From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: libc-stable@sourceware.org
Subject: [2.31 COMMITTED] arm: Fix softp-fp Implies (BZ #25635)
Date: Tue, 10 Mar 2020 10:34:53 -0300 [thread overview]
Message-ID: <20200310133454.20821-4-adhemerval.zanella@linaro.org> (raw)
In-Reply-To: <20200310133454.20821-1-adhemerval.zanella@linaro.org>
The commit "arm: Split BE/LE abilist"
(1673ba87fefe019c834c09d33673d1d453ea698d) changed the soft-fp order for
ARM selection when __SOFTFP__ is defined by the compiler.
On 2.30 the sysdeps order is:
2.30
sysdeps/unix/sysv/linux/arm
sysdeps/arm/nptl
sysdeps/unix/sysv/linux
sysdeps/nptl
sysdeps/pthread
sysdeps/gnu
sysdeps/unix/inet
sysdeps/unix/sysv
sysdeps/unix/arm
sysdeps/unix
sysdeps/posix
sysdeps/arm/nofpu
sysdeps/ieee754/soft-fp
sysdeps/arm
sysdeps/wordsize-32
sysdeps/ieee754/flt-32
sysdeps/ieee754/dbl-64
sysdeps/ieee754
sysdeps/generic
While on master is:
sysdeps/unix/sysv/linux/arm/le
sysdeps/unix/sysv/linux/arm
sysdeps/arm/nptl
sysdeps/unix/sysv/linux
sysdeps/nptl
sysdeps/pthread
sysdeps/gnu
sysdeps/unix/inet
sysdeps/unix/sysv
sysdeps/unix/arm
sysdeps/unix
sysdeps/posix
sysdeps/arm/le
sysdeps/arm
sysdeps/wordsize-32
sysdeps/ieee754/flt-32
sysdeps/ieee754/dbl-64
sysdeps/arm/nofpu
sysdeps/ieee754/soft-fp
sysdeps/ieee754
sysdeps/generic
It make the build select some routines (fadd, fdiv, fmul, fsub, and fma)
on ieee754/flt-32 and ieee754/dbl-64 that requires fenv support to be
correctly rounded which in turns lead to math failures since the
__SOFTFP__ does not have fenv support.
With this patch the order is now:
sysdeps/unix/sysv/linux/arm/le
sysdeps/unix/sysv/linux/arm
sysdeps/arm/nptl
sysdeps/unix/sysv/linux
sysdeps/nptlsysdeps/pthread
sysdeps/gnu
sysdeps/unix/inet
sysdeps/unix/sysv
sysdeps/unix/arm
sysdeps/unix
sysdeps/posix
sysdeps/arm/le/nofpu
sysdeps/arm/nofpu
sysdeps/ieee754/soft-fp
sysdeps/arm/le
sysdeps/arm
sysdeps/wordsize-32
sysdeps/ieee754/flt-32
sysdeps/ieee754/dbl-64
sysdeps/ieee754
sysdeps/generic
Checked on arm-linux-gnuaebi.
(cherry picked from commit af09e5e5d9ec3ca20891e61a6922eac984fcbdc4)
---
sysdeps/arm/be/nofpu/Implies | 1 +
sysdeps/arm/le/nofpu/Implies | 1 +
2 files changed, 2 insertions(+)
create mode 100644 sysdeps/arm/be/nofpu/Implies
create mode 100644 sysdeps/arm/le/nofpu/Implies
diff --git a/sysdeps/arm/be/nofpu/Implies b/sysdeps/arm/be/nofpu/Implies
new file mode 100644
index 0000000000..c90dd7fd5c
--- /dev/null
+++ b/sysdeps/arm/be/nofpu/Implies
@@ -0,0 +1 @@
+arm/nofpu
diff --git a/sysdeps/arm/le/nofpu/Implies b/sysdeps/arm/le/nofpu/Implies
new file mode 100644
index 0000000000..c90dd7fd5c
--- /dev/null
+++ b/sysdeps/arm/le/nofpu/Implies
@@ -0,0 +1 @@
+arm/nofpu
--
2.17.1
next prev parent reply other threads:[~2020-03-10 13:35 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-10 13:34 [2.31 COMMITTED] i386: Use comdat instead of .gnu.linkonce for i386 setup pic register (BZ #20543) Adhemerval Zanella
2020-03-10 13:34 ` [2.31 COMMITTED] linux: Clear mode_t padding bits (BZ#25623) Adhemerval Zanella
2020-03-10 13:34 ` [2.31 COMMITTED] linux/sysipc: Include linux/posix_types.h for __kernel_mode_t Adhemerval Zanella
2020-03-10 13:34 ` Adhemerval Zanella [this message]
2020-03-10 13:34 ` [2.31 COMMITTED] sparc: Move sigreturn stub to assembly Adhemerval Zanella
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200310133454.20821-4-adhemerval.zanella@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=libc-stable@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).