[PATCHv2] gdb/fortran: Use floatformats_ia64_quad for fortran 16-byte floats
Commit Message
This new version moves the use of floatformats_ia64_quad out of
f-lang.c, which I think is an improvement. With this done I plan to
push this patch shortly unless anyone complains.
Thanks,
Andrew
---
PR gdb/18644 is caused by GDB using the wrong floating point format
for gfortran's 16-byte floating point type, including when the 16-byte
float is used as the component of a 32-byte complex type.
This commit addresses the issue in two places, first in i386-tdep.c,
there is already some code to force the use of floatformats_ia64_quad
for specific named types, this is extended to include the type names
that gfortran uses for its 16-byte floats.
Second, the builtin 16-byte float type (in f-lang.c) is changed so it
no longer uses gdbarch_long_double_format. On i386 this type is not
16-bytes, but is smaller, this is not what gfortran is expecting.
Instead we now use gdbarch_floatformat_for_type and ask for a
16-byte (128 bit) type using the common gfortran type name. This is
then spotted in i386-tdep.c (thanks to the first change above) and we
again get floatformats_ia64_quad returned.
This patch was tested on X86-64/GNU-Linux using '--target_board=unix'
and '--target_board=unix/-m32', and resolves all of the known failures
associated with PR gdb/18644. I've also added the test case from the
original bug report.
gdb/ChangeLog:
PR gdb/18644:
* f-lang.c (build_fortran_types): Use floatformats_ia64_quad for
16-byte floats.
* i386-tdep.c (i386_floatformat_for_type): Use
floatformats_ia64_quad for the 16-byte floating point component
within a fortran 32-byte complex number.
gdb/testsuite/ChangeLog:
PR gdb/18644
* gdb.fortran/complex.exp: Remove setup_kfail calls.
* gdb.fortran/printing-types.exp: Add new test.
* gdb.fortran/printing-types.f90: Add 16-byte real variable for
testing.
* gdb.fortran/type-kinds.exp (test_cast_1_to_type_kind): Remove
setup_kfail call.
---
gdb/ChangeLog | 9 +++++++++
gdb/f-lang.c | 4 ++--
gdb/i386-tdep.c | 4 +++-
gdb/testsuite/ChangeLog | 10 ++++++++++
gdb/testsuite/gdb.fortran/complex.exp | 2 --
gdb/testsuite/gdb.fortran/printing-types.exp | 1 +
gdb/testsuite/gdb.fortran/printing-types.f90 | 2 ++
gdb/testsuite/gdb.fortran/type-kinds.exp | 6 ------
8 files changed, 27 insertions(+), 11 deletions(-)
Comments
* Andrew Burgess <andrew.burgess@embecosm.com> [2019-05-16 17:00:58 +0100]:
> This new version moves the use of floatformats_ia64_quad out of
> f-lang.c, which I think is an improvement. With this done I plan to
> push this patch shortly unless anyone complains.
This has now been pushed.
Thanks,
Andrew
>
> Thanks,
> Andrew
>
>
> ---
>
> PR gdb/18644 is caused by GDB using the wrong floating point format
> for gfortran's 16-byte floating point type, including when the 16-byte
> float is used as the component of a 32-byte complex type.
>
> This commit addresses the issue in two places, first in i386-tdep.c,
> there is already some code to force the use of floatformats_ia64_quad
> for specific named types, this is extended to include the type names
> that gfortran uses for its 16-byte floats.
>
> Second, the builtin 16-byte float type (in f-lang.c) is changed so it
> no longer uses gdbarch_long_double_format. On i386 this type is not
> 16-bytes, but is smaller, this is not what gfortran is expecting.
> Instead we now use gdbarch_floatformat_for_type and ask for a
> 16-byte (128 bit) type using the common gfortran type name. This is
> then spotted in i386-tdep.c (thanks to the first change above) and we
> again get floatformats_ia64_quad returned.
>
> This patch was tested on X86-64/GNU-Linux using '--target_board=unix'
> and '--target_board=unix/-m32', and resolves all of the known failures
> associated with PR gdb/18644. I've also added the test case from the
> original bug report.
>
> gdb/ChangeLog:
>
> PR gdb/18644:
> * f-lang.c (build_fortran_types): Use floatformats_ia64_quad for
> 16-byte floats.
> * i386-tdep.c (i386_floatformat_for_type): Use
> floatformats_ia64_quad for the 16-byte floating point component
> within a fortran 32-byte complex number.
>
> gdb/testsuite/ChangeLog:
>
> PR gdb/18644
> * gdb.fortran/complex.exp: Remove setup_kfail calls.
> * gdb.fortran/printing-types.exp: Add new test.
