Attachment 19720 Details for Bug 18106 – rename the current ARM code to ARM SIMD

[patch] rename the current ARM code to ARM SIMD

patch-pixman-arm-simd (text/plain), 31.24 KB, created by Adrian Bunk on 2008-10-17 07:42:35 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Adrian Bunk

Created: 2008-10-17 07:42:35 UTC

Size: 31.24 KB

patch

obsolete

>
>
> configure.ac                               |   32 ++++++++++++++--------------
> pixman/Makefile.am                         |   16 +++++++-------
> pixman/{pixman-arm.c => pixman-arm-simd.c} |    2 +-
> pixman/{pixman-arm.h => pixman-arm-simd.h} |    8 +++---
> pixman/pixman-pict.c                       |   12 +++++-----
> 5 files changed, 35 insertions(+), 35 deletions(-)
>
>diff --git a/configure.ac b/configure.ac
>index dd095ff..7937f95 100644
>--- a/configure.ac
>+++ b/configure.ac
>@@ -277,42 +277,42 @@ AC_SUBST(VMX_CFLAGS)
> 
> AM_CONDITIONAL(USE_VMX, test $have_vmx_intrinsics = yes)
> 
>-dnl Check for ARM
>+dnl Check for ARM SIMD instructions
> 
>-have_armv6_simd=no
>-AC_MSG_CHECKING(whether to use ARM assembler)
>+have_arm_simd=no
>+AC_MSG_CHECKING(whether to use ARM SIMD assembler)
> xserver_save_CFLAGS=$CFLAGS
> CFLAGS="$CFLAGS $ARM_CFLAGS"
> AC_COMPILE_IFELSE([
> int main () {
>     asm("uqadd8 r1, r1, r2");
>     return 0;
>-}], have_armv6_simd=yes)
>+}], have_arm_simd=yes)
> CFLAGS=$xserver_save_CFLAGS
> 
>-AC_ARG_ENABLE(arm,
>-   [AC_HELP_STRING([--disable-arm],
>-                   [disable ARM fast paths])],
>-   [enable_arm=$enableval], [enable_arm=auto])
>+AC_ARG_ENABLE(arm-simd,
>+   [AC_HELP_STRING([--disable-arm-simd],
>+                   [disable ARM SIMD fast paths])],
>+   [enable_arm_simd=$enableval], [enable_arm_simd=auto])
> 
>-if test $enable_arm = no ; then
>-   have_armv6_simd=disabled
>+if test $enable_arm_simd = no ; then
>+   have_arm_simd=disabled
> fi
> 
>-if test $have_armv6_simd = yes ; then
>-   AC_DEFINE(USE_ARM, 1, [use ARM compiler intrinsics])
>+if test $have_arm_simd = yes ; then
>+   AC_DEFINE(USE_ARM_SIMD, 1, [use ARM SIMD compiler intrinsics])
> else
>    ARM_CFLAGS=
> fi
> 
>-AC_MSG_RESULT($have_armv6_simd)
>-if test $enable_arm = yes && test $have_armv6_simd = no ; then
>-   AC_MSG_ERROR([ARM intrinsics not detected])
>+AC_MSG_RESULT($have_arm_simd)
>+if test $enable_arm_simd = yes && test $have_arm_simd = no ; then
>+   AC_MSG_ERROR([ARM SIMD intrinsics not detected])
> fi
> 
> AC_SUBST(ARM_CFLAGS)
> 
>-AM_CONDITIONAL(USE_ARM, test $have_armv6_simd = yes)
>+AM_CONDITIONAL(USE_ARM_SIMD, test $have_arm_simd = yes)
> 
> 
> AC_ARG_ENABLE(gtk,
>diff --git a/pixman/Makefile.am b/pixman/Makefile.am
>index 339e01e..6d5a643 100644
>--- a/pixman/Makefile.am
>+++ b/pixman/Makefile.am
>@@ -80,14 +80,14 @@ libpixman_1_la_LIBADD += libpixman-sse2.la
> endif
> 
> # arm code
>-if USE_ARM
>-noinst_LTLIBRARIES += libpixman-arm.la
>-libpixman_arm_la_SOURCES = \
>-	pixman-arm.c \
>-	pixman-arm.h
>-libpixman_arm_la_CFLAGS = $(DEP_CFLAGS) $(ARM_CFLAGS)
>-libpixman_arm_la_LIBADD = $(DEP_LIBS)
>-libpixman_1_la_LIBADD += libpixman-arm.la
>+if USE_ARM_SIMD
>+noinst_LTLIBRARIES += libpixman-arm-simd.la
>+libpixman_arm_simd_la_SOURCES = \
>+	pixman-arm-simd.c \
>+	pixman-arm-simd.h
>+libpixman_arm_simd_la_CFLAGS = $(DEP_CFLAGS) $(ARM_CFLAGS)
>+libpixman_arm_simd_la_LIBADD = $(DEP_LIBS)
>+libpixman_1_la_LIBADD += libpixman-arm-simd.la
> endif
> 
> 
>diff --git a/pixman/pixman-arm-simd.c b/pixman/pixman-arm-simd.c
>new file mode 100644
>index 0000000..c7851cb
>--- /dev/null
>+++ b/pixman/pixman-arm-simd.c
>@@ -0,0 +1,407 @@
>+/*
>+ * Copyright Â© 2008 Mozilla Corporation
>+ *
>+ * Permission to use, copy, modify, distribute, and sell this software and its
>+ * documentation for any purpose is hereby granted without fee, provided that
>+ * the above copyright notice appear in all copies and that both that
>+ * copyright notice and this permission notice appear in supporting
>+ * documentation, and that the name of Mozilla Corporation not be used in
>+ * advertising or publicity pertaining to distribution of the software without
>+ * specific, written prior permission.  Mozilla Corporation makes no
>+ * representations about the suitability of this software for any purpose.  It
>+ * is provided "as is" without express or implied warranty.
