vasilito/RedBear-OS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,583 Commits 19 Branches 0 Tags
0cf88bf1cbdbbbfcb623db8331dc2dd94d2aae9c
Commit Graph

156 Commits

Author SHA1 Message Date
vasilito 0cf88bf1cb intel: GEM performance counters, TTM power state, code comments 
gem_perf.rs (130 lines):
  PerfCounters: 22 atomic counters for all GEM operations
    create/close/mmap/execbuffer/tiling/domain/wait/pin/unpin
    export/import, context, request submit/complete
    eviction, shrink, throttle, fence, ttm_move, clflush
  PerfSnapshot: cloneable point-in-time counter snapshot
  TtmPowerState: suspend/resume with region saving

gem_object.rs: added documentation comments
  MemoryRegionType: System=DRAM, LocalMemory=VRAM, Stolen=BIOS
  CacheLevel: GPU cacheability (affects MOCS/PAT)

Ported from Linux 7.1:
  i915_perf.c counters → PerfCounters
  i915_gem_ttm_pm.c → TtmPowerState

GEM subdirectory: 25 files, 2,280 lines, 0 errors
 2026-06-02 10:34:27 +03:00
vasilito 76610fc8d0 intel: GEM ioctl dispatch + buffer validation 
gem_dispatch.rs (120 lines):
  GemIoctlDispatch: thread-safe ioctl router
    All GEM managers behind Arc<Mutex<>> for shared access
    create/close/pin/unpin/cache/gtt_offset operations
    context_create/destroy delegation
    set_tiling/get_tiling with lock poisoning handling

  BufferValidator: static validation helpers
    validate_size: 0 < size <= 4GB check
    validate_offset_length: bounds + DWORD alignment
    validate_alignment: 4K alignment requirement
    validate_handle_list: batch validation of all handles

Ported from Linux 7.1:
  i915_gem_ioctls.h → GemIoctlDispatch dispatch constants
  i915_gem.c validate helpers → BufferValidator

GEM subdirectory: 24 files, 2,150 lines, 0 errors
 2026-06-02 10:14:16 +03:00
vasilito eff3e6a850 intel: GEM render state, fence registers, FBC tracking 
gem_state.rs (170 lines):
  RenderState: gen-specific golden context initialization
    init_gen9: STATE_BASE_ADDRESS + CC_STATE_POINTERS programming
    init_gen8: simplified context image
    GGTT-backed with proper page mapping

  FenceRegisterState: 32-slot fence register pool
    allocate/release with tiling + pitch tracking
    find_by_handle reverse lookup
    used_count query

  FrameBufferCompressionState:
    Register/unregister scanout buffers
    mark_compressed/uncompressed with CFB offset tracking
    Per-buffer compressed byte accounting

Ported from Linux 7.1:
  i915_gem_render_state.c → RenderState
  i915_gem_fence.c → FenceRegisterState
  i915_gem_gtt.c (partial) → FBC tracking

GEM subdirectory: 23 files, 2,030 lines, 0 errors
 2026-06-02 10:11:01 +03:00
vasilito f904f59b68 intel: GEM eviction manager, fence objects, wound/wait mutex 
gem_evict.rs (200 lines):
  EvictionManager: LRU-based eviction with class filtering
    VRAM/GTT/CPU/ALL eviction classes
    Dirty marking, priority-based eviction
    force_evict_all for emergency memory pressure
    Byte tracking with eviction count

  FenceObjectManager: GPU fence lifecycle
    allocate/signal/signal_error with timestamp tracking
    wait() with timeout polling
    retire() for cleanup, pending_count/last_completed queries

  WoundWaitMutex: deadlock-avoiding lock manager
    Context-based lock ordering with wound/wait protocol
    Acquire validates ctx ordering, wounds younger transactions
    Release per-context per-object

Ported from Linux 7.1:
  i915_gem_evict.c → EvictionManager
  i915_gem_fence.c → FenceObjectManager
  i915_gem_ww.c → WoundWaitMutex

