- Stencil-related functions now work in display lists
Changes:
- renamed aux.h as glaux.h (MS-DOS names can't start with aux)
- - most filenames are in 8.3 format to accomodate MS-DOS
+ - most filenames are in 8.3 format to accommodate MS-DOS
- use GLubytes to store arrays of colors instead of GLints
1.2.2 August 2, 1995
- glGetTexImage was using pixel unpacking instead of packing params
- auto-mipmap generation for cube maps was incorrect
Changes:
- - max texture units reduced to six to accomodate texture rectangles
+ - max texture units reduced to six to accommodate texture rectangles
- removed unfinished GL_MESA_sprite_point extension code
- glGetTexImage was using pixel unpacking instead of packing params
- auto-mipmap generation for cube maps was incorrect
Changes:
- - max texture units reduced to six to accomodate texture rectangles
+ - max texture units reduced to six to accommodate texture rectangles
- removed unfinished GL_MESA_sprite_point extension code
</pre>
libGL.so is thread safe. The overhead of thread safety for common,
single-thread clients is negligible. However, the overhead of thread
safety for multi-threaded clients is significant. Each GL API call
-requires two calls to pthread_get_specific() which can noticably
+requires two calls to pthread_get_specific() which can noticeably
impact performance. Warning: libGL.so is thread safe but individual
DRI drivers may not be. Please consult the documentation for a driver
to learn if it is thread safe.
<li>Add md5sums for 10.0.2. release.</li>
<li>cherry-ignore: Ignore several patches not yet ready for the stable branch</li>
<li>Drop another couple of patches.</li>
- <li>cherry-ignore: Ignore 4 patches at teh request of the author, (Anuj).</li>
+ <li>cherry-ignore: Ignore 4 patches at the request of the author, (Anuj).</li>
<li>Update version to 10.0.3</li>
</ul>
<li>llvmpipe: Special case complementary and identify blend factors in SoA.</li>
<li>llvmpipe: Make rgb/alpha bland func/factors match, when there is no alpha.</li>
<li>draw: Prevent clipped vertices overflow.</li>
- <li>draw: Fullfil the new min_lod/max_lod/lod_bias/border_color dynamic state</li>
+ <li>draw: Fulfil the new min_lod/max_lod/lod_bias/border_color dynamic state</li>
<li>gallivm: Fetch the lod from the dynamic state when min_lod == max_lod.</li>
<li>gallivm: Remove dead experimental code.</li>
<li>llvmpipe: Decouple sampler view and sampler state updates.</li>
Changes:
<ul>
<li>renamed aux.h as glaux.h (MS-DOS names can't start with aux)
-<li>most filenames are in 8.3 format to accomodate MS-DOS
+<li>most filenames are in 8.3 format to accommodate MS-DOS
<li>use GLubytes to store arrays of colors instead of GLints
</ul>
</ul>
Changes:
<ul>
-<li>max texture units reduced to six to accomodate texture rectangles
+<li>max texture units reduced to six to accommodate texture rectangles
<li>removed unfinished GL_MESA_sprite_point extension code
</ul>
/**
* Wait for the fence to expire, and remove it from the fenced list.
*
- * This function will release and re-aquire the mutex, so any copy of mutable
+ * This function will release and re-acquire the mutex, so any copy of mutable
* state must be discarded after calling it.
*/
static INLINE enum pipe_error
/**
* Internal function used by rbug_send_* functions.
*
- * Finish writeing data to the socket.
+ * Finish writing data to the socket.
* Ups the send_serial and sets the serial argument if supplied.
*/
int rbug_connection_send_finish(struct rbug_connection *c, uint32_t *serial)
struct i915_context *i915 = i915_render->i915;
if (i915->vbo_flushed)
- debug_printf("%s bad vbo flush occured stalling on hw\n", __FUNCTION__);
+ debug_printf("%s bad vbo flush occurred stalling on hw\n", __FUNCTION__);
#ifdef VBUF_MAP_BUFFER
return (unsigned char *)i915_render->vbo_ptr + i915_render->vbo_sw_offset;
tex->stride = align(util_format_get_stride(pt->format, width), 4);
- /* May need to adjust pitch to accomodate the placement of
+ /* May need to adjust pitch to accommodate the placement of
* the 2nd mipmap level. This occurs when the alignment
* constraints of mipmap placement push the right edge of the
* 2nd mipmap level out past the width of its parent.
struct i915_tracked_state i915_hw_dynamic = {
"dynamic",
update_dynamic,
- ~0 /* all state atoms, becuase we do internal checking */
+ ~0 /* all state atoms, because we do internal checking */
};
struct i915_tracked_state i915_hw_immediate = {
"immediate",
update_immediate,
- ~0 /* all state atoms, becuase we do internal checking */
+ ~0 /* all state atoms, because we do internal checking */
};
/**
- * Begining rasterization of a tile.
