[PATCH v5 08/12] benchtests: Use __f128 on ilogbf128-inputs constants

[Adhemerval Zanella]

The f128 is not a valid floating constant suffix on clang.
---
 benchtests/ilogbf128-inputs | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/benchtests/ilogbf128-inputs b/benchtests/ilogbf128-inputs
index bfbfc93714..596ab4f513 100644
--- a/benchtests/ilogbf128-inputs
+++ b/benchtests/ilogbf128-inputs
@@ -3,9 +3,9 @@
 ## includes: math.h
 
 ## name: subnormal
-6.47517511943802511092443895822764655e-4966f128
-0x1.fffffffffffffff8p-16383f128
+__f128(6.47517511943802511092443895822764655e-4966)
+__f128(0x1.fffffffffffffff8p-16383)
 
 ## name: normal
-1.0
--0x8.2faf442f390a9211f5af128673fp+0L
+__f128(1.0)
+__f128(-0x8.2faf442f390a9211f5af128673fp+0)
-- 
2.43.0