diff mbox series

math: Update i686 ulps

Message ID 20180601124724.05B5541548878@oldenburg.str.redhat.com
State New
Headers show
Series math: Update i686 ulps | expand

Commit Message

Florian Weimer June 1, 2018, 12:47 p.m. UTC 
2018-06-01  Florian Weimer  <fweimer@redhat.com>

	* sysdeps/i386/i686/fpu/multiarch/libm-test-ulps: Update with
	results from building with “-march=x86-64 -mtune=generic
	-mfpmath=sse” and running on a Haswell-era CPU.

Comments

Florian Weimer June 1, 2018, 12:48 p.m. UTC | #1 
On 06/01/2018 02:47 PM, Florian Weimer wrote:
>   Function: "y1_upward":
> -double: 1
> +double: 7
>   float: 3
>   float128: 5
> -idouble: 1
> +idouble: 7
>   ifloat: 3
>   ifloat128: 5
>   ildouble: 7

I'm posting this for review because some of the changes (like the one 
above) are fairly large.

Thanks,
Florian
Joseph Myers June 1, 2018, 4:58 p.m. UTC | #2 
On Fri, 1 Jun 2018, Florian Weimer wrote:

> On 06/01/2018 02:47 PM, Florian Weimer wrote:
> >   Function: "y1_upward":
> > -double: 1
> > +double: 7
> >   float: 3
> >   float128: 5
> > -idouble: 1
> > +idouble: 7
> >   ifloat: 3
> >   ifloat128: 5
> >   ildouble: 7
> 
> I'm posting this for review because some of the changes (like the one above)
> are fairly large.

How do they compare to ulps for these functions on e.g. x86_64?  My 
presumption is that they arise from use of -mfpmath=sse (thus, meaning no 
excess precision, so results should mostly be the same as on other 
architectures).
Florian Weimer June 1, 2018, 5:25 p.m. UTC | #3 
On 06/01/2018 06:58 PM, Joseph Myers wrote:
> On Fri, 1 Jun 2018, Florian Weimer wrote:
> 
>> On 06/01/2018 02:47 PM, Florian Weimer wrote:
>>>    Function: "y1_upward":
>>> -double: 1
>>> +double: 7
>>>    float: 3
>>>    float128: 5
>>> -idouble: 1
>>> +idouble: 7
>>>    ifloat: 3
>>>    ifloat128: 5
>>>    ildouble: 7
>>
>> I'm posting this for review because some of the changes (like the one above)
>> are fairly large.
> 
> How do they compare to ulps for these functions on e.g. x86_64?  My
> presumption is that they arise from use of -mfpmath=sse (thus, meaning no
> excess precision, so results should mostly be the same as on other
> architectures).

Hmm.  Right.  sysdeps/x86_64/fpu/libm-test-ulps gas this:

Function: "y1_upward":
double: 7
float: 2
float128: 5
idouble: 7
ifloat: 2
ifloat128: 5
ildouble: 7
ldouble: 7

So they are not substantially worse.

(If I copy over the x86-64 ulp baseline to the i686 directory, I still 
get many testsuite failures, though.)

Thanks,
Florian
Joseph Myers June 1, 2018, 5:31 p.m. UTC | #4 
On Fri, 1 Jun 2018, Florian Weimer wrote:

> > How do they compare to ulps for these functions on e.g. x86_64?  My
> > presumption is that they arise from use of -mfpmath=sse (thus, meaning no
> > excess precision, so results should mostly be the same as on other
> > architectures).
> 
> Hmm.  Right.  sysdeps/x86_64/fpu/libm-test-ulps gas this:
> 
> Function: "y1_upward":
> double: 7
> float: 2
> float128: 5
> idouble: 7
> ifloat: 2
> ifloat128: 5
> ildouble: 7
> ldouble: 7
> 
> So they are not substantially worse.

In that case I think you should commit the patch.

