[Bug build/27811] New: [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[Bug build/27811] New: [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

            Bug ID: 27811
           Summary: [11 Regression] arm-linux-gnueabihf and
                    aarch64-linux-gnu fail to configure in sim,
                    sim/aarch64
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: build
          Assignee: unassigned at sourceware dot org
          Reporter: doko at debian dot org
  Target Milestone: ---

trunk 20210502 fails to configure on arm-linux-gnueabihf and aarch64-linux-gnu.

[...]
checking whether C11 is supported by default... yes
checking dependency style of aarch64-linux-gnu-gcc... (cached) none
configure: error: no usable dependency style found
configure: error:
/home/ubuntu/gdb/snap/gdb-11.0.50.20210502/sim/aarch64/configure failed for
aarch64
make[2]: *** [Makefile:8334: configure-sim] Error 1

gdb 10.2 succeeds to configure in the same build environment.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Matthias Klose <doko at debian dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |arm-linux-gnueabihf
                   |                            |aarch64-linux-gnu
                 CC|                            |ktkachov at sourceware dot org

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Matthias Klose <doko at debian dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tromey at sourceware dot org

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #1 from Matthias Klose <doko at debian dot org> ---
this looks like some fall-out from the sim/*/Makefile reorg.

on powerpc64le-linux-gnu I also see much later in the build:

Making info in readline
make[4]: Entering directory '/<<PKGBUILDDIR>>/build/default/readline/readline'
( cd doc && /usr/bin/make -w info )
make[5]: Entering directory
'/<<PKGBUILDDIR>>/build/default/readline/readline/doc'
make[5]: Nothing to be done for 'info'.
make[5]: Leaving directory
'/<<PKGBUILDDIR>>/build/default/readline/readline/doc'
make[4]: Leaving directory '/<<PKGBUILDDIR>>/build/default/readline/readline'
make[4]: Entering directory '/<<PKGBUILDDIR>>/build/default/readline'
make[4]: Nothing to be done for 'info-am'.
make[4]: Leaving directory '/<<PKGBUILDDIR>>/build/default/readline'
make[3]: Leaving directory '/<<PKGBUILDDIR>>/build/default/readline'
Doing info in sim
make[3]: Entering directory '/<<PKGBUILDDIR>>/build/default/sim'
Making info in ppc
make[4]: Entering directory '/<<PKGBUILDDIR>>/build/default/sim/ppc'
make[4]: *** No rule to make target 'info'.  Stop.
make[4]: Leaving directory '/<<PKGBUILDDIR>>/build/default/sim/ppc'
make[3]: *** [Makefile:928: info-recursive] Error 1
make[3]: Leaving directory '/<<PKGBUILDDIR>>/build/default/sim'
make[2]: *** [Makefile:8422: info-sim] Error 1
make[2]: Leaving directory '/<<PKGBUILDDIR>>/build/default'
make[1]: *** [Makefile:984: do-info] Error 2
make[1]: Leaving directory '/<<PKGBUILDDIR>>/build/default'
make: *** [debian/rules:240: stamps/build-default] Error 2

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[schwab@linux-m68k.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #2 from Andreas Schwab <schwab@linux-m68k.org> ---
> checking dependency style of aarch64-linux-gnu-gcc... (cached) none

Where is the cached value set?

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #3 from Matthias Klose <doko at debian dot org> ---
looking at the log, I see the message
  checking dependency style of aarch64-linux-gnu-gcc... none
while configuring the sim subdir (and also before when configuring the bfd and
opcodes subdirs.

[...]
checking whether /usr/bin/make supports nested variables... yes
checking dependency style of aarch64-linux-gnu-gcc... none
configure: updating cache ./config.cache
checking that generated files are newer than configure... done
configure: creating ./config.status
config.status: creating Makefile
config.status: executing depfiles commands
=== configuring in aarch64
(/home/ubuntu/gdb/snap/gdb-11.0.50.20210502/build/default/sim/aarch64)
[...]


Forgot to mention, these are native builds.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #4 from Matthias Klose <doko at debian dot org> ---
so sim/config.cache has gcc3 in 10, but none in 11, looks like the value in
other caches is none in both versions.

gdb-10.2 configury:

$ grep 'am_cv_CC_dependencies_compiler_type' $(find build/default -name
config.cache)
build/default/sim/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=gcc3}
build/default/gdbsupport/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}
build/default/bfd/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}
build/default/opcodes/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}
build/default/gnulib/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}

gdb-11 configury:

$ grep 'am_cv_CC_dependencies_compiler_type' $(find build/default -name
config.cache)
build/default/sim/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}
build/default/bfd/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}
build/default/opcodes/config.cache:am_cv_CC_dependencies_compiler_type=${am_cv_CC_dependencies_compiler_type=none}

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #5 from Matthias Klose <doko at debian dot org> ---
also fails on riscv64-linux-gnu,  seems to succeed only on architectures which
don't have an arch specific subdir in sim.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[tromey at sourceware dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #6 from Tom Tromey <tromey at sourceware dot org> ---
What options did you pass to configure?