+ * Beginning rasterization of a tile.
* \param x window X position of the tile, in pixels
* \param y window Y position of the tile, in pixels
*/
#define LP_SCENE_MAX_SIZE (9*1024*1024)
/* The maximum amount of texture storage referenced by a scene is
- * clamped ot this size:
+ * clamped to this size:
*/
#define LP_SCENE_MAX_RESOURCE_SIZE (64*1024*1024)
/**
* Generic helper for growing an array that has separate size/count
- * and reserved counters to accomodate up to num new element.
+ * and reserved counters to accommodate up to num new element.
*
* type * Array;
* unsigned int Size;
/* Reuse the source index of the source that
* was just deleted and set its register
* index. We can't use rc_pair_alloc_source
- * for this becuase it might return a source
+ * for this because it might return a source
* index that is already being used. */
if (src_type & RC_SOURCE_RGB) {
pair_inst->RGB.Src[src_index]
OUT_CS_RELOC(r300->query_current);
case 2:
/* pipe 1 only */
- /* As mentioned above, accomodate RV380 and older. */
+ /* As mentioned above, accommodate RV380 and older. */
OUT_CS_REG(R300_SU_REG_DEST,
1 << (caps->high_second_pipe ? 3 : 1));
OUT_CS_REG(R300_ZB_ZPASS_ADDR, (query->num_results + 1) * 4);
#define R300_GA_SOFT_RESET 0x429c
/* Not sure why there are duplicate of factor and constant values.
- * My best guess so far is that there are seperate zbiases for test and write.
+ * My best guess so far is that there are separate zbiases for test and write.
* Ordering might be wrong.
* Some of the tests indicate that fgl has a fallback implementation of zbias
* via pixel shaders.
* until rc grows proper support, indicate if they're safe to use. */
boolean use_half_swizzles;
- /* If an error occured. */
+ /* If an error occurred. */
boolean error;
};
r600_need_dma_space(&rctx->b, ncopy * 5);
for (i = 0; i < ncopy; i++) {
csize = size < EG_DMA_COPY_MAX_SIZE ? size : EG_DMA_COPY_MAX_SIZE;
- /* emit reloc before writting cs so that cs is always in consistent state */
+ /* emit reloc before writing cs so that cs is always in consistent state */
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, rsrc, RADEON_USAGE_READ,
RADEON_PRIO_MIN);
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, rdst, RADEON_USAGE_WRITE,
cheight = (EG_DMA_COPY_MAX_SIZE * 4) / pitch;
}
size = (cheight * pitch) / 4;
- /* emit reloc before writting cs so that cs is always in consistent state */
+ /* emit reloc before writing cs so that cs is always in consistent state */
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, &rsrc->resource,
RADEON_USAGE_READ, RADEON_PRIO_MIN);
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, &rdst->resource,
r600_need_dma_space(&rctx->b, ncopy * 5);
for (i = 0; i < ncopy; i++) {
csize = size < R600_DMA_COPY_MAX_SIZE_DW ? size : R600_DMA_COPY_MAX_SIZE_DW;
- /* emit reloc before writting cs so that cs is always in consistent state */
+ /* emit reloc before writing cs so that cs is always in consistent state */
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, rsrc, RADEON_USAGE_READ,
RADEON_PRIO_MIN);
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, rdst, RADEON_USAGE_WRITE,
for (i = 0; i < ncopy; i++) {
cheight = cheight > copy_height ? copy_height : cheight;
size = (cheight * pitch) / 4;
- /* emit reloc before writting cs so that cs is always in consistent state */
+ /* emit reloc before writing cs so that cs is always in consistent state */
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, &rsrc->resource, RADEON_USAGE_READ,
RADEON_PRIO_MIN);
r600_context_bo_reloc(&rctx->b, &rctx->b.rings.dma, &rdst->resource, RADEON_USAGE_WRITE,
if (rctx->zwritemask != dsa->zwritemask) {
rctx->zwritemask = dsa->zwritemask;
if (rctx->b.chip_class >= EVERGREEN) {
- /* work around some issue when not writting to zbuffer
+ /* work around some issue when not writing to zbuffer
* we are having lockup on evergreen so do not enable
- * hyperz when not writting zbuffer
+ * hyperz when not writing zbuffer
*/
rctx->db_misc_state.atom.dirty = true;
}
}
/* Save the packed value for logic ops. Can't reuse r4
- * becuase other things might smash it (like sRGB)
+ * because other things might smash it (like sRGB)
*/
packed_dst_color = qir_MOV(c, r4);
}
* Do this by calling XGetWindowAttributes() for the window and
* checking if we catch an X error.
