libFLAC : Add asm versions for two _wide() functions.

GCC generates slow ia32 code for FLAC__lpc_restore_signal_wide() and
FLAC__lpc_compute_residual_from_qlp_coefficients_wide() so 24-bit
encoding/decoding is slower for GCC compile than for MSVS or ICC
compile. This patch adds ia32 asm versions of these functions.

Patch-from: lvqcl <lvqcl.mail@gmail.com>
This commit is contained in:
Erik de Castro Lopo
2014-01-07 21:35:03 +11:00
parent 8e4a45ac86
commit 7e9278934e
4 changed files with 568 additions and 1 deletions

View File

@@ -152,6 +152,7 @@ void FLAC__lpc_compute_residual_from_qlp_coefficients_wide(const FLAC__int32 *da
# ifdef FLAC__HAS_NASM
void FLAC__lpc_compute_residual_from_qlp_coefficients_asm_ia32(const FLAC__int32 *data, unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 residual[]);
void FLAC__lpc_compute_residual_from_qlp_coefficients_asm_ia32_mmx(const FLAC__int32 *data, unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 residual[]);
void FLAC__lpc_compute_residual_from_qlp_coefficients_wide_asm_ia32(const FLAC__int32 *data, unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 residual[]);
# endif
# endif
# if (defined FLAC__CPU_IA32 || defined FLAC__CPU_X86_64) && defined FLAC__HAS_X86INTRIN
@@ -187,6 +188,7 @@ void FLAC__lpc_restore_signal_wide(const FLAC__int32 residual[], unsigned data_l
# ifdef FLAC__HAS_NASM
void FLAC__lpc_restore_signal_asm_ia32(const FLAC__int32 residual[], unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 data[]);
void FLAC__lpc_restore_signal_asm_ia32_mmx(const FLAC__int32 residual[], unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 data[]);
void FLAC__lpc_restore_signal_wide_asm_ia32(const FLAC__int32 residual[], unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 data[]);
# endif /* FLAC__HAS_NASM */
# elif defined FLAC__CPU_PPC
void FLAC__lpc_restore_signal_asm_ppc_altivec_16(const FLAC__int32 residual[], unsigned data_len, const FLAC__int32 qlp_coeff[], unsigned order, int lp_quantization, FLAC__int32 data[]);