vpx/vp8/common/generic/systemdependent.c

150 lines
4.8 KiB
C
Raw Normal View History

2010-05-18 17:58:33 +02:00
/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
2010-05-18 17:58:33 +02:00
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
2010-05-18 17:58:33 +02:00
*/
#include "vpx_ports/config.h"
#include "vp8/common/g_common.h"
#include "vp8/common/subpixel.h"
#include "vp8/common/loopfilter.h"
#include "vp8/common/recon.h"
#include "vp8/common/idct.h"
#include "vp8/common/onyxc_int.h"
2010-05-18 17:58:33 +02:00
#if CONFIG_MULTITHREAD
#if HAVE_UNISTD_H
#include <unistd.h>
#elif defined(_WIN32)
#include <windows.h>
typedef void (WINAPI *PGNSI)(LPSYSTEM_INFO);
#endif
#endif
2010-05-18 17:58:33 +02:00
extern void vp8_arch_x86_common_init(VP8_COMMON *ctx);
Add runtime CPU detection support for ARM. The primary goal is to allow a binary to be built which supports NEON, but can fall back to non-NEON routines, since some Android devices do not have NEON, even if they are otherwise ARMv7 (e.g., Tegra). The configure-generated flags HAVE_ARMV7, etc., are used to decide which versions of each function to build, and when CONFIG_RUNTIME_CPU_DETECT is enabled, the correct version is chosen at run time. In order for this to work, the CFLAGS must be set to something appropriate (e.g., without -mfpu=neon for ARMv7, and with appropriate -march and -mcpu for even earlier configurations), or the native C code will not be able to run. The ASFLAGS must remain set for the most advanced instruction set required at build time, since the ARM assembler will refuse to emit them otherwise. I have not attempted to make any changes to configure to do this automatically. Doing so will probably require the addition of new configure options. Many of the hooks for RTCD on ARM were already there, but a lot of the code had bit-rotted, and a good deal of the ARM-specific code is not integrated into the RTCD structs at all. I did not try to resolve the latter, merely to add the minimal amount of protection around them to allow RTCD to work. Those functions that were called based on an ifdef at the calling site were expanded to check the RTCD flags at that site, but they should be added to an RTCD struct somewhere in the future. The functions invoked with global function pointers still are, but these should be moved into an RTCD struct for thread safety (I believe every platform currently supported has atomic pointer stores, but this is not guaranteed). The encoder's boolhuff functions did not even have _c and armv7 suffixes, and the correct version was resolved at link time. The token packing functions did have appropriate suffixes, but the version was selected with a define, with no associated RTCD struct. However, for both of these, the only armv7 instruction they actually used was rbit, and this was completely superfluous, so I reworked them to avoid it. The only non-ARMv4 instruction remaining in them is clz, which is ARMv5 (not even ARMv5TE is required). Considering that there are no ARM-specific configs which are not at least ARMv5TE, I did not try to detect these at runtime, and simply enable them for ARMv5 and above. Finally, the NEON register saving code was completely non-reentrant, since it saved the registers to a global, static variable. I moved the storage for this onto the stack. A single binary built with this code was tested on an ARM11 (ARMv6) and a Cortex A8 (ARMv7 w/NEON), for both the encoder and decoder, and produced identical output, while using the correct accelerated functions on each. I did not test on any earlier processors. Change-Id: I45cbd63a614f4554c3b325c45d46c0806f009eaa
2010-10-21 00:39:11 +02:00
extern void vp8_arch_arm_common_init(VP8_COMMON *ctx);
2010-05-18 17:58:33 +02:00
#if CONFIG_MULTITHREAD
static int get_cpu_count()
{
int core_count = 16;
#if HAVE_UNISTD_H
#if defined(_SC_NPROCESSORS_ONLN)
core_count = sysconf(_SC_NPROCESSORS_ONLN);
#elif defined(_SC_NPROC_ONLN)
core_count = sysconf(_SC_NPROC_ONLN);
#endif
#elif defined(_WIN32)
{
PGNSI pGNSI;
SYSTEM_INFO sysinfo;
/* Call GetNativeSystemInfo if supported or
* GetSystemInfo otherwise. */
