[Qemu-devel] [PATCH 0/7] POWER9 TCG and softfloat enablements - part16 + s390 float fixes

Bharata B Rao posted 7 patches 164 weeks ago
Failed in applying to current master (apply log)
fpu/softfloat.c                     | 125 +++++++++++++++++++++++++++++++++++-
include/fpu/softfloat.h             |   6 ++
target/ppc/fpu_helper.c             |  44 ++++++-------
target/ppc/helper.h                 |   2 +
target/ppc/translate/vsx-impl.inc.c |   2 +
target/ppc/translate/vsx-ops.inc.c  |   4 +-
target/s390x/fpu_helper.c           |   6 +-
7 files changed, 161 insertions(+), 28 deletions(-)

[Qemu-devel] [PATCH 0/7] POWER9 TCG and softfloat enablements - part16 + s390 float fixes

Posted by Bharata B Rao 164 weeks ago
This series contains 2 new instructions (xscvqpudz, xscvqpuwz) and
round-to-odd variants of 6 instructions (xsaddqpo, xsmulqpo, xsdivqpo,
xscvqpdpo, xssqrtqpo, xssubqpo)
 
This also includes softfloat enhancements that are needed by
the above instructions. These softfloat patches have been posted
separately earlier and reviewed. Including them here as the new
instructions here depend on them.

I also saw that target-s390 also needs float128_to_uint64() and
float128_to_uint32(). Implemented the latter in this series and
made fixes in s390 code to use them. Note that s390 changes have
been compile-tested only. The last two patches ideally needn't
be part of this series, but included them here as s390 fixes
depend on the softfloat changes that are part of this patchset.

Bharata B Rao (7):
  softfloat: Add round-to-odd rounding mode
  softfloat: Add float128_to_uint64_round_to_zero()
  softfloat: Add float128_to_uint32_round_to_zero()
  target-ppc: Implement round to odd variants of quad FP instructions
  target-ppc: Add xscvqpudz and xscvqpuwz instructions
  softfloat: Add float128_to_uint32()
  target-s390: Use float128_to_uint[64/32] where required

 fpu/softfloat.c                     | 125 +++++++++++++++++++++++++++++++++++-
 include/fpu/softfloat.h             |   6 ++
 target/ppc/fpu_helper.c             |  44 ++++++-------
 target/ppc/helper.h                 |   2 +
 target/ppc/translate/vsx-impl.inc.c |   2 +
 target/ppc/translate/vsx-ops.inc.c  |   4 +-
 target/s390x/fpu_helper.c           |   6 +-
 7 files changed, 161 insertions(+), 28 deletions(-)

-- 
2.7.4


Re: [Qemu-devel] [PATCH 0/7] POWER9 TCG and softfloat enablements - part16 + s390 float fixes

Posted by David Gibson 164 weeks ago
On Fri, Feb 10, 2017 at 12:53:04PM +0530, Bharata B Rao wrote:
> This series contains 2 new instructions (xscvqpudz, xscvqpuwz) and
> round-to-odd variants of 6 instructions (xsaddqpo, xsmulqpo, xsdivqpo,
> xscvqpdpo, xssqrtqpo, xssubqpo)
>  
> This also includes softfloat enhancements that are needed by
> the above instructions. These softfloat patches have been posted
> separately earlier and reviewed. Including them here as the new
> instructions here depend on them.
> 
> I also saw that target-s390 also needs float128_to_uint64() and
> float128_to_uint32(). Implemented the latter in this series and
> made fixes in s390 code to use them. Note that s390 changes have
> been compile-tested only. The last two patches ideally needn't
> be part of this series, but included them here as s390 fixes
> depend on the softfloat changes that are part of this patchset.

4 & 5 also applied to my tree.  6 & 7 I'll leave to go through the
s390 tree.

> 
> Bharata B Rao (7):
>   softfloat: Add round-to-odd rounding mode
>   softfloat: Add float128_to_uint64_round_to_zero()
>   softfloat: Add float128_to_uint32_round_to_zero()
>   target-ppc: Implement round to odd variants of quad FP instructions
>   target-ppc: Add xscvqpudz and xscvqpuwz instructions
>   softfloat: Add float128_to_uint32()
>   target-s390: Use float128_to_uint[64/32] where required
> 
>  fpu/softfloat.c                     | 125 +++++++++++++++++++++++++++++++++++-
>  include/fpu/softfloat.h             |   6 ++
>  target/ppc/fpu_helper.c             |  44 ++++++-------
>  target/ppc/helper.h                 |   2 +
>  target/ppc/translate/vsx-impl.inc.c |   2 +
>  target/ppc/translate/vsx-ops.inc.c  |   4 +-
>  target/s390x/fpu_helper.c           |   6 +-
>  7 files changed, 161 insertions(+), 28 deletions(-)
> 

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Re: [Qemu-devel] [PATCH 0/7] POWER9 TCG and softfloat enablements - part16 + s390 float fixes

Posted by David Gibson 164 weeks ago
On Fri, Feb 10, 2017 at 12:53:04PM +0530, Bharata B Rao wrote:
1;4601;0c> This series contains 2 new instructions (xscvqpudz, xscvqpuwz) and
> round-to-odd variants of 6 instructions (xsaddqpo, xsmulqpo, xsdivqpo,
> xscvqpdpo, xssqrtqpo, xssubqpo)
>  
> This also includes softfloat enhancements that are needed by
> the above instructions. These softfloat patches have been posted
> separately earlier and reviewed. Including them here as the new
> instructions here depend on them.
> 
> I also saw that target-s390 also needs float128_to_uint64() and
> float128_to_uint32(). Implemented the latter in this series and
> made fixes in s390 code to use them. Note that s390 changes have
> been compile-tested only. The last two patches ideally needn't
> be part of this series, but included them here as s390 fixes
> depend on the softfloat changes that are part of this patchset.

I've applied 1-3 to ppc-for-2.9, still looking at the rest.

> 
> Bharata B Rao (7):
>   softfloat: Add round-to-odd rounding mode
>   softfloat: Add float128_to_uint64_round_to_zero()
>   softfloat: Add float128_to_uint32_round_to_zero()
>   target-ppc: Implement round to odd variants of quad FP instructions
>   target-ppc: Add xscvqpudz and xscvqpuwz instructions
>   softfloat: Add float128_to_uint32()
>   target-s390: Use float128_to_uint[64/32] where required
> 
>  fpu/softfloat.c                     | 125 +++++++++++++++++++++++++++++++++++-
>  include/fpu/softfloat.h             |   6 ++
>  target/ppc/fpu_helper.c             |  44 ++++++-------
>  target/ppc/helper.h                 |   2 +
>  target/ppc/translate/vsx-impl.inc.c |   2 +
>  target/ppc/translate/vsx-ops.inc.c  |   4 +-
>  target/s390x/fpu_helper.c           |   6 +-
>  7 files changed, 161 insertions(+), 28 deletions(-)
> 

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson