Displaying 20 results from an estimated 24 matches for "dst_stride".
2013 Jul 08
1
patch to fix error in src/opus_multistream_encoder.c when DISABLE_FLOAT_API is defined
...AnalysisInfo analysis_info;
+#endif
const CELTMode *celt_mode;
opus_int32 bitrates[256];
opus_val16 bandLogE[42];
@@ -629,27 +631,6 @@ static int opus_multistream_encode_native
}
-#if !defined(DISABLE_FLOAT_API)
-static void opus_copy_channel_in_float(
- opus_val16 *dst,
- int dst_stride,
- const void *src,
- int src_stride,
- int src_channel,
- int frame_size
-)
-{
- const float *float_src;
- opus_int32 i;
- float_src = (const float *)src;
- for (i=0;i<frame_size;i++)
-#if defined(FIXED_POINT)
- dst[i*dst_stride] = FLOAT2INT16(float_src[i*src_stride+src_channel...
2017 Nov 20
2
[PATCH] Fix memory issue in Projection API
Hello,
Attached is a patch to resolve a memory issue using the Projection API when
compiling using a psuedo-stack / limited memory.
Please let me any ?s you might have.
Cheers,
Drew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xiph.org/pipermail/opus/attachments/20171120/84e36e1b/attachment-0001.html>
-------------- next part --------------
2017 Nov 22
2
[PATCH] Fix memory issue in Projection API
...t; Please check you haven't made that same mistake elsewhere.
>
> 3) In opus_projection_copy_channel_out_float() and
> opus_projection_copy_channel_out_short(), you're zeroing the values with
> this loop:
> for (i=0;i<frame_size;i++)
> {
> for (j=0;j<dst_stride;j++)
> {
> float_dst[i*dst_stride+j] = 0;
> }
> }
> That looks wrong, since you'll be overwriting all channels instead of
> just the one you're copying. That should have come up in testing, no?
>
> 4) You might want to have a look at
> mapp...
2009 Aug 17
2
[PATCH] kms: Fix <nv11 hardware cursor.
...S (NV10_CURSOR_SIZE * NV10_CURSOR_SIZE)
-
#define SOURCE_MASK_INTERLEAVE 32
#define TRANSPARENT_PIXEL 0
-static void
-nv_cursor_convert_cursor(int px, uint32_t *src, uint16_t *dst, int bpp, uint32_t fg, uint32_t bg)
+void
+nv_cursor_convert_cursor(uint32_t *src, void *dst, int src_stride, int dst_stride,
+ int bpp, uint32_t fg, uint32_t bg)
{
+ int width = min(src_stride, dst_stride);
uint32_t b, m, pxval;
- int i, j;
+ int i, j, k;
+
+ for (i = 0; i < width; i++) {
+ for (j = 0; j < width / SOURCE_MASK_INTERLEAVE; j++) {
+ int src_off = i*src_stride/SOURCE_MASK_INTERLEAVE + j;
+...
2017 Nov 23
2
[PATCH] Fix memory issue in Projection API
....
> >
> > 3) In opus_projection_copy_channel_out_float() and
> > opus_projection_copy_channel_out_short(), you're zeroing the values
> with
> > this loop:
> > for (i=0;i<frame_size;i++)
> > {
> > for (j=0;j<dst_stride;j++)
> > {
> > float_dst[i*dst_stride+j] = 0;
> > }
> > }
> > That looks wrong, since you'll be overwriting all channels instead of
> > just the one you're copying. That should have come up in testing, no?
&...
2017 Nov 23
2
[PATCH] Fix memory issue in Projection API
...hannel_out_float() and
> > > opus_projection_copy_channel_out_short(), you're zeroing the
> > values with
> > > this loop:
> > > for (i=0;i<frame_size;i++)
> > > {
> > > for (j=0;j<dst_stride;j++)
> > > {
> > > float_dst[i*dst_stride+j] = 0;
> > > }
> > > }
> > > That looks wrong, since you'll be overwriting all channels
> > instead of
> > > just th...
2005 Mar 23
3
[PATCH] promised MMX patches rc1
...st=frag->buffer[dst_framei];
+ src=frag->buffer[src_framei];
+ __asm__ __volatile__ (
+ " lea (%3, %3, 2), %%esi \n\t" /* esi=src_stride*3 */
+ " movq (%1), %%mm0 \n\t" /* src */
+ " lea (%2, %2, 2), %%edi \n\t" /* edi=dst_stride*3 */
+ " movq (%1, %3), %%mm1 \n\t" /* src+1x stride */
+ " movq (%1, %3, 2), %%mm2 \n\t" /* src+2x stride */
+ " movq (%1, %%esi), %%mm3 \n\t" /* src+3x stride */
+ " movq %%mm0, (%0) \n\t" /* dst */...
2017 Nov 24
3
[PATCH] Fix memory issue in Projection API
..._projection_copy_channel_out_short(), you're zeroing
> the
> > > values with
> > > > this loop:
> > > > for (i=0;i<frame_size;i++)
> > > > {
> > > > for (j=0;j<dst_stride;j++)
> > > > {
> > > > float_dst[i*dst_stride+j] = 0;
> > > > }
> > > > }
> > > > That looks wrong, since you'll be overwriting all
> channels
>...
2017 Nov 22
0
[PATCH] Fix memory issue in Projection API
...problems for the same reasons as 1).
Please check you haven't made that same mistake elsewhere.
