summaryrefslogtreecommitdiff
path: root/libavcodec/dcadsp.c
diff options
context:
space:
mode:
authorJames Almer <jamrial@gmail.com>2016-02-05 11:17:04 -0300
committerJames Almer <jamrial@gmail.com>2016-02-05 11:17:04 -0300
commit16af350ac5b19ee464121a0545f100f6c0063296 (patch)
tree4d980ddbe636aa952c17451212a45728f72ce6eb /libavcodec/dcadsp.c
parent6632802aa00c8fe2126697d9eb8e498c600aee79 (diff)
downloadffmpeg-16af350ac5b19ee464121a0545f100f6c0063296.tar.gz
avcodec/dcadsp: replace intptr_t with ptrdiff_t
Reviewed-by: Hendrik Leppkes <h.leppkes@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
Diffstat (limited to 'libavcodec/dcadsp.c')
-rw-r--r--libavcodec/dcadsp.c46
1 files changed, 23 insertions, 23 deletions
diff --git a/libavcodec/dcadsp.c b/libavcodec/dcadsp.c
index cee3d608b3..787d7b827a 100644
--- a/libavcodec/dcadsp.c
+++ b/libavcodec/dcadsp.c
@@ -27,8 +27,8 @@ static void decode_hf_c(int32_t **dst,
const int32_t *vq_index,
const int8_t hf_vq[1024][32],
int32_t scale_factors[32][2],
- intptr_t sb_start, intptr_t sb_end,
- intptr_t ofs, intptr_t len)
+ ptrdiff_t sb_start, ptrdiff_t sb_end,
+ ptrdiff_t ofs, ptrdiff_t len)
{
int i, j;
@@ -42,8 +42,8 @@ static void decode_hf_c(int32_t **dst,
static void decode_joint_c(int32_t **dst, int32_t **src,
const int32_t *scale_factors,
- intptr_t sb_start, intptr_t sb_end,
- intptr_t ofs, intptr_t len)
+ ptrdiff_t sb_start, ptrdiff_t sb_end,
+ ptrdiff_t ofs, ptrdiff_t len)
{
int i, j;
@@ -55,7 +55,7 @@ static void decode_joint_c(int32_t **dst, int32_t **src,
}
static void lfe_fir_float_c(float *pcm_samples, int32_t *lfe_samples,
- const float *filter_coeff, intptr_t npcmblocks,
+ const float *filter_coeff, ptrdiff_t npcmblocks,
int dec_select)
{
// Select decimation factor
@@ -85,19 +85,19 @@ static void lfe_fir_float_c(float *pcm_samples, int32_t *lfe_samples,
}
static void lfe_fir1_float_c(float *pcm_samples, int32_t *lfe_samples,
- const float *filter_coeff, intptr_t npcmblocks)
+ const float *filter_coeff, ptrdiff_t npcmblocks)
{
lfe_fir_float_c(pcm_samples, lfe_samples, filter_coeff, npcmblocks, 0);
}
static void lfe_fir2_float_c(float *pcm_samples, int32_t *lfe_samples,
- const float *filter_coeff, intptr_t npcmblocks)
+ const float *filter_coeff, ptrdiff_t npcmblocks)
{
lfe_fir_float_c(pcm_samples, lfe_samples, filter_coeff, npcmblocks, 1);
}
static void lfe_x96_float_c(float *dst, const float *src,
- float *hist, intptr_t len)
+ float *hist, ptrdiff_t len)
{
float prev = *hist;
int i;
@@ -119,7 +119,7 @@ static void sub_qmf32_float_c(SynthFilterContext *synth,
int32_t **subband_samples_lo,
int32_t **subband_samples_hi,
float *hist1, int *offset, float *hist2,
- const float *filter_coeff, intptr_t npcmblocks,
+ const float *filter_coeff, ptrdiff_t npcmblocks,
float scale)
{
LOCAL_ALIGNED(32, float, input, [32]);
@@ -148,7 +148,7 @@ static void sub_qmf64_float_c(SynthFilterContext *synth,
int32_t **subband_samples_lo,
int32_t **subband_samples_hi,
float *hist1, int *offset, float *hist2,
- const float *filter_coeff, intptr_t npcmblocks,
+ const float *filter_coeff, ptrdiff_t npcmblocks,
float scale)
{
LOCAL_ALIGNED(32, float, input, [64]);
@@ -192,7 +192,7 @@ static void sub_qmf64_float_c(SynthFilterContext *synth,
}
static void lfe_fir_fixed_c(int32_t *pcm_samples, int32_t *lfe_samples,
- const int32_t *filter_coeff, intptr_t npcmblocks)
+ const int32_t *filter_coeff, ptrdiff_t npcmblocks)
{
// Select decimation factor
