From: Maarten Bosmans <[email protected]>
---
sbc/sbc_primitives_armv6.c | 4 ++--
sbc/sbc_primitives_iwmmxt.c | 4 ++--
sbc/sbc_primitives_mmx.c | 14 +++++++-------
sbc/sbc_primitives_neon.c | 28 ++++++++++++++--------------
4 files changed, 25 insertions(+), 25 deletions(-)
diff --git a/sbc/sbc_primitives_armv6.c b/sbc/sbc_primitives_armv6.c
index 9586098..b321272 100644
--- a/sbc/sbc_primitives_armv6.c
+++ b/sbc/sbc_primitives_armv6.c
@@ -41,7 +41,7 @@
static void __attribute__((naked)) sbc_analyze_four_armv6()
{
/* r0 = in, r1 = out, r2 = consts */
- asm volatile (
+ __asm__ volatile (
"push {r1, r4-r7, lr}\n"
"push {r8-r11}\n"
"ldrd r4, r5, [r0, #0]\n"
@@ -112,7 +112,7 @@ static void __attribute__((naked)) sbc_analyze_four_armv6()
static void __attribute__((naked)) sbc_analyze_eight_armv6()
{
/* r0 = in, r1 = out, r2 = consts */
- asm volatile (
+ __asm__ volatile (
"push {r1, r4-r7, lr}\n"
"push {r8-r11}\n"
"ldrd r4, r5, [r0, #24]\n"
diff --git a/sbc/sbc_primitives_iwmmxt.c b/sbc/sbc_primitives_iwmmxt.c
index 213967e..e0bd060 100644
--- a/sbc/sbc_primitives_iwmmxt.c
+++ b/sbc/sbc_primitives_iwmmxt.c
@@ -42,7 +42,7 @@
static inline void sbc_analyze_four_iwmmxt(const int16_t *in, int32_t *out,
const FIXED_T *consts)
{
- asm volatile (
+ __asm__ volatile (
"wldrd wr0, [%0]\n"
"tbcstw wr4, %2\n"
"wldrd wr2, [%1]\n"
@@ -115,7 +115,7 @@ static inline void sbc_analyze_four_iwmmxt(const int16_t *in, int32_t *out,
static inline void sbc_analyze_eight_iwmmxt(const int16_t *in, int32_t *out,
const FIXED_T *consts)
{
- asm volatile (
+ __asm__ volatile (
"wldrd wr0, [%0]\n"
"tbcstw wr15, %2\n"
"wldrd wr1, [%0, #8]\n"
diff --git a/sbc/sbc_primitives_mmx.c b/sbc/sbc_primitives_mmx.c
index 7f2fbc3..27e9a56 100644
--- a/sbc/sbc_primitives_mmx.c
+++ b/sbc/sbc_primitives_mmx.c
@@ -45,7 +45,7 @@ static inline void sbc_analyze_four_mmx(const int16_t *in, int32_t *out,
1 << (SBC_PROTO_FIXED4_SCALE - 1),
1 << (SBC_PROTO_FIXED4_SCALE - 1),
};
- asm volatile (
+ __asm__ volatile (
"movq (%0), %%mm0\n"
"movq 8(%0), %%mm1\n"
"pmaddwd (%1), %%mm0\n"
@@ -111,7 +111,7 @@ static inline void sbc_analyze_eight_mmx(const int16_t *in, int32_t *out,
1 << (SBC_PROTO_FIXED8_SCALE - 1),
1 << (SBC_PROTO_FIXED8_SCALE - 1),
};
- asm volatile (
+ __asm__ volatile (
"movq (%0), %%mm0\n"
"movq 8(%0), %%mm1\n"
"movq 16(%0), %%mm2\n"
@@ -258,7 +258,7 @@ static inline void sbc_analyze_4b_4s_mmx(int16_t *x, int32_t *out,
out += out_stride;
sbc_analyze_four_mmx(x + 0, out, analysis_consts_fixed4_simd_even);
- asm volatile ("emms\n");
+ __asm__ volatile ("emms\n");
}
static inline void sbc_analyze_4b_8s_mmx(int16_t *x, int32_t *out,
@@ -273,7 +273,7 @@ static inline void sbc_analyze_4b_8s_mmx(int16_t *x, int32_t *out,
out += out_stride;
sbc_analyze_eight_mmx(x + 0, out, analysis_consts_fixed8_simd_even);
- asm volatile ("emms\n");
+ __asm__ volatile ("emms\n");
}
static void sbc_calc_scalefactors_mmx(
@@ -291,7 +291,7 @@ static void sbc_calc_scalefactors_mmx(
for (sb = 0; sb < subbands; sb += 2) {
blk = (blocks - 1) * (((char *) &sb_sample_f[1][0][0] -
(char *) &sb_sample_f[0][0][0]));
- asm volatile (
+ __asm__ volatile (
"movq (%4), %%mm0\n"
"1:\n"
"movq (%1, %0), %%mm1\n"
@@ -326,7 +326,7 @@ static void sbc_calc_scalefactors_mmx(
: "cc", "memory");
}
}
- asm volatile ("emms\n");
+ __asm__ volatile ("emms\n");
}
static int check_mmx_support(void)
@@ -335,7 +335,7 @@ static int check_mmx_support(void)
return 1; /* We assume that all 64-bit processors have MMX support */
#else
int cpuid_feature_information;
- asm volatile (
+ __asm__ volatile (
/* According to Intel manual, CPUID instruction is supported
* if the value of ID bit (bit 21) in EFLAGS can be modified */
"pushf\n"
diff --git a/sbc/sbc_primitives_neon.c b/sbc/sbc_primitives_neon.c
index 0572158..5d4d0e3 100644
--- a/sbc/sbc_primitives_neon.c
+++ b/sbc/sbc_primitives_neon.c
@@ -44,7 +44,7 @@ static inline void _sbc_analyze_four_neon(const int16_t *in, int32_t *out,
/* TODO: merge even and odd cases (or even merge all four calls to this
* function) in order to have only aligned reads from 'in' array
* and reduce number of load instructions */