> (If I copy over the x86-64 ulp baseline to the i686 directory, I still get
> many testsuite failures, though.)

x86_64 and i686 both have some architecture-specific function 
implementations, so one can't expect to have exactly the same ulps for 
them.
diff mbox series

Patch

diff --git a/sysdeps/i386/i686/fpu/multiarch/libm-test-ulps b/sysdeps/i386/i686/fpu/multiarch/libm-test-ulps
index da2b5fd8bc..8a51503bc5 100644
--- a/sysdeps/i386/i686/fpu/multiarch/libm-test-ulps
+++ b/sysdeps/i386/i686/fpu/multiarch/libm-test-ulps
@@ -281,20 +281,20 @@  ldouble: 1
 
 Function: Real part of "cacos":
 double: 1
-float: 1
+float: 2
 float128: 2
 idouble: 1
-ifloat: 1
+ifloat: 2
 ifloat128: 2
 ildouble: 1
 ldouble: 1
 
 Function: Imaginary part of "cacos":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
@@ -360,21 +360,21 @@  ildouble: 7
 ldouble: 7
 
 Function: Real part of "cacosh":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: Imaginary part of "cacosh":
 double: 1
-float: 1
+float: 2
 float128: 2
 idouble: 1
-ifloat: 1
+ifloat: 2
 ifloat128: 2
 ildouble: 1
 ldouble: 1
@@ -420,10 +420,10 @@  ildouble: 2
 ldouble: 2
 
 Function: Real part of "cacosh_upward":
-double: 4
+double: 5
 float: 4
 float128: 6
-idouble: 4
+idouble: 5
 ifloat: 4
 ifloat128: 6
 ildouble: 5
@@ -488,11 +488,11 @@  ildouble: 1
 ldouble: 1
 
 Function: Imaginary part of "casin":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
@@ -558,11 +558,11 @@  ildouble: 7
 ldouble: 7
 
 Function: Real part of "casinh":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
@@ -774,11 +774,11 @@  ildouble: 1
 ldouble: 1
 
 Function: Real part of "catanh_upward":
-double: 2
-float: 2
+double: 4
+float: 4
 float128: 4
-idouble: 2
-ifloat: 2
+idouble: 4
+ifloat: 4
 ifloat128: 4
 ildouble: 4
 ldouble: 4
@@ -875,10 +875,10 @@  ldouble: 3
 
 Function: Real part of "ccos_towardzero":
 double: 1
-float: 1
+float: 2
 float128: 2
 idouble: 1
-ifloat: 1
+ifloat: 2
 ifloat128: 2
 ildouble: 3
 ldouble: 3
@@ -934,10 +934,10 @@  ildouble: 1
 ldouble: 1
 
 Function: Real part of "ccosh_downward":
-double: 1
+double: 2
 float: 2
 float128: 2
-idouble: 1
+idouble: 2
 ifloat: 2
 ifloat128: 2
 ildouble: 3
@@ -954,11 +954,11 @@  ildouble: 3
 ldouble: 3
 
 Function: Real part of "ccosh_towardzero":
-double: 1
-float: 2
+double: 2
+float: 3
 float128: 2
-idouble: 1
-ifloat: 2
+idouble: 2
+ifloat: 3
 ifloat128: 2
 ildouble: 3
 ldouble: 3
@@ -1075,10 +1075,10 @@  ldouble: 3
 
 Function: Real part of "clog":
 double: 2
-float: 1
+float: 3
 float128: 2
 idouble: 2
-ifloat: 1
+ifloat: 3
 ifloat128: 2
 ildouble: 3
 ldouble: 3
@@ -1092,79 +1092,81 @@  ildouble: 1
 ldouble: 1
 
 Function: Real part of "clog10":
-double: 2
-float: 2
+double: 3
+float: 4
 float128: 2
-idouble: 2
-ifloat: 2
+idouble: 3
+ifloat: 4
 ifloat128: 2
 ildouble: 4
 ldouble: 4
 
 Function: Imaginary part of "clog10":
-double: 1
+double: 2
+float: 1
 float128: 2
-idouble: 1
+idouble: 2
+ifloat: 1
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: Real part of "clog10_downward":
-double: 3
-float: 3
+double: 4
+float: 4
 float128: 3
-idouble: 3
-ifloat: 3
+idouble: 4
+ifloat: 4
 ifloat128: 3
 ildouble: 8
 ldouble: 8
 
 Function: Imaginary part of "clog10_downward":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 3
 ildouble: 3
 ldouble: 3
 