int nlfesamples = npcmblocks >> 1;
@@ -219,7 +219,7 @@ static void lfe_fir_fixed_c(int32_t *pcm_samples, int32_t *lfe_samples,
}
static void lfe_x96_fixed_c(int32_t *dst, const int32_t *src,
- int32_t *hist, intptr_t len)
+ int32_t *hist, ptrdiff_t len)
{
int32_t prev = *hist;
int i;
@@ -241,7 +241,7 @@ static void sub_qmf32_fixed_c(SynthFilterContext *synth,
int32_t **subband_samples_lo,
int32_t **subband_samples_hi,
int32_t *hist1, int *offset, int32_t *hist2,
- const int32_t *filter_coeff, intptr_t npcmblocks)
+ const int32_t *filter_coeff, ptrdiff_t npcmblocks)
{
LOCAL_ALIGNED(32, int32_t, input, [32]);
int i, j;
@@ -265,7 +265,7 @@ static void sub_qmf64_fixed_c(SynthFilterContext *synth,
int32_t **subband_samples_lo,
int32_t **subband_samples_hi,
int32_t *hist1, int *offset, int32_t *hist2,
- const int32_t *filter_coeff, intptr_t npcmblocks)
+ const int32_t *filter_coeff, ptrdiff_t npcmblocks)
{
LOCAL_ALIGNED(32, int32_t, input, [64]);
int i, j;
@@ -295,7 +295,7 @@ static void sub_qmf64_fixed_c(SynthFilterContext *synth,
}
}
-static void decor_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_t len)
+static void decor_c(int32_t *dst, const int32_t *src, int coeff, ptrdiff_t len)
{
int i;
@@ -304,7 +304,7 @@ static void decor_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_t l
}
static void dmix_sub_xch_c(int32_t *dst1, int32_t *dst2,
- const int32_t *src, intptr_t len)
+ const int32_t *src, ptrdiff_t len)
{
int i;
@@ -315,7 +315,7 @@ static void dmix_sub_xch_c(int32_t *dst1, int32_t *dst2,
}
}
-static void dmix_sub_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_t len)
+static void dmix_sub_c(int32_t *dst, const int32_t *src, int coeff, ptrdiff_t len)
{
int i;
@@ -323,7 +323,7 @@ static void dmix_sub_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_
dst[i] -= mul15(src[i], coeff);
}
-static void dmix_add_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_t len)
+static void dmix_add_c(int32_t *dst, const int32_t *src, int coeff, ptrdiff_t len)
{
int i;
@@ -331,7 +331,7 @@ static void dmix_add_c(int32_t *dst, const int32_t *src, intptr_t coeff, intptr_
dst[i] += mul15(src[i], coeff);
}
-static void dmix_scale_c(int32_t *dst, intptr_t scale, intptr_t len)
+static void dmix_scale_c(int32_t *dst, int scale, ptrdiff_t len)
{
int i;
@@ -339,7 +339,7 @@ static void dmix_scale_c(int32_t *dst, intptr_t scale, intptr_t len)
dst[i] = mul15(dst[i], scale);
}
-static void dmix_scale_inv_c(int32_t *dst, intptr_t scale_inv, intptr_t len)
+static void dmix_scale_inv_c(int32_t *dst, int scale_inv, ptrdiff_t len)
{
int i;
@@ -347,7 +347,7 @@ static void dmix_scale_inv_c(int32_t *dst, intptr_t scale_inv, intptr_t len)
dst[i] = mul16(dst[i], scale_inv);
}
-static void filter0(int32_t *dst, const int32_t *src, int32_t coeff, intptr_t len)
+static void filter0(int32_t *dst, const int32_t *src, int32_t coeff, ptrdiff_t len)
{
int i;
@@ -355,7 +355,7 @@ static void filter0(int32_t *dst, const int32_t *src, int32_t coeff, intptr_t le
dst[i] -= mul22(src[i], coeff);
}
-static void filter1(int32_t *dst, const int32_t *src, int32_t coeff, intptr_t len)
+static void filter1(int32_t *dst, const int32_t *src, int32_t coeff, ptrdiff_t len)
{
int i;
@@ -364,7 +364,7 @@ static void filter1(int32_t *dst, const int32_t *src, int32_t coeff, intptr_t le
}
static void assemble_freq_bands_c(int32_t *dst, int32_t *src0, int32_t *src1,
- const int32_t *coeff, intptr_t len)
+ const int32_t *coeff, ptrdiff_t len)
{
int i;