GEM subdirectory: 22 files, 1,860 lines, 0 errors
 2026-06-02 10:00:44 +03:00
vasilito 6ede7a045c intel: GEM request tracking, scheduler, statistics 
gem_request.rs (150 lines):
  RequestManager: fence-based GPU request lifecycle
    create/complete/retire_completed with inflight limit
    pending_for_context query
  Scheduler: priority-based runqueue with dequeue
    i32 priority levels, highest-first dispatch
    QueuedRequest with fence/ctx/engine/priority/timestamp
  GemStatistics: global counters
    create/close/submit/complete/retire counts
    peak inflight/objects, byte allocation tracking
  EngineClass enum: Render/Blitter/Video/VideoEnhance/Compute

Ported from Linux 7.1:
  i915_request.c → RequestManager + GemRequest
  i915_scheduler.c → Scheduler
  i915_gem.c stats → GemStatistics

GEM subdirectory: 20 files, 1,660 lines, 0 errors
 2026-06-02 09:46:35 +03:00
vasilito fc63a47c05 intel: GEM final phases — init, busy tracking, shrinker, throttle, VMA resources 
gem_init.rs (200 lines):
  GemInitManager: top-level GEM initialization
    System/LMEM/Stolen region management
    object_manager + vma_manager as Arcs for shared access
  BusyTracker: per-handle engine busy state with event history
    mark_busy/idle, engine_mask, submission count
  ShrinkerWithEviction: LRU-based eviction under memory pressure
    Last-used timestamp ordering, pinned object protection
    try_shrink() with target-based eviction
  VmaResourceManager: VMA resource lifecycle tracking
    track/release with released state
  RingThrottle: per-ring submission limits with global cap
    submit/retire with fence-based completion

Ported from Linux 7.1:
  i915_gem.c → GemInitManager
  i915_gem_busy.c → BusyTracker
  i915_gem_shrinker.c → ShrinkerWithEviction
  i915_vma_resource.c → VmaResourceManager
  i915_gem_throttle.c → RingThrottle

GEM subdirectory: 18 files, 1,510 lines, 0 errors — complete port
 2026-06-02 09:28:26 +03:00
vasilito ca8f583a23 intel: GEM Phase 13-21 — shmem, TTM, userptr, wait, frontbuffer, clflush 
gem_backend.rs (80 lines):
  ShmemBackend: DMA-backed page allocation with byte tracking
  InternalBackend: heap-backed buffer pool for kernel-internal use
  PhysBackend: physically contiguous DmaBuffer allocation
  ClflushManager: cache flush counter

gem_ttm.rs (60 lines):
  TtmManager: placement tracking + migration lifecycle
  Migration struct: src/dst offset, size, completed, timestamp
  PowerManager: suspend/resume with frozen state tracking

gem_ioctl.rs (120 lines):
  UserptrManager: user pointer registration with GEM binding
  WaitManager: per-handle waiter queue with timeout/signal
  FrontbufferTracker: scanout buffer dirty rect tracking
  FrontbufferState: dirty/scanout flags + rect coordinate list

Modules ported from Linux 7.1:
  gem/i915_gem_shmem.c → ShmemBackend
  gem/i915_gem_ttm.c → TtmManager + PowerManager
  gem/i915_gem_ttm_move.c → Migration
  gem/i915_gem_userptr.c → UserptrManager
  gem/i915_gem_wait.c → WaitManager
  gem/i915_gem_clflush.c → ClflushManager
  gem/i915_gem_internal.c → InternalBackend
  gem/i915_gem_phys.c → PhysBackend
  gem/i915_gem_object_frontbuffer.c → FrontbufferTracker
  gem/i915_gem_pm.c → PowerManager

GEM subdirectory: 17 files, 1,310 lines, 0 errors
 2026-06-02 09:16:48 +03:00
vasilito 27e5326ee0 intel: GEM Phase 10-12 — DMA-BUF, create params, LMEM allocator 
gem_dmabuf.rs (65 lines):
  DmaBufManager: export/import/release with fd→handle mapping
  DmaBufExport/DmaBufImport structs with size tracking
  Per-handle export ref-counting through GemObjectManager

gem_create.rs (40 lines):
  CreateParams: size, region, flags, alignment, name
  CreateManager::create() with param validation
  create_lmem/create_named convenience constructors

gem_lmem.rs (60 lines):
  LmemAllocator: linear allocator for discrete GPU VRAM
  write_region/read_region with DWORD-aligned MMIO access
  64KB alignment, used_bytes tracking, out-of-bounds checking

