This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Use of "SUSPENDED" in Bugzilla
From: David Miller <davem@davemloft.net>
Date: Tue, 21 Feb 2012 15:37:54 -0500 (EST)
> From: "Joseph S. Myers" <joseph@codesourcery.com>
> Date: Mon, 20 Feb 2012 18:07:26 +0000 (UTC)
>
>> On Mon, 20 Feb 2012, Aurelien Jarno wrote:
>>
>>> Ah indeed. I actually got this issue, that is atan2 was moved when
>>> running gen-libm-test.pl. I decided to remove the corresponding hunk to
>>> avoid having unrelated changes (I also thought it was locale related).
>>> That said the problem was already there on the previous version of the
>>> ulps file, nothing has changed with my patch.
>>
>> I've applied the patch resulting from passing all these files through
>> gen-libm-test.pl (sorting, removing duplicate entries, etc.).
>
> Aurelien, I'm still getting ULP failures for the long double cases
> on both 32-bit and 64-bit sparc. In what kind of environment did
> you generate these? For example on sparcv9-linux with multiarch
> enabled I get the following for math/test-ldouble.out:
The following patch seems to clear all the remaining failures
for me.
diff --git a/ChangeLog b/ChangeLog
index 73e51dc..f7a8a96 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+2012-02-21 David S. Miller <davem@davemloft.net>
+
+ * sysdeps/sparc/sparc32/fpu/libm-test-ulps: More jn test ULP updates.
+ * sysdeps/sparc/sparc64/fpu/libm-test-ulps: Likewise.
+
2012-02-20 David S. Miller <davem@davemloft.net>
* sysdeps/sparc/sparc32/__longjmp.S: Unwind in the 'thread' path
diff --git a/sysdeps/sparc/sparc32/fpu/libm-test-ulps b/sysdeps/sparc/sparc32/fpu/libm-test-ulps
index d75c97d..91e6760 100644
--- a/sysdeps/sparc/sparc32/fpu/libm-test-ulps
+++ b/sysdeps/sparc/sparc32/fpu/libm-test-ulps
@@ -659,6 +659,8 @@ double: 2
float: 1
idouble: 2
ifloat: 1
+ildouble: 1
+ldouble: 1
Test "jn (3, 0.125) == 0.406503832554912875023029337653442868e-4":
double: 1
float: 1
@@ -684,34 +686,48 @@ ifloat: 2
Test "jn (3, 2.4048255576957729) == 0.19899990535769083404042146764530813":
double: 3
idouble: 3
+ildouble: 1
+ldouble: 1
Test "jn (4, 2.4048255576957729) == 0.647466661641779720084932282551219891E-1":
double: 1
idouble: 1
+ildouble: 1
+ldouble: 1
Test "jn (5, 2.4048255576957729) == 0.163892432048058525099230549946147698E-1":
double: 3
float: 1
idouble: 3
ifloat: 1
+ildouble: 2
+ldouble: 2
Test "jn (6, 2.4048255576957729) == 0.34048184720278336646673682895929161E-2":
double: 4
float: 3
idouble: 4
ifloat: 3
+ildouble: 5
+ldouble: 5
Test "jn (7, 2.4048255576957729) == 0.60068836573295394221291569249883076E-3":
double: 3
float: 5
idouble: 3
ifloat: 5
+ildouble: 3
+ldouble: 3
Test "jn (8, 2.4048255576957729) == 0.92165786705344923232879022467054148E-4":
double: 3
float: 2
idouble: 3
ifloat: 2
+ildouble: 8
+ldouble: 8
Test "jn (9, 2.4048255576957729) == 0.12517270977961513005428966643852564E-4":
double: 1
float: 2
idouble: 1
ifloat: 2
+ildouble: 3
+ldouble: 3
# lgamma
Test "lgamma (-0.5) == log(2*sqrt(pi))":
@@ -1276,8 +1292,8 @@ double: 4
float: 5
idouble: 4
ifloat: 5
-ildouble: 4
-ldouble: 4
+ildouble: 8
+ldouble: 8
Function: "lgamma":
double: 1
diff --git a/sysdeps/sparc/sparc64/fpu/libm-test-ulps b/sysdeps/sparc/sparc64/fpu/libm-test-ulps
index d75c97d..91e6760 100644
--- a/sysdeps/sparc/sparc64/fpu/libm-test-ulps
+++ b/sysdeps/sparc/sparc64/fpu/libm-test-ulps
@@ -659,6 +659,8 @@ double: 2
float: 1
idouble: 2
ifloat: 1
+ildouble: 1
+ldouble: 1
Test "jn (3, 0.125) == 0.406503832554912875023029337653442868e-4":
double: 1
float: 1
@@ -684,34 +686,48 @@ ifloat: 2
Test "jn (3, 2.4048255576957729) == 0.19899990535769083404042146764530813":
double: 3
idouble: 3
+ildouble: 1
+ldouble: 1
Test "jn (4, 2.4048255576957729) == 0.647466661641779720084932282551219891E-1":
double: 1
idouble: 1
+ildouble: 1
+ldouble: 1
Test "jn (5, 2.4048255576957729) == 0.163892432048058525099230549946147698E-1":
double: 3
float: 1
idouble: 3
ifloat: 1
+ildouble: 2
+ldouble: 2
Test "jn (6, 2.4048255576957729) == 0.34048184720278336646673682895929161E-2":
double: 4
float: 3
idouble: 4
ifloat: 3
+ildouble: 5
+ldouble: 5
Test "jn (7, 2.4048255576957729) == 0.60068836573295394221291569249883076E-3":
double: 3
float: 5
idouble: 3
ifloat: 5
+ildouble: 3
+ldouble: 3
Test "jn (8, 2.4048255576957729) == 0.92165786705344923232879022467054148E-4":
double: 3
float: 2
idouble: 3
ifloat: 2
+ildouble: 8
+ldouble: 8
Test "jn (9, 2.4048255576957729) == 0.12517270977961513005428966643852564E-4":
double: 1
float: 2
idouble: 1
ifloat: 2
+ildouble: 3
+ldouble: 3
# lgamma
Test "lgamma (-0.5) == log(2*sqrt(pi))":
@@ -1276,8 +1292,8 @@ double: 4
float: 5
idouble: 4
ifloat: 5
-ildouble: 4
-ldouble: 4
+ildouble: 8
+ldouble: 8
Function: "lgamma":
double: 1