mirror of
https://github.com/signalwire/freeswitch.git
synced 2026-11-17 23:08:35 +00:00
178 lines
8.9 KiB
Markdown
178 lines
8.9 KiB
Markdown
# Gemini Project Context: libyuv Row Functions
|
|
|
|
This file provides context for the core row-processing architecture of
|
|
libyuv. Use these guidelines when refactoring, reviewing, or generating
|
|
code within the `row_*.cc` files.
|
|
|
|
## Architectural Overview
|
|
|
|
Libyuv uses a dispatch system where high-level conversion functions call
|
|
optimized "Row" functions. These functions are categorized by SIMD architecture
|
|
and compiler compatibility.
|
|
|
|
## Source File Map
|
|
|
|
### x86 Architectures (32-bit and 64-bit)
|
|
|
|
* **row_gcc.cc**: **Master copy.** Contains inline assembly in GCC syntax for
|
|
GCC and Clang. Supports AVX, and AVX512. AVX512 implementations are strictly
|
|
for 64-bit targets.
|
|
* **row_win.cc**: Derivative of `row_gcc.cc`. Contains C++ intrinsics
|
|
specifically for Visual C++ (MSVC). Can be tested with Clang using
|
|
`-DLIBYUV_ENABLE_ROWWIN`.
|
|
* **Note**: Use either `row_gcc` or `row_win`, never both.
|
|
|
|
### Arm Architecture (32-bit)
|
|
|
|
* **row_neon.cc**: 32-bit Arm. Written entirely in inline assembly for
|
|
GCC/Clang.
|
|
|
|
### Arm Architecture (AArch64) (64-bit)
|
|
|
|
* **row_neon64.cc**: 64-bit Arm (AArch64) Neon. Written entirely in inline
|
|
assembly for GCC/Clang.
|
|
* **row_sve.cc**: Armv9 Scalable Vector Extension (SVE2).
|
|
* **row_sme.cc**: Armv9 Scalable Matrix Extension (SME) and Streaming SVE
|
|
(SSVE).
|
|
* **row_sve.h**: Streaming-compatible implementations shared across
|
|
**row_sve.cc** and **row_sme.cc**.
|
|
|
|
### Other Architectures
|
|
|
|
* **row_rvv.cc**: RISC-V Vector (RVV). Implemented using intrinsics. Optimized
|
|
for SiFive X280.
|
|
* **row_hvx.cc**: Qualcomm Hexagon Vector Extensions (HVX). Implemented using
|
|
intrinsics.
|
|
* **row_lsx.cc / row_lasx.cc**: Loongarch MIPS-like extensions.
|
|
|
|
### Utility and Fallbacks
|
|
|
|
* **row_common.cc**: Portable C/C++ versions and multi-step `ARGB` wrapper
|
|
row functions. This is the reference implementation.
|
|
* **row_any.cc**: Handles "remainder" pixels for widths not multiples of SIMD
|
|
register size. Used for SSE/AVX2, Neon, and LoongArch (LSX/LASX). Not
|
|
required for AVX-512, SVE2, SME, RVV, or HVX, which handle tail masking
|
|
directly inside the Row function.
|
|
|
|
## Deprecated and Lower-Priority Formats
|
|
|
|
The following datatypes and functions are deprecated or lower priority and are
|
|
not always fully hand-optimized across every SIMD extension:
|
|
|
|
* **16-bit RGB formats** (deprecated): `RGB565` (`FOURCC_RGBP`,
|
|
`FOURCC_L565`), `ARGB1555` (`FOURCC_RGBO`, `FOURCC_L555`, `FOURCC_5551`),
|
|
and `ARGB4444` (`FOURCC_R444`).
|
|
* **Legacy fixed-matrix RGB-to-YUV row functions** (deprecated in favor of
|
|
`*MatrixRow` variants): `RGB565ToYRow`, `RGB565ToUVRow`, `ARGB1555ToYRow`,
|
|
`ARGB1555ToUVRow`, `ARGB4444ToYRow`, `ARGB4444ToUVRow`.