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d4, d5}, [%0, :64]!\n"
"vld1.16 {d8, d9}, [%1, :128]!\n"
@@ -104,7 +104,7 @@ static inline void _sbc_analyze_eight_neon(const int16_t *in, int32_t *out,
/* TODO: merge even and odd cases (or even merge all four calls to this
* function) in order to have only aligned reads from 'in' array
* and reduce number of load instructions */
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d4, d5}, [%0, :64]!\n"
"vld1.16 {d8, d9}, [%1, :128]!\n"
@@ -247,7 +247,7 @@ static void sbc_calc_scalefactors_neon(
for (sb = 0; sb < subbands; sb += 4) {
int blk = blocks;
int32_t *in = &sb_sample_f[0][ch][sb];
- asm volatile (
+ __asm__ volatile (
"vmov.s32 q0, #0\n"
"vmov.s32 q1, %[c1]\n"
"vmov.s32 q14, #1\n"
@@ -306,7 +306,7 @@ int sbc_calc_scalefactors_j_neon(
i = subbands;
- asm volatile (
+ __asm__ volatile (
/*
* constants: q13 = (31 - SCALE_OUT_BITS), q14 = 1
* input: q0 = ((1 << SCALE_OUT_BITS) + 1)
@@ -561,7 +561,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_4s_neon_internal(
if (position < nsamples) {
int16_t *dst = &X[0][SBC_X_BUFFER_SIZE - 40];
int16_t *src = &X[0][position];
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
"vst1.16 {d0, d1, d2, d3}, [%[dst], :128]!\n"
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
@@ -575,7 +575,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_4s_neon_internal(
if (nchannels > 1) {
dst = &X[1][SBC_X_BUFFER_SIZE - 40];
src = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
"vst1.16 {d0, d1, d2, d3}, [%[dst], :128]!\n"
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
@@ -594,7 +594,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_4s_neon_internal(
/* poor 'pcm' alignment */
int16_t *x = &X[0][position];
int16_t *y = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #16\n"
@@ -628,7 +628,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_4s_neon_internal(
/* proper 'pcm' alignment */
int16_t *x = &X[0][position];
int16_t *y = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #16\n"
@@ -658,7 +658,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_4s_neon_internal(
"d20", "d21", "d22", "d23");
} else {
int16_t *x = &X[0][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #16\n"
@@ -703,7 +703,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_8s_neon_internal(
if (position < nsamples) {
int16_t *dst = &X[0][SBC_X_BUFFER_SIZE - 72];
int16_t *src = &X[0][position];
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
"vst1.16 {d0, d1, d2, d3}, [%[dst], :128]!\n"
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
@@ -721,7 +721,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_8s_neon_internal(
if (nchannels > 1) {
dst = &X[1][SBC_X_BUFFER_SIZE - 72];
src = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
"vst1.16 {d0, d1, d2, d3}, [%[dst], :128]!\n"
"vld1.16 {d0, d1, d2, d3}, [%[src], :128]!\n"
@@ -744,7 +744,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_8s_neon_internal(
/* poor 'pcm' alignment */
int16_t *x = &X[0][position];
int16_t *y = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1, d2, d3}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #32\n"
@@ -782,7 +782,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_8s_neon_internal(
/* proper 'pcm' alignment */
int16_t *x = &X[0][position];
int16_t *y = &X[1][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1, d2, d3}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #32\n"
@@ -816,7 +816,7 @@ static SBC_ALWAYS_INLINE int sbc_enc_process_input_8s_neon_internal(
"d20", "d21", "d22", "d23");
} else {
int16_t *x = &X[0][position];
- asm volatile (
+ __asm__ volatile (
"vld1.8 {d0, d1, d2, d3}, [%[perm], :128]\n"
"1:\n"
"sub %[x], %[x], #32\n"
--
1.7.6.1
Hi,
On Tue, Sep 06, 2011, Luiz Augusto von Dentz wrote:
> From: Maarten Bosmans <[email protected]>
>
> ---
> sbc/sbc_primitives_armv6.c | 4 ++--
> sbc/sbc_primitives_iwmmxt.c | 4 ++--
> sbc/sbc_primitives_mmx.c | 14 +++++++-------
> sbc/sbc_primitives_neon.c | 28 ++++++++++++++--------------
> 4 files changed, 25 insertions(+), 25 deletions(-)
Applied. Thanks.