> * gdb.fortran/printing-types.f90: Add 16-byte real variable for
> testing.
> * gdb.fortran/type-kinds.exp (test_cast_1_to_type_kind): Remove
> setup_kfail call.
> ---
> gdb/ChangeLog | 9 +++++++++
> gdb/f-lang.c | 4 ++--
> gdb/i386-tdep.c | 4 +++-
> gdb/testsuite/ChangeLog | 10 ++++++++++
> gdb/testsuite/gdb.fortran/complex.exp | 2 --
> gdb/testsuite/gdb.fortran/printing-types.exp | 1 +
> gdb/testsuite/gdb.fortran/printing-types.f90 | 2 ++
> gdb/testsuite/gdb.fortran/type-kinds.exp | 6 ------
> 8 files changed, 27 insertions(+), 11 deletions(-)
>
> diff --git a/gdb/f-lang.c b/gdb/f-lang.c
> index 9da6fdb3e1c..5855c68b38c 100644
> --- a/gdb/f-lang.c
> +++ b/gdb/f-lang.c
> @@ -727,9 +727,9 @@ build_fortran_types (struct gdbarch *gdbarch)
> builtin_f_type->builtin_real_s8
> = arch_float_type (gdbarch, gdbarch_double_bit (gdbarch),
> "real*8", gdbarch_double_format (gdbarch));
> + auto fmt = gdbarch_floatformat_for_type (gdbarch, "real(kind=16)", 128);
> builtin_f_type->builtin_real_s16
> - = arch_float_type (gdbarch, gdbarch_long_double_bit (gdbarch),
> - "real*16", gdbarch_long_double_format (gdbarch));
> + = arch_float_type (gdbarch, 128, "real*16", fmt);
>
> builtin_f_type->builtin_complex_s8
> = arch_complex_type (gdbarch, "complex*8",
> diff --git a/gdb/i386-tdep.c b/gdb/i386-tdep.c
> index 54d9dd873b8..66fc6d679e2 100644
> --- a/gdb/i386-tdep.c
> +++ b/gdb/i386-tdep.c
> @@ -8158,7 +8158,9 @@ i386_floatformat_for_type (struct gdbarch *gdbarch,
> if (len == 128 && name)
> if (strcmp (name, "__float128") == 0
> || strcmp (name, "_Float128") == 0
> - || strcmp (name, "complex _Float128") == 0)
> + || strcmp (name, "complex _Float128") == 0
> + || strcmp (name, "complex(kind=16)") == 0
> + || strcmp (name, "real(kind=16)") == 0)
> return floatformats_ia64_quad;
>
> return default_floatformat_for_type (gdbarch, name, len);
> diff --git a/gdb/testsuite/gdb.fortran/complex.exp b/gdb/testsuite/gdb.fortran/complex.exp
> index 136f1c4df79..94ac53afc70 100644
> --- a/gdb/testsuite/gdb.fortran/complex.exp
> +++ b/gdb/testsuite/gdb.fortran/complex.exp
> @@ -33,7 +33,6 @@ gdb_test "print c4" " = \\(1000,-50\\)"
> gdb_test "print c8" " = \\(321,-22\\)"
> gdb_test "print dc" " = \\(321,-22\\)"
>
> -setup_kfail gdb/18644 "*-*-*"
> gdb_test "print c16" " = \\(-874,19\\)"
>
> gdb_test "whatis c" "type = complex\\(kind=4\\)"
> @@ -53,7 +52,6 @@ gdb_test "print \$_creal (dc)" " = 321"
> gdb_test "whatis \$" " = real\\*8"
>
> gdb_test "whatis c16" "type = complex\\(kind=16\\)"
> -setup_kfail gdb/18644 "*-*-*"
> gdb_test "print \$_creal (c16)" " = -874"
> gdb_test "whatis \$" " = real\\*16"
>
> diff --git a/gdb/testsuite/gdb.fortran/printing-types.exp b/gdb/testsuite/gdb.fortran/printing-types.exp
> index 2f6be4ec249..6394e45f4c3 100644
> --- a/gdb/testsuite/gdb.fortran/printing-types.exp
> +++ b/gdb/testsuite/gdb.fortran/printing-types.exp
> @@ -33,3 +33,4 @@ gdb_test "print oneByte" " = 1"
> gdb_test "print twobytes" " = 2"
> gdb_test "print chvalue" " = \'a\'"
> gdb_test "print logvalue" " = \.TRUE\."