|
|
* **Row-biplanar YUV** (deprecated): `M420` (`FOURCC_M420`).
|
|
|
|
When adding SIMD coverage for deprecated or less relevant formats, prefer
|
|
two-step wrapper implementations in `source/row_common.cc` that convert to or
|
|
from an intermediate stack-allocated `ARGB` row buffer (or delegate to existing
|
|
`*MatrixRow` functions) rather than writing large bespoke assembly kernels.
|
|
|
|
## Coding Guidelines
|
|
|
|
1. **Feature Macros**: Use the `HAS_` macros in `include/libyuv/row.h` to
|
|
enable or disable specific instruction set extensions.
|
|
2. **Inline asm clobbers (all architectures)**: Every register written by an
|
|
inline asm block must be listed in its clobber list: vector registers,
|
|
general purpose registers, mask/predicate registers, `"cc"` when flags
|
|
are set (`subs`, `cmp`, `whilelt`, ...) and `"memory"` when the block
|
|
stores. Declare them even if the ABI treats them as caller-saved: with
|
|
LTO/LTCG the compiler may keep values live in any register across the
|
|
asm, and other OS ABIs (e.g. Windows) differ in which registers are
|
|
callee-saved. Never remove a clobber because it "is not needed by the
|
|
ABI". Use real register names, not numbers (`"v22"`, not `"22"`). Order
|
|
the list `"memory"` first (when stores occur), then `"cc"` (on
|
|
architectures with flags), then registers.
|
|
|
|
### x86 Architectures (32-bit and 64-bit)
|
|
|
|
1. **AVX512 Logic**: AVX512 row functions are strictly enabled for **64-bit x86
|
|
only**. Do not create `_Any_` wrappers for AVX512 row functions; handle
|
|
arbitrary tail widths directly inside the Row function using `k1`-`k7`
|
|
opmask registers.
|
|
2. **MSVC Intrinsics (`row_win.cc`)**: When converting `row_gcc.cc` to
|
|
`row_win.cc`, use `_mm512_permutex2var_epi8`, not `_mm512_permi2var_epi8`
|
|
intrinsic.
|
|
3. **Clobber names**: Always declare vector clobbers as `"xmmN"`, never
|
|
`"ymmN"` or `"zmmN"`, including for AVX2 and AVX512 code. Clang for
|
|
Windows must preserve xmm6-xmm15 and does not save them when they are
|
|
declared as ymm or zmm.
|
|
4. **MSan Intrinsics (`*_win.cc`)**: Clang builds with MemorySanitizer
|
|
(`-fsanitize=memory`) define `LIBYUV_ENABLE_ROWWIN` and compile `*_win.cc`
|
|
(`row_win.cc`, `scale_win.cc`, `compare_win.cc`, `rotate_win.cc`) instead of
|
|
inline assembly (`*_gcc.cc`). When porting `*_gcc.cc` to `*_win.cc`, never
|
|
use AVX-512 intrinsics that LLVM's MSan pass does not model:
|
|
* **Down-converting masked stores**: Never use
|
|
`_mm512_mask_cvtepi32_storeu_epi8`, `_mm512_mask_cvtepi64_storeu_epi8`,
|
|
`_mm512_mask_cvtepi16_storeu_epi8`, or
|
|
`_mm512_mask_cvtepi32_storeu_epi16`. They lower to target-specific
|
|
`@llvm.x86.avx512.mask.pmov.*.mem.*` intrinsics that LLVM MSan does not
|
|
model, leaving the destination shadow memory poisoned (uninitialized).
|
|
Instead, convert to a narrower vector first and use a standard masked
|
|
store (e.g. `_mm_mask_storeu_epi8(dst, mask, _mm512_cvtepi32_epi8(v))`),
|
|
which lowers to `@llvm.masked.store` for MSan and still compiles to a
|
|
single `vpmovdb` instruction at `-O2`.