Ported from Linux 7.1:
  gem/i915_gem_dmabuf.c → DmaBufManager
  gem/i915_gem_create.c → CreateManager + CreateParams
  gem/i915_gem_lmem.c → LmemAllocator

GEM subdirectory: 14 files, 1050 lines, 0 errors
 2026-06-02 08:47:27 +03:00
vasilito 7b42abeec9 intel: GEM Phase 5-9 — context, mmap, tiling, domains, stolen memory 
gem_context.rs (90 lines):
  GemContext: handle, priority, PPGTT flag, VMA manager per-context
  ContextManager: create/destroy/activate/deactivate
  create_default_context() with PPGTT enabled

gem_mmap.rs (60 lines):
  MmapManager: create/unmap with offset-based handle lookup
  MmapType: WriteCombine/WriteBack/Uncached
  MMAP_OFFSET_SHIFT for page-aligned offset allocation

gem_tiling.rs (80 lines):
  TilingManager: set/get_tiling per handle
  FenceRegisterManager: 32 fence register pool with alloc/free
  TilingMode: None/X/Y/Yf/Ys with stride + fence reg binding

gem_domain.rs (80 lines):
  DomainManager: read/write domain tracking with clflush flag
  BusyManager: per-handle engine busy state
  ThrottleManager: pending submission limit enforcement

gem_stolen.rs (60 lines):
  StolenMemoryManager: BIOS stolen memory region with reserve
  ShrinkerManager: memory pressure tracking with shrink attempts

Ported from Linux 7.1:
  gem/i915_gem_context.c → ContextManager + GemContext
  gem/i915_gem_mman.c → MmapManager + MmapEntry
  gem/i915_gem_tiling.c → TilingManager + FenceRegisterManager
  gem/i915_gem_domain.c → DomainManager
  gem/i915_gem_busy.c → BusyManager
  gem/i915_gem_stolen.c → StolenMemoryManager
  gem/i915_gem_shrinker.c → ShrinkerManager

GEM subdirectory: 11 files, 880 lines, 0 errors
 2026-06-02 08:43:08 +03:00
vasilito 9c609a8389 intel: GEM Phase 3-4 — execbuffer submission + page management 
gem_execbuffer.rs (80 lines):
  ExecObject: batch object with offset, length, gtt_offset, binding flag
  RelocationEntry: target handle, delta, offset validation
  ExecbufferSubmission: batch objects, relocations, fence tracking
  ExecbufferManager: validate → reloc_process → bind → fence_out
  AtomicU64 fence counter with per-submission increment

gem_pages.rs (65 lines):
  PageManager: BTreeMap<phys_addr, Page> with refcount allocation/free
  Page struct: phys_addr, refcount, dirty flag
  TtmMoveManager: source→destination region migration
  Bump allocator with total/free byte tracking per region

Ported from Linux 7.1:
  gem/i915_gem_execbuffer.c → ExecbufferManager + ExecObject
  gem/i915_gem_ttm.c → TtmMoveManager
  gem/i915_gem_pages.c → PageManager

GEM subdirectory now: 6 files, 515 lines, covering object lifecycle,
memory regions, VMA management, execbuffer, and page management.
 2026-06-02 08:19:48 +03:00
vasilito 7053990358 intel: GEM Phase 1-2 — core object lifecycle + memory regions + VMA 
gem/gem_object.rs (175 lines):
  GemObject struct: handle, size, region, cache_level, gtt/vram offset
  GemObjectManager: BTreeMap registry with create/close/pin/unpin
  Memory tracking: system/vram byte counters with limits
  Export tracking, per-object cache level, name support

gem/gem_region.rs (85 lines):
  MemoryRegion: System/LMEM/Stolen types with alloc/free
  IO-mapped vs CPU-visible region properties
  Min page size per region (4K system, 64K LMEM)
  Bump allocator with free tracking

gem/gem_vma.rs (100 lines):
  GemVma: virtual address binding with address space type
  VmaManager: BTreeMap registry with overlap detection
  Bind/unbind with bound byte tracking
  Per-object VMA query