pGNSI = (PGNSI) GetProcAddress(
GetModuleHandle(TEXT("kernel32.dll")), "GetNativeSystemInfo");
if (pGNSI != NULL)
pGNSI(&sysinfo);
else
GetSystemInfo(&sysinfo);
core_count = sysinfo.dwNumberOfProcessors;
}
#else
/* other platforms */
#endif
return core_count > 0 ? core_count : 1;
}
#endif
2010-05-18 17:58:33 +02:00
void vp8_machine_specific_config(VP8_COMMON *ctx)
{
#if CONFIG_RUNTIME_CPU_DETECT
VP8_COMMON_RTCD *rtcd = &ctx->rtcd;
rtcd->idct.idct1 = vp8_short_idct4x4llm_1_c;
rtcd->idct.idct16 = vp8_short_idct4x4llm_c;
rtcd->idct.idct1_scalar_add = vp8_dc_only_idct_add_c;
2010-05-18 17:58:33 +02:00
rtcd->idct.iwalsh1 = vp8_short_inv_walsh4x4_1_c;
rtcd->idct.iwalsh16 = vp8_short_inv_walsh4x4_c;
rtcd->recon.copy16x16 = vp8_copy_mem16x16_c;
rtcd->recon.copy8x8 = vp8_copy_mem8x8_c;
rtcd->recon.copy8x4 = vp8_copy_mem8x4_c;
rtcd->recon.recon = vp8_recon_b_c;
2010-05-18 17:58:33 +02:00
rtcd->recon.recon2 = vp8_recon2b_c;
rtcd->recon.recon4 = vp8_recon4b_c;
rtcd->recon.recon_mb = vp8_recon_mb_c;
rtcd->recon.recon_mby = vp8_recon_mby_c;
rtcd->recon.build_intra_predictors_mby =
vp8_build_intra_predictors_mby;
rtcd->recon.build_intra_predictors_mby_s =
vp8_build_intra_predictors_mby_s;
rtcd->recon.build_intra_predictors_mbuv =
vp8_build_intra_predictors_mbuv;
rtcd->recon.build_intra_predictors_mbuv_s =
vp8_build_intra_predictors_mbuv_s;
rtcd->recon.intra4x4_predict =
vp8_intra4x4_predict;
2010-05-18 17:58:33 +02:00
rtcd->subpix.sixtap16x16 = vp8_sixtap_predict16x16_c;
rtcd->subpix.sixtap8x8 = vp8_sixtap_predict8x8_c;
rtcd->subpix.sixtap8x4 = vp8_sixtap_predict8x4_c;
rtcd->subpix.sixtap4x4 = vp8_sixtap_predict_c;
rtcd->subpix.bilinear16x16 = vp8_bilinear_predict16x16_c;
rtcd->subpix.bilinear8x8 = vp8_bilinear_predict8x8_c;
rtcd->subpix.bilinear8x4 = vp8_bilinear_predict8x4_c;
rtcd->subpix.bilinear4x4 = vp8_bilinear_predict4x4_c;
rtcd->loopfilter.normal_mb_v = vp8_loop_filter_mbv_c;
rtcd->loopfilter.normal_b_v = vp8_loop_filter_bv_c;
rtcd->loopfilter.normal_mb_h = vp8_loop_filter_mbh_c;
rtcd->loopfilter.normal_b_h = vp8_loop_filter_bh_c;
rtcd->loopfilter.simple_mb_v = vp8_loop_filter_simple_vertical_edge_c;
2010-05-18 17:58:33 +02:00
rtcd->loopfilter.simple_b_v = vp8_loop_filter_bvs_c;
rtcd->loopfilter.simple_mb_h = vp8_loop_filter_simple_horizontal_edge_c;
2010-05-18 17:58:33 +02:00
rtcd->loopfilter.simple_b_h = vp8_loop_filter_bhs_c;
#if CONFIG_POSTPROC || (CONFIG_VP8_ENCODER && CONFIG_INTERNAL_STATS)
rtcd->postproc.down = vp8_mbpost_proc_down_c;
rtcd->postproc.across = vp8_mbpost_proc_across_ip_c;
rtcd->postproc.downacross = vp8_post_proc_down_and_across_c;
rtcd->postproc.addnoise = vp8_plane_add_noise_c;
rtcd->postproc.blend_mb_inner = vp8_blend_mb_inner_c;
rtcd->postproc.blend_mb_outer = vp8_blend_mb_outer_c;
rtcd->postproc.blend_b = vp8_blend_b_c;
2010-05-18 17:58:33 +02:00
#endif
#endif
#if ARCH_X86 || ARCH_X86_64
vp8_arch_x86_common_init(ctx);
#endif
experiment extending the quantizer range Prior to this change, VP8 min quantizer is 4, which caps the highest quality around 51DB. This experimental change extends the min quantizer to 1, removes the cap and allows the highest quality to be around ~73DB, consistent with the fdct/idct round trip error. To test this change, at configure time use options: --enable-experimental --enable-extend_qrange The following is a brief log of changes in each of the patch sets patch set 1: In this commit, the quantization/dequantization constants are kept unchanged, instead scaling factor 4 is rolled into fdct/idct. Fixed Q0 encoding tests on mobile: Before: 9560.567kbps Overall PSNR:50.255DB VPXSSIM:98.288 Now: 18035.774kbps Overall PSNR:73.022DB VPXSSIM:99.991 patch set 2: regenerated dc/ac quantizer lookup tables based on the scaling factor rolled in the fdct/idct. Also slightly extended the range towards the high quantizer end. patch set 3: slightly tweaked the quantizer tables and generated bits_per_mb table based on Paul's suggestions. patch set 4: fix a typo in idct, re-calculated tables relating active max Q to active min Q patch set 5: added rdmult lookup table based on Q patch set 6: fix rdmult scale: dct coefficient has scaled up by 4 patch set 7: make transform coefficients to be within 16bits patch set 8: normalize 2nd order quantizers patch set 9: fix mis-spellings patch set 10: change the configure script and macros to allow experimental code to be enabled at configure time with --enable-extend_qrange patch set 11: rebase for merge Change-Id: Ib50641ddd44aba2a52ed890222c309faa31cc59c