* Input: dpy - the display
- * win - the window to check for existance
+ * win - the window to check for existence
* Return: GL_TRUE - window exists
* GL_FALSE - window doesn't exist
*/
This->mode = NULL;
}
- /* Note: It is the role of the backend to fill if neccessary
+ /* Note: It is the role of the backend to fill if necessary
* BackBufferWidth and BackBufferHeight */
ID3DPresent_SetPresentParameters(This->present, pParams, This->mode);
if (!vlGetFuncFTAB(function_id, function_pointer))
return VDP_STATUS_INVALID_FUNC_ID;
- VDPAU_MSG(VDPAU_TRACE, "[VDPAU] Got proc adress %p for id %d\n", *function_pointer, function_id);
+ VDPAU_MSG(VDPAU_TRACE, "[VDPAU] Got proc address %p for id %d\n", *function_pointer, function_id);
return VDP_STATUS_OK;
}
* acquiring the stw_framebuffer::mutex of the framebuffer to be deleted.
* This ensures that nobody else is reading/writing to the.
*
- * It is not necessary to aquire the mutex above to navigate the linked list
+ * It is not necessary to acquire the mutex above to navigate the linked list
* given that deletions are done with stw_device::fb_mutex held, so no other
* thread can delete.
*/
/**
* Wait for the fence to expire, and remove it from the fenced list.
*
- * This function will release and re-aquire the mutex, so any copy of mutable
+ * This function will release and re-acquire the mutex, so any copy of mutable
* state must be discarded after calling it.
*/
static INLINE enum pipe_error
/*
* Only proceed if the fence object didn't change in the meanwhile.
* Otherwise assume the work has been already carried out by another
- * thread that re-aquired the lock before us.
+ * thread that re-acquired the lock before us.
*/
proceed = fence == fenced_buf->fence ? TRUE : FALSE;
*
* - Construct a matrix from an arbirary combination of vectors and
* scalars. The components of the constructor parameters are assigned
- * to the matrix in colum-major order until the matrix is full.
+ * to the matrix in column-major order until the matrix is full.
*
* - Construct a matrix from a single matrix. The source matrix is copied
* to the upper left portion of the constructed matrix, and the remaining
else {
if (argument->head == NULL) {
/* Don't treat initial whitespace as
- * part of the arguement. */
+ * part of the argument. */
if (node->token->type == SPACE)
continue;
}
*
* As uniforms are added to the active set the number of active uniforms and
* the storage requirements for those uniforms are accumulated. The active
- * uniforms are added the the hash table supplied to the constructor.
+ * uniforms are added to the hash table supplied to the constructor.
*
* If the same uniform is added multiple times (i.e., once for each shader
* target), it will only be accounted once.
* return pack_uvec2_to_uint(uvec2(
* round(clamp(VEC2_RVALUE, 0.0f, 1.0f) * 65535.0f)));
*
- * Here it is safe to directly convert the vec2 to uvec2 because the the
- * vec2 has been clamped to a non-negative range.
+ * Here it is safe to directly convert the vec2 to uvec2 because the vec2
+ * has been clamped to a non-negative range.
*/
assert(vec2_rval->type == glsl_type::vec2_type);
* return pack_uvec4_to_uint(uvec4(
* round(clamp(VEC2_RVALUE, 0.0f, 1.0f) * 255.0f)));
*
- * Here it is safe to directly convert the vec4 to uvec4 because the the
- * vec4 has been clamped to a non-negative range.