>+ *
>+ * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS
>+ * SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
>+ * FITNESS, IN NO EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY
>+ * SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
>+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
>+ * AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
>+ * OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
>+ * SOFTWARE.
>+ *
>+ * Author:  Jeff Muizelaar (jeff@infidigm.net)
>+ *
>+ */
>+#ifdef HAVE_CONFIG_H
>+#include <config.h>
>+#endif
>+
>+#include "pixman-arm-simd.h"
>+
>+void
>+fbCompositeSrcAdd_8000x8000arm (pixman_op_t op,
>+				pixman_image_t * pSrc,
>+				pixman_image_t * pMask,
>+				pixman_image_t * pDst,
>+				int16_t      xSrc,
>+				int16_t      ySrc,
>+				int16_t      xMask,
>+				int16_t      yMask,
>+				int16_t      xDst,
>+				int16_t      yDst,
>+				uint16_t     width,
>+				uint16_t     height)
>+{
>+    uint8_t	*dstLine, *dst;
>+    uint8_t	*srcLine, *src;
>+    int	dstStride, srcStride;
>+    uint16_t	w;
>+    uint8_t	s, d;
>+
>+    fbComposeGetStart (pSrc, xSrc, ySrc, uint8_t, srcStride, srcLine, 1);
>+    fbComposeGetStart (pDst, xDst, yDst, uint8_t, dstStride, dstLine, 1);
>+
>+    while (height--)
>+    {
>+	dst = dstLine;
>+	dstLine += dstStride;
>+	src = srcLine;
>+	srcLine += srcStride;
>+	w = width;
>+
>+	while (w && (unsigned long)dst & 3)
>+	{
>+	    s = *src;
>+	    d = *dst;
>+	    asm("uqadd8 %0, %1, %2" : "+r"(d) : "r"(s));
>+	    *dst = d;
>+
>+	    dst++;
>+	    src++;
>+	    w--;
>+	}
>+
>+	while (w >= 4)
>+	{
>+	    asm("uqadd8 %0, %1, %2" : "=r"(*(uint32_t*)dst) : "r"(*(uint32_t*)src), "r"(*(uint32_t*)dst));
>+	    dst += 4;
>+	    src += 4;
>+	    w -= 4;
>+	}
>+
>+	while (w)
>+	{
>+	    s = *src;
>+	    d = *dst;
>+	    asm("uqadd8 %0, %1, %2" : "+r"(d) : "r"(s));
>+	    *dst = d;
>+
>+	    dst++;
>+	    src++;
>+	    w--;
>+	}
>+    }
>+
>+}
>+
>+void
>+fbCompositeSrc_8888x8888arm (pixman_op_t op,
>+			 pixman_image_t * pSrc,
>+			 pixman_image_t * pMask,
>+			 pixman_image_t * pDst,
>+			 int16_t      xSrc,
>+			 int16_t      ySrc,
>+			 int16_t      xMask,
>+			 int16_t      yMask,
>+			 int16_t      xDst,
>+			 int16_t      yDst,
>+			 uint16_t     width,
>+			 uint16_t     height)
>+{
>+    uint32_t	*dstLine, *dst;
>+    uint32_t	*srcLine, *src;
>+    int	dstStride, srcStride;
>+    uint16_t	w;
>+    uint32_t component_half = 0x800080;
>+    uint32_t upper_component_mask = 0xff00ff00;
>+    uint32_t alpha_mask = 0xff;
>+
>+    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>+    fbComposeGetStart (pSrc, xSrc, ySrc, uint32_t, srcStride, srcLine, 1);
>+
>+    while (height--)
>+    {
>+	dst = dstLine;
>+	dstLine += dstStride;
>+	src = srcLine;
>+	srcLine += srcStride;
>+	w = width;
>+
>+//#define inner_branch
>+	asm volatile (
>+			"cmp %[w], #0\n\t"
>+			"beq 2f\n\t"
>+			"1:\n\t"
>+			/* load src */
>+			"ldr r5, [%[src]], #4\n\t"
>+#ifdef inner_branch
>+			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>+			 * The 0x0 case also allows us to avoid doing an unecessary data
>+			 * write which is more valuable so we only check for that */
>+			"cmp r5, #0\n\t"
>+			"beq 3f\n\t"
>+
>+			/* = 255 - alpha */
>+			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>+
>+			"ldr r4, [%[dest]] \n\t"
>+
>+#else
>+			"ldr r4, [%[dest]] \n\t"
>+
>+			/* = 255 - alpha */
>+			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>+#endif
>+			"uxtb16 r6, r4\n\t"
>+			"uxtb16 r7, r4, ror #8\n\t"
>+
>+			/* multiply by 257 and divide by 65536 */
>+			"mla r6, r6, r8, %[component_half]\n\t"
>+			"mla r7, r7, r8, %[component_half]\n\t"
>+
>+			"uxtab16 r6, r6, r6, ror #8\n\t"
>+			"uxtab16 r7, r7, r7, ror #8\n\t"
>+
>+			/* recombine the 0xff00ff00 bytes of r6 and r7 */
>+			"and r7, %[upper_component_mask]\n\t"
>+			"uxtab16 r6, r7, r6, ror #8\n\t"
