mathops.patch

From d7caa351e2dada0d493fd745bbd6a00f0e3f2d66 Mon Sep 17 00:00:00 2001
From: Aaron Harvey <10888871+aaron-harvey@users.noreply.github.com>
Date: Sun, 13 Oct 2024 23:55:13 -0400
Subject: [PATCH] patch mathops

---
 libavcodec/x86/mathops.h | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/libavcodec/x86/mathops.h b/libavcodec/x86/mathops.h
index 6298f5ed19..9411da2184 100644
--- a/libavcodec/x86/mathops.h
+++ b/libavcodec/x86/mathops.h
@@ -35,12 +35,20 @@
 static av_always_inline av_const int MULL(int a, int b, unsigned shift)
 {
     int rt, dummy;
+    if (__builtin_constant_p(shift))
     __asm__ (
         "imull %3               \n\t"
         "shrdl %4, %%edx, %%eax \n\t"
         :"=a"(rt), "=d"(dummy)
-        :"a"(a), "rm"(b), "ci"((uint8_t)shift)
+        :"a"(a), "rm"(b), "i"(shift & 0x1F)
     );
+    else
+        __asm__ (
+            "imull %3               \n\t"
+            "shrdl %4, %%edx, %%eax \n\t"
+            :"=a"(rt), "=d"(dummy)
+            :"a"(a), "rm"(b), "c"((uint8_t)shift)
+        );
     return rt;
 }
 
@@ -113,19 +121,31 @@ __asm__ volatile(\
 // avoid +32 for shift optimization (gcc should do that ...)
 #define NEG_SSR32 NEG_SSR32
 static inline  int32_t NEG_SSR32( int32_t a, int8_t s){
+    if (__builtin_constant_p(s))
     __asm__ ("sarl %1, %0\n\t"
          : "+r" (a)
-         : "ic" ((uint8_t)(-s))
+         : "i" (-s & 0x1F)
     );
+    else
+        __asm__ ("sarl %1, %0\n\t"
+               : "+r" (a)
+               : "c" ((uint8_t)(-s))
+        );
     return a;
 }
 
 #define NEG_USR32 NEG_USR32
 static inline uint32_t NEG_USR32(uint32_t a, int8_t s){
+    if (__builtin_constant_p(s))
     __asm__ ("shrl %1, %0\n\t"
          : "+r" (a)
-         : "ic" ((uint8_t)(-s))
+         : "i" (-s & 0x1F)
     );
+    else
+        __asm__ ("sarl %1, %0\n\t"
+               : "+r" (a)
+               : "c" ((uint8_t)(-s))
+        );
     return a;
 }
 
-- 
2.47.1