[v4,09/10] RISC-V: Fix llrint and llround missing exceptions on RV32
Commit Message
Similar to the recent fix for MIPS, ARM and S/390, RV32 is missing
correct exception on overflow from llrint and llround functions because
cast from floating-point types to long long do not result in correct
exceptions on overflow.
2018-11-29 Zong Li <zong@andestech.com>
* sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h: New file.
---
ChangeLog | 1 +
sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h | 38 ++++++++++++++++++++++++
2 files changed, 39 insertions(+)
create mode 100644 sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h
Comments
On Sat, 1 Dec 2018, Zong Li wrote:
> Similar to the recent fix for MIPS, ARM and S/390, RV32 is missing
That's not "recent" now.
Joseph Myers <joseph@codesourcery.com> 於 2018年12月4日 週二 上午2:05寫道:
>
> On Sat, 1 Dec 2018, Zong Li wrote:
>
> > Similar to the recent fix for MIPS, ARM and S/390, RV32 is missing
>
> That's not "recent" now.
>
Ok, I would modify the commit message.
DJ Delorie <dj@redhat.com> 於 2018年12月8日 週六 上午8:08寫道:
>
>
> Zong Li <zongbox@gmail.com> writes:
> > + * sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h: New file.
> >
> > +/* Fix for conversion of floating point to integer overflow. ARM version.
>
> RISC-V version, you mean.
Thanks for pointing out the mistake.
@@ -42,6 +42,7 @@
* sysdeps/unix/sysv/linux/riscv/configure.ac: Likewise.
* sysdeps/unix/sysv/linux/riscv/shlib-versions: Likewise.
* sysdeps/riscv/preconfigure: Likewise.
+ * sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h: New file.
2018-11-28 Florian Weimer <fweimer@redhat.com>
new file mode 100644
@@ -0,0 +1,38 @@
+/* Fix for conversion of floating point to integer overflow. ARM version.
+ Copyright (C) 2015-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; if not, see
+ <http://www.gnu.org/licenses/>. */
+
+#ifndef FIX_FP_INT_CONVERT_OVERFLOW_H
+#define FIX_FP_INT_CONVERT_OVERFLOW_H 1
+
+/* As of GCC 5, the generic libgcc2.c conversions from floating point
+ to long long may not raise the correct exceptions on overflow (and
+ may raise spurious "inexact" exceptions even in non-overflow cases,
+ see <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59412>). */
+#define FIX_FLT_LONG_CONVERT_OVERFLOW 0
+#define FIX_FLT_LLONG_CONVERT_OVERFLOW 1
+
+#define FIX_DBL_LONG_CONVERT_OVERFLOW 0
+#define FIX_DBL_LLONG_CONVERT_OVERFLOW 1
+
+#define FIX_LDBL_LONG_CONVERT_OVERFLOW 0
+#define FIX_LDBL_LLONG_CONVERT_OVERFLOW 0
+
+#define FIX_FLT128_LONG_CONVERT_OVERFLOW 0
+#define FIX_FLT128_LLONG_CONVERT_OVERFLOW 0
+
+#endif /* fix-fp-int-convert-overflow.h */