+ * Here it is safe to directly convert the vec4 to uvec4 because the vec4
+ * has been clamped to a non-negative range.
*/
assert(vec4_rval->type == glsl_type::vec4_type);
*
* The algorithm works by playing this little shell game with the values.
* We start by recording where every source value is and which source value
- * each destination value should recieve. We then grab any copy whose
+ * each destination value should receive. We then grab any copy whose
* destination is "empty", i.e. not used as a source, and do the following:
* - Find where its source value currently lives
* - Emit the move instruction
* the built-in varyings have pre-assigned locations. Also, the elimination
* of unused gl_TexCoord elements requires its own lowering pass anyway.
*
- * It's implemented by replacing all occurences of dead varyings with
+ * It's implemented by replacing all occurrences of dead varyings with
* temporary variables, which creates dead code. It is recommended to run
* a dead-code elimination pass after this.
*
*
* We're going to break down the gl_TexCoord array into separate
* variables. First, add declarations of the new variables all
- * occurences of gl_TexCoord will be replaced with.
+ * occurrences of gl_TexCoord will be replaced with.
*/
if (info->lower_texcoord_array) {
prepare_array(ir, this->new_texcoord, ARRAY_SIZE(this->new_texcoord),
* variable dereference representing gl_TexCoord[i].
*/
if (this->info->lower_texcoord_array) {
- /* gl_TexCoord[i] occurence */
+ /* gl_TexCoord[i] occurrence */
ir_dereference_array *const da = (*rvalue)->as_dereference_array();
if (da && da->variable_referenced() ==
/* Same for gl_FragData. */
if (this->info->lower_fragdata_array) {
- /* gl_FragData[i] occurence */
+ /* gl_FragData[i] occurrence */
ir_dereference_array *const da = (*rvalue)->as_dereference_array();
if (da && da->variable_referenced() == this->info->fragdata_array) {
parameters[i] = sig_param->clone(ctx, ht);
parameters[i]->data.mode = ir_var_auto;
- /* Remove the read-only decoration becuase we're going to write
+ /* Remove the read-only decoration because we're going to write
* directly to this variable. If the cloned variable is left
* read-only and the inlined function is inside a loop, the loop
* analysis code will get confused.
* destroyDrawable(drawable);
*
* which is a textbook race condition - the window may disappear
- * from the server between checking for its existance and
+ * from the server between checking for its existence and
* destroying it. Instead we change the semantics of
* __DRIinterfaceMethodsRec::destroyDrawable() to succeed even if
* the windows is gone, by wrapping the destroy call in an error
return id;
}
-/* Return true if an error occured. */
+/* Return true if an error occurred. */
bool
apple_init_glx(Display * dpy)
{
}
-/* Return true if an error occured. */
+/* Return true if an error occurred. */
bool
apple_glx_make_current_context(Display * dpy, void *oldptr, void *ptr,
GLXDrawable drawable)
}
}
-/* Return true if an error occured. */
+/* Return true if an error occurred. */
static bool
create_surface(Display * dpy, int screen, struct apple_glx_drawable *d)
{
return true; /* unable to create a surface. */
}
-/* Return true if an error occured. */
+/* Return true if an error occurred. */
/* This returns a referenced object via resultptr. */
bool
apple_glx_surface_create(Display * dpy, int screen,
}
/*
-** Query the existance of the GLX extension
+** Query the existence of the GLX extension
*/
_X_EXPORT Bool
glXQueryExtension(Display * dpy, int *errorBase, int *eventBase)
i += vertex_program_attribs;
- /* Vertex array *must* be last becuase of the way that
+ /* Vertex array *must* be last because of the way that
* emit_DrawArrays_none works.
*/
if (compsize == 0) {
/*
- ** Error occured; don't modify user's buffer.
+ ** Error occurred; don't modify user's buffer.
*/
}
else {
if (compsize == 0) {
/*
- ** Error occured; don't modify user's buffer.
+ ** Error occurred; don't modify user's buffer.
*/
}
else {
if (compsize == 0) {
/*
- ** Error occured; don't modify user's buffer.
+ ** Error occurred; don't modify user's buffer.
*/
}
else {
if (compsize == 0) {
/*
- ** Error occured; don't modify user's buffer.
+ ** Error occurred; don't modify user's buffer.