>+
>+			"uqadd8 r5, r6, r5\n\t"
>+
>+#ifdef inner_branch
>+			"3:\n\t"
>+
>+#endif
>+			"str r5, [%[dest]], #4\n\t"
>+			/* increment counter and jmp to top */
>+			"subs	%[w], %[w], #1\n\t"
>+			"bne	1b\n\t"
>+			"2:\n\t"
>+			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src)
>+			: [component_half] "r" (component_half), [upper_component_mask] "r" (upper_component_mask),
>+			  [alpha_mask] "r" (alpha_mask)
>+			: "r4", "r5", "r6", "r7", "r8", "cc", "memory"
>+			);
>+    }
>+}
>+
>+void
>+fbCompositeSrc_8888x8x8888arm (pixman_op_t op,
>+			       pixman_image_t * pSrc,
>+			       pixman_image_t * pMask,
>+			       pixman_image_t * pDst,
>+			       int16_t	xSrc,
>+			       int16_t	ySrc,
>+			       int16_t      xMask,
>+			       int16_t      yMask,
>+			       int16_t      xDst,
>+			       int16_t      yDst,
>+			       uint16_t     width,
>+			       uint16_t     height)
>+{
>+    uint32_t	*dstLine, *dst;
>+    uint32_t	*srcLine, *src;
>+    uint32_t	mask;
>+    int	dstStride, srcStride;
>+    uint16_t	w;
>+    uint32_t component_half = 0x800080;
>+    uint32_t alpha_mask = 0xff;
>+
>+    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>+    fbComposeGetStart (pSrc, xSrc, ySrc, uint32_t, srcStride, srcLine, 1);
>+
>+    fbComposeGetSolid (pMask, mask, pDst->bits.format);
>+    mask = (mask) >> 24;
>+
>+    while (height--)
>+    {
>+	dst = dstLine;
>+	dstLine += dstStride;
>+	src = srcLine;
>+	srcLine += srcStride;
>+	w = width;
>+
>+//#define inner_branch
>+	asm volatile (
>+			"cmp %[w], #0\n\t"
>+			"beq 2f\n\t"
>+			"1:\n\t"
>+			/* load src */
>+			"ldr r5, [%[src]], #4\n\t"
>+#ifdef inner_branch
>+			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>+			 * The 0x0 case also allows us to avoid doing an unecessary data
>+			 * write which is more valuable so we only check for that */
>+			"cmp r5, #0\n\t"
>+			"beq 3f\n\t"
>+
>+#endif
>+			"ldr r4, [%[dest]] \n\t"
>+
>+			"uxtb16 r6, r5\n\t"
>+			"uxtb16 r7, r5, ror #8\n\t"
>+
>+			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>+			"mla r6, r6, %[mask_alpha], %[component_half]\n\t"
>+			"mla r7, r7, %[mask_alpha], %[component_half]\n\t"
>+
>+			"uxtab16 r6, r6, r6, ror #8\n\t"
>+			"uxtab16 r7, r7, r7, ror #8\n\t"
>+
>+			"uxtb16 r6, r6, ror #8\n\t"
>+			"uxtb16 r7, r7, ror #8\n\t"
>+
>+			/* recombine */
>+			"orr r5, r6, r7, lsl #8\n\t"
>+
>+			"uxtb16 r6, r4\n\t"
>+			"uxtb16 r7, r4, ror #8\n\t"
>+
>+			/* 255 - alpha */
>+			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>+
>+			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>+			"mla r6, r6, r8, %[component_half]\n\t"
>+			"mla r7, r7, r8, %[component_half]\n\t"
>+
>+			"uxtab16 r6, r6, r6, ror #8\n\t"
>+			"uxtab16 r7, r7, r7, ror #8\n\t"
>+
>+			"uxtb16 r6, r6, ror #8\n\t"
>+			"uxtb16 r7, r7, ror #8\n\t"
>+
>+			/* recombine */
>+			"orr r6, r6, r7, lsl #8\n\t"
>+
>+			"uqadd8 r5, r6, r5\n\t"
>+
>+#ifdef inner_branch
>+			"3:\n\t"
>+
>+#endif
>+			"str r5, [%[dest]], #4\n\t"
>+			/* increment counter and jmp to top */
>+			"subs	%[w], %[w], #1\n\t"
>+			"bne	1b\n\t"
>+			"2:\n\t"
>+			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src)
>+			: [component_half] "r" (component_half), [mask_alpha] "r" (mask),
>+			  [alpha_mask] "r" (alpha_mask)
>+			: "r4", "r5", "r6", "r7", "r8", "r9", "cc", "memory"
>+			);
>+    }
>+}
>+
>+void
>+fbCompositeSolidMask_nx8x8888arm (pixman_op_t      op,
>+			       pixman_image_t * pSrc,
>+			       pixman_image_t * pMask,
>+			       pixman_image_t * pDst,
>+			       int16_t      xSrc,
>+			       int16_t      ySrc,
>+			       int16_t      xMask,
>+			       int16_t      yMask,
>+			       int16_t      xDst,
>+			       int16_t      yDst,
>+			       uint16_t     width,
>+			       uint16_t     height)
>+{
>+    uint32_t	 src, srca;
>+    uint32_t	*dstLine, *dst;
>+    uint8_t	*maskLine, *mask;