 Function: Real part of "clog10_towardzero":
-double: 3
-float: 3
+double: 5
+float: 5
 float128: 4
-idouble: 3
-ifloat: 3
+idouble: 5
+ifloat: 5
 ifloat128: 4
 ildouble: 8
 ldouble: 8
 
 Function: Imaginary part of "clog10_towardzero":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 3
 ildouble: 3
 ldouble: 3
 
 Function: Real part of "clog10_upward":
-double: 3
-float: 3
+double: 4
+float: 5
 float128: 4
-idouble: 3
-ifloat: 3
+idouble: 4
+ifloat: 5
 ifloat128: 4
 ildouble: 8
 ldouble: 8
 
 Function: Imaginary part of "clog10_upward":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 3
 ildouble: 3
 ldouble: 3
@@ -1191,10 +1193,10 @@  ldouble: 1
 
 Function: Real part of "clog_towardzero":
 double: 3
-float: 3
+float: 4
 float128: 3
 idouble: 3
-ifloat: 3
+ifloat: 4
 ifloat128: 3
 ildouble: 5
 ldouble: 5
@@ -1480,10 +1482,10 @@  ildouble: 1
 ldouble: 1
 
 Function: Real part of "csinh_downward":
-double: 1
+double: 2
 float: 2
 float128: 2
-idouble: 1
+idouble: 2
 ifloat: 2
 ifloat128: 2
 ildouble: 3
@@ -1500,10 +1502,10 @@  ildouble: 3
 ldouble: 3
 
 Function: Real part of "csinh_towardzero":
-double: 1
+double: 2
 float: 2
 float128: 2
-idouble: 1
+idouble: 2
 ifloat: 2
 ifloat128: 2
 ildouble: 3
@@ -1540,79 +1542,81 @@  ildouble: 3
 ldouble: 3
 
 Function: Real part of "csqrt":
-double: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: Imaginary part of "csqrt":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: Real part of "csqrt_downward":
-double: 1
-float: 1
+double: 4
+float: 4
 float128: 4
-idouble: 1
-ifloat: 1
+idouble: 4
+ifloat: 4
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: Imaginary part of "csqrt_downward":
-double: 1
-float: 1
+double: 3
+float: 3
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 4
 ldouble: 4
 
 Function: Real part of "csqrt_towardzero":
-double: 1
-float: 1
+double: 3
+float: 3
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 4
 ldouble: 4
 
 Function: Imaginary part of "csqrt_towardzero":
-double: 1
-float: 1
+double: 3
+float: 3
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 4
 ldouble: 4
 
 Function: Real part of "csqrt_upward":
-double: 1
-float: 1
+double: 4
+float: 4
 float128: 4
-idouble: 1
-ifloat: 1
+idouble: 4
+ifloat: 4
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: Imaginary part of "csqrt_upward":
-double: 1
-float: 2
+double: 3
+float: 3
 float128: 3
-idouble: 1
-ifloat: 2
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 4
 ldouble: 4
@@ -1629,20 +1633,20 @@  ldouble: 2
 
 Function: Imaginary part of "ctan":
 double: 2
-float: 1
+float: 2
 float128: 3
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 3
 ildouble: 1
 ldouble: 1
 
 Function: Real part of "ctan_downward":
-double: 1
-float: 2
+double: 6
+float: 5
 float128: 4
-idouble: 1
-ifloat: 2
+idouble: 6
+ifloat: 5
 ifloat128: 4
 ildouble: 5
 ldouble: 5
@@ -1658,31 +1662,31 @@  ildouble: 4
 ldouble: 4
 
 Function: Real part of "ctan_towardzero":
-double: 3
-float: 1
+double: 5
+float: 3
 float128: 4
-idouble: 3
-ifloat: 1
+idouble: 5
+ifloat: 3
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: Imaginary part of "ctan_towardzero":
 double: 2
-float: 1
+float: 2
 float128: 5
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 5
 ildouble: 4
 ldouble: 4
 
 Function: Real part of "ctan_upward":
 double: 3
-float: 2
+float: 4
 float128: 5
 idouble: 3
-ifloat: 2
+ifloat: 4
 ifloat128: 5
 ildouble: 3
 ldouble: 3
@@ -1699,20 +1703,20 @@  ldouble: 3
 
