mbox series

[v2,0/4] More fixes for building tests with clang

Message ID 20250110181508.350378-1-adhemerval.zanella@linaro.org (mailing list archive)
Headers
Series More fixes for building tests with clang |

Message

Adhemerval Zanella Netto Jan. 10, 2025, 6:12 p.m. UTC 
  On aarch64 with clang 18.1.3 (from ubuntu 22i), there still some tests
that fails to build.  With this patchset applied, building and running
the tests with clang shows:

FAIL: conform/ISO11/tgmath.h/conform
FAIL: conform/ISO99/tgmath.h/conform
FAIL: conform/POSIX2008/tgmath.h/conform
FAIL: conform/XOPEN2K/tgmath.h/conform
FAIL: conform/XOPEN2K8/tgmath.h/conform
FAIL: elf/ifuncmain1static
FAIL: elf/ifuncmain4
FAIL: elf/ifuncmain4static
FAIL: elf/ifuncmain6pie
FAIL: elf/ifuncmain7
FAIL: elf/ifuncmain7pic
FAIL: elf/ifuncmain7picstatic
FAIL: elf/ifuncmain7pie
FAIL: elf/ifuncmain7static
FAIL: gmon/tst-gmon-gprof
FAIL: gmon/tst-gmon-pie-gprof
FAIL: gmon/tst-gmon-static-gprof
FAIL: gmon/tst-gmon-static-pie-gprof
FAIL: gmon/tst-mcount-overflow-check
FAIL: math/test-float128-isgreater
FAIL: math/test-float128-isgreaterequal
FAIL: math/test-float128-isless
FAIL: math/test-float128-islessequal
FAIL: math/test-float64x-isgreater
FAIL: math/test-float64x-isgreaterequal
FAIL: math/test-float64x-isless
FAIL: math/test-float64x-islessequal
FAIL: math/test-ldouble-isgreater
FAIL: math/test-ldouble-isgreaterequal
FAIL: math/test-ldouble-isless
FAIL: math/test-ldouble-islessequal
FAIL: math/test-tgmath
FAIL: math/test-tgmath2
FAIL: stdio-common/tst-vfprintf-user-type

Adhemerval Zanella (4):
  aarch64: Use 64-bit variable to access the special registers
  gmon: Disable tst-profile-static with clang
  elf: Suppress unused function clang for __ifunc_resolver
  stdio-common: Suppress Clang warnings on scanf13.c with fortify enable

 elf/ifuncmain9.c                              |  4 +++
 elf/tst-ifunc-fault-lazy.c                    |  4 +++
 gmon/Makefile                                 |  3 +-
 stdio-common/scanf13.c                        |  5 +++
 sysdeps/aarch64/fpu/fpu_control.h             | 36 +++++++++++++------
 sysdeps/aarch64/fpu/fraiseexcpt.c             |  3 +-
 sysdeps/aarch64/sfp-machine.h                 |  2 +-
 .../unix/sysv/linux/aarch64/cpu-features.c    |  2 +-
 sysdeps/unix/sysv/linux/aarch64/sysconf.c     |  2 +-
 9 files changed, 45 insertions(+), 16 deletions(-)

Comments

Adhemerval Zanella Netto Jan. 10, 2025, 6:54 p.m. UTC | #1 
Hi Andreas,

Are these patches ok for master?

On 10/01/25 15:12, Adhemerval Zanella wrote:
> On aarch64 with clang 18.1.3 (from ubuntu 22i), there still some tests
> that fails to build.  With this patchset applied, building and running
> the tests with clang shows:
> 
> FAIL: conform/ISO11/tgmath.h/conform
> FAIL: conform/ISO99/tgmath.h/conform
> FAIL: conform/POSIX2008/tgmath.h/conform
> FAIL: conform/XOPEN2K/tgmath.h/conform
> FAIL: conform/XOPEN2K8/tgmath.h/conform
> FAIL: elf/ifuncmain1static
> FAIL: elf/ifuncmain4
> FAIL: elf/ifuncmain4static
> FAIL: elf/ifuncmain6pie
> FAIL: elf/ifuncmain7
> FAIL: elf/ifuncmain7pic
> FAIL: elf/ifuncmain7picstatic
> FAIL: elf/ifuncmain7pie
> FAIL: elf/ifuncmain7static
> FAIL: gmon/tst-gmon-gprof
> FAIL: gmon/tst-gmon-pie-gprof
> FAIL: gmon/tst-gmon-static-gprof
> FAIL: gmon/tst-gmon-static-pie-gprof
> FAIL: gmon/tst-mcount-overflow-check
> FAIL: math/test-float128-isgreater
> FAIL: math/test-float128-isgreaterequal
> FAIL: math/test-float128-isless
> FAIL: math/test-float128-islessequal
> FAIL: math/test-float64x-isgreater
> FAIL: math/test-float64x-isgreaterequal
> FAIL: math/test-float64x-isless
> FAIL: math/test-float64x-islessequal
> FAIL: math/test-ldouble-isgreater
> FAIL: math/test-ldouble-isgreaterequal
> FAIL: math/test-ldouble-isless
> FAIL: math/test-ldouble-islessequal
> FAIL: math/test-tgmath
> FAIL: math/test-tgmath2
> FAIL: stdio-common/tst-vfprintf-user-type
> 
> Adhemerval Zanella (4):
>   aarch64: Use 64-bit variable to access the special registers
>   gmon: Disable tst-profile-static with clang
>   elf: Suppress unused function clang for __ifunc_resolver
>   stdio-common: Suppress Clang warnings on scanf13.c with fortify enable
> 
>  elf/ifuncmain9.c                              |  4 +++
>  elf/tst-ifunc-fault-lazy.c                    |  4 +++
>  gmon/Makefile                                 |  3 +-
>  stdio-common/scanf13.c                        |  5 +++
>  sysdeps/aarch64/fpu/fpu_control.h             | 36 +++++++++++++------
>  sysdeps/aarch64/fpu/fraiseexcpt.c             |  3 +-
>  sysdeps/aarch64/sfp-machine.h                 |  2 +-
>  .../unix/sysv/linux/aarch64/cpu-features.c    |  2 +-
>  sysdeps/unix/sysv/linux/aarch64/sysconf.c     |  2 +-
>  9 files changed, 45 insertions(+), 16 deletions(-)
>
Andreas K. Huettel Jan. 11, 2025, 1:52 p.m. UTC | #2 
Am Freitag, 10. Januar 2025, 19:54:56 Mitteleuropäische Normalzeit schrieb Adhemerval Zanella Netto:
> Hi Andreas,
> 
> Are these patches ok for master?