2010-12-02 00:50:14 +01:00
Add runtime CPU detection support for ARM. The primary goal is to allow a binary to be built which supports NEON, but can fall back to non-NEON routines, since some Android devices do not have NEON, even if they are otherwise ARMv7 (e.g., Tegra). The configure-generated flags HAVE_ARMV7, etc., are used to decide which versions of each function to build, and when CONFIG_RUNTIME_CPU_DETECT is enabled, the correct version is chosen at run time. In order for this to work, the CFLAGS must be set to something appropriate (e.g., without -mfpu=neon for ARMv7, and with appropriate -march and -mcpu for even earlier configurations), or the native C code will not be able to run. The ASFLAGS must remain set for the most advanced instruction set required at build time, since the ARM assembler will refuse to emit them otherwise. I have not attempted to make any changes to configure to do this automatically. Doing so will probably require the addition of new configure options. Many of the hooks for RTCD on ARM were already there, but a lot of the code had bit-rotted, and a good deal of the ARM-specific code is not integrated into the RTCD structs at all. I did not try to resolve the latter, merely to add the minimal amount of protection around them to allow RTCD to work. Those functions that were called based on an ifdef at the calling site were expanded to check the RTCD flags at that site, but they should be added to an RTCD struct somewhere in the future. The functions invoked with global function pointers still are, but these should be moved into an RTCD struct for thread safety (I believe every platform currently supported has atomic pointer stores, but this is not guaranteed). The encoder's boolhuff functions did not even have _c and armv7 suffixes, and the correct version was resolved at link time. The token packing functions did have appropriate suffixes, but the version was selected with a define, with no associated RTCD struct. However, for both of these, the only armv7 instruction they actually used was rbit, and this was completely superfluous, so I reworked them to avoid it. The only non-ARMv4 instruction remaining in them is clz, which is ARMv5 (not even ARMv5TE is required). Considering that there are no ARM-specific configs which are not at least ARMv5TE, I did not try to detect these at runtime, and simply enable them for ARMv5 and above. Finally, the NEON register saving code was completely non-reentrant, since it saved the registers to a global, static variable. I moved the storage for this onto the stack. A single binary built with this code was tested on an ARM11 (ARMv6) and a Cortex A8 (ARMv7 w/NEON), for both the encoder and decoder, and produced identical output, while using the correct accelerated functions on each. I did not test on any earlier processors. Change-Id: I45cbd63a614f4554c3b325c45d46c0806f009eaa
2010-10-21 00:39:11 +02:00
#if ARCH_ARM
vp8_arch_arm_common_init(ctx);
#endif
experiment extending the quantizer range Prior to this change, VP8 min quantizer is 4, which caps the highest quality around 51DB. This experimental change extends the min quantizer to 1, removes the cap and allows the highest quality to be around ~73DB, consistent with the fdct/idct round trip error. To test this change, at configure time use options: --enable-experimental --enable-extend_qrange The following is a brief log of changes in each of the patch sets patch set 1: In this commit, the quantization/dequantization constants are kept unchanged, instead scaling factor 4 is rolled into fdct/idct. Fixed Q0 encoding tests on mobile: Before: 9560.567kbps Overall PSNR:50.255DB VPXSSIM:98.288 Now: 18035.774kbps Overall PSNR:73.022DB VPXSSIM:99.991 patch set 2: regenerated dc/ac quantizer