Ported from Linux 7.1:
  gem/i915_gem_object_types.h → GemObject
  gem/i915_gem_object.c → GemObjectManager
  gem/i915_gem_region.c → MemoryRegion
  i915_vma.c → VmaManager + GemVma

This is a new gem/ subdirectory under the Intel driver — the foundation
for the full 27,472-line GEM subsystem port from Linux i915.
 2026-06-02 08:14:14 +03:00
vasilito e7ed83144e intel: complete DPLL reimplementation from Linux 7.1 
Ported skl_ddi_calculate_wrpll() computation engine:
- DCO central frequency table (8.4/9.0/9.6 GHz)
- 43 divider values (36 even + 7 odd)
- Deviation-minimizing search across all dividers × central freqs
- skl_wrpll_multipliers() → p0×p1×p2 decomposition
- skl_wrpll_params_populate() → dco_integer/fraction with 0x8000 precision

Per-platform PLL programming:
- SKL: CFGCR0+CFGCR1+CTL at 0x164284/0x164288/0x16428C
  with qdiv_mode, kdiv/pdiv, posting reads, lock polling
- ICL: same as SKL register set
- TGL: TGL-specific CFGCR1 at 0x164298
- MTL: DPLL_FREQ + DPLL_CTRL1 with PLL_POWER_ENABLE
- Xe2: DPLL_CTRL1/CTRL2 with power+enable+lock

162 → 250 lines of actual computation and programming logic
 2026-06-02 06:42:55 +03:00
vasilito c9b723c941 intel: reimplement DPLL from Linux 7.1 — per-platform PLL management 
163 → 300 lines covering all supported generations:

SKL (Gen9):
  LCPLL1/LCPLL2 + WRPLL1/WRPLL2 enable with lock polling
  DCO integer/fraction programming via WRPLL_CTL registers

ICL (Gen11):
  DPLL_CFGCR0/CFGCR1 with frequency_enable + DCO/QDIV/KDIV/PDIV
  Dedicated CFGCR registers at 0x164284/0x164288

TGL/ADL/DG2 (Gen12):
  TGL-specific DPLL_CFGCR0/CFGCR1 at 0x164294/0x164298
  Link rate selection (HBR2/HBR3)

MTL (Gen12.7):
  DPLL_CTRL1/DPLL_FREQ register programming
  DCO integer/fraction packed into single frequency register

Xe2 (ARL/LNL/BMG):
  DPLL_CTRL1/CTRL2 with POWER_ENABLE + ENABLE bits

New DpllConfig with: pll_id, dco_int, dco_frac, pdiv/qdiv/kdiv
  vco_khz() computed from refclk * kdiv / (pdiv * qdiv)
  get_pll_for_clock() with config search + program + active tracking
  release_pll() for connector hot-unplug cleanup
  next_available_pll() automatic allocation across 4+ PLLs
 2026-06-02 06:38:28 +03:00
vasilito 9e4bf89d24 intel: reimplement CDCLK from Linux 7.1 — full per-gen support 
Replaces 155-line Gen9/Xe2-only CDCLK with 300-line comprehensive
implementation covering all supported generations:

SKL/KBL/CFL (Gen9):
  CDCLK programming via CDCLK_CTL register with freq_select + decimal
  4 frequency options: 308.57/337.5/450/432/540/675/617.14 MHz
  VCO decode: 3.2GHz to 6.75GHz per frequency/decimal combination
  Squash vs crawl waveform detection for seamless transitions

TGL/ADL/DG2 (Gen12):
  CDCLK programming via CDCLK_FREQ register
  7 frequency options with 38.4 MHz refclk
  VCO computation: 14.7GHz to 17.3GHz