Yes as fixup, with the two comments (HJ's on patch 1 re long int, mine on patch 4)

> 
> On 10/01/25 15:12, Adhemerval Zanella wrote:
> > On aarch64 with clang 18.1.3 (from ubuntu 22i), there still some tests
> > that fails to build.  With this patchset applied, building and running
> > the tests with clang shows:
> > 
> > FAIL: conform/ISO11/tgmath.h/conform
> > FAIL: conform/ISO99/tgmath.h/conform
> > FAIL: conform/POSIX2008/tgmath.h/conform
> > FAIL: conform/XOPEN2K/tgmath.h/conform
> > FAIL: conform/XOPEN2K8/tgmath.h/conform
> > FAIL: elf/ifuncmain1static
> > FAIL: elf/ifuncmain4
> > FAIL: elf/ifuncmain4static
> > FAIL: elf/ifuncmain6pie
> > FAIL: elf/ifuncmain7
> > FAIL: elf/ifuncmain7pic
> > FAIL: elf/ifuncmain7picstatic
> > FAIL: elf/ifuncmain7pie
> > FAIL: elf/ifuncmain7static
> > FAIL: gmon/tst-gmon-gprof
> > FAIL: gmon/tst-gmon-pie-gprof
> > FAIL: gmon/tst-gmon-static-gprof
> > FAIL: gmon/tst-gmon-static-pie-gprof
> > FAIL: gmon/tst-mcount-overflow-check
> > FAIL: math/test-float128-isgreater
> > FAIL: math/test-float128-isgreaterequal
> > FAIL: math/test-float128-isless
> > FAIL: math/test-float128-islessequal
> > FAIL: math/test-float64x-isgreater
> > FAIL: math/test-float64x-isgreaterequal
> > FAIL: math/test-float64x-isless
> > FAIL: math/test-float64x-islessequal
> > FAIL: math/test-ldouble-isgreater
> > FAIL: math/test-ldouble-isgreaterequal
> > FAIL: math/test-ldouble-isless
> > FAIL: math/test-ldouble-islessequal
> > FAIL: math/test-tgmath
> > FAIL: math/test-tgmath2
> > FAIL: stdio-common/tst-vfprintf-user-type
> > 
> > Adhemerval Zanella (4):
> >   aarch64: Use 64-bit variable to access the special registers
> >   gmon: Disable tst-profile-static with clang
> >   elf: Suppress unused function clang for __ifunc_resolver
> >   stdio-common: Suppress Clang warnings on scanf13.c with fortify enable
> > 
> >  elf/ifuncmain9.c                              |  4 +++
> >  elf/tst-ifunc-fault-lazy.c                    |  4 +++
> >  gmon/Makefile                                 |  3 +-
> >  stdio-common/scanf13.c                        |  5 +++
> >  sysdeps/aarch64/fpu/fpu_control.h             | 36 +++++++++++++------
> >  sysdeps/aarch64/fpu/fraiseexcpt.c             |  3 +-
> >  sysdeps/aarch64/sfp-machine.h                 |  2 +-
> >  .../unix/sysv/linux/aarch64/cpu-features.c    |  2 +-
> >  sysdeps/unix/sysv/linux/aarch64/sysconf.c     |  2 +-
> >  9 files changed, 45 insertions(+), 16 deletions(-)
> > 
> 
>