pixman-arm: Use unified syntax

Allows us to use the same assembly without a bunch of #ifdef __clang__.

8 files changed, 74 insertions, 105 deletions

diff --git a/meson.build b/meson.build
index 4337f93..438e6cf 100644
--- a/meson.build
+++ b/meson.build
@@ -252,6 +252,13 @@ if cc.compiles('''
     config.set('ASM_HAVE_FUNC_DIRECTIVE', 1)    
 endif
 
+if cc.compiles('''
+    __asm__ (
+    ".syntax unified\n"
+    );''',
+    name : 'test for ASM .syntax unified directive')
+    config.set('ASM_HAVE_SYNTAX_UNIFIED', 1)
+endif
 
 if cc.links('''
     #include <stdint.h>
diff --git a/pixman/pixman-arm-asm.h b/pixman/pixman-arm-asm.h
index 8253906..edf8e82 100644
--- a/pixman/pixman-arm-asm.h
+++ b/pixman/pixman-arm-asm.h
@@ -50,6 +50,12 @@
 #endif
 .endm
 
+.macro pixman_syntax_unified
+#ifdef ASM_HAVE_SYNTAX_UNIFIED
+	.syntax unified
+#endif
+.endm
+
 .macro pixman_end_asm_function
 #ifdef ASM_HAVE_FUNC_DIRECTIVE
 	.endfunc
diff --git a/pixman/pixman-arm-neon-asm-bilinear.S b/pixman/pixman-arm-neon-asm-bilinear.S
index ce4d5f8..6bd2736 100644
--- a/pixman/pixman-arm-neon-asm-bilinear.S
+++ b/pixman/pixman-arm-neon-asm-bilinear.S
@@ -68,6 +68,8 @@
 #include "pixman-arm-asm.h"
 #include "pixman-arm-neon-asm.h"
 
+pixman_syntax_unified
+
 /*
  * Bilinear macros from pixman-arm-neon-asm.S
  */
diff --git a/pixman/pixman-arm-neon-asm.S b/pixman/pixman-arm-neon-asm.S
index 7025eba..0e09257 100644
--- a/pixman/pixman-arm-neon-asm.S
+++ b/pixman/pixman-arm-neon-asm.S
@@ -34,12 +34,6 @@
  *  - pixman_composite_over_n_8_0565_asm_neon
  */
 
-#ifdef __clang__
-#define ldrgeb ldrbge
-#define subges subsge
-#define subpls subspl
-#endif
-
 /* Prevent the stack from becoming executable for no reason... */
 #if defined(__linux__) && defined(__ELF__)
 .section .note.GNU-stack,"",%progbits
@@ -59,6 +53,8 @@
 #include "pixman-arm-asm.h"
 #include "pixman-arm-neon-asm.h"
 
+    pixman_syntax_unified
+
 /* Global configuration options and preferences */
 
 /*
@@ -287,12 +283,12 @@
                                     PF subge, PF_X, PF_X, ORIG_W
     vrshr.u16   q3, q11, #8
     vrshr.u16   q15, q12, #8
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
         vsri.u16    q14, q9, #11
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
     vraddhn.u16 d20, q10, q13
     vraddhn.u16 d23, q11, q3
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vraddhn.u16 d22, q12, q15
         vst1.16     {d28, d29}, [DST_W, :128]!
 .endm
@@ -451,9 +447,9 @@ generate_composite_function \
     vshll.u8    q8, d1, #8
         vst1.16     {d28, d29}, [DST_W, :128]!
                                     PF subge, PF_X, PF_X, ORIG_W
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vshll.u8    q14, d2, #8
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
     vshll.u8    q9, d0, #8
 .endm
 
@@ -525,10 +521,10 @@ generate_composite_function \
                                     PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift]
                                     PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift]
                                     PF subge, PF_X, PF_X, ORIG_W
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vqadd.u8    q14, q0, q2
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vqadd.u8    q15, q1, q3
 .endm
 