>+    int		 dstStride, maskStride;
>+    uint16_t	 w;
>+
>+    fbComposeGetSolid(pSrc, src, pDst->bits.format);
>+
>+    srca = src >> 24;
>+    if (src == 0)
>+	return;
>+
>+    uint32_t component_mask = 0xff00ff;
>+    uint32_t component_half = 0x800080;
>+
>+    uint32_t src_hi = (src >> 8) & component_mask;
>+    uint32_t src_lo = src & component_mask;
>+
>+    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>+    fbComposeGetStart (pMask, xMask, yMask, uint8_t, maskStride, maskLine, 1);
>+
>+    while (height--)
>+    {
>+	dst = dstLine;
>+	dstLine += dstStride;
>+	mask = maskLine;
>+	maskLine += maskStride;
>+	w = width;
>+
>+//#define inner_branch
>+	asm volatile (
>+			"cmp %[w], #0\n\t"
>+			"beq 2f\n\t"
>+			"1:\n\t"
>+			/* load mask */
>+			"ldrb r5, [%[mask]], #1\n\t"
>+#ifdef inner_branch
>+			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>+			 * The 0x0 case also allows us to avoid doing an unecessary data
>+			 * write which is more valuable so we only check for that */
>+			"cmp r5, #0\n\t"
>+			"beq 3f\n\t"
>+
>+#endif
>+			"ldr r4, [%[dest]] \n\t"
>+
>+			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>+			"mla r6, %[src_lo], r5, %[component_half]\n\t"
>+			"mla r7, %[src_hi], r5, %[component_half]\n\t"
>+
>+			"uxtab16 r6, r6, r6, ror #8\n\t"
>+			"uxtab16 r7, r7, r7, ror #8\n\t"
>+
>+			"uxtb16 r6, r6, ror #8\n\t"
>+			"uxtb16 r7, r7, ror #8\n\t"
>+
>+			/* recombine */
>+			"orr r5, r6, r7, lsl #8\n\t"
>+
>+			"uxtb16 r6, r4\n\t"
>+			"uxtb16 r7, r4, ror #8\n\t"
>+
>+			/* we could simplify this to use 'sub' if we were
>+			 * willing to give up a register for alpha_mask */
>+			"mvn r8, r5\n\t"
>+			"mov r8, r8, lsr #24\n\t"
>+
>+			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>+			"mla r6, r6, r8, %[component_half]\n\t"
>+			"mla r7, r7, r8, %[component_half]\n\t"
>+
>+			"uxtab16 r6, r6, r6, ror #8\n\t"
>+			"uxtab16 r7, r7, r7, ror #8\n\t"
>+
>+			"uxtb16 r6, r6, ror #8\n\t"
>+			"uxtb16 r7, r7, ror #8\n\t"
>+
>+			/* recombine */
>+			"orr r6, r6, r7, lsl #8\n\t"
>+
>+			"uqadd8 r5, r6, r5\n\t"
>+
>+#ifdef inner_branch
>+			"3:\n\t"
>+
>+#endif
>+			"str r5, [%[dest]], #4\n\t"
>+			/* increment counter and jmp to top */
>+			"subs	%[w], %[w], #1\n\t"
>+			"bne	1b\n\t"
>+			"2:\n\t"
>+			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src), [mask] "+r" (mask)
>+			: [component_half] "r" (component_half),
>+			  [src_hi] "r" (src_hi), [src_lo] "r" (src_lo)
>+			: "r4", "r5", "r6", "r7", "r8", "cc", "memory"
>+			);
>+    }
>+}
>diff --git a/pixman/pixman-arm-simd.h b/pixman/pixman-arm-simd.h
>new file mode 100644
>index 0000000..ecaace5
>--- /dev/null
>+++ b/pixman/pixman-arm-simd.h
>@@ -0,0 +1,94 @@
>+/*
>+ * Copyright Â© 2008 Mozilla Corporation
>+ *
>+ * Permission to use, copy, modify, distribute, and sell this software and its
>+ * documentation for any purpose is hereby granted without fee, provided that
>+ * the above copyright notice appear in all copies and that both that
>+ * copyright notice and this permission notice appear in supporting
>+ * documentation, and that the name of Mozilla Corporation not be used in
>+ * advertising or publicity pertaining to distribution of the software without
>+ * specific, written prior permission.  Mozilla Corporation makes no
>+ * representations about the suitability of this software for any purpose.  It
>+ * is provided "as is" without express or implied warranty.
>+ *
>+ * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS
>+ * SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
>+ * FITNESS, IN NO EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY
>+ * SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
>+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
>+ * AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
>+ * OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
>+ * SOFTWARE.