Also, it may be worth looking in the config.log to see
why the check is getting the wrong answer.
Any gcc should support the "gcc3" dependency style.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #7 from Matthias Klose <doko at debian dot org> ---
for aarch64:

configured with: --build=aarch64-linux-gnu
         --host=aarch64-linux-gnu
         --prefix=/usr
         --libexecdir=${prefix}/lib/gdb
         --disable-werror
         --disable-maintainer-mode
         --disable-dependency-tracking
         --disable-silent-rules
         --disable-gdbtk
         --disable-shared
         --without-system-readline
         --srcdir=/<<PKGBUILDDIR>>
         --with-expat
         --with-system-zlib
         --without-guile
         --without-babeltrace
         --with-debuginfod
         --with-babeltrace
         --with-system-gdbinit=/etc/gdb/gdbinit
         --with-system-gdbinit-dir=/etc/gdb/gdbinit.d
         --enable-tui
         --with-lzma
         --with-python=python3
         --with-xxhash
         --with-mpfr

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[simark at simark dot ca]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Simon Marchi <simark at simark dot ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |simark at simark dot ca

--- Comment #8 from Simon Marchi <simark at simark dot ca> ---
(In reply to Matthias Klose from comment #7)
> for aarch64:
> 
> configured with: --build=aarch64-linux-gnu
> 	 --host=aarch64-linux-gnu
> 	 --prefix=/usr
> 	 --libexecdir=${prefix}/lib/gdb
> 	 --disable-werror
> 	 --disable-maintainer-mode
> 	 --disable-dependency-tracking
> 	 --disable-silent-rules
> 	 --disable-gdbtk
> 	 --disable-shared
> 	 --without-system-readline
> 	 --srcdir=/<<PKGBUILDDIR>>
> 	 --with-expat
> 	 --with-system-zlib
> 	 --without-guile
> 	 --without-babeltrace
> 	 --with-debuginfod
> 	 --with-babeltrace
> 	 --with-system-gdbinit=/etc/gdb/gdbinit
> 	 --with-system-gdbinit-dir=/etc/gdb/gdbinit.d
> 	 --enable-tui
> 	 --with-lzma
> 	 --with-python=python3
> 	 --with-xxhash
> 	 --with-mpfr

Side-note: it's interesting that configure complains about not finding a
suitable dependency tracking method, even if the build is with
--disable-dependency-tracking.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[wilson at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Jim Wilson <wilson at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |wilson at gcc dot gnu.org

--- Comment #9 from Jim Wilson <wilson at gcc dot gnu.org> ---
I can reproduce if I use --disable-dependency-tracking.  The problem appears to
be the sim aclocal.m4 files which include $top/config/depstand.m4 which does
not respect --disable-dependency-tracking and forces an error if no dependency
tracking is found.  So what happens is that the sim dir is configured first,
respects --disable-dependency-tracking, and sets config.cache entry to none,
then we configure any sim/$target dir, which ignores
--disable-dependency-tracking, sees the config.cache entry, and gives an error.

Though I haven't figured out why sim/configure respects
--disable-dependency-tracking, as it appears to include config/depstand.m4
also.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

--- Comment #10 from Matthias Klose <doko at debian dot org> ---
configuring without --disable-dependency-tracking doesn't make a difference for
me. Rechecked with the 20210629 trunk.

Doing info in sim
make[3]: Entering directory '/<<PKGBUILDDIR>>/build/default/sim'
Making info in ppc
make[4]: Entering directory '/<<PKGBUILDDIR>>/build/default/sim/ppc'
make[4]: *** No rule to make target 'info'.  Stop.
make[4]: Leaving directory '/<<PKGBUILDDIR>>/build/default/sim/ppc'
make[3]: *** [Makefile:1256: info-recursive] Error 1
make[3]: Leaving directory '/<<PKGBUILDDIR>>/build/default/sim'
make[2]: *** [Makefile:8088: info-sim] Error 1
make[2]: Leaving directory '/<<PKGBUILDDIR>>/build/default'
make[1]: *** [Makefile:984: do-info] Error 2

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug build/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Matthias Klose <doko at debian dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|arm-linux-gnueabihf         |powerpc64le-linux-gnu
                   |aarch64-linux-gnu           |

--- Comment #11 from Matthias Klose <doko at debian dot org> ---
looks like the configury works with --disable-dependency-tracking, on
arm-linux-gnueabihf, aarch64-linux-gnu and riscv64-linux-gnu.  However it still
fails with that error on powerpc64le-linux-gnu

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug sim/27811] [11 Regression] arm-linux-gnueabihf and aarch64-linux-gnu fail to configure in sim, sim/aarch64

[vapier at gentoo dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=27811

Mike Frysinger <vapier at gentoo dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|build                       |sim
                 CC|                            |vapier at gentoo dot org
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #12 from Mike Frysinger <vapier at gentoo dot org> ---
> make[4]: *** No rule to make target 'info'.  Stop.

this has been fixed in master:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=acbf56d780b77c30276431474b41defc868e2e5c

i don't normally pay attention to backports to gdb branches since they get cut
at a regular cadence.

-- 
You are receiving this mail because:
You are on the CC list for the bug.