@@ -557,10 +553,10 @@ generate_composite_function \
                                     PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift]
                                     PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift]
                                     PF subge, PF_X, PF_X, ORIG_W
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vqadd.u8    q14, q0, q2
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vqadd.u8    q15, q1, q3
 .endm
 
@@ -631,9 +627,9 @@ generate_composite_function_single_scanline \
     vmull.u8    q8, d22, d4
                                     PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q9, d22, d5
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
     vmull.u8    q10, d22, d6
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vmull.u8    q11, d22, d7
 .endm
 
@@ -683,11 +679,11 @@ generate_composite_function_single_scanline \
         vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q8, d22, d4
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q9, d22, d5
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
     vmull.u8    q10, d22, d6
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vmull.u8    q11, d22, d7
 .endm
 
@@ -759,9 +755,9 @@ generate_composite_function_single_scanline \
     vmull.u8    q9, d24, d5
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q10, d24, d6
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q11, d24, d7
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
         vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
 .endm
 
@@ -810,10 +806,10 @@ generate_composite_function \
         vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q8, d22, d4
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q9, d22, d5
     vmull.u8    q10, d22, d6
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
     vmull.u8    q11, d22, d7
 .endm
 
@@ -1265,9 +1261,9 @@ generate_composite_function \
     vmull.u8    q9, d24, d1
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q10, d24, d2
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q11, d24, d3
-                                    PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
         vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
     vrsra.u16   q8, q8, #8
     vrsra.u16   q9, q9, #8
@@ -1334,9 +1330,9 @@ generate_composite_function \
     vmull.u8    q1,  d25, d16
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q2,  d26, d16
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q3,  d27, d16
-                                    PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
         vst1.8      {d28, d29, d30, d31}, [DST_W, :128]!
     vrsra.u16   q0, q0,  #8
     vrsra.u16   q1, q1,  #8
@@ -1430,11 +1426,11 @@ generate_composite_function \
     vmull.u8    q7, d24, d9
                                     PF subge, PF_X, PF_X, ORIG_W
     vmull.u8    q8, d24, d10
-                                    PF subges, PF_CTL, PF_CTL, #0x10
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
     vmull.u8    q9, d24, d11
-                                    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
         vqadd.u8    q14, q0, q14
-                                    PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
+                                    PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
         vqadd.u8    q15, q1, q15
     vrshr.u16   q10, q6, #8
     vrshr.u16   q11, q7, #8
@@ -2444,8 +2440,8 @@ generate_composite_function \
                                     PF cmp, PF_X, ORIG_W
                                     PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift]
                                     PF subge, PF_X, PF_X, ORIG_W
-                                    PF subges, PF_CTL, PF_CTL, #0x10
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
 .endm
 
 generate_composite_function \
@@ -2501,8 +2497,8 @@ generate_composite_function \
                                     PF cmp, PF_X, ORIG_W
                                     PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift]
                                     PF subge, PF_X, PF_X, ORIG_W
-                                    PF subges, PF_CTL, PF_CTL, #0x10
-                                    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+                                    PF subsge, PF_CTL, PF_CTL, #0x10
+                                    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
 .endm
 