3) In opus_projection_copy_channel_out_float() and
opus_projection_copy_channel_out_short(), you're zeroing the values with
this loop:
for (i=0;i<frame_size;i++)
{
for (j=0;j<dst_stride;j++)
{
float_dst[i*dst_stride+j] = 0;
}
}
That looks wrong, since you'll be overwriting all channels instead of
just the one you're copying. That should have come up in testing, no?
4) You might want to have a look at
mapping_matrix_multiply_channel_out_short(). I c...
2017 Nov 23
0
[PATCH] Fix memory issue in Projection API
...'t made that same mistake elsewhere.
>
> 3) In opus_projection_copy_channel_out_float() and
> opus_projection_copy_channel_out_short(), you're zeroing the values with
> this loop:
> for (i=0;i<frame_size;i++)
> {
> for (j=0;j<dst_stride;j++)
> {
> float_dst[i*dst_stride+j] = 0;
> }
> }
> That looks wrong, since you'll be overwriting all channels instead of
> just the one you're copying. That should have come up in testing, no?
>
> 4) You might wan...
2017 Dec 04
3
[PATCH] Fix memory issue in Projection API
...> zeroing the
> > > > values with
> > > > > this loop:
> > > > > for (i=0;i<frame_size;i++)
> > > > > {
> > > > > for (j=0;j<dst_stride;j++)
> > > > > {
> > > > > float_dst[i*dst_stride+j] = 0;
> > > > > }
> > > > > }
> > > > > That looks wrong, since you&...
2017 Nov 24
2
[PATCH] Fix memory issue in Projection API
...> zeroing the
> > > > values with
> > > > > this loop:
> > > > > for (i=0;i<frame_size;i++)
> > > > > {
> > > > > for (j=0;j<dst_stride;j++)
> > > > > {
> > > > > float_dst[i*dst_stride+j] = 0;
> > > > > }
> > > > > }
> > > > > That looks wrong, since you&...
2017 Nov 23
0
[PATCH] Fix memory issue in Projection API
... 3) In opus_projection_copy_channel_out_float() and
> > opus_projection_copy_channel_out_short(), you're zeroing the
> values with
> > this loop:
> > for (i=0;i<frame_size;i++)
> > {
> > for (j=0;j<dst_stride;j++)
> > {
> > float_dst[i*dst_stride+j] = 0;
> > }
> > }
> > That looks wrong, since you'll be overwriting all channels
> instead of
> > just the one you're copying. That shou...
2005 Mar 23
0
[PATCH]
...st=frag->buffer[dst_framei];
+ src=frag->buffer[src_framei];
+ __asm__ __volatile__ (
+ " lea (%3, %3, 2), %%esi \n\t" /* esi=src_stride*3 */
+ " movq (%1), %%mm0 \n\t" /* src */
+ " lea (%2, %2, 2), %%edi \n\t" /* edi=dst_stride*3 */
+ " movq (%1, %3), %%mm1 \n\t" /* src+1x stride */
+ " movq (%1, %3, 2), %%mm2 \n\t" /* src+2x stride */
+ " movq (%1, %%esi), %%mm3 \n\t" /* src+3x stride */
+ " movq %%mm0, (%0) \n\t" /* dst */...
2017 Nov 23
0
[PATCH] Fix memory issue in Projection API
...gt; > > opus_projection_copy_channel_out_short(), you're zeroing the
> > values with
> > > this loop:
> > > for (i=0;i<frame_size;i++)
> > > {
> > > for (j=0;j<dst_stride;j++)
> > > {
> > > float_dst[i*dst_stride+j] = 0;
> > > }
> > > }
> > > That looks wrong, since you'll be overwriting all channels
> > instead of
>...
2017 Nov 24
0
[PATCH] Fix memory issue in Projection API
..._out_short(), you're
> zeroing the
> > > values with
> > > > this loop:
> > > > for (i=0;i<frame_size;i++)
> > > > {
> > > > for (j=0;j<dst_stride;j++)
> > > > {
> > > > float_dst[i*dst_stride+j] = 0;
> > > > }
> > > > }
> > > > That looks wrong, since you'll be overwriting all...
2017 Nov 28
2
[PATCH] Fix memory issue in Projection API
... > > values with
> > > > > this loop:
> > > > > for (i=0;i<frame_size;i++)
> > > > > {
> > > > > for (j=0;j<dst_stride;j++)
> > > > > {
> > > > > float_dst[i*dst_stride+j] = 0;
> > > > > }
> > > > > }
> > > > ...
2017 Nov 27
0
[PATCH] Fix memory issue in Projection API
...>> > > > values with
>> > > > > this loop:
>> > > > > for (i=0;i<frame_size;i++)
>> > > > > {
>> > > > > for (j=0;j<dst_stride;j++)
>> > > > > {
>> > > > > float_dst[i*dst_stride+j] = 0;
>> > > > > }
>> > > > > }
>> > > > > That loo...
2010 Jan 06
3
Removal of Non-KMS support
I did a very quick pass at removing all the non-KMS support from the
DDX. It's tested on G80 but nowhere else currently, I thought some
discussion would be a good idea rather than just ripping it out :)
The non-KMS paths are messy, and lets face it, rotting badly. IMO the
KMS code is stable enough now that we can continue without the UMS
crutch, and indeed, the KMS code supports a lot more
2017 Dec 07
0
[PATCH] Fix memory issue in Projection API
...> > > > values with
> > > > > this loop:
> > > > > for (i=0;i<frame_size;i++)
> > > > > {
> > > > > for (j=0;j<dst_stride;j++)
> > > > > {
> > > > > float_dst[i*dst_stride+j] = 0;
> > > > > }
> > > > > }
> > > > > That loo...