aboutsummaryrefslogtreecommitdiff
path: root/mesalib/src/mesa/swrast/s_logic.c
diff options
context:
space:
mode:
Diffstat (limited to 'mesalib/src/mesa/swrast/s_logic.c')
-rw-r--r--mesalib/src/mesa/swrast/s_logic.c438
1 files changed, 219 insertions, 219 deletions
diff --git a/mesalib/src/mesa/swrast/s_logic.c b/mesalib/src/mesa/swrast/s_logic.c
index 2f0c0ea7c..80ee46c24 100644
--- a/mesalib/src/mesa/swrast/s_logic.c
+++ b/mesalib/src/mesa/swrast/s_logic.c
@@ -1,219 +1,219 @@
-/*
- * Mesa 3-D graphics library
- * Version: 6.5.2
- *
- * Copyright (C) 1999-2006 Brian Paul All Rights Reserved.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
- * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
- * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
- */
-
-
-#include "main/glheader.h"
-#include "main/context.h"
-#include "main/imports.h"
-#include "main/macros.h"
-
-#include "s_context.h"
-#include "s_logic.h"
-#include "s_span.h"
-
-
-/**
- * We do all logic ops on 4-byte GLuints.
- * Depending on bytes per pixel, the mask array elements correspond to
- * 1, 2 or 4 GLuints.
- */
-#define LOGIC_OP_LOOP(MODE, MASKSTRIDE) \
-do { \
- GLuint i; \
- switch (MODE) { \
- case GL_CLEAR: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = 0; \
- } \
- } \
- break; \
- case GL_SET: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~0; \
- } \
- } \
- break; \
- case GL_COPY: \
- /* do nothing */ \
- break; \
- case GL_COPY_INVERTED: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~src[i]; \
- } \
- } \
- break; \
- case GL_NOOP: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = dest[i]; \
- } \
- } \
- break; \
- case GL_INVERT: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~dest[i]; \
- } \
- } \
- break; \
- case GL_AND: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] &= dest[i]; \
- } \
- } \
- break; \
- case GL_NAND: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~(src[i] & dest[i]); \
- } \
- } \
- break; \
- case GL_OR: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] |= dest[i]; \
- } \
- } \
- break; \
- case GL_NOR: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~(src[i] | dest[i]); \
- } \
- } \
- break; \
- case GL_XOR: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] ^= dest[i]; \
- } \
- } \
- break; \
- case GL_EQUIV: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~(src[i] ^ dest[i]); \
- } \
- } \
- break; \
- case GL_AND_REVERSE: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = src[i] & ~dest[i]; \
- } \
- } \
- break; \
- case GL_AND_INVERTED: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~src[i] & dest[i]; \
- } \
- } \
- break; \
- case GL_OR_REVERSE: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = src[i] | ~dest[i]; \
- } \
- } \
- break; \
- case GL_OR_INVERTED: \
- for (i = 0; i < n; i++) { \
- if (mask[i / MASKSTRIDE]) { \
- src[i] = ~src[i] | dest[i]; \
- } \
- } \
- break; \
- default: \
- _mesa_problem(ctx, "bad logicop mode");\
- } \
-} while (0)
-
-
-
-static INLINE void
-logicop_uint1(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
- const GLubyte mask[])
-{
- LOGIC_OP_LOOP(ctx->Color.LogicOp, 1);
-}
-
-
-static INLINE void
-logicop_uint2(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
- const GLubyte mask[])
-{
- LOGIC_OP_LOOP(ctx->Color.LogicOp, 2);
-}
-
-
-static INLINE void
-logicop_uint4(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
- const GLubyte mask[])
-{
- LOGIC_OP_LOOP(ctx->Color.LogicOp, 4);
-}
-
-
-
-/**
- * Apply the current logic operator to a span of RGBA pixels.
- * We can handle horizontal runs of pixels (spans) or arrays of x/y
- * pixel coordinates.