 generate_composite_function \
diff --git a/pixman/pixman-arm-neon-asm.h b/pixman/pixman-arm-neon-asm.h
index e85526d..06318d9 100644
--- a/pixman/pixman-arm-neon-asm.h
+++ b/pixman/pixman-arm-neon-asm.h
@@ -213,24 +213,24 @@
 .if \elem_size == 16
     mov     TMP1, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP1, \mem_operand, TMP1, asl #1
     mov     TMP2, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP2, \mem_operand, TMP2, asl #1
     vld1.16 {d\()\reg1\()[0]}, [TMP1, :16]
     mov     TMP1, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP1, \mem_operand, TMP1, asl #1
     vld1.16 {d\()\reg1\()[1]}, [TMP2, :16]
     mov     TMP2, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP2, \mem_operand, TMP2, asl #1
     vld1.16 {d\()\reg1\()[2]}, [TMP1, :16]
@@ -238,12 +238,12 @@
 .elseif \elem_size == 32
     mov     TMP1, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP1, \mem_operand, TMP1, asl #2
     mov     TMP2, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP2, \mem_operand, TMP2, asl #2
     vld1.32 {d\()\reg1\()[0]}, [TMP1, :32]
@@ -281,14 +281,14 @@
 .if \elem_size == 16
     mov     TMP1, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP1, \mem_operand, TMP1, asl #1
     vld1.16 {d\()\reg1\()[\idx]}, [TMP1, :16]
 .elseif \elem_size == 32
     mov     TMP1, VX, asr #16
     adds    VX, VX, UNIT_X
-5:  subpls  VX, VX, SRC_WIDTH_FIXED
+5:  subspl  VX, VX, SRC_WIDTH_FIXED
     bpl     5b
     add     TMP1, \mem_operand, TMP1, asl #2
     vld1.32 {d\()\reg1\()[\idx]}, [TMP1, :32]
@@ -420,15 +420,15 @@
     PF pld, [PF_MASK, PF_X, lsl #mask_bpp_shift]
 .endif
     PF subge, PF_X, PF_X, ORIG_W
-    PF subges, PF_CTL, PF_CTL, #0x10
+    PF subsge, PF_CTL, PF_CTL, #0x10
 .if src_bpp_shift >= 0
-    PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
+    PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]!
 .endif
 .if dst_r_bpp != 0
-    PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+    PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
 .endif
 .if mask_bpp_shift >= 0
-    PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
+    PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
 .endif
 .endif
 .endm
diff --git a/pixman/pixman-arm-simd-asm-scaled.S b/pixman/pixman-arm-simd-asm-scaled.S
index 25445b8..cc62c81 100644
--- a/pixman/pixman-arm-simd-asm-scaled.S
+++ b/pixman/pixman-arm-simd-asm-scaled.S
@@ -25,10 +25,6 @@
  *
  */
 
-#ifdef __clang__
-#define subpls subspl
-#endif
-
 /* Prevent the stack from becoming executable */
 #if defined(__linux__) && defined(__ELF__)
 .section .note.GNU-stack,"",%progbits
@@ -43,6 +39,8 @@
 
 #include "pixman-arm-asm.h"
 
+	pixman_syntax_unified
+
 /*
  * Note: This code is only using armv5te instructions (not even armv6),
  *       but is scheduled for ARM Cortex-A8 pipeline. So it might need to
@@ -89,21 +87,21 @@ pixman_asm_function \fname
 		and	TMP2, VXMASK, VX, asr #(16 - \bpp_shift)
 		adds	VX, VX, UNIT_X
 		str\()\t	TMP1, [DST], #(1 << \bpp_shift)
-9:		subpls	VX, VX, SRC_WIDTH_FIXED
+9:		subspl	VX, VX, SRC_WIDTH_FIXED
 		bpl	9b
 
 		ldr\()\t	TMP2, [SRC, TMP2]
 		and	TMP1, VXMASK, VX, asr #(16 - \bpp_shift)
 		adds	VX, VX, UNIT_X
 		str\()\t	TMP2, [DST], #(1 << \bpp_shift)
-9:		subpls	VX, VX, SRC_WIDTH_FIXED
+9:		subspl	VX, VX, SRC_WIDTH_FIXED
 		bpl	9b
 	.endm
 