MTL (Gen12.7):
  4 frequency options: 172.8/307.2/556.8/652.8 MHz

Xe2 (ARL/LNL/BMG):
  DE_CAP register primary, CDCLK_FREQ fallback
  4 frequency options: 307.2/384/556.8/652.8 MHz

CdclkState struct now carries: frequency_khz, vco_khz, refclk_khz,
  voltage_level, waveform
set_frequency() with wait_cdclk_change() polling
required_cdclk() for mode-based frequency selection
 2026-06-02 06:32:47 +03:00
vasilito dc50d019e0 intel: cargo fix — auto-clean unused imports (10→3 warnings) 2026-06-02 06:26:03 +03:00
vasilito 5fe927e506 intel: audio/ELD, RPS/RC6, GPU reset — final subsystems 
audio_eld.rs: EDID-Like Data + N/CTS computation
  EldData parser from CEA-861 EDID extension block
  HDMI OUI (0x000C03) Short Audio Descriptor extraction
  Speaker allocation string mapping (stereo → 7.1)
  compute_n_cts for 7 sample rates (32/44.1/48/88.2/96/176.4/192 kHz)

rps_rc6.rs: full Render Power State + RC6 management
  Gen9 freq table: 20 entries (100-1050 MHz)
  Gen12 freq table: 17 entries (100-1500 MHz)
  RPS up/down with evaluation interval + timeout gating
  RC6 enable/disable with HW-managed transitions
  Interrupt limits + up/down threshold programming

gpu_reset.rs: per-engine + global GPU reset recovery
  reset_engine with RESET_CTL request → ready → clear sequence
  reset_render convenience wrapper
  reset_gpu with GEN6_GDRST full domain register
  Per-domain force_reset for render/media/blitter/vecs/GUC
  Reset count + recovery success counter tracking
 2026-06-02 06:23:41 +03:00
vasilito a52ffc5ac6 intel: color pipeline, DMC DC5/6, PSR full, GuC submission — remaining MAJOR 
color_pipeline.rs: CSC/CTM coefficient encoding with precision
  encode_csc_coefficient: 12-bit fixed point with sign bit
  encode_ctm_coefficient: 64-bit FP with mantissa + exponent
  compute_hdr_metadata: ST.2086 HDR static metadata block
  ColorPipelineState struct (degamma/CSC/CTM/gamma enables)

dmc_power.rs: DC5/DC6 deep power states
  allow_dc5/allow_dc6 with DMC firmware handshake
  disallow_dc5/disallow_dc6 for display active prevention
  DC state register controls at 0x45400/0x45404/0x45504

psr_full.rs: complete PSR sink+source communication
  DPCD PSR_STATUS/ERROR_STATUS/SINK_STATUS monitoring
  PSR exit request via sink DPCD write
  Source PSR state polling (SRDENT/SRDONACK/IDLE)
  Error/entry/exit counter tracking

guc_submission.rs: GuC work queue submission protocol
  WQ head/tail ring buffer management
  doorbell trigger with per-context ID assignment
  CT message: context register/deregister, sched policy
  Timeslice + preemption timeout configuration
 2026-06-02 06:19:57 +03:00
vasilito 3f4b7074ed intel: full workarounds, display IRQ handler — remaining CRITICAL 
workarounds.rs: per-generation workaround tables (40+ register writes)
  wa_gen9: 7 workarounds (HDC, half-slice, cache, L3, sampler, row)
  wa_gen9_5: 8 workarounds (ICL/EHL additional L3 + common slice)
  wa_gen12: 7 workarounds with stepping A0 gating
  wa_gen12_7: 6 workarounds (MTL additional common slice bits)
  wa_xe2: 6 workarounds with stepping A0 gating (BMG-specific)
  apply_full_workarounds() dispatches per-generation

display_irq.rs: full display + GT interrupt handler
  DE pipe A/B/C vblank + vsync enable
  PORT hotplug + PCH hotplug enable
  FIFO underrun + PSR interrupt tracking
  GT render user + CSB + GuC interrupt enable
  DisplayIrqEvents struct with typed event fields
 2026-06-02 06:15:10 +03:00
vasilito e2253df39a intel: bandwidth calc, ALPM, panel fitter — remaining MINOR items 
bandwidth.rs: link bandwidth computation
  compute_data_rate per mode+bpp
  compute_required_lanes for DP link negotiation
  compute_dbuf_blocks per display buffer configuration