- */
-void
-_swrast_logicop_rgba_span(struct gl_context *ctx, struct gl_renderbuffer *rb,
- SWspan *span)
-{
- void *rbPixels;
-
- ASSERT(span->end < MAX_WIDTH);
- ASSERT(span->arrayMask & SPAN_RGBA);
- ASSERT(rb->DataType == span->array->ChanType);
-
- rbPixels = _swrast_get_dest_rgba(ctx, rb, span);
-
- if (span->array->ChanType == GL_UNSIGNED_BYTE) {
- /* treat 4*GLubyte as GLuint */
- logicop_uint1(ctx, span->end,
- (GLuint *) span->array->rgba8,
- (const GLuint *) rbPixels, span->array->mask);
- }
- else if (span->array->ChanType == GL_UNSIGNED_SHORT) {
- /* treat 2*GLushort as GLuint */
- logicop_uint2(ctx, 2 * span->end,
- (GLuint *) span->array->rgba16,
- (const GLuint *) rbPixels, span->array->mask);
- }
- else {
- logicop_uint4(ctx, 4 * span->end,
- (GLuint *) span->array->attribs[FRAG_ATTRIB_COL0],
- (const GLuint *) rbPixels, span->array->mask);
- }
-}
+/*
+ * Mesa 3-D graphics library
+ * Version: 6.5.2
+ *
+ * Copyright (C) 1999-2006 Brian Paul All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+
+#include "main/glheader.h"
+#include "main/context.h"
+#include "main/imports.h"
+#include "main/macros.h"
+
+#include "s_context.h"
+#include "s_logic.h"
+#include "s_span.h"
+
+
+/**
+ * We do all logic ops on 4-byte GLuints.
+ * Depending on bytes per pixel, the mask array elements correspond to
+ * 1, 2 or 4 GLuints.
+ */
+#define LOGIC_OP_LOOP(MODE, MASKSTRIDE) \
+do { \
+ GLuint i; \
+ switch (MODE) { \
+ case GL_CLEAR: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = 0; \
+ } \
+ } \
+ break; \
+ case GL_SET: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~0; \
+ } \
+ } \
+ break; \
+ case GL_COPY: \
+ /* do nothing */ \
+ break; \
+ case GL_COPY_INVERTED: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~src[i]; \
+ } \
+ } \
+ break; \
+ case GL_NOOP: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_INVERT: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_AND: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] &= dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_NAND: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~(src[i] & dest[i]); \
+ } \
+ } \
+ break; \
+ case GL_OR: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] |= dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_NOR: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~(src[i] | dest[i]); \
+ } \
+ } \
+ break; \
+ case GL_XOR: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] ^= dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_EQUIV: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~(src[i] ^ dest[i]); \
+ } \
+ } \
+ break; \
+ case GL_AND_REVERSE: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = src[i] & ~dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_AND_INVERTED: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~src[i] & dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_OR_REVERSE: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = src[i] | ~dest[i]; \
+ } \
+ } \
+ break; \
+ case GL_OR_INVERTED: \
+ for (i = 0; i < n; i++) { \
+ if (mask[i / MASKSTRIDE]) { \
+ src[i] = ~src[i] | dest[i]; \
+ } \
+ } \
+ break; \
+ default: \
+ _mesa_problem(ctx, "bad logicop mode");\
+ } \
+} while (0)
+
+
+
+static inline void
+logicop_uint1(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
+ const GLubyte mask[])
+{
+ LOGIC_OP_LOOP(ctx->Color.LogicOp, 1);
+}
+
+
+static inline void
+logicop_uint2(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
+ const GLubyte mask[])
+{
+ LOGIC_OP_LOOP(ctx->Color.LogicOp, 2);
+}
+
+
+static inline void
+logicop_uint4(struct gl_context *ctx, GLuint n, GLuint src[], const GLuint dest[],
+ const GLubyte mask[])
+{
+ LOGIC_OP_LOOP(ctx->Color.LogicOp, 4);
+}
+
+
+
+/**
+ * Apply the current logic operator to a span of RGBA pixels.
+ * We can handle horizontal runs of pixels (spans) or arrays of x/y
+ * pixel coordinates.
+ */
+void
+_swrast_logicop_rgba_span(struct gl_context *ctx, struct gl_renderbuffer *rb,
+ SWspan *span)
+{
+ void *rbPixels;
+
+ ASSERT(span->end < MAX_WIDTH);
+ ASSERT(span->arrayMask & SPAN_RGBA);
+ ASSERT(rb->DataType == span->array->ChanType);
+
+ rbPixels = _swrast_get_dest_rgba(ctx, rb, span);
+
+ if (span->array->ChanType == GL_UNSIGNED_BYTE) {
+ /* treat 4*GLubyte as GLuint */
+ logicop_uint1(ctx, span->end,
+ (GLuint *) span->array->rgba8,
+ (const GLuint *) rbPixels, span->array->mask);
+ }
+ else if (span->array->ChanType == GL_UNSIGNED_SHORT) {
+ /* treat 2*GLushort as GLuint */
+ logicop_uint2(ctx, 2 * span->end,
+ (GLuint *) span->array->rgba16,
+ (const GLuint *) rbPixels, span->array->mask);
+ }
+ else {
+ logicop_uint4(ctx, 4 * span->end,
+ (GLuint *) span->array->attribs[FRAG_ATTRIB_COL0],
+ (const GLuint *) rbPixels, span->array->mask);
+ }
+}