 Function: Real part of "ctanh":
 double: 2
-float: 1
+float: 2
 float128: 3
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 3
 ildouble: 1
 ldouble: 1
 
 Function: Imaginary part of "ctanh":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 3
 ildouble: 2
 ldouble: 2
@@ -1728,51 +1732,51 @@  ildouble: 4
 ldouble: 4
 
 Function: Imaginary part of "ctanh_downward":
-double: 2
-float: 1
+double: 6
+float: 5
 float128: 4
-idouble: 2
-ifloat: 1
+idouble: 6
+ifloat: 5
 ifloat128: 4
 ildouble: 4
 ldouble: 4
 
 Function: Real part of "ctanh_towardzero":
 double: 2
-float: 1
+float: 2
 float128: 5
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 5
 ildouble: 4
 ldouble: 4
 
 Function: Imaginary part of "ctanh_towardzero":
-double: 2
-float: 1
+double: 5
+float: 3
 float128: 3
-idouble: 2
-ifloat: 1
+idouble: 5
+ifloat: 3
 ifloat128: 3
 ildouble: 3
 ldouble: 3
 
 Function: Real part of "ctanh_upward":
 double: 2
-float: 1
+float: 2
 float128: 5
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 5
 ildouble: 3
 ldouble: 3
 
 Function: Imaginary part of "ctanh_upward":
 double: 3
-float: 2
+float: 3
 float128: 5
 idouble: 3
-ifloat: 2
+ifloat: 3
 ifloat128: 5
 ildouble: 3
 ldouble: 3
@@ -1818,41 +1822,41 @@  ildouble: 1
 ldouble: 1
 
 Function: "erfc":
-double: 1
-float: 1
+double: 3
+float: 2
 float128: 2
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 2
 ifloat128: 2
 ildouble: 3
 ldouble: 3
 
 Function: "erfc_downward":
-double: 2
-float: 3
+double: 5
+float: 6
 float128: 5
-idouble: 2
-ifloat: 3
+idouble: 5
+ifloat: 6
 ifloat128: 5
 ildouble: 4
 ldouble: 4
 
 Function: "erfc_towardzero":
-double: 2
-float: 2
+double: 3
+float: 4
 float128: 4
-idouble: 2
-ifloat: 2
+idouble: 3
+ifloat: 4
 ifloat128: 4
 ildouble: 4
 ldouble: 4
 
 Function: "erfc_upward":
-double: 2
-float: 3
+double: 5
+float: 6
 float128: 5
-idouble: 2
-ifloat: 3
+idouble: 5
+ifloat: 6
 ifloat128: 5
 ildouble: 5
 ldouble: 5
@@ -1996,34 +2000,34 @@  ildouble: 4
 ldouble: 4
 
 Function: "gamma":
-double: 3
+double: 4
 float: 3
-idouble: 3
+idouble: 4
 ifloat: 3
 ildouble: 4
 ldouble: 4
 
 Function: "gamma_downward":
-double: 4
+double: 5
 float: 5
-idouble: 4
+idouble: 5
 ifloat: 5
 ildouble: 7
 ldouble: 7
 
 Function: "gamma_towardzero":
-double: 4
+double: 5
 float: 4
-idouble: 4
+idouble: 5
 ifloat: 4
 ildouble: 7
 ldouble: 7
 
 Function: "gamma_upward":
-double: 3
-float: 4
-idouble: 3
-ifloat: 4
+double: 5
+float: 5
+idouble: 5
+ifloat: 5
 ildouble: 6
 ldouble: 6
 
@@ -2061,39 +2065,39 @@  ldouble: 1
 
 Function: "j0":
 double: 2
-float: 1
+float: 2
 float128: 2
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: "j0_downward":
-double: 1
-float: 3
+double: 2
+float: 4
 float128: 4
-idouble: 1
-ifloat: 3
+idouble: 2
+ifloat: 4
 ifloat128: 4
 ildouble: 4
 ldouble: 4
 
 Function: "j0_towardzero":
-double: 2
-float: 1
+double: 3
+float: 2
 float128: 2
-idouble: 2
-ifloat: 1
+idouble: 3
+ifloat: 2
 ifloat128: 2
 ildouble: 5
 ldouble: 5
 