lookup tables based on the scaling factor rolled in the fdct/idct. Also slightly extended the range towards the high quantizer end. patch set 3: slightly tweaked the quantizer tables and generated bits_per_mb table based on Paul's suggestions. patch set 4: fix a typo in idct, re-calculated tables relating active max Q to active min Q patch set 5: added rdmult lookup table based on Q patch set 6: fix rdmult scale: dct coefficient has scaled up by 4 patch set 7: make transform coefficients to be within 16bits patch set 8: normalize 2nd order quantizers patch set 9: fix mis-spellings patch set 10: change the configure script and macros to allow experimental code to be enabled at configure time with --enable-extend_qrange patch set 11: rebase for merge Change-Id: Ib50641ddd44aba2a52ed890222c309faa31cc59c
2010-12-02 00:50:14 +01:00
#if CONFIG_EXTEND_QRANGE
rtcd->idct.idct1 = vp8_short_idct4x4llm_1_c;
rtcd->idct.idct16 = vp8_short_idct4x4llm_c;
rtcd->idct.idct1_scalar_add = vp8_dc_only_idct_add_c;
rtcd->idct.iwalsh1 = vp8_short_inv_walsh4x4_1_c;
rtcd->idct.iwalsh16 = vp8_short_inv_walsh4x4_c;
#endif
Add runtime CPU detection support for ARM. The primary goal is to allow a binary to be built which supports NEON, but can fall back to non-NEON routines, since some Android devices do not have NEON, even if they are otherwise ARMv7 (e.g., Tegra). The configure-generated flags HAVE_ARMV7, etc., are used to decide which versions of each function to build, and when CONFIG_RUNTIME_CPU_DETECT is enabled, the correct version is chosen at run time. In order for this to work, the CFLAGS must be set to something appropriate (e.g., without -mfpu=neon for ARMv7, and with appropriate -march and -mcpu for even earlier configurations), or the native C code will not be able to run. The ASFLAGS must remain set for the most advanced instruction set required at build time, since the ARM assembler will refuse to emit them otherwise. I have not attempted to make any changes to configure to do this automatically. Doing so will probably require the addition of new configure options. Many of the hooks for RTCD on ARM were already there, but a lot of the code had bit-rotted, and a good deal of the ARM-specific code is not integrated into the RTCD structs at all. I did not try to resolve the latter, merely to add the minimal amount of protection around them to allow RTCD to work. Those functions that were called based on an ifdef at the calling site were expanded to check the RTCD flags at that site, but they should be added to an RTCD struct somewhere in the future. The functions invoked with global function pointers still are, but these should be moved into an RTCD struct for thread safety (I believe every platform currently supported has atomic pointer stores, but this is not guaranteed). The encoder's boolhuff functions did not even have _c and armv7 suffixes, and the correct version was resolved at link time. The token packing functions did have appropriate suffixes, but the version was selected with a define, with no associated RTCD struct. However, for both of these, the only armv7 instruction they actually used was rbit, and this was completely superfluous, so I reworked them to avoid it. The only non-ARMv4 instruction remaining in them is clz, which is ARMv5 (not even ARMv5TE is required). Considering that there are no ARM-specific configs which are not at least ARMv5TE, I did not try to detect these at runtime, and simply enable them for ARMv5 and above. Finally, the NEON register saving code was completely non-reentrant, since it saved the registers to a global, static variable. I moved the storage for this onto the stack. A single binary built with this code was tested on an ARM11 (ARMv6) and a Cortex A8 (ARMv7 w/NEON), for both the encoder and decoder, and produced identical output, while using the correct accelerated functions on each. I did not test on any earlier processors. Change-Id: I45cbd63a614f4554c3b325c45d46c0806f009eaa
2010-10-21 00:39:11 +02:00
#if CONFIG_MULTITHREAD
ctx->processor_core_count = get_cpu_count();
#endif /* CONFIG_MULTITHREAD */
2010-05-18 17:58:33 +02:00
}