From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: Justin Forbes <jmforbes@linuxtx.org>,
Zwane Mwaikambo <zwane@arm.linux.org.uk>,
"Theodore Ts'o" <tytso@mit.edu>,
Randy Dunlap <rdunlap@xenotime.net>,
Dave Jones <davej@redhat.com>,
Chuck Wolber <chuckw@quantumlinux.com>,
Chris Wedgwood <reviews@ml.cw.f00f.org>,
Michael Krufky <mkrufky@linuxtv.org>,
Chuck Ebbert <cebbert@redhat.com>,
Domenico Andreoli <cavokz@gmail.com>,
torvalds@linux-foundation.org, akpm@linux-foundation.org,
alan@lxorguk.ukuu.org.uk, bunk@stusta.de,
"David S. Miller" <davem@davemloft.net>
Subject: [patch 11/28] Fix soft-fp underflow handling.
Date: Thu, 23 Aug 2007 15:26:48 -0700 [thread overview]
Message-ID: <20070823222648.GL18559@kroah.com> (raw)
In-Reply-To: <20070823221811.GA18559@kroah.com>
[-- Attachment #1: fix-soft-fp-underflow-handling.patch --]
[-- Type: text/plain, Size: 3935 bytes --]
-stable review patch. If anyone has any objections, please let us know.
------------------
From: David Miller <davem@davemloft.net>
The underflow exception cases were wrong.
This is one weird area of ieee1754 handling in that the underflow
behavior changes based upon whether underflow is enabled in the trap
enable mask of the FPU control register. As a specific case the Sparc
V9 manual gives us the following description:
--------------------
If UFM = 0: Underflow occurs if a nonzero result is tiny and a
loss of accuracy occurs. Tininess may be detected
before or after rounding. Loss of accuracy may be
either a denormalization loss or an inexact result.
If UFM = 1: Underflow occurs if a nonzero result is tiny.
Tininess may be detected before or after rounding.
--------------------
What this amounts to in the packing case is if we go subnormal,
we set underflow if any of the following are true:
1) rounding sets inexact
2) we ended up rounding back up to normal (this is the case where
we set the exponent to 1 and set the fraction to zero), this
should set inexact too
3) underflow is set in FPU control register trap-enable mask
The initially discovered example was "DBL_MIN / 16.0" which
incorrectly generated an underflow. It should not, unless underflow
is set in the trap-enable mask of the FPU csr.
Another example, "0x0.0000000000001p-1022 / 16.0", should signal both
inexact and underflow. The cpu implementations and ieee1754
literature is very clear about this. This is case #2 above.
However, if underflow is set in the trap enable mask, only underflow
should be set and reported as a trap. That is handled properly by the
prioritization logic in
arch/sparc{,64}/math-emu/math.c:record_exception().
Based upon a report and test case from Jakub Jelinek.
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
include/asm-sparc/sfp-machine.h | 6 ++++++
include/asm-sparc64/sfp-machine.h | 2 ++
include/math-emu/op-common.h | 5 ++++-
include/math-emu/soft-fp.h | 7 +++++++
4 files changed, 19 insertions(+), 1 deletion(-)
--- a/include/asm-sparc/sfp-machine.h
+++ b/include/asm-sparc/sfp-machine.h
@@ -203,4 +203,10 @@ extern struct task_struct *last_task_use
#define FP_INHIBIT_RESULTS ((last_task_used_math->thread.fsr >> 23) & _fex)
#endif
+#ifdef CONFIG_SMP
+#define FP_TRAPPING_EXCEPTIONS ((current->thread.fsr >> 23) & 0x1f)
+#else
+#define FP_TRAPPING_EXCEPTIONS ((last_task_used_math->thread.fsr >> 23) & 0x1f)
+#endif
+
#endif
--- a/include/asm-sparc64/sfp-machine.h
+++ b/include/asm-sparc64/sfp-machine.h
@@ -88,4 +88,6 @@
#define FP_INHIBIT_RESULTS ((current_thread_info()->xfsr[0] >> 23) & _fex)
+#define FP_TRAPPING_EXCEPTIONS ((current_thread_info()->xfsr[0] >> 23) & 0x1f)
+
#endif
--- a/include/math-emu/op-common.h
+++ b/include/math-emu/op-common.h
@@ -145,13 +145,16 @@ do { \
{ \
X##_e = 1; \
_FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc); \
+ FP_SET_EXCEPTION(FP_EX_INEXACT); \
} \
else \
{ \
X##_e = 0; \
_FP_FRAC_SRL_##wc(X, _FP_WORKBITS); \
- FP_SET_EXCEPTION(FP_EX_UNDERFLOW); \
} \
+ if ((FP_CUR_EXCEPTIONS & FP_EX_INEXACT) || \
+ (FP_TRAPPING_EXCEPTIONS & FP_EX_UNDERFLOW)) \