|
|
* **Ternary logic (`_mm*_ternarylogic_epi32` / `epi64`)**: Never use
|
|
`_mm512_ternarylogic_epi32`/`epi64` (or `_mm256_`/`_mm_` variants)
|
|
anywhere in `*_win.cc` when porting `vpternlogd`/`vpternlogq` from
|
|
`*_gcc.cc`. They lower to `@llvm.x86.avx512.pternlog.*`, which LLVM MSan
|
|
does not model bitwise—`handleUnknownIntrinsic` traps immediately with
|
|
`use-of-uninitialized-value` if *any* bit of any of the three inputs is
|
|
uninitialized (such as the unused alpha byte of an `RGBX` buffer or tail
|
|
padding), even if masked off by `imm8` or merely copied to memory.
|
|
* For boolean blends and inversions (e.g. `ARGBGrayRow_AVX512VNNI`,
|
|
`I422ToAR30Row_AVX512BW`, `ARGBToUV444MatrixRow_AVX512VNNIVBMI`),
|
|
use standard bitwise intrinsics (`_mm512_and_si512`,
|
|
`_mm512_andnot_si512`, `_mm512_or_si512`, `_mm512_xor_si512`), which
|
|
LLVM MSan models bit-by-bit and Clang `-O2` folds back into
|
|
`vpternlog` in non-MSan builds.
|
|
* For write-only partial channel updates into `dst` (e.g. writing
|
|
alpha in `ARGBCopyYToAlphaRow_AVX512BW`), use byte-masked stores
|
|
(`_mm512_mask_storeu_epi8`) instead of reading `dst` and blending.
|
|
* **Single-source variable permutes (`_mm512_permutexvar_epi*` /
|
|
`_mm512_mask_permutexvar_epi*`)**: Never use
|
|
`_mm512_permutexvar_epi{8,16,32,64}` or
|
|
`_mm512_mask_permutexvar_epi{8,16,32,64}` directly. They lower to
|
|
`@llvm.x86.avx512.permvar.*`, which LLVM MSan does not model
|
|
lane-by-lane—instead it ORs the input shadows without permuting them,
|
|
poisoning initialized output lanes whenever *any* lane of the input
|
|
vector is uninitialized (e.g. `_mm512_castsi128_si512`,
|
|
`_mm512_castsi256_si512`, unused `RGBX` alpha bytes, or uninitialized
|
|
tail lanes). Use the `PermutexvarEpi*` and `MaskPermutexvarEpi*`
|
|
helpers in `row_win.cc` / `scale_win.cc`, which use portable
|
|
lane-by-lane C loops under `LIBYUV_MSAN` and the native intrinsics
|
|
otherwise.
|
|
* **512-bit float-to-int truncation (`_mm512_cvttps_epi32`)**: Never use
|
|
`_mm512_cvttps_epi32` directly; it lowers to
|
|
`@llvm.x86.avx512.mask.cvttps2dq.512`, which LLVM MSan does not model
|
|
lane-by-lane. Use the `CvttpsEpi32` helper in `row_win.cc`, which uses
|
|
a portable lane-by-lane C loop under `LIBYUV_MSAN` and
|
|
`_mm512_cvttps_epi32` otherwise.
|
|
|
|
### AArch64
|
|
|
|
1. For SVE2/SME, aim to use full predicates (`ptrue`) and full vectors for the
|
|
main loop body and use predication only to mask the tail loop as needed.
|
|
Pointer updates should be done with `inc[bhwd]` or `add`, not by `incp`.
|
|
|
|
## Changelist (CL) Format Guidelines
|
|
|
|
When adding new code, remove trailing spaces.
|
|
When generating descriptions, follow the Chromium standard format. Wrap
|
|
commit message text at 72 characters
|
|
|
|
### Changelist (CL) Description Example
|
|
|
|
\[libyuv] Optimized ARGBToRGB24 for AVX2
|
|
|
|
Detailed technical explanation of the change. Describe the SIMD
|
|
optimization logic and any platform-specific constraints. Mention
|
|
if this impacts specific row files like row_gcc.cc.
|
|
|
|
Test: libyuv_unittest --gunit_filter=*ARGBToRGB24*
|
|
Bug: libyuv:12345, b/67890
|