexNDArray>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../dev/libinterp/octave-value/ov-base-mat-inst.cc:46:30: warning: type
attributes ignored after type is already defined [-Wattributes]
   46 | template class OCTINTERP_API octave_base_matrix<FloatNDArray>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../dev/libinterp/octave-value/ov-base-mat-inst.cc:47:30: warning: type
attributes ignored after type is already defined [-Wattributes]
   47 | template class OCTINTERP_API octave_base_matrix<NDArray>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~
../dev/libinterp/octave-value/ov-base-mat-inst.cc:129:30: warning: type
attributes ignored after type is already defined [-Wattributes]
  129 | template class OCTINTERP_API octave_base_matrix<Cell>;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~

but crossbuilding the resulting dist archive leads to a failure.

I asked on Discourse about buildbots for default Octave, but AFAIU there are
no buildbots for the bytecode-interpreter branch. So it seems
bytecode-interpreter build failures go largely unnoticed then, am I right?

I'll cc Petter (creator of the bytecode-interpreter branch).

Compressed buildlog of mxe-octave attached.







    _______________________________________________________
File Attachments:


-------------------------------------------------------
Name: default-octave-BI.zip  Size: 207KiB
<https://file.savannah.gnu.org/file/default-octave-BI.zip?file_id=55929>

    AGPL NOTICE

These attachments are served by Savane. You can download the corresponding
source code of Savane at
https://git.savannah.nongnu.org/cgit/administration/savane.git/snapshot/savane-319f5d616b046e3879e4959f590f30fc8cc4e6d6.tar.gz

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?65573>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]