Johan
On Tue, Sep 6, 2011 at 1:02 PM, Maarten Bosmans <[email protected]> wrote:
> 2011/9/6 Siarhei Siamashka <[email protected]>:
>> On Tue, Sep 6, 2011 at 10:40 AM, Luiz Augusto von Dentz
>> <[email protected]> wrote:
>>> From: Maarten Bosmans <[email protected]>
>>>
>>> ---
>>> sbc/sbc_primitives_armv6.c | 4 ++--
>>> sbc/sbc_primitives_iwmmxt.c | 4 ++--
>>> sbc/sbc_primitives_mmx.c | 14 +++++++-------
>>> sbc/sbc_primitives_neon.c | 28 ++++++++++++++--------------
>>> 4 files changed, 25 insertions(+), 25 deletions(-)
>>
>> Not that I'm opposing this change, but having some rationale and
>> explanations in the commit message would be nice.
>
> First: consistency. __asm__ was already used elsewhere in the files,
> so using that throughout is cleaner.
This argument alone is only about cosmetics. It would be not very good
to end up flipping between '__asm__' vs. 'asm' back and forth if
somebody aesthetically does not like the underscores, or some legacy
version of gcc or the other compiler fails to work with '__asm__' but
works with 'asm', etc.
> Second: both asm and __asm__ are GCC-specific extensions, not defined
> in the C standard. When compiling with --std=gnu99 both are
> recognized, but when using --std=c99 only __asm__ is recognized to
> make it perfectly clear that you're not using some standard C99
> construct, but a GCC-extension.
This whole inline assembly code in question is not defined in the C
standard. It is gcc specific, or at least specific for the use with
the compilers which claim to be gcc compatible and provide __GNUC_
define. Does the use of 'asm' instead of '__asm__' already cause some
problems somewhere? Like has somebody switched to using '--std=c99'
option or maybe going to use it soon?
Right now it is not obvious whether this is a bugfix or cosmetics by
just looking at the commit message. That's why I wanted to clarify
these additional details. The informative commit message could provide
a great deal of information to the one who might need to change this
code in the future or would like to back out your patch for whatever
reason.
--
Best regards,
Siarhei Siamashka
2011/9/6 Siarhei Siamashka <[email protected]>:
> On Tue, Sep 6, 2011 at 10:40 AM, Luiz Augusto von Dentz
> <[email protected]> wrote:
>> From: Maarten Bosmans <[email protected]>
>>
>> ---
>> sbc/sbc_primitives_armv6.c | 4 ++--
>> sbc/sbc_primitives_iwmmxt.c | 4 ++--
>> sbc/sbc_primitives_mmx.c | 14 +++++++-------
>> sbc/sbc_primitives_neon.c | 28 ++++++++++++++--------------
>> 4 files changed, 25 insertions(+), 25 deletions(-)
>
> Not that I'm opposing this change, but having some rationale and
> explanations in the commit message would be nice.
First: consistency. __asm__ was already used elsewhere in the files,
so using that throughout is cleaner.
Second: both asm and __asm__ are GCC-specific extensions, not defined
in the C standard. When compiling with --std=gnu99 both are
recognized, but when using --std=c99 only __asm__ is recognized to
make it perfectly clear that you're not using some standard C99
construct, but a GCC-extension.
Maarten
Maarten
On Tue, Sep 6, 2011 at 10:40 AM, Luiz Augusto von Dentz
<[email protected]> wrote:
> From: Maarten Bosmans <[email protected]>
>
> ---
> sbc/sbc_primitives_armv6.c | 4 ++--
> sbc/sbc_primitives_iwmmxt.c | 4 ++--
> sbc/sbc_primitives_mmx.c | 14 +++++++-------
> sbc/sbc_primitives_neon.c | 28 ++++++++++++++--------------
> 4 files changed, 25 insertions(+), 25 deletions(-)
Not that I'm opposing this change, but having some rationale and
explanations in the commit message would be nice.
--
Best regards,
Siarhei Siamashka