mknative-gcc for most targets and enabling isl.

mknative-gcc for GCC 12.3.0 on arm64*, arm* besides armv4, ppc, alpha,
sh, sparc*, mips*, riscv*, vax, hppa, ia64, and m68*.

earmv4 didn't start properly, and powerpc64 failed to build, but all
the rest seem to be OK now.

mknative-gcc for GCC 10.5.0.

branches: 1.38.2;
regen everything for GCC 10.4.0.

mknative-gcc for GCC 10 and almost all platforms. one of the
myriad of arm platforms, and the powerpc* platforms are not
yet done.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for GCC 12.3.0 on arm64*, arm* besides armv4, ppc, alpha,
sh, sparc*, mips*, riscv*, vax, hppa, ia64, and m68*.

earmv4 didn't start properly, and powerpc64 failed to build, but all
the rest seem to be OK now.

mknative-gcc for GCC 10.5.0.

branches: 1.38.2;
regen everything for GCC 10.4.0.

mknative-gcc for GCC 10 and almost all platforms. one of the
myriad of arm platforms, and the powerpc* platforms are not
yet done.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for GCC 10.5.0.

regen everything for GCC 10.4.0.

mknative-gcc for GCC 10 and almost all platforms. one of the
myriad of arm platforms, and the powerpc* platforms are not
yet done.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

regen everything for GCC 10.4.0.

mknative-gcc for GCC 10 and almost all platforms. one of the
myriad of arm platforms, and the powerpc* platforms are not
yet done.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for GCC 10 and almost all platforms. one of the
myriad of arm platforms, and the powerpc* platforms are not
yet done.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for gcc 9.3.0 and arm64, alpha, hppa, ia64, m68*, mipseb,
powerpc, riscv*, sh3*, sparc*, vax.

move data from README.gcc8 to README.gcc9.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

branches: 1.30.2;
mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for all platforms for GCC 8.4.0.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

Enable EOWNERDEAD & ENOTRECOVERABLE in libstdc++

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

regen for mknative and all platforms. there are a few things
fixed in this, mostly related to tgmath.h or frexp, and some
generated vs hand-edited white space diffs.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

Switch the iconv(3) prototype to the POSIX conformant variation

Remove const from the 2nd argument.

const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).

Keep the const form for the internal purposes inside citrus and rump.

Address the build breakage fallout in the same change.

There are no ABI changes.

Change accepted by core@.

mknative-gcc for GCC 8.3 and all ports.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for GCC 8.3 and all ports.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2; 1.25.4;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it

regenerate mknative-gcc files for most platforms, post skrll@ fix
and joerg@'s enable tls change.

missing: m68010 (coming shortly) and m68k (breaks the build)

XXX: didn't test i re-merged the powerpc by-hand changes properly

branches: 1.4.2;
re-generate various files after dl_iterate_phdr has been enabled.

regenerate

various (re-)generations. most of these sort of entirely work now.

(re-)generate mknative-gcc files for many platforms.
(re-)generate the gmp files for many platforms.

mknative-gcc for most platforms again after recent tools/gcc/Makefile change.

regen mknative with symver no longer disabled.

mknative-gcc for alpha, hpaa, ia64, mips*, sh3*, and vax.
re- for powerpc, sparc, i386, and amd64.

regenerate everything for GCC 6.5.

sh3eb and sh3el had a problem wtih libstdc++ i had to work around for now.

redo mknative-gcc for all ports. main changes include:
- "#define HAVE_CC_TLS 1" for most/all ports, thanks maya@
- "#define _GLIBCXX_HAVE_LDEXPL 1" and "#define _GLIBCXX_HAVE_TGMATH_H 1"
for many ports
- arm64 and amd64 had a broken c++config.h that disabled many things
- configargs.h has more normalisation
- ppc64 has a few things fixed, must have missed several mknative rounds

branches: 1.25.2;
more mknative-gcc 6.4 for m68k, mips*, powerpc, sh*, sparc* and amd64.

rebuild mknative GCC 6.4 for these targets:
arm armeb earmeb earmhf earmhfeb earmv4eb
earmv6 earmv6eb earmv6hf earmv6hfeb
earmv7 earmv7eb earmv7hf earmv7hfeb
hppa ia64 m68000 m68k
mips64eb mips64el mipseb mipsel
powerpc64 sh3eb sh3el vax

merge mknative files from gcc.old back into gcc.
they're back to being the same again now.

regen for gcc 5.5.

Implement __cxa_thread_atexit and __cxa_thread_atexit_impl. This
functions are used for destructors of thread_local objects.

If a pending destructor exists, prevent unloading of shared objects.
Introduce __dl_cxa_refcount interface for this purpose. When the last
reference is gone and the object has been dlclose'd before, the
unloading is finalized.

Ideally, __cxa_thread_atexit_impl wouldn't exist, but libstdc++ insists
on providing __cxa_thread_atexit as direct wrapper without further
patching.

branches: 1.20.6;
regen mknative for all ports and GCC 5.4 and -current. many ports
were using 5.3 data, and none of them had <stdalign.h>.

regen for alpha and sh now that they have fenv.h

branches: 1.18.2;
mknative-gcc for all but sparc64, vax and sun2 and GCC 5.4.

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

regen mknative files for GCC 4.8.5.

rerun mknative for GCC 4.8.4 and all platforms.

m68000 needs -fPIC removed (again), and hppa c++config.h turned off
several things present in -lm, fixed by hand.

regen mknative for gcc 4.8-2014-10-09.

regen gcc mknative files for new complex and more.

branches: 1.12.2;
re-run mknative-gcc after threading fixes by skrll and myself.

rerun mknative-gcc for GCC 4.8.3 release.

XXX: still not 100% right, but we're getting close.

branches: 1.10.2;
mknative-gcc 4.8 for most platforms. missing: ppc64, m68010, ia64, coldfire.

merge GCC 4.8 build.

regen mknative files for GCC 4.5.4.

re-run mknative.

- i386 tested
- m68k gomp tls reverted by hand
- powerpc softfloat added (un-reverted) by hand

branches: 1.6.2;
rerun mknative-gcc:
- turn off <fenv.h> on platforms that don't have it