alpm.rs: Adaptive Link Power Management
  ALPM_CTL enable/disable per DP port
  Link power state monitoring via ALPM_STATUS

panel_fitter.rs: Panel fitting / scaling mode
  compute_panel_fitter for non-native resolution handling
  ScalingMode enum: None/FullScreen/Center/FullAspect
  Aspect ratio-aware destination rectangle computation
 2026-06-02 06:12:38 +03:00
vasilito 483a40d318 intel: full color management, CDCLK tables, LMEM page migration 
color_lmem.rs: combined color + LMEM management
  CscCoefficients struct with 12 matrix elements
  compute_csc_identity/BT601/BT709/BT2020 standard matrices
  LmemManager with alloc_region/free_region/page_migrate
  Discrete GPU detection and BAR-based address management

cdclk_tables.rs: per-generation CDCLK frequency tables
  ICL (Ice Lake): 6 entries (172-652 MHz)
  TGL (Tiger Lake): 12 entries with 19.2/38.4 MHz refclks
  MTL (Meteor Lake): 8 entries
  BMG (Battlemage): 4 entries (38400 refclk only)
  cdclk_table_for_device() auto-selects per platform
 2026-06-02 06:10:16 +03:00
vasilito a9cec3954d intel: VRR + DSC — Variable Refresh Rate and Display Stream Compression 
vrr.rs: Adaptive Sync / VRR support
  VRR_CTL enable with flip line configuration
  Min/max vtotal frame time programming
  Transcoder-based VRR status monitoring

dsc.rs: DSC 1.2a compression encoder
  PPS (Picture Parameter Set) computation per mode
  Slice count (1/2/4/8) and BPC (8/10/12) configuration
  DPCD sink DSC enable/disable communication
  probe_sink_caps via DP_DSC_SUPPORT register
 2026-06-02 06:07:50 +03:00
vasilito fc7cceaa6b intel: DSB, watermarks, PCH — remaining CRITICAL infrastructure 
dsb.rs: Display State Buffer for atomic commit batching
  Hardware batch programming via MMIO_TRIGGER
  Non-blocking commit with vblank synchronization
  MMIO write + wait_us + wait_vblank opcodes

watermark.rs: display buffer watermark computation
  Per-generation latency (3.5us Gen12+, 5us older)
  DBUF block count per mode configuration

mod.rs: wired DisplayStateBuffer alongside FBC/DRRS/PSR
  DSB available for glitch-free atomic modeset commits
 2026-06-02 06:05:32 +03:00
vasilito b85f07ad22 intel: universal plane programming + DP MST topology manager 
plane_universal.rs: per-plane enable/disable with format+rotation
  XRGB8888, ARGB8888, NV12 format support
  0/90/180/270 rotation via PLANE_CTL bits
  Position, stride, surface address, colorkey programming

dp_mst.rs: Multi-Stream Transport topology management
  DPCD MST capability probe (DPCD 0x0021)
  Upstream MST enable (DPCD 0x0111)
  VCPI allocation/deallocation (payload table + allocate set)
  MstBranchDevice tracking for branch device enumeration

mod.rs: registered plane_universal and dp_mst modules
 2026-06-02 05:56:09 +03:00
vasilito cf38cff205 intel: PCH detection module — platform configuration hub enumeration 
pch.rs: PchType enum spanning 12 PCH generations
  from_generation(): maps IntelGeneration to PchType automatically
  display_well_base(): per-PCH power well register base
  has_separate_ddi_wells(): Gen9+ PCH feature flag
  gmbus_base(): 0x5100 (pre-PCH) vs 0xC5100 (PCH-based)
  ddi_port_count(): 2-6 ports per PCH generation

This gates correct DDI buffer, power well, and GMBUS configuration
per platform. Previously all platforms used Gen9 hardcoded defaults.
 2026-06-02 05:46:53 +03:00
vasilito a839be7d6c intel: fix dead code — replace ASLS tuple literal with debug message 
Was: let _ = (ASLS_PCI_OFFSET, OPREGION_SIG, OPREGION_VBT_OFFSET);
Now: debug! with note that OpRegion VBT lookup is not yet implemented