>+ *
>+ * Author:  Jeff Muizelaar (jeff@infidigm.net)
>+ *
>+ */
>+
>+#include "pixman-private.h"
>+
>+#ifdef USE_ARM_SIMD
>+
>+static inline pixman_bool_t pixman_have_arm_simd(void) { return TRUE; }
>+
>+#else
>+#define pixman_have_arm_simd() FALSE
>+#endif
>+
>+#ifdef USE_ARM_SIMD
>+
>+void
>+fbCompositeSrcAdd_8000x8000arm (pixman_op_t op,
>+				pixman_image_t * pSrc,
>+				pixman_image_t * pMask,
>+				pixman_image_t * pDst,
>+				int16_t      xSrc,
>+				int16_t      ySrc,
>+				int16_t      xMask,
>+				int16_t      yMask,
>+				int16_t      xDst,
>+				int16_t      yDst,
>+				uint16_t     width,
>+				uint16_t     height);
>+void
>+fbCompositeSrc_8888x8888arm (pixman_op_t op,
>+			 pixman_image_t * pSrc,
>+			 pixman_image_t * pMask,
>+			 pixman_image_t * pDst,
>+			 int16_t      xSrc,
>+			 int16_t      ySrc,
>+			 int16_t      xMask,
>+			 int16_t      yMask,
>+			 int16_t      xDst,
>+			 int16_t      yDst,
>+			 uint16_t     width,
>+			 uint16_t     height);
>+
>+void
>+fbCompositeSrc_8888x8x8888arm (pixman_op_t op,
>+			 pixman_image_t * pSrc,
>+			 pixman_image_t * pMask,
>+			 pixman_image_t * pDst,
>+			 int16_t      xSrc,
>+			 int16_t      ySrc,
>+			 int16_t      xMask,
>+			 int16_t      yMask,
>+			 int16_t      xDst,
>+			 int16_t      yDst,
>+			 uint16_t     width,
>+			 uint16_t     height);
>+void
>+fbCompositeSolidMask_nx8x8888arm (pixman_op_t op,
>+			 pixman_image_t * pSrc,
>+			 pixman_image_t * pMask,
>+			 pixman_image_t * pDst,
>+			 int16_t      xSrc,
>+			 int16_t      ySrc,
>+			 int16_t      xMask,
>+			 int16_t      yMask,
>+			 int16_t      xDst,
>+			 int16_t      yDst,
>+			 uint16_t     width,
>+			 uint16_t     height);
>+
>+
>+#endif /* USE_ARM */
>diff --git a/pixman/pixman-arm.c b/pixman/pixman-arm.c
>deleted file mode 100644
>index b73134c..0000000
>--- a/pixman/pixman-arm.c
>+++ /dev/null
>@@ -1,407 +0,0 @@
>-/*
>- * Copyright Â© 2008 Mozilla Corporation
>- *
>- * Permission to use, copy, modify, distribute, and sell this software and its
>- * documentation for any purpose is hereby granted without fee, provided that
>- * the above copyright notice appear in all copies and that both that
>- * copyright notice and this permission notice appear in supporting
>- * documentation, and that the name of Mozilla Corporation not be used in
>- * advertising or publicity pertaining to distribution of the software without
>- * specific, written prior permission.  Mozilla Corporation makes no
>- * representations about the suitability of this software for any purpose.  It
>- * is provided "as is" without express or implied warranty.
>- *
>- * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS
>- * SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
>- * FITNESS, IN NO EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY
>- * SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
>- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
>- * AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
>- * OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
>- * SOFTWARE.
>- *
>- * Author:  Jeff Muizelaar (jeff@infidigm.net)
>- *
>- */
>-#ifdef HAVE_CONFIG_H
>-#include <config.h>
>-#endif
>-
>-#include "pixman-arm.h"
>-
>-void
>-fbCompositeSrcAdd_8000x8000arm (pixman_op_t op,
>-				pixman_image_t * pSrc,
>-				pixman_image_t * pMask,
>-				pixman_image_t * pDst,
>-				int16_t      xSrc,
>-				int16_t      ySrc,
>-				int16_t      xMask,
>-				int16_t      yMask,
>-				int16_t      xDst,
>-				int16_t      yDst,
>-				uint16_t     width,
>-				uint16_t     height)
>-{
>-    uint8_t	*dstLine, *dst;
>-    uint8_t	*srcLine, *src;
>-    int	dstStride, srcStride;
>-    uint16_t	w;
>-    uint8_t	s, d;
>-
>-    fbComposeGetStart (pSrc, xSrc, ySrc, uint8_t, srcStride, srcLine, 1);
>-    fbComposeGetStart (pDst, xDst, yDst, uint8_t, dstStride, dstLine, 1);
>-
>-    while (height--)
>-    {
>-	dst = dstLine;
>-	dstLine += dstStride;
>-	src = srcLine;
>-	srcLine += srcStride;
>-	w = width;
>-
>-	while (w && (unsigned long)dst & 3)
>-	{
>-	    s = *src;
>-	    d = *dst;
>-	    asm("uqadd8 %0, %1, %2" : "+r"(d) : "r"(s));
>-	    *dst = d;
>-
>-	    dst++;
>-	    src++;
>-	    w--;
>-	}
>-
>-	while (w >= 4)
>-	{
>-	    asm("uqadd8 %0, %1, %2" : "=r"(*(uint32_t*)dst) : "r"(*(uint32_t*)src), "r"(*(uint32_t*)dst));
>-	    dst += 4;
>-	    src += 4;
>-	    w -= 4;
>-	}
>-
>-	while (w)
>-	{
>-	    s = *src;
>-	    d = *dst;
>-	    asm("uqadd8 %0, %1, %2" : "+r"(d) : "r"(s));
>-	    *dst = d;
>-
>-	    dst++;
>-	    src++;
>-	    w--;
>-	}
>-    }
>-
>-}
>-
>-void
>-fbCompositeSrc_8888x8888arm (pixman_op_t op,