*/
}
else {
if (reply.newMode != mode) {
/*
** Switch to new mode did not take effect, therefore an error
- ** occured. When an error happens the server won't send us any
+ ** occurred. When an error happens the server won't send us any
** other data.
*/
}
void Normalize();
- // Returns a pointer to the last occurence of a valid path separator in
+ // Returns a pointer to the last occurrence of a valid path separator in
// the FilePath. On Windows, for example, both '/' and '\' are valid path
// separators. Returns NULL if no path separator was found.
const char* FindLastPathSeparator() const;
return *this;
}
-// Returns a pointer to the last occurence of a valid path separator in
+// Returns a pointer to the last occurrence of a valid path separator in
// the FilePath. On Windows, for example, both '/' and '\' are valid path
// separators. Returns NULL if no path separator was found.
const char* FilePath::FindLastPathSeparator() const {
/* Function to get a different device than the one we are to use by default,
* if the user requests so and it is possible. The initial fd will be closed
- * if neccessary. The returned fd is potentially a render-node.
+ * if necessary. The returned fd is potentially a render-node.
*/
int
print ' struct glx_context * const gc = __glXGetCurrentContext();'
# The only reason that single and vendor private commands need
- # a variable called 'dpy' is becuase they use the SyncHandle
+ # a variable called 'dpy' is because they use the SyncHandle
# macro. For whatever brain-dead reason, that macro is hard-
# coded to use a variable called 'dpy' instead of taking a
# parameter.
# Fill self.count and self.enums using the dictionary of enums
# that was passed in. The generic Get functions (e.g.,
# GetBooleanv and friends) are handled specially here. In
- # the data the generic Get functions are refered to as "Get".
+ # the data the generic Get functions are referred to as "Get".
if func_name in ["GetIntegerv", "GetBooleanv", "GetFloatv", "GetDoublev"]:
match_name = "Get"
# The x86-64 ABI divides function parameters into a couple
# classes. For the OpenGL interface, the only ones that are
- # relevent are INTEGER and SSE. Basically, the first 8
+ # relevant are INTEGER and SSE. Basically, the first 8
# GLfloat or GLdouble parameters are placed in %xmm0 - %xmm7,
# the first 6 non-GLfloat / non-GLdouble parameters are placed
# in registers listed in int_parameters.
}
/**
- * Makes sure that the proper enables are set for LogicOp, Independant Alpha
+ * Makes sure that the proper enables are set for LogicOp, Independent Alpha
* Blend, and Blending. It needs to be called from numerous places where we
- * could change the LogicOp or Independant Alpha Blend without subsequent
+ * could change the LogicOp or Independent Alpha Blend without subsequent
* calls to glEnable.
*
* \todo
}
/* This will catch a logicop blend equation. It will also ensure
- * independant alpha blend is really in the correct state (either enabled
+ * independent alpha blend is really in the correct state (either enabled
* or disabled) if blending is already enabled.
*/
}
/* This function makes sure that the proper enables are
- * set for LogicOp, Independant Alpha Blend, and Blending.
+ * set for LogicOp, Independent Alpha Blend, and Blending.
* It needs to be called from numerous places where we
- * could change the LogicOp or Independant Alpha Blend without subsequent
+ * could change the LogicOp or Independent Alpha Blend without subsequent
* calls to glEnable.
*/
static void
intel_tri_func draw_tri;
/**
- * Set if rendering has occured to the drawable's front buffer.
+ * Set if rendering has occurred to the drawable's front buffer.
*
* This is used in the DRI2 case to detect that glFlush should also copy
* the contents of the fake front buffer to the real front buffer.
mt->total_width = ALIGN(mt->physical_width0, mt->align_w);
}
- /* May need to adjust width to accomodate the placement of
+ /* May need to adjust width to accommodate the placement of
* the 2nd mipmap. This occurs when the alignment
* constraints of mipmap placement push the right edge of the
* 2nd mipmap out past the width of its parent.
brw_clip_have_varying(c, VARYING_SLOT_BFC1)))
return;
- /* In some wierd degnerate cases we can end up testing the
+ /* In some weird degenerate cases we can end up testing the
* direction twice, once for culling and once for bfc copying. Oh
- * well, that's what you get for setting wierd GL state.
+ * well, that's what you get for setting weird GL state.
