Do not include <machine/frame.h> in libc sparc64 fpu routines, they don't
need any of its contents.
ok claudio@ kettenis@

sys/param.h is not needed in these files

Fix conversions to long double on sparc64

Bug exposed by erratic sqlite3 behavior used in ports/devel/proj,
as pointed out by landry@. Richard Hipps (SQLite) pointed at the
culprit (_Qp_div), many thanks.

Adapted from FreeBSD revision 146673 by Stephen Paskaluk and
stefanf@FreeBSD. FreeBSD commit message:

"""
Fix long (and long long) to long double, unsigned to long double and
unsigned long (and unsigned long long) to long double conversions.
- Add a parameter that specifies the position of the sign bit to the _QP_TTOQ
macro, previously it always looked at bit 31. Pass a negative number to
disable sign inspection for unsigned types. This fixes _Qp_xtoq(),
_Qp_uitoq() and _Qp_uxtoq().
- In the functions __fpu_itof() and __fpu_xtof(), look at the sign
bit to decide whether we're doing a conversion from an unsigned type. If so, don't
negate the mantissa if the integer exceeds the biggest signed number.
"""

ok deraadt@

Remove FBSDID.

ok deraadt@

Hide __fpu_* and make internal _Qp_* calls go direct.

ok kettenis@

Remove excessive sys/cdefs.h inclusion
ok guenther millert kettenis

please gcc

Fix unsigned {int, long, long long} to long double conversions for
values with the high bit set. ok kettenis@ jason@

From the how many ways can you screw up replacing one line of code
department: cast 'i' to u_int64_t so the sign comparison actually
makes sense. Any other bugs in here just ship in the release, I'm done.

use absolute value like rev 1.1 did; pointed out by tom and ok deraadt.

gcc3 doesn't align things the same as gcc2. Casting upwards is dangerous
to a program's health, avoid it. ok deraadt

quad float emulation stuff. Mostly based on freebsd (which is based on
NetBSD's kernel emulation stuff). Not enabled yet.

sys/param.h is not needed in these files

Fix conversions to long double on sparc64

Bug exposed by erratic sqlite3 behavior used in ports/devel/proj,
as pointed out by landry@. Richard Hipps (SQLite) pointed at the
culprit (_Qp_div), many thanks.

Adapted from FreeBSD revision 146673 by Stephen Paskaluk and
stefanf@FreeBSD. FreeBSD commit message:

"""
Fix long (and long long) to long double, unsigned to long double and
unsigned long (and unsigned long long) to long double conversions.
- Add a parameter that specifies the position of the sign bit to the _QP_TTOQ
macro, previously it always looked at bit 31. Pass a negative number to
disable sign inspection for unsigned types. This fixes _Qp_xtoq(),
_Qp_uitoq() and _Qp_uxtoq().
- In the functions __fpu_itof() and __fpu_xtof(), look at the sign
bit to decide whether we're doing a conversion from an unsigned type. If so, don't
negate the mantissa if the integer exceeds the biggest signed number.
"""

ok deraadt@

Remove FBSDID.

ok deraadt@

Hide __fpu_* and make internal _Qp_* calls go direct.

ok kettenis@

Remove excessive sys/cdefs.h inclusion
ok guenther millert kettenis

please gcc

Fix unsigned {int, long, long long} to long double conversions for
values with the high bit set. ok kettenis@ jason@

From the how many ways can you screw up replacing one line of code
department: cast 'i' to u_int64_t so the sign comparison actually
makes sense. Any other bugs in here just ship in the release, I'm done.

use absolute value like rev 1.1 did; pointed out by tom and ok deraadt.

gcc3 doesn't align things the same as gcc2. Casting upwards is dangerous
to a program's health, avoid it. ok deraadt

quad float emulation stuff. Mostly based on freebsd (which is based on
NetBSD's kernel emulation stuff). Not enabled yet.

Fix conversions to long double on sparc64

Bug exposed by erratic sqlite3 behavior used in ports/devel/proj,
as pointed out by landry@. Richard Hipps (SQLite) pointed at the
culprit (_Qp_div), many thanks.

Adapted from FreeBSD revision 146673 by Stephen Paskaluk and
stefanf@FreeBSD. FreeBSD commit message:

"""
Fix long (and long long) to long double, unsigned to long double and
unsigned long (and unsigned long long) to long double conversions.
- Add a parameter that specifies the position of the sign bit to the _QP_TTOQ
macro, previously it always looked at bit 31. Pass a negative number to
disable sign inspection for unsigned types. This fixes _Qp_xtoq(),
_Qp_uitoq() and _Qp_uxtoq().
- In the functions __fpu_itof() and __fpu_xtof(), look at the sign
bit to decide whether we're doing a conversion from an unsigned type. If so, don't
negate the mantissa if the integer exceeds the biggest signed number.
"""

ok deraadt@

Remove FBSDID.

ok deraadt@

Hide __fpu_* and make internal _Qp_* calls go direct.

ok kettenis@

Remove excessive sys/cdefs.h inclusion
ok guenther millert kettenis

please gcc

Fix unsigned {int, long, long long} to long double conversions for
values with the high bit set. ok kettenis@ jason@

From the how many ways can you screw up replacing one line of code
department: cast 'i' to u_int64_t so the sign comparison actually
makes sense. Any other bugs in here just ship in the release, I'm done.

use absolute value like rev 1.1 did; pointed out by tom and ok deraadt.

gcc3 doesn't align things the same as gcc2. Casting upwards is dangerous
to a program's health, avoid it. ok deraadt

quad float emulation stuff. Mostly based on freebsd (which is based on
NetBSD's kernel emulation stuff). Not enabled yet.

Remove FBSDID.

ok deraadt@

Hide __fpu_* and make internal _Qp_* calls go direct.

ok kettenis@

Remove excessive sys/cdefs.h inclusion
ok guenther millert kettenis

please gcc

Fix unsigned {int, long, long long} to long double conversions for
values with the high bit set. ok kettenis@ jason@

From the how many ways can you screw up replacing one line of code
department: cast 'i' to u_int64_t so the sign comparison actually
makes sense. Any other bugs in here just ship in the release, I'm done.

use absolute value like rev 1.1 did; pointed out by tom and ok deraadt.

gcc3 doesn't align things the same as gcc2. Casting upwards is dangerous
to a program's health, avoid it. ok deraadt

quad float emulation stuff. Mostly based on freebsd (which is based on
NetBSD's kernel emulation stuff). Not enabled yet.

Hide __fpu_* and make internal _Qp_* calls go direct.

ok kettenis@

Remove excessive sys/cdefs.h inclusion
ok guenther millert kettenis

please gcc

Fix unsigned {int, long, long long} to long double conversions for
values with the high bit set. ok kettenis@ jason@

From the how many ways can you screw up replacing one line of code
department: cast 'i' to u_int64_t so the sign comparison actually
makes sense. Any other bugs in here just ship in the release, I'm done.

use absolute value like rev 1.1 did; pointed out by tom and ok deraadt.

gcc3 doesn't align things the same as gcc2. Casting upwards is dangerous
to a program's health, avoid it. ok deraadt

quad float emulation stuff. Mostly based on freebsd (which is based on
NetBSD's kernel emulation stuff). Not enabled yet.