> +gdb_test "print rVal" " = 2000"
> diff --git a/gdb/testsuite/gdb.fortran/printing-types.f90 b/gdb/testsuite/gdb.fortran/printing-types.f90
> index b4ff928604f..36b63532c8e 100644
> --- a/gdb/testsuite/gdb.fortran/printing-types.f90
> +++ b/gdb/testsuite/gdb.fortran/printing-types.f90
> @@ -18,10 +18,12 @@ program prog
> integer(2) :: twoBytes
> character :: chValue
> logical(1) :: logValue
> + real(kind=16) :: rVal
>
> oneByte = 1
> twoBytes = 2
> chValue = 'a'
> logValue = .true.
> + rVal = 2000
> write(*,*) s
> end
> diff --git a/gdb/testsuite/gdb.fortran/type-kinds.exp b/gdb/testsuite/gdb.fortran/type-kinds.exp
> index 1ae15b96f1a..9d19a9ceb39 100644
> --- a/gdb/testsuite/gdb.fortran/type-kinds.exp
> +++ b/gdb/testsuite/gdb.fortran/type-kinds.exp
> @@ -27,12 +27,6 @@ if { [skip_fortran_tests] } { continue }
> proc test_cast_1_to_type_kind {base_type type_kind cast_result size_result} {
> set type_string "$base_type (kind=$type_kind)"
> gdb_test "p (($type_string) 1)" " = $cast_result"
> -
> - if {($base_type == "real" || $base_type == "complex")
> - && $type_kind == 16} {
> - setup_kfail gdb/18644 "*-*-*"
> - }
> -
> gdb_test "p sizeof (($type_string) 1)" " = $size_result"
> }
>
> --
> 2.14.5
>
@@ -727,9 +727,9 @@ build_fortran_types (struct gdbarch *gdbarch)
builtin_f_type->builtin_real_s8
= arch_float_type (gdbarch, gdbarch_double_bit (gdbarch),
"real*8", gdbarch_double_format (gdbarch));
+ auto fmt = gdbarch_floatformat_for_type (gdbarch, "real(kind=16)", 128);
builtin_f_type->builtin_real_s16
- = arch_float_type (gdbarch, gdbarch_long_double_bit (gdbarch),
- "real*16", gdbarch_long_double_format (gdbarch));
+ = arch_float_type (gdbarch, 128, "real*16", fmt);
builtin_f_type->builtin_complex_s8
= arch_complex_type (gdbarch, "complex*8",
@@ -8158,7 +8158,9 @@ i386_floatformat_for_type (struct gdbarch *gdbarch,
if (len == 128 && name)
if (strcmp (name, "__float128") == 0
|| strcmp (name, "_Float128") == 0
- || strcmp (name, "complex _Float128") == 0)
+ || strcmp (name, "complex _Float128") == 0
+ || strcmp (name, "complex(kind=16)") == 0
+ || strcmp (name, "real(kind=16)") == 0)
return floatformats_ia64_quad;
return default_floatformat_for_type (gdbarch, name, len);
@@ -33,7 +33,6 @@ gdb_test "print c4" " = \\(1000,-50\\)"
gdb_test "print c8" " = \\(321,-22\\)"
gdb_test "print dc" " = \\(321,-22\\)"
-setup_kfail gdb/18644 "*-*-*"
gdb_test "print c16" " = \\(-874,19\\)"
gdb_test "whatis c" "type = complex\\(kind=4\\)"
@@ -53,7 +52,6 @@ gdb_test "print \$_creal (dc)" " = 321"
gdb_test "whatis \$" " = real\\*8"
gdb_test "whatis c16" "type = complex\\(kind=16\\)"
-setup_kfail gdb/18644 "*-*-*"
gdb_test "print \$_creal (c16)" " = -874"
gdb_test "whatis \$" " = real\\*16"
@@ -33,3 +33,4 @@ gdb_test "print oneByte" " = 1"
gdb_test "print twobytes" " = 2"
gdb_test "print chvalue" " = \'a\'"
gdb_test "print logvalue" " = \.TRUE\."
+gdb_test "print rVal" " = 2000"
@@ -18,10 +18,12 @@ program prog
integer(2) :: twoBytes
character :: chValue
logical(1) :: logValue
+ real(kind=16) :: rVal
oneByte = 1
twoBytes = 2
chValue = 'a'
logValue = .true.
+ rVal = 2000
write(*,*) s
end
@@ -27,12 +27,6 @@ if { [skip_fortran_tests] } { continue }
proc test_cast_1_to_type_kind {base_type type_kind cast_result size_result} {
set type_string "$base_type (kind=$type_kind)"
gdb_test "p (($type_string) 1)" " = $cast_result"
-
- if {($base_type == "real" || $base_type == "complex")
- && $type_kind == 16} {
- setup_kfail gdb/18644 "*-*-*"
- }
-
gdb_test "p sizeof (($type_string) 1)" " = $size_result"
}