>-			 pixman_image_t * pSrc,
>-			 pixman_image_t * pMask,
>-			 pixman_image_t * pDst,
>-			 int16_t      xSrc,
>-			 int16_t      ySrc,
>-			 int16_t      xMask,
>-			 int16_t      yMask,
>-			 int16_t      xDst,
>-			 int16_t      yDst,
>-			 uint16_t     width,
>-			 uint16_t     height)
>-{
>-    uint32_t	*dstLine, *dst;
>-    uint32_t	*srcLine, *src;
>-    int	dstStride, srcStride;
>-    uint16_t	w;
>-    uint32_t component_half = 0x800080;
>-    uint32_t upper_component_mask = 0xff00ff00;
>-    uint32_t alpha_mask = 0xff;
>-
>-    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>-    fbComposeGetStart (pSrc, xSrc, ySrc, uint32_t, srcStride, srcLine, 1);
>-
>-    while (height--)
>-    {
>-	dst = dstLine;
>-	dstLine += dstStride;
>-	src = srcLine;
>-	srcLine += srcStride;
>-	w = width;
>-
>-//#define inner_branch
>-	asm volatile (
>-			"cmp %[w], #0\n\t"
>-			"beq 2f\n\t"
>-			"1:\n\t"
>-			/* load src */
>-			"ldr r5, [%[src]], #4\n\t"
>-#ifdef inner_branch
>-			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>-			 * The 0x0 case also allows us to avoid doing an unecessary data
>-			 * write which is more valuable so we only check for that */
>-			"cmp r5, #0\n\t"
>-			"beq 3f\n\t"
>-
>-			/* = 255 - alpha */
>-			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>-
>-			"ldr r4, [%[dest]] \n\t"
>-
>-#else
>-			"ldr r4, [%[dest]] \n\t"
>-
>-			/* = 255 - alpha */
>-			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>-#endif
>-			"uxtb16 r6, r4\n\t"
>-			"uxtb16 r7, r4, ror #8\n\t"
>-
>-			/* multiply by 257 and divide by 65536 */
>-			"mla r6, r6, r8, %[component_half]\n\t"
>-			"mla r7, r7, r8, %[component_half]\n\t"
>-
>-			"uxtab16 r6, r6, r6, ror #8\n\t"
>-			"uxtab16 r7, r7, r7, ror #8\n\t"
>-
>-			/* recombine the 0xff00ff00 bytes of r6 and r7 */
>-			"and r7, %[upper_component_mask]\n\t"
>-			"uxtab16 r6, r7, r6, ror #8\n\t"
>-
>-			"uqadd8 r5, r6, r5\n\t"
>-
>-#ifdef inner_branch
>-			"3:\n\t"
>-
>-#endif
>-			"str r5, [%[dest]], #4\n\t"
>-			/* increment counter and jmp to top */
>-			"subs	%[w], %[w], #1\n\t"
>-			"bne	1b\n\t"
>-			"2:\n\t"
>-			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src)
>-			: [component_half] "r" (component_half), [upper_component_mask] "r" (upper_component_mask),
>-			  [alpha_mask] "r" (alpha_mask)
>-			: "r4", "r5", "r6", "r7", "r8", "cc", "memory"
>-			);
>-    }
>-}
>-
>-void
>-fbCompositeSrc_8888x8x8888arm (pixman_op_t op,
>-			       pixman_image_t * pSrc,
>-			       pixman_image_t * pMask,
>-			       pixman_image_t * pDst,
>-			       int16_t	xSrc,
>-			       int16_t	ySrc,
>-			       int16_t      xMask,
>-			       int16_t      yMask,
>-			       int16_t      xDst,
>-			       int16_t      yDst,
>-			       uint16_t     width,
>-			       uint16_t     height)
>-{
>-    uint32_t	*dstLine, *dst;
>-    uint32_t	*srcLine, *src;
>-    uint32_t	mask;
>-    int	dstStride, srcStride;
>-    uint16_t	w;
>-    uint32_t component_half = 0x800080;
>-    uint32_t alpha_mask = 0xff;
>-
>-    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>-    fbComposeGetStart (pSrc, xSrc, ySrc, uint32_t, srcStride, srcLine, 1);
>-
>-    fbComposeGetSolid (pMask, mask, pDst->bits.format);
>-    mask = (mask) >> 24;
>-
>-    while (height--)
>-    {
>-	dst = dstLine;
>-	dstLine += dstStride;
>-	src = srcLine;
>-	srcLine += srcStride;
>-	w = width;
>-
>-//#define inner_branch
>-	asm volatile (
>-			"cmp %[w], #0\n\t"
>-			"beq 2f\n\t"
>-			"1:\n\t"
>-			/* load src */
>-			"ldr r5, [%[src]], #4\n\t"
>-#ifdef inner_branch
>-			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>-			 * The 0x0 case also allows us to avoid doing an unecessary data
>-			 * write which is more valuable so we only check for that */
>-			"cmp r5, #0\n\t"
>-			"beq 3f\n\t"
>-
>-#endif
>-			"ldr r4, [%[dest]] \n\t"
>-
>-			"uxtb16 r6, r5\n\t"
>-			"uxtb16 r7, r5, ror #8\n\t"
>-
>-			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>-			"mla r6, r6, %[mask_alpha], %[component_half]\n\t"
>-			"mla r7, r7, %[mask_alpha], %[component_half]\n\t"
>-
>-			"uxtab16 r6, r6, r6, ror #8\n\t"
>-			"uxtab16 r7, r7, r7, ror #8\n\t"
>-
>-			"uxtb16 r6, r6, ror #8\n\t"
>-			"uxtb16 r7, r7, ror #8\n\t"
>-
>-			/* recombine */
>-			"orr r5, r6, r7, lsl #8\n\t"
>-
>-			"uxtb16 r6, r4\n\t"
>-			"uxtb16 r7, r4, ror #8\n\t"
>-
>-			/* 255 - alpha */
>-			"sub r8, %[alpha_mask], r5, lsr #24\n\t"
>-
>-			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>-			"mla r6, r6, r8, %[component_half]\n\t"
>-			"mla r7, r7, r8, %[component_half]\n\t"
>-
>-			"uxtab16 r6, r6, r6, ror #8\n\t"