Removes dead code while documenting the feature gap for future work.
 2026-06-02 05:36:23 +03:00
vasilito 442d450ae7 intel: DP HBR2/HBR3/UHBR rates, HDMI vs DP paths, multi-engine init 
dp_link.rs: expanded rate table to 7 entries
  HBR2 (8.1Gbps), HBR3 (8.1Gbps), UHBR10/13.5/20 for DP 2.0
  rate_to_khz handles all 7 rates including UHBR (1-2M kHz)

display.rs: distinct HDMI vs DP enable paths in set_mode
  connector_type parameter branches DP link retrain vs HDMI
  HDMI paths skip link training (uses TMDS clock instead)

mod.rs: multi-engine ring initialization
  Blitter (BCS) and VideoEnhance (VECS) rings alongside Render (RCS)
  Optional init — gracefully handled if ring creation fails
  Stored as Mutex<Option<IntelRing>> for lazy access
 2026-06-02 05:23:27 +03:00
vasilito 7a5a881615 kms: full property registration infrastructure 
property.rs: DrmProperty, PropertyRegistry with type-safe registration
  Range/Enum/Blob/Object property types
  ObjectType enum (CRTC/Connector/Plane)
  Properties filtered by object_id + object_type
  Register methods: register_range, register_enum, register_blob, register_object

This replaces the hardcoded synthetic property arrays in scheme.rs.
Compositors can now query properties dynamically per-object.
 2026-06-02 05:07:18 +03:00
vasilito 7f8f93146d intel: FBC, DRRS, PSR2, Gen4-7 regs — comprehensive subsystems 
fbc.rs: Frame Buffer Compression
  FBC_CTL enable/compression/fence programming (Gen7+)
  FBC_STATUS compressed buffer tracking
  nuke() for frontbuffer invalidation on render

drrs.rs: Display Refresh Rate Switching
  Transitions high→low refresh rate on idle timeout
  DRRS_CTL with idle frame count configuration
  mark_active() for compositor interaction

psr2.rs: Panel Self Refresh v2
  Selective update via sink DPCD capability probe
  PSR2_MAN_TRK_CTL for partial frame update tracking
  DP_PSR2_SUPPORT/EN_CFG2 sink communication

regs_gen4_7.rs: Pre-Gen9 FDI register definitions
  Gen4-7 pipe/plane/DDI registers (PIPEACONF, DSPACNTR)
  GMBUS at 0x5100 base (pre-PCH offset)
  No DDI_BUF_CTL — these platforms use FDI display

mod.rs: Wired FbcState, DrrsState alongside existing PsrState
 2026-06-02 05:06:16 +03:00
vasilito d5636ae1de intel: P0/P1 — LRC context state, MI commands, DDI sequences, pipe scaler 
context.rs: proper LRC image population
  Removed context restore inhibit — GPU now saves/restores state on preempt
  Added PDP register initialization in LRC image (4 PDP entries)
  Added FAULT_AND_STREAM_CTL, BB_STATE initialization
  set_pdp_registers() for per-context PPGTT configuration

ring.rs: MI command emission helpers
  emit_bb_start/emit_bb_end: batch buffer chaining
  emit_load_register_imm: LRI for workaround application
  emit_store_data_imm: store-to-memory for fence signaling
  emit_arb_check: preemption point insertion
  emit_semaphore_wait: inter-engine synchronization
  emit_user_interrupt: explicit interrupt generation

display.rs: DDI pre-enable/post-disable sequences
  ddi_pre_enable: DP link retrain + DDI_BUF_CTL enable with pipe routing
  ddi_post_disable: DDI_BUF_CTL, PIPECONF, DSPCNTR disable + pipe update
  Pipe scaler PS_CTRL/PS_WIN_POS/PS_WIN_SIZE programming in set_mode

