mips/acelp_filters: fix incorrect register constraint

Change register constraint on the v variable from = to +. This was causing GCC
to think that the v variable was never read and therefore not initialize it.

This fixes about 20 fate failures on mips64el.

Signed-off-by: James Cowgill <james410@cowgill.org.uk>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit b9de1303a6414174ab2f3bccefa801bfabcf0f88)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

diff --git a/libavcodec/mips/acelp_filters_mips.c b/libavcodec/mips/acelp_filters_mips.c
index c8d980a..ffc0fe6 100644 (file)
--- a/libavcodec/mips/acelp_filters_mips.c
+++ b/libavcodec/mips/acelp_filters_mips.c
@@ -89,7 +89,7 @@ static void ff_acelp_interpolatef_mips(float *out, const float *in,
                 "addu   %[p_filter_coeffs_m],  %[p_filter_coeffs_m],   %[prec] \n\t"
                 "madd.s %[v],%[v],%[in_val_m], %[fc_val_m]                     \n\t"
 
-                : [v] "=&f" (v),[p_in_p] "+r" (p_in_p), [p_in_m] "+r" (p_in_m),
+                : [v] "+&f" (v),[p_in_p] "+r" (p_in_p), [p_in_m] "+r" (p_in_m),
                   [p_filter_coeffs_p] "+r" (p_filter_coeffs_p),
                   [in_val_p] "=&f" (in_val_p), [in_val_m] "=&f" (in_val_m),
                   [fc_val_p] "=&f" (fc_val_p), [fc_val_m] "=&f" (fc_val_m),