*/
if (c->key.copy_bfc_ccw)
conditional = BRW_CONDITIONAL_GE;
struct brw_indirect v0ptr = brw_indirect(2, 0);
struct brw_indirect v1ptr = brw_indirect(3, 0);
- /* Need a seperate loop for offset:
+ /* Need a separate loop for offset:
*/
if (do_offset) {
brw_MOV(p, c->reg.loopcount, c->reg.nr_verts);
brw_MOV(p, get_element_ud(c->reg.R0, 2), brw_imm_ud(header));
- /* Send each vertex as a seperate write to the urb. This
+ /* Send each vertex as a separate write to the urb. This
* is different to the concept in brw_sf_emit.c, where
* subsequent writes are used to build up a single urb
- * entry. Each of these writes instantiates a seperate
+ * entry. Each of these writes instantiates a separate
* urb entry - (I think... what about 'allocate'?)
*/
brw_urb_WRITE(p,
ctx->Const.MaxViewports = GEN7_NUM_VIEWPORTS;
ctx->Const.ViewportSubpixelBits = 0;
- /* Cast to float before negating becuase MaxViewportWidth is unsigned.
+ /* Cast to float before negating because MaxViewportWidth is unsigned.
*/
ctx->Const.ViewportBounds.Min = -(float)ctx->Const.MaxViewportWidth;
ctx->Const.ViewportBounds.Max = ctx->Const.MaxViewportWidth;
* sRGB encode if the renderbuffer can handle it. You can ask specifically
* for a visual where you're guaranteed to be capable, but it turns out that
* everyone just makes all their ARGB8888 visuals capable and doesn't offer
- * incapable ones, becuase there's no difference between the two in resources
+ * incapable ones, because there's no difference between the two in resources
* used. Applications thus get built that accidentally rely on the default
* visual choice being sRGB, so we make ours sRGB capable. Everything sounds
* great...
* Number of resets observed in the system at context creation.
*
* This is tracked in the context so that we can determine that another
- * reset has occured.
+ * reset has occurred.
*/
uint32_t reset_count;
} upload;
/**
- * Set if rendering has occured to the drawable's front buffer.
+ * Set if rendering has occurred to the drawable's front buffer.
*
* This is used in the DRI2 case to detect that glFlush should also copy
* the contents of the fake front buffer to the real front buffer.
*
* When the matching 'else' instruction is reached (presumably by
* countdown of the instruction count patched in by our ELSE/ENDIF
- * functions), the relevent flags are inverted.
+ * functions), the relevant flags are inverted.
*
* When the matching 'endif' instruction is reached, the flags are
* popped off. If the stack is now empty, normal execution resumes.
emit_endif = false;
/*
- * A single next_insn() may change the base adress of instruction store
+ * A single next_insn() may change the base address of instruction store
* memory(p->store), so call it first before referencing the instruction
* store pointer from an index
*/
mt->total_width = ALIGN(mt->physical_width0, mt->align_w);
}
- /* May need to adjust width to accomodate the placement of
+ /* May need to adjust width to accommodate the placement of
* the 2nd mipmap. This occurs when the alignment
* constraints of mipmap placement push the right edge of the
* 2nd mipmap out past the width of its parent.
/* _NEW_POLYGON */
if (ctx->Polygon.OffsetFill) {
wm->wm5.depth_offset = 1;
- /* Something wierd going on with legacy_global_depth_bias,
+ /* Something weird going on with legacy_global_depth_bias,
* offset_constant, scaling and MRD. This value passes glean
* but gives some odd results elsewere (eg. the
* quad-offset-units test).
* for fake "hole" components, rather than simply taking the offset
* for each real varying. Each hole can have size 1, 2, 3, or 4; we
* program as many size = 4 holes as we can, then a final hole to
- * accomodate the final 1, 2, or 3 remaining.
+ * accommodate the final 1, 2, or 3 remaining.
*/
int skip_components =
linked_xfb_info->Outputs[i].DstOffset - next_offset[buffer];
rmesa->radeon.dma.flush( &rmesa->radeon.glCtx ); \
} while (0)
-/* Can accomodate several state changes and primitive changes without
+/* Can accommodate several state changes and primitive changes without
* actually firing the buffer.