+ FP_SET_EXCEPTION(FP_EX_UNDERFLOW); \
} \
else \
{ \
--- a/include/math-emu/soft-fp.h
+++ b/include/math-emu/soft-fp.h
@@ -97,12 +97,19 @@
#define FP_INHIBIT_RESULTS 0
#endif
+#ifndef FP_TRAPPING_EXCEPTIONS
+#define FP_TRAPPING_EXCEPTIONS 0
+#endif
+
#define FP_SET_EXCEPTION(ex) \
_fex |= (ex)
#define FP_UNSET_EXCEPTION(ex) \
_fex &= ~(ex)
+#define FP_CUR_EXCEPTIONS \
+ (_fex)
+
#define FP_CLEAR_EXCEPTIONS \
_fex = 0
--
next prev parent reply other threads:[~2007-08-23 22:32 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20070823220656.101239233@mini.kroah.org>
2007-08-23 22:18 ` [patch 00/28] 2.6.22-stable review cycle again Greg KH
2007-08-23 22:25 ` [patch 01/28] ocfs2: Fix bad source start calculation during kernel writes Greg KH
2007-08-23 22:25 ` [patch 02/28] NET: Share correct feature code between bridging and bonding Greg KH
2007-08-23 22:26 ` [patch 03/28] sky2: dont clear phy power bits Greg KH
2007-08-23 22:26 ` [patch 04/28] uml: fix previous request size limit fix Greg KH
2007-08-23 22:26 ` [patch 05/28] i386: fix lazy mode vmalloc synchronization for paravirt Greg KH
2007-08-23 22:26 ` [patch 06/28] signalfd: fix interaction with posix-timers Greg KH
2007-08-23 22:26 ` [patch 07/28] signalfd: make it group-wide, fix posix-timers scheduling Greg KH
2007-08-23 22:26 ` [patch 08/28] DCCP: Fix DCCP GFP_KERNEL allocation in atomic context Greg KH
2007-08-23 22:26 ` [patch 09/28] IPV6: Fix kernel panic while send SCTP data with IP fragments Greg KH
2007-08-23 22:26 ` [patch 10/28] IPv6: Invalid semicolon after if statement Greg KH
2007-08-23 22:26 ` Greg KH [this message]
2007-08-23 22:26 ` [patch 12/28] Netfilter: Missing Kbuild entry for netfilter Greg KH
2007-08-23 22:26 ` [patch 13/28] SNAP: Fix SNAP protocol header accesses Greg KH
2007-08-23 22:27 ` [patch 14/28] NET: Fix missing rcu unlock in __sock_create() Greg KH
2007-08-23 22:27 ` [patch 15/28] SPARC64: Fix sparc64 task stack traces Greg KH
2007-08-23 22:27 ` [patch 16/28] SPARC64: Fix sparc64 PCI config accesses on sun4u Greg KH
2007-08-23 22:27 ` [patch 18/28] TCP: Fix TCP rate-halving on bidirectional flows Greg KH
2007-08-23 22:27 ` [patch 19/28] TCP: Fix TCP handling of SACK in " Greg KH
2007-08-23 22:27 ` [patch 20/28] PPP: Fix PPP buffer sizing Greg KH
2007-08-23 22:27 ` [patch 21/28] PCI: lets kill the PCI hidden behind bridge message Greg KH
2007-08-23 22:27 ` [patch 22/28] PCI: disable MSI on RS690 Greg KH
2007-08-23 22:27 ` [patch 23/28] PCI: disable MSI on RD580 Greg KH
2007-08-23 22:27 ` [patch 24/28] PCI: disable MSI on RX790 Greg KH
2007-08-23 22:27 ` [patch 25/28] USB: cdc-acm: fix sysfs attribute registration bug Greg KH
2007-08-24 13:59 ` Alan Stern
2007-08-24 15:49 ` [stable] " Greg KH
2007-08-24 17:59 ` Alan Stern
2007-08-24 18:04 ` Greg KH
2007-08-29 18:48 ` Chuck Ebbert
2007-08-29 23:33 ` Chuck Ebbert
2007-08-31 5:10 ` Greg KH
2007-08-23 22:27 ` [patch 26/28] USB: allow retry on descriptor fetch errors Greg KH
2007-08-23 22:27 ` [patch 27/28] USB: fix DoS in pwc USB video driver Greg KH
2007-08-23 22:28 ` [patch 28/28] usb: add PRODUCT, TYPE to usb-interface events Greg KH
2007-08-23 22:29 ` [patch 17/28] TCP: Do not autobind ports for TCP sockets Greg KH
2007-08-23 22:32 ` [patch 00/28] 2.6.22-stable review cycle again Greg KH
2007-08-29 19:43 ` Thomas Backlund
2007-08-29 20:03 ` Willy Tarreau
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=20070823222648.GL18559@kroah.com \
--to=gregkh@suse.de \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=bunk@stusta.de \
--cc=cavokz@gmail.com \
--cc=cebbert@redhat.com \
--cc=chuckw@quantumlinux.com \
--cc=davej@redhat.com \
--cc=davem@davemloft.net \
--cc=jmforbes@linuxtx.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mkrufky@linuxtv.org \
--cc=rdunlap@xenotime.net \
--cc=reviews@ml.cw.f00f.org \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=tytso@mit.edu \
--cc=zwane@arm.linux.org.uk \
/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