This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
[COMMITTED] Update powerpc libm-test ULPs
- From: Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>
- To: "GNU C. Library" <libc-alpha at sourceware dot org>
- Date: Tue, 28 May 2013 10:45:23 -0300
- Subject: [COMMITTED] Update powerpc libm-test ULPs
I'm seeing the long double tests now fails with the following message:
ulp (0x0.0p0) is not FP_SUBNORMAL!
I believe it might be related to the ulp() fixes from Carlos and I see a note from
Andreas saying the code might be wrong for IBM long double. Is that the case?
---
ChangeLog | 4 +++
sysdeps/powerpc/fpu/libm-test-ulps | 37 ++++++++++++++++++++++++++++++++++++
2 files changed, 41 insertions(+), 0 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 61d8a40..f1afcf0 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,7 @@
+2013-05-28 Adhemerval Zanella <azanella@linux.vnet.ibm.com>
+
+ * sysdeps/powerpc/fpu/libm-test-ulps: Update.
+
2013-05-26 Thomas Schwinge <thomas@codesourcery.com>
* stdio-common/bug26.c (main): Correct fscanf template.
diff --git a/sysdeps/powerpc/fpu/libm-test-ulps b/sysdeps/powerpc/fpu/libm-test-ulps
index 73b83b4..5475b51 100644
--- a/sysdeps/powerpc/fpu/libm-test-ulps
+++ b/sysdeps/powerpc/fpu/libm-test-ulps
@@ -5840,6 +5840,18 @@ Test "expm1 (500.0)":
double: 1
idouble: 1
+# gamma
+Test "gamma (0.7)":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "gamma (1.2)":
+double: 1
+float: 2
+idouble: 1
+ifloat: 2
+
# hypot
Test "hypot (-0.7, -12.4)":
double: 1
@@ -6170,6 +6182,23 @@ Test "pow (0x1.000002p0, 0x1p24)":
float: 1
ifloat: 1
+# pow10
+Test "pow10 (-1)":
+double: 1
+idouble: 1
+Test "pow10 (-305)":
+double: 1
+idouble: 1
+Test "pow10 (-36)":
+double: 1
+idouble: 1
+Test "pow10 (3)":
+double: 1
+idouble: 1
+Test "pow10 (36)":
+double: 1
+idouble: 1
+
# pow_downward
Test "pow_downward (1.0625, 1.125)":
ildouble: 1
@@ -7774,6 +7803,10 @@ ildouble: 1
ldouble: 1
Function: "gamma":
+double: 1
+float: 2
+idouble: 1
+ifloat: 2
ildouble: 1
ldouble: 1
@@ -7845,6 +7878,10 @@ ifloat: 1
ildouble: 1
ldouble: 1
+Function: "pow10":
+double: 1
+idouble: 1
+
Function: "pow_downward":
float: 1
ifloat: 1
--
1.7.1