*/
#define R200_STATECHANGE( rmesa, ATOM ) \
*/
t->pp_txformat_x |= R200_TEXCOORD_PROJ;
}
- /* FIXME: NPOT sizes, Is it correct realy? */
+ /* FIXME: NPOT sizes, is it correct really? */
t->pp_txsize = (((firstImage->Width - 1) << R200_PP_TX_WIDTHMASK_SHIFT)
| ((firstImage->Height - 1) << R200_PP_TX_HEIGHTMASK_SHIFT));
GLboolean front_cliprects;
/**
- * Set if rendering has occured to the drawable's front buffer.
+ * Set if rendering has occurred to the drawable's front buffer.
*
* This is used in the DRI2 case to detect that glFlush should also copy
* the contents of the fake front buffer to the real front buffer.
rmesa->radeon.dma.flush( &rmesa->radeon.glCtx ); \
} while (0)
-/* Can accomodate several state changes and primitive changes without
+/* Can accommodate several state changes and primitive changes without
* actually firing the buffer.
*/
* Do this by calling XGetWindowAttributes() for the window and
* checking if we catch an X error.
* Input: dpy - the display
- * win - the window to check for existance
+ * win - the window to check for existence
* Return: GL_TRUE - window exists
* GL_FALSE - window doesn't exist
*/
{
/* Find the larger of Mesa's dispatch table and libGL's dispatch table.
* In practice, this'll be the same for stand-alone Mesa. But for DRI
- * Mesa we do this to accomodate different versions of libGL and various
+ * Mesa we do this to accommodate different versions of libGL and various
* DRI drivers.
*/
GLint numEntries = MAX2(_glapi_get_dispatch_table_size(), _gloffset_COUNT);
struct gl_vertex_program *prog;
struct state_key key;
- /* Grab all the relevent state and put it in a single structure:
+ /* Grab all the relevant state and put it in a single structure:
*/
make_state_key(ctx, &key);
/**
* Same as _mesa_format_image_size() but returns a 64-bit value to
- * accomodate very large textures.
+ * accommodate very large textures.
*/
uint64_t
_mesa_format_image_size64(mesa_format format, GLsizei width,
/*@{*/
#define MAT_FLAG_IDENTITY 0 /**< is an identity matrix flag.
* (Not actually used - the identity
- * matrix is identified by the absense
+ * matrix is identified by the absence
* of all other flags.)
*/
#define MAT_FLAG_GENERAL 0x1 /**< is a general matrix flag */
* Add an element to a hash table with replacement
*
* \return
- * 1 if it did replace the the value (in which case the old key is kept), 0 if
- * it did not replace the value (in which case the new key is kept).
+ * 1 if it did replace the value (in which case the old key is kept), 0 if it
+ * did not replace the value (in which case the new key is kept).
*
* \warning
* If \c key is already in the hash table, \c data will \b replace the most
/**
- * Rebuild/expand the hash table to accomodate more entries
+ * Rebuild/expand the hash table to accommodate more entries
*/
static void
rehash(struct gl_program_cache *cache)
* Finish() is identical to Flush() in all cases - no differences in
* rendering or ReadPixels are visible if we opt not to wait here.
*
- * Only set this up on windows to avoid suprise elsewhere.
+ * Only set this up on Windows to avoid surprise elsewhere.
*/
#ifdef PIPE_OS_WINDOWS
functions->Finish = st_glFlush;
/*
* Must do all texture sampling before combining in order to
- * accomodate GL_ARB_texture_env_crossbar.
+ * accommodate GL_ARB_texture_env_crossbar.
*/
for (unit = 0; unit < ctx->Const.MaxTextureUnits; unit++) {
const struct gl_texture_unit *texUnit = &ctx->Texture.Unit[unit];
/* It will probably be necessary to allow drivers to specify new
- * emit-styles to cover all the wierd and wacky things out there.
+ * emit-styles to cover all the weird and wacky things out there.
*/
enum tnl_attr_format {
EMIT_1F,
return;
}
- /* Heuristic: attempt to isolate attributes occuring outside
+ /* Heuristic: attempt to isolate attributes occurring outside
* begin/end pairs.
*/
if (exec->vtx.vertex_size && !exec->vtx.attrsz[0])
goto end;
}
else if (save->replay_flags) {
- /* Various degnerate cases: translate into immediate mode
+ /* Various degenerate cases: translate into immediate mode
* calls rather than trying to execute in place.
*/
vbo_save_loopback_vertex_list( ctx, node );
projects / mesa.git / commitdiff