 	/* now do the scaling */
 	and	TMP1, VXMASK, VX, asr #(16 - \bpp_shift)
 	adds	VX, VX, UNIT_X
-9:	subpls	VX, VX, SRC_WIDTH_FIXED
+9:	subspl	VX, VX, SRC_WIDTH_FIXED
 	bpl	9b
 	subs	W, W, #(8 + \prefetch_braking_distance)
 	blt	2f
@@ -112,7 +110,7 @@ pixman_asm_function \fname
 	mla	PF_OFFS, UNIT_X, PF_OFFS, VX
 1:	/* main loop, process 8 pixels per iteration with prefetch */
 	pld	[SRC, PF_OFFS, asr #(16 - \bpp_shift)]
-	add	PF_OFFS, UNIT_X, lsl #3
+	add	PF_OFFS, PF_OFFS, UNIT_X, lsl #3
 	scale_2_pixels
 	scale_2_pixels
 	scale_2_pixels
@@ -133,13 +131,8 @@ pixman_asm_function \fname
 	scale_2_pixels
 2:
 	tst	W, #1
-#ifdef __clang__
 	ldr\()\t\()ne	TMP1, [SRC, TMP1]
 	str\()\t\()ne	TMP1, [DST]
-#else
-	ldrne\()\t	TMP1, [SRC, TMP1]
-	strne\()\t	TMP1, [DST]
-#endif
 	/* cleanup helper macro */
 	.purgem	scale_2_pixels
 	.unreq	DST
diff --git a/pixman/pixman-arm-simd-asm.S b/pixman/pixman-arm-simd-asm.S
index 0c93ef4..34d38f1 100644
--- a/pixman/pixman-arm-simd-asm.S
+++ b/pixman/pixman-arm-simd-asm.S
@@ -25,11 +25,6 @@
  *
  */
 
-#ifdef __clang__
-#define adceqs adcseq
-#define ldmnedb ldmdbne
-#endif
-
 /* Prevent the stack from becoming executable */
 #if defined(__linux__) && defined(__ELF__)
 .section .note.GNU-stack,"",%progbits
@@ -45,6 +40,8 @@
 #include "pixman-arm-asm.h"
 #include "pixman-arm-simd-asm.h"
 
+	pixman_syntax_unified
+
 /* A head macro should do all processing which results in an output of up to
  * 16 bytes, as far as the final load instruction. The corresponding tail macro
  * should complete the processing of the up-to-16 bytes. The calling macro will
@@ -127,7 +124,7 @@ generate_composite_function \
 
 .macro src_n_0565_init
         ldrh    SRC, [sp, #ARGS_STACK_OFFSET]
-        orr     SRC, SRC, lsl #16
+        orr     SRC, SRC, SRC, lsl #16
         mov     STRIDE_S, SRC
         mov     MASK, SRC
         mov     STRIDE_M, SRC
@@ -135,8 +132,8 @@ generate_composite_function \
 
