diff --git a/libswscale/x86/output.asm b/libswscale/x86/output.asm
index 3b8e56d2bd..646a11dbab 100644
--- a/libswscale/x86/output.asm
+++ b/libswscale/x86/output.asm
@@ -59,7 +59,7 @@ SECTION .text
 %macro yuv2planeX_fn 3
 
 %if ARCH_X86_32
-%define cntr_reg filterq
+%define cntr_reg fltsizeq
 %define movsx mov
 %else
 %define cntr_reg r11
@@ -146,7 +146,7 @@ cglobal yuv2planeX_%1, %3, 7, %2, filter, fltsize, src, dst, w, dither, offset
     mova            m1, [yuv2yuvX_%1_start]
     mova            m2,  m1
 %endif ; %1 == 8/9/10/16
-    movsx     cntr_reg,  fltsizem
+    movsx     cntr_reg,  r1m ; FIXME should be fltsizem, but the assembler does the wrong thing b/c of SUB above
 .filterloop_ %+ %%i:
     ; input pixels
     mov             r6, [srcq+gprsize*cntr_reg-2*gprsize]
diff --git a/tests/fate/image.mak b/tests/fate/image.mak
index f5b648e3b9..60c6cb26ef 100644
--- a/tests/fate/image.mak
+++ b/tests/fate/image.mak
@@ -16,6 +16,9 @@ fate-sunraster-1bit-rle: CMD = framecrc -i $(SAMPLES)/sunraster/lena-1bit-rle.su
 FATE_SUNRASTER += fate-sunraster-8bit-raw
 fate-sunraster-8bit-raw: CMD = framecrc -i $(SAMPLES)/sunraster/lena-8bit-raw.sun -pix_fmt rgb24
 
+FATE_SUNRASTER += fate-sunraster-8bit_gray-raw
+fate-sunraster-8bit_gray-raw: CMD = framecrc -i $(SAMPLES)/sunraster/gray.ras
+
 FATE_SUNRASTER += fate-sunraster-8bit-rle
 fate-sunraster-8bit-rle: CMD = framecrc -i $(SAMPLES)/sunraster/lena-8bit-rle.sun -pix_fmt rgb24
 
diff --git a/tests/ref/fate/sunraster-8bit_gray-raw b/tests/ref/fate/sunraster-8bit_gray-raw
new file mode 100644
index 0000000000..455af5f612
--- /dev/null
+++ b/tests/ref/fate/sunraster-8bit_gray-raw
@@ -0,0 +1,2 @@
+#tb 0: 1/25
+0,          0,          0,        1,   786432, 0x5859b48e