This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
[RFC v2 19/20] RISC-V: Fix llrint and llround missing exceptions on RV32
- From: Alistair Francis <alistair dot francis at wdc dot com>
- To: libc-alpha at sourceware dot org
- Cc: arnd at arndb dot de, adhemerval dot zanella at linaro dot org, fweimer at redhat dot com, palmer at sifive dot com, macro at wdc dot com, zongbox at gmail dot com, zong at andestech dot com, alistair dot francis at wdc dot com, alistair23 at gmail dot com
- Date: Mon, 24 Jun 2019 17:09:41 -0700
- Subject: [RFC v2 19/20] RISC-V: Fix llrint and llround missing exceptions on RV32
- Ironport-sdr: UBUgZmef2w3Icuw9c9c/BbTtUeKpWORhVIFbM34wt9L7wD3F1jL/hEHc/uY5tI+S56yx2Zx9lY fWSm3Ul91oJVOG0UQQYR37Wn4Xl1mhKNs3ywqDCyfMUjGe2rNsP3fCj7mWJbxOXvDCOz8kVCqA 8w2riXM69VNYCw6fu76QtP0ISDwgMDj0TF1rBkFE1xrrlRrMaLPQos/vjN6FKUXGaDWYQJeMVv KnRjLrvhisG24HIyYRhL52e3oUQd2VKkSXDqdNrnfwt0FRwQvqYhzh8ewP0CczBCGLry6do1GY UtrftV6mAx0Fbbip5mWsl2W9
- Ironport-sdr: KnRPFYyjEa/SaB5Wd/EH799lWXhr0yPD6hsuop/O7R+XFKHmpgh03j/GgtyKJW+H8LSM3LQjcH kUYfffawovg0ZQXY7WcDxlQPQE9Ufr7L0ShS5W0UQhDme9HQ8r8fhijF43MB2jI9Z8LKV3l0iW 5dbPiMIczjfcsrjSsyNPG+zvTeIoo3T80gzVZppuMId3/loB1t+fLoA5HrioJXhyCwDvH3CTxB c1H2nhZ8fcLEO81kCmfbf/plMMIkBBiU8wl2jE+fR0Q67gRKXkQTYOfsitFrQcXHfSngNvwP6L tok=
- References: <cover.1561421042.git.alistair.francis@wdc.com>
From: Zong Li <zongbox@gmail.com>
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 | 10 +++++
.../riscv/rv32/fix-fp-int-convert-overflow.h | 38 +++++++++++++++++++
2 files changed, 48 insertions(+)
create mode 100644 sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h
diff --git a/ChangeLog b/ChangeLog
index 54966f45b8..f731e94fc9 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -91,6 +91,16 @@
* sysdeps/unix/sysv/linux/riscv/configure.ac: Likewise.
* sysdeps/unix/sysv/linux/riscv/shlib-versions: Likewise.
* sysdeps/riscv/preconfigure: Likewise.
+ * sysdeps/riscv/rv32/Implies-after: New file.
+ * sysdeps/riscv/rv32/rvd/Implies: Likewise.
+ * sysdeps/riscv/rv32/rvf/Implies: Likewise.
+ * sysdeps/unix/sysv/linux/riscv/rv32/Implies: Likewise.
+ * sysdeps/unix/sysv/linux/riscv/Makefile: Support rv32.
+ * sysdeps/unix/sysv/linux/riscv/configure: Likewise.
+ * 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.
2019-06-20 Dmitry V. Levin <ldv@altlinux.org>
Florian Weimer <fweimer@redhat.com>
diff --git a/sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h b/sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h
new file mode 100644
index 0000000000..6b34a8415e
--- /dev/null
+++ b/sysdeps/riscv/rv32/fix-fp-int-convert-overflow.h
@@ -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 */
--
2.22.0