 Function: "j0_upward":
-double: 2
+double: 3
 float: 3
 float128: 5
-idouble: 2
+idouble: 3
 ifloat: 3
 ifloat128: 5
 ildouble: 4
@@ -2101,120 +2105,120 @@  ldouble: 4
 
 Function: "j1":
 double: 2
-float: 1
+float: 2
 float128: 4
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 4
 ildouble: 1
 ldouble: 1
 
 Function: "j1_downward":
-double: 2
-float: 2
+double: 3
+float: 3
 float128: 4
-idouble: 2
-ifloat: 2
+idouble: 3
+ifloat: 3
 ifloat128: 4
 ildouble: 4
 ldouble: 4
 
 Function: "j1_towardzero":
-double: 2
+double: 3
 float: 2
 float128: 4
-idouble: 2
+idouble: 3
 ifloat: 2
 ifloat128: 4
 ildouble: 4
 ldouble: 4
 
 Function: "j1_upward":
-double: 2
-float: 3
+double: 3
+float: 5
 float128: 3
-idouble: 2
-ifloat: 3
+idouble: 3
+ifloat: 5
 ifloat128: 3
 ildouble: 3
 ldouble: 3
 
 Function: "jn":
-double: 3
-float: 3
+double: 4
+float: 4
 float128: 7
-idouble: 3
-ifloat: 3
+idouble: 4
+ifloat: 4
 ifloat128: 7
 ildouble: 4
 ldouble: 4
 
 Function: "jn_downward":
-double: 2
-float: 3
+double: 5
+float: 5
 float128: 8
-idouble: 2
-ifloat: 3
+idouble: 5
+ifloat: 5
 ifloat128: 8
 ildouble: 4
 ldouble: 4
 
 Function: "jn_towardzero":
-double: 2
-float: 3
+double: 5
+float: 5
 float128: 8
-idouble: 2
-ifloat: 3
+idouble: 5
+ifloat: 5
 ifloat128: 8
 ildouble: 5
 ldouble: 5
 
 Function: "jn_upward":
-double: 3
-float: 3
+double: 5
+float: 5
 float128: 7
-idouble: 3
-ifloat: 3
+idouble: 5
+ifloat: 5
 ifloat128: 7
 ildouble: 5
 ldouble: 5
 
 Function: "lgamma":
-double: 3
+double: 4
 float: 3
 float128: 5
-idouble: 3
+idouble: 4
 ifloat: 3
 ifloat128: 5
 ildouble: 4
 ldouble: 4
 
 Function: "lgamma_downward":
-double: 4
+double: 5
 float: 5
 float128: 8
-idouble: 4
+idouble: 5
 ifloat: 5
 ifloat128: 8
 ildouble: 7
 ldouble: 7
 
 Function: "lgamma_towardzero":
-double: 4
+double: 5
 float: 4
 float128: 5
-idouble: 4
+idouble: 5
 ifloat: 4
 ifloat128: 5
 ildouble: 7
 ldouble: 7
 
 Function: "lgamma_upward":
-double: 3
-float: 4
+double: 5
+float: 5
 float128: 8
-idouble: 3
-ifloat: 4
+idouble: 5
+ifloat: 5
 ifloat128: 8
 ildouble: 6
 ldouble: 6
@@ -2468,39 +2472,41 @@  ildouble: 3
 ldouble: 3
 
 Function: "sinh":
-double: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 2
 ldouble: 2
 
 Function: "sinh_downward":
-double: 2
-float: 1
+double: 3
+float: 3
 float128: 3
-idouble: 2
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 5
 ldouble: 5
 
 Function: "sinh_towardzero":
 double: 2
-float: 1
+float: 2
 float128: 3
 idouble: 2
-ifloat: 1
+ifloat: 2
 ifloat128: 3
 ildouble: 4
 ldouble: 4
 
 Function: "sinh_upward":
 double: 4
-float: 2
+float: 3
 float128: 4
 idouble: 4
-ifloat: 2
+ifloat: 3
 ifloat128: 4
 ildouble: 5
 ldouble: 5
@@ -2544,119 +2550,121 @@  ildouble: 2
 ldouble: 2
 