>-			"uxtab16 r7, r7, r7, ror #8\n\t"
>-
>-			"uxtb16 r6, r6, ror #8\n\t"
>-			"uxtb16 r7, r7, ror #8\n\t"
>-
>-			/* recombine */
>-			"orr r6, r6, r7, lsl #8\n\t"
>-
>-			"uqadd8 r5, r6, r5\n\t"
>-
>-#ifdef inner_branch
>-			"3:\n\t"
>-
>-#endif
>-			"str r5, [%[dest]], #4\n\t"
>-			/* increment counter and jmp to top */
>-			"subs	%[w], %[w], #1\n\t"
>-			"bne	1b\n\t"
>-			"2:\n\t"
>-			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src)
>-			: [component_half] "r" (component_half), [mask_alpha] "r" (mask),
>-			  [alpha_mask] "r" (alpha_mask)
>-			: "r4", "r5", "r6", "r7", "r8", "r9", "cc", "memory"
>-			);
>-    }
>-}
>-
>-void
>-fbCompositeSolidMask_nx8x8888arm (pixman_op_t      op,
>-			       pixman_image_t * pSrc,
>-			       pixman_image_t * pMask,
>-			       pixman_image_t * pDst,
>-			       int16_t      xSrc,
>-			       int16_t      ySrc,
>-			       int16_t      xMask,
>-			       int16_t      yMask,
>-			       int16_t      xDst,
>-			       int16_t      yDst,
>-			       uint16_t     width,
>-			       uint16_t     height)
>-{
>-    uint32_t	 src, srca;
>-    uint32_t	*dstLine, *dst;
>-    uint8_t	*maskLine, *mask;
>-    int		 dstStride, maskStride;
>-    uint16_t	 w;
>-
>-    fbComposeGetSolid(pSrc, src, pDst->bits.format);
>-
>-    srca = src >> 24;
>-    if (src == 0)
>-	return;
>-
>-    uint32_t component_mask = 0xff00ff;
>-    uint32_t component_half = 0x800080;
>-
>-    uint32_t src_hi = (src >> 8) & component_mask;
>-    uint32_t src_lo = src & component_mask;
>-
>-    fbComposeGetStart (pDst, xDst, yDst, uint32_t, dstStride, dstLine, 1);
>-    fbComposeGetStart (pMask, xMask, yMask, uint8_t, maskStride, maskLine, 1);
>-
>-    while (height--)
>-    {
>-	dst = dstLine;
>-	dstLine += dstStride;
>-	mask = maskLine;
>-	maskLine += maskStride;
>-	w = width;
>-
>-//#define inner_branch
>-	asm volatile (
>-			"cmp %[w], #0\n\t"
>-			"beq 2f\n\t"
>-			"1:\n\t"
>-			/* load mask */
>-			"ldrb r5, [%[mask]], #1\n\t"
>-#ifdef inner_branch
>-			/* We can avoid doing the multiplication in two cases: 0x0 or 0xff.
>-			 * The 0x0 case also allows us to avoid doing an unecessary data
>-			 * write which is more valuable so we only check for that */
>-			"cmp r5, #0\n\t"
>-			"beq 3f\n\t"
>-
>-#endif
>-			"ldr r4, [%[dest]] \n\t"
>-
>-			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>-			"mla r6, %[src_lo], r5, %[component_half]\n\t"
>-			"mla r7, %[src_hi], r5, %[component_half]\n\t"
>-
>-			"uxtab16 r6, r6, r6, ror #8\n\t"
>-			"uxtab16 r7, r7, r7, ror #8\n\t"
>-
>-			"uxtb16 r6, r6, ror #8\n\t"
>-			"uxtb16 r7, r7, ror #8\n\t"
>-
>-			/* recombine */
>-			"orr r5, r6, r7, lsl #8\n\t"
>-
>-			"uxtb16 r6, r4\n\t"
>-			"uxtb16 r7, r4, ror #8\n\t"
>-
>-			/* we could simplify this to use 'sub' if we were
>-			 * willing to give up a register for alpha_mask */
>-			"mvn r8, r5\n\t"
>-			"mov r8, r8, lsr #24\n\t"
>-
>-			/* multiply by alpha (r8) then by 257 and divide by 65536 */
>-			"mla r6, r6, r8, %[component_half]\n\t"
>-			"mla r7, r7, r8, %[component_half]\n\t"
>-
>-			"uxtab16 r6, r6, r6, ror #8\n\t"
>-			"uxtab16 r7, r7, r7, ror #8\n\t"
>-
>-			"uxtb16 r6, r6, ror #8\n\t"
>-			"uxtb16 r7, r7, ror #8\n\t"
>-
>-			/* recombine */
>-			"orr r6, r6, r7, lsl #8\n\t"
>-
>-			"uqadd8 r5, r6, r5\n\t"
>-
>-#ifdef inner_branch
>-			"3:\n\t"
>-
>-#endif
>-			"str r5, [%[dest]], #4\n\t"
>-			/* increment counter and jmp to top */
>-			"subs	%[w], %[w], #1\n\t"
>-			"bne	1b\n\t"
>-			"2:\n\t"
>-			: [w] "+r" (w), [dest] "+r" (dst), [src] "+r" (src), [mask] "+r" (mask)
>-			: [component_half] "r" (component_half),
>-			  [src_hi] "r" (src_hi), [src_lo] "r" (src_lo)
>-			: "r4", "r5", "r6", "r7", "r8", "cc", "memory"
>-			);
>-    }
>-}
>diff --git a/pixman/pixman-arm.h b/pixman/pixman-arm.h
>deleted file mode 100644
>index 258054a..0000000
>--- a/pixman/pixman-arm.h
>+++ /dev/null
>@@ -1,94 +0,0 @@
>-/*
>- * Copyright Â© 2008 Mozilla Corporation
>- *
>- * Permission to use, copy, modify, distribute, and sell this software and its
>- * documentation for any purpose is hereby granted without fee, provided that
>- * the above copyright notice appear in all copies and that both that
>- * copyright notice and this permission notice appear in supporting
>- * documentation, and that the name of Mozilla Corporation not be used in
>- * advertising or publicity pertaining to distribution of the software without
>- * specific, written prior permission.  Mozilla Corporation makes no
>- * representations about the suitability of this software for any purpose.  It
>- * is provided "as is" without express or implied warranty.
