~ubuntu-branches/ubuntu/raring/libav/raring-security

« back to all changes in this revision

Viewing changes to debian/patches/post-0.7.1/0010-ARM-workaround-for-bug-in-GNU-assembler.patch

  • Committer: Package Import Robot
  • Author(s): Reinhard Tartler
  • Date: 2011-09-28 09:18:34 UTC
  • mfrom: (1.3.7 sid)
  • Revision ID: package-import@ubuntu.com-20110928091834-w415mnuh06h4zpvc
Tags: 4:0.7.1-7ubuntu2
Revert "Convert package to include multiarch support."

Show diffs side-by-side

added added

removed removed

Lines of Context:
 
1
From 15355f9af2a415bbfbffdace04a4341c8af050e7 Mon Sep 17 00:00:00 2001
 
2
From: Mans Rullgard <mans@mansr.com>
 
3
Date: Tue, 5 Jul 2011 18:29:35 +0100
 
4
Subject: [PATCH 10/70] ARM: workaround for bug in GNU assembler
 
5
 
 
6
Some versions of the GNU assembler do not handle 64-bit
 
7
immediate operands containing arithmetic.  Writing the
 
8
value out in full works correctly.
 
9
 
 
10
Signed-off-by: Mans Rullgard <mans@mansr.com>
 
11
(cherry picked from commit fce1e43410bdc032c4cf2b1c66166a9ed99cc8f1)
 
12
 
 
13
Signed-off-by: Anton Khirnov <anton@khirnov.net>
 
14
---
 
15
 libavcodec/arm/fft_fixed_neon.S |    2 +-
 
16
 1 files changed, 1 insertions(+), 1 deletions(-)
 
17
 
 
18
diff --git a/libavcodec/arm/fft_fixed_neon.S b/libavcodec/arm/fft_fixed_neon.S
 
19
index 14884d3..63d8159 100644
 
20
--- a/libavcodec/arm/fft_fixed_neon.S
 
21
+++ b/libavcodec/arm/fft_fixed_neon.S
 
22
@@ -56,7 +56,7 @@
 
23
         vhsub.s16       \r0, \d0, \d1           @ t3, t4, t8, t7
 
24
         vhsub.s16       \r1, \d1, \d0
 
25
         vhadd.s16       \d0, \d0, \d1           @ t1, t2, t6, t5
 
26
-        vmov.i64        \d1, #0xffff<<32
 
27
+        vmov.i64        \d1, #0xffff00000000
 
28
         vbit            \r0, \r1, \d1
 
29
         vrev64.16       \r1, \r0                @ t7, t8, t4, t3
 
30
         vtrn.32         \r0, \r1                @ t3, t4, t7, t8
 
31
-- 
 
32
1.7.4.1
 
33