[FFmpeg-devel] [PATCH 2/4] lavu/riscv: use Zbb REV8 at run-time

[Rémi Denis-Courmont]

This adds runtime support to use Zbb REV8 for 32- and 64-bit byte-wise
swaps. The result is about five times slower than if targetting Zbb
statically, but still a lot faster than the default bespoke C code or a
call to GCC run-time functions.

For 16-bit swap, this is however unsurprisingly a lot worse, and so this
sticks to the baseline. In fact, even using REV8 statically does not
seem to be beneficial in that case.

         Zbb static    Zbb dynamic   I baseline
bswap16:  0.668184765   3.340764069   0.668029012
bswap32:  0.668174014   3.340763319   9.353855435
bswap64:  0.668221765   3.340496313  14.698672283
(seconds for 1 billion iterations on a SiFive-U74 core)
---
 libavutil/riscv/bswap.h | 44 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 42 insertions(+), 2 deletions(-)

diff --git a/libavutil/riscv/bswap.h b/libavutil/riscv/bswap.h
index ce75de974e..886893e241 100644
--- a/libavutil/riscv/bswap.h
+++ b/libavutil/riscv/bswap.h
@@ -22,11 +22,51 @@
 #include <stdint.h>
 #include "config.h"
 #include "libavutil/attributes.h"
+#include "libavutil/riscv/cpu.h"
 
 #if defined (__GNUC__) || defined (__clang__)
 #define av_bswap16 __builtin_bswap16
-#define av_bswap32 __builtin_bswap32
-#define av_bswap64 __builtin_bswap64
+
+static av_always_inline av_const uint32_t av_bswap32_rv(uint32_t x)
+{
+#if HAVE_RV && !defined(__riscv_zbb)
+    if (!__builtin_constant_p(x) &&
+        __builtin_expect(ff_rv_zbb_support(), 1)) {
+        uintptr_t y;
+
+        __asm__ (
+            ".option push\n"
+            ".option arch, +zbb\n"
+            "rev8    %0, %1\n"
+            ".option pop" : "=r" (y) : "r" (x));
+        return y >> (__riscv_xlen - 32);
+    }
+#endif
+    return __builtin_bswap32(x);
+}
+#define av_bswap32 av_bswap32_rv
+
+#if __riscv_xlen >= 64
+static av_always_inline av_const uint64_t av_bswap64_rv(uint64_t x)
+{
+#if HAVE_RV && !defined(__riscv_zbb)
+    if (!__builtin_constant_p(x) &&
+        __builtin_expect(ff_rv_zbb_support(), 1)) {
+        uintptr_t y;
+
+        __asm__ (
+            ".option push\n"
+            ".option arch, +zbb\n"
+            "rev8    %0, %1\n"
+            ".option pop" : "=r" (y) : "r" (x));
+        return y >> (__riscv_xlen - 64);
+    }
+#endif
+    return __builtin_bswap64(x);
+}
+#define av_bswap64 av_bswap64_rv
+#endif
+
 #endif
 
 #endif /* AVUTIL_RISCV_BSWAP_H */
-- 
2.45.1