>- *
>- * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS
>- * SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
>- * FITNESS, IN NO EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY
>- * SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
>- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
>- * AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
>- * OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
>- * SOFTWARE.
>- *
>- * Author:  Jeff Muizelaar (jeff@infidigm.net)
>- *
>- */
>-
>-#include "pixman-private.h"
>-
>-#ifdef USE_ARM
>-
>-static inline pixman_bool_t pixman_have_arm(void) { return TRUE; }
>-
>-#else
>-#define pixman_have_arm() FALSE
>-#endif
>-
>-#ifdef USE_ARM
>-
>-void
>-fbCompositeSrcAdd_8000x8000arm (pixman_op_t op,
>-				pixman_image_t * pSrc,
>-				pixman_image_t * pMask,
>-				pixman_image_t * pDst,
>-				int16_t      xSrc,
>-				int16_t      ySrc,
>-				int16_t      xMask,
>-				int16_t      yMask,
>-				int16_t      xDst,
>-				int16_t      yDst,
>-				uint16_t     width,
>-				uint16_t     height);
>-void
>-fbCompositeSrc_8888x8888arm (pixman_op_t op,
>-			 pixman_image_t * pSrc,
>-			 pixman_image_t * pMask,
>-			 pixman_image_t * pDst,
>-			 int16_t      xSrc,
>-			 int16_t      ySrc,
>-			 int16_t      xMask,
>-			 int16_t      yMask,
>-			 int16_t      xDst,
>-			 int16_t      yDst,
>-			 uint16_t     width,
>-			 uint16_t     height);
>-
>-void
>-fbCompositeSrc_8888x8x8888arm (pixman_op_t op,
>-			 pixman_image_t * pSrc,
>-			 pixman_image_t * pMask,
>-			 pixman_image_t * pDst,
>-			 int16_t      xSrc,
>-			 int16_t      ySrc,
>-			 int16_t      xMask,
>-			 int16_t      yMask,
>-			 int16_t      xDst,
>-			 int16_t      yDst,
>-			 uint16_t     width,
>-			 uint16_t     height);
>-void
>-fbCompositeSolidMask_nx8x8888arm (pixman_op_t op,
>-			 pixman_image_t * pSrc,
>-			 pixman_image_t * pMask,
>-			 pixman_image_t * pDst,
>-			 int16_t      xSrc,
>-			 int16_t      ySrc,
>-			 int16_t      xMask,
>-			 int16_t      yMask,
>-			 int16_t      xDst,
>-			 int16_t      yDst,
>-			 uint16_t     width,
>-			 uint16_t     height);
>-
>-
>-#endif /* USE_ARM */
>diff --git a/pixman/pixman-pict.c b/pixman/pixman-pict.c
>index 070b190..6dbc009 100644
>--- a/pixman/pixman-pict.c
>+++ b/pixman/pixman-pict.c
>@@ -34,7 +34,7 @@
> #include "pixman-mmx.h"
> #include "pixman-vmx.h"
> #include "pixman-sse2.h"
>-#include "pixman-arm.h"
>+#include "pixman-arm-simd.h"
> #include "pixman-combine32.h"
> 
> #ifdef __GNUC__
>@@ -1522,8 +1522,8 @@ static const FastPathInfo vmx_fast_paths[] =
> };
> #endif
> 
>-#ifdef USE_ARM
>-static const FastPathInfo arm_fast_paths[] =
>+#ifdef USE_ARM_SIMD
>+static const FastPathInfo arm_simd_fast_paths[] =
> {
>     { PIXMAN_OP_OVER, PIXMAN_a8r8g8b8, PIXMAN_null,     PIXMAN_a8r8g8b8, fbCompositeSrc_8888x8888arm,      0 },
>     { PIXMAN_OP_OVER, PIXMAN_a8r8g8b8, PIXMAN_null,	PIXMAN_x8r8g8b8, fbCompositeSrc_8888x8888arm,	   0 },
>@@ -1897,9 +1897,9 @@ pixman_image_composite (pixman_op_t      op,
> 	    info = get_fast_path (vmx_fast_paths, op, pSrc, pMask, pDst, pixbuf);
> #endif
> 
>-#ifdef USE_ARM
>-	if (!info && pixman_have_arm())
>-	    info = get_fast_path (arm_fast_paths, op, pSrc, pMask, pDst, pixbuf);
>+#ifdef USE_ARM_SIMD
>+	if (!info && pixman_have_arm_simd())
>+	    info = get_fast_path (arm_simd_fast_paths, op, pSrc, pMask, pDst, pixbuf);
> #endif
> 
>         if (!info)

Actions: View | Splinter Review

Attachments on bug 18106: 19720

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.