 Function: "tanh":
-double: 1
+double: 2
+float: 2
 float128: 2
-idouble: 1
+idouble: 2
+ifloat: 2
 ifloat128: 2
 ildouble: 3
 ldouble: 3
 
 Function: "tanh_downward":
-double: 1
-float: 1
+double: 3
+float: 3
 float128: 4
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 4
 ildouble: 7
 ldouble: 4
 
 Function: "tanh_towardzero":
-double: 1
-float: 1
+double: 2
+float: 2
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 2
+ifloat: 2
 ifloat128: 3
 ildouble: 3
 ldouble: 3
 
 Function: "tanh_upward":
-double: 1
-float: 1
+double: 3
+float: 3
 float128: 3
-idouble: 1
-ifloat: 1
+idouble: 3
+ifloat: 3
 ifloat128: 3
 ildouble: 5
 ldouble: 4
 
 Function: "tgamma":
-double: 3
-float: 3
+double: 5
+float: 4
 float128: 4
-idouble: 3
-ifloat: 3
+idouble: 5
+ifloat: 4
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: "tgamma_downward":
-double: 3
+double: 6
 float: 5
 float128: 5
-idouble: 3
+idouble: 6
 ifloat: 5
 ifloat128: 5
 ildouble: 5
 ldouble: 5
 
 Function: "tgamma_towardzero":
-double: 4
+double: 6
 float: 5
 float128: 5
-idouble: 4
+idouble: 6
 ifloat: 5
 ifloat128: 5
 ildouble: 5
 ldouble: 5
 
 Function: "tgamma_upward":
-double: 4
+double: 5
 float: 6
 float128: 4
-idouble: 4
+idouble: 5
 ifloat: 6
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: "y0":
-double: 1
+double: 2
 float: 1
 float128: 3
-idouble: 1
+idouble: 2
 ifloat: 1
 ifloat128: 3
 ildouble: 1
 ldouble: 1
 
 Function: "y0_downward":
-double: 2
-float: 2
+double: 3
+float: 4
 float128: 4
-idouble: 2
-ifloat: 2
+idouble: 3
+ifloat: 4
 ifloat128: 4
 ildouble: 5
 ldouble: 5
 
 Function: "y0_towardzero":
-double: 2
+double: 3
 float: 3
 float128: 3
-idouble: 2
+idouble: 3
 ifloat: 3
 ifloat128: 3
 ildouble: 5
 ldouble: 5
 
 Function: "y0_upward":
-double: 1
-float: 3
+double: 3
+float: 5
 float128: 3
-idouble: 1
-ifloat: 3
+idouble: 3
+ifloat: 5
 ifloat128: 3
 ildouble: 3
 ldouble: 3
@@ -2672,30 +2680,30 @@  ildouble: 2
 ldouble: 2
 
 Function: "y1_downward":
-double: 2
-float: 2
+double: 3
+float: 4
 float128: 4
-idouble: 2
-ifloat: 2
+idouble: 3
+ifloat: 4
 ifloat128: 4
 ildouble: 7
 ldouble: 7
 
 Function: "y1_towardzero":
-double: 2
+double: 3
 float: 2
 float128: 2
-idouble: 2
+idouble: 3
 ifloat: 2
 ifloat128: 2
 ildouble: 5
 ldouble: 5
 
 Function: "y1_upward":
-double: 1
+double: 7
 float: 3
 float128: 5
-idouble: 1
+idouble: 7
 ifloat: 3
 ifloat128: 5
 ildouble: 7
@@ -2713,10 +2721,10 @@  ldouble: 4
 
 Function: "yn_downward":
 double: 3
-float: 2
+float: 4
 float128: 5
 idouble: 3
-ifloat: 2
+ifloat: 4
 ifloat128: 5
 ildouble: 5
 ldouble: 5
@@ -2733,10 +2741,10 @@  ldouble: 5
 
 Function: "yn_upward":
 double: 4
-float: 3
+float: 5
 float128: 5
 idouble: 4
-ifloat: 3
+ifloat: 5
 ifloat128: 5
 ildouble: 4
 ldouble: 4