 .macro src_n_8_init
         ldrb    SRC, [sp, #ARGS_STACK_OFFSET]
-        orr     SRC, SRC, lsl #8
-        orr     SRC, SRC, lsl #16
+        orr     SRC, SRC, SRC, lsl #8
+        orr     SRC, SRC, SRC, lsl #16
         mov     STRIDE_S, SRC
         mov     MASK, SRC
         mov     STRIDE_M, SRC
@@ -1098,13 +1095,13 @@ generate_composite_function \
  .elseif \numbytes == 8
         teq     ORIG_W, WK\()\reg1
         teqeq   ORIG_W, ORIG_W, asr #32  /* all 0 or all -1? */
-        ldmnedb DST, {WK\()\reg1-WK\()\reg2}
+        ldmdbne DST, {WK\()\reg1-WK\()\reg2}
  .else
         teq     ORIG_W, WK\()\reg1
         teqeq   ORIG_W, WK\()\reg2
         teqeq   ORIG_W, WK\()\reg3
         teqeq   ORIG_W, ORIG_W, asr #32  /* all 0 or all -1? */
-        ldmnedb DST, {WK\()\reg1-WK\()\reg4}
+        ldmdbne DST, {WK\()\reg1-WK\()\reg4}
  .endif
         cmnne   DST, #0   /* clear C if NE */
         bcs     49f       /* no writes to dest if source all -1 */
diff --git a/pixman/pixman-arm-simd-asm.h b/pixman/pixman-arm-simd-asm.h
index 3e78e8a..5ec19e0 100644
--- a/pixman/pixman-arm-simd-asm.h
+++ b/pixman/pixman-arm-simd-asm.h
@@ -119,37 +119,21 @@
         \op\()r\()\cond    WK\()\reg2, [\base], #4
         \op\()r\()\cond    WK\()\reg3, [\base], #4
   .else
-#ifdef __clang__
         \op\()mia\()\cond  \base!, {WK\()\reg0,WK\()\reg1,WK\()\reg2,WK\()\reg3}
-#else
-        \op\()m\()\cond\()ia  \base!, {WK\()\reg0,WK\()\reg1,WK\()\reg2,WK\()\reg3}
-#endif
   .endif
  .elseif \numbytes == 8
   .if \unaligned == 1
         \op\()r\()\cond    WK\()\reg0, [\base], #4
         \op\()r\()\cond    WK\()\reg1, [\base], #4
   .else
-#ifdef __clang__
         \op\()mia\()\cond  \base!, {WK\()\reg0,WK\()\reg1}
-#else
-        \op\()m\()\cond\()ia  \base!, {WK\()\reg0,WK\()\reg1}
-#endif
   .endif
  .elseif \numbytes == 4
         \op\()r\()\cond    WK\()\reg0, [\base], #4
  .elseif \numbytes == 2
-#ifdef __clang__
         \op\()rh\()\cond   WK\()\reg0, [\base], #2
-#else
-        \op\()r\()\cond\()h   WK\()\reg0, [\base], #2
-#endif
  .elseif \numbytes == 1
-#ifdef __clang__
         \op\()rb\()\cond   WK\()\reg0, [\base], #1
-#else
-        \op\()r\()\cond\()b   WK\()\reg0, [\base], #1
-#endif
  .else
   .error "unsupported size: \numbytes"
  .endif
@@ -157,31 +141,15 @@
 
 .macro pixst_baseupdated cond, numbytes, reg0, reg1, reg2, reg3, base
  .if \numbytes == 16
-#ifdef __clang__
         stm\()\cond\()db \base, {WK\()\reg0,WK\()\reg1,WK\()\reg2,WK\()\reg3}
-#else
-        stmdb\()\cond \base, {WK\()\reg0,WK\()\reg1,WK\()\reg2,WK\()\reg3}
-#endif
  .elseif \numbytes == 8
-#ifdef __clang__
         stmdb\()\cond \base, {WK\()\reg0,WK\()\reg1}
-#else
-        stm\()\cond\()db \base, {WK\()\reg0,WK\()\reg1}
-#endif
  .elseif \numbytes == 4
         str\()\cond    WK\()\reg0, [\base, #-4]
  .elseif \numbytes == 2
-#ifdef __clang__
         strh\()\cond   WK\()\reg0, [\base, #-2]
-#else
-        str\()\cond\()h   WK\()\reg0, [\base, #-2]
-#endif
  .elseif \numbytes == 1
-#ifdef __clang__
         strb\()\cond   WK\()\reg0, [\base, #-1]
-#else
-        str\()\cond\()b   WK\()\reg0, [\base, #-1]
-#endif
  .else
   .error "unsupported size: \numbytes"
  .endif
@@ -291,7 +259,7 @@
         /* If exactly one fetch per block, then we need either 0, 1 or 2 extra preloads */
         PF  mov,    SCRATCH, \base, lsl #32-5
         PF  adds,   SCRATCH, SCRATCH, X, lsl #32-5+\bpp_shift
-        PF  adceqs, SCRATCH, SCRATCH, #0
+        PF  adcseq, SCRATCH, SCRATCH, #0
         /* The instruction above has two effects: ensures Z is only
          * set if C was clear (so Z indicates that both shifted quantities
          * were 0), and clears C if Z was set (so C indicates that the sum