regs.rs + regs_gen9/gen12/xe2: scaler register trait methods
  ps_ctrl, ps_win_pos, ps_win_size at 0x68180/0x68170/0x68174 per pipe
 2026-06-01 23:58:13 +03:00
vasilito aa3257c6eb intel: P0 — DP retrain at modeset, sync_file fd export/import 
display.rs: DP link retraining at set_mode time
  Before enabling DDI_BUF_CTL, retrain DP link for the active port
  Uses display's own dp_aux vector for DPCD communication
  Fixes link recovery after mode changes

syncobj.rs: sync_file fd infrastructure
  export_sync_file(): generate fd token, map syncobj → fd
  import_sync_file(): resolve fd token → syncobj handle
  close_sync_file(): remove fd from table

driver.rs: GpuDriver trait methods for sync_file
  syncobj_export_fd() / syncobj_import_fd() with default Unsupported

mod.rs: IntelDriver implementation delegates to SyncobjManager

scheme.rs: DRM_IOCTL_REDOX_SYNCOBJ_HANDLE_TO_FD (0x73)
  DRM_IOCTL_REDOX_SYNCOBJ_FD_TO_HANDLE (0x74)
  Wire types with proper #[repr(C)] Copy+Clone attributes
 2026-06-01 23:42:19 +03:00
vasilito bd6e9cd70c intel: P0 bugfixes — MI_USER_INTERRUPT, MI_FLUSH_DW, SETPLANE 
ring.rs: fix MI_USER_INTERRUPT value
  Was 0x0200_0000 (same as MI_FLUSH_DW) — GPU never generated interrupts.
  Correct Gen7+ value: 0x6200_0000 (MI_USER_INTERRUPT with proper encoding)
  MI_FLUSH_DW in flush() now uses proper DWord length encoding (1<<22)

scheme.rs: SETPLANE now forwards primary plane to page_flip
  Primary plane (id=3) flips via driver.page_flip()
  Overlay/cursor planes get silent no-op (KWin falls back to primary)
  Removes EOPNOTSUPP blocker for KWin Wayland compositor
 2026-06-01 23:14:36 +03:00
vasilito 018b173320 intel: expand KMS properties — DEGAMMA_LUT, CTM, VRR, link-status 
scheme.rs: CRTC properties 20→27 (8 total):
  + DEGAMMA_LUT_SIZE (256 entries, immutable range)
  + DEGAMMA_LUT (atomic blob, wires to gamma.rs degamma LUT)
  + CTM (atomic blob, wires to gamma.rs identity CTM)
  + VRR_ENABLED (atomic range 0-1, variable refresh rate)
  Connector properties 30→33 (4 total):
  + link-status (atomic enum Good/Bad, for DP link health)

These expose hardware capabilities already present in gamma.rs
but previously invisible to userspace (KWin/night color/SDDM).
 2026-06-01 23:08:30 +03:00
vasilito b19dd74f39 intel: fix pre-Gen9 per-gen flags, enable Gen8 PPGTT, expand plan 
info.rs:
- Gen8 now has has_ddi/has_dp_aux: true (Broadwell uses DDI display engine)
- Gen7+ now has has_gmbus: true (Ivy Bridge introduced GMBUS at 0xC5100)
- Gen4-Gen7 pre-Gen8: num_ports=3 (3 display ports, not 4 DDI ports)
- Added is_gen8_or_later() for PPGTT gate

mod.rs: PPGTT gate extended from is_gen9_or_later() to is_gen8_or_later()
  Broadwell (Gen8) supports 48-bit PPGTT

INTEL-DRIVER-FULL-IMPLEMENTATION-PLAN.md: comprehensive pre-Gen9 gap catalog
  FDI vs DDI register table for all generations
  Per-generation forcewake, power well, PLL, interrupt differences
  Implementation priority: P0 (Gen8 flags) done, P1 (FDI) documented
 2026-06-01 22:59:09 +03:00
vasilito da6b7685a7 intel: enable Gen4-Gen8 support — 56 pre-Gen9 device IDs, remove probe gate 
drivers/mod.rs: remove Gen8+ gate in is_supported_intel_generation()
  All pre-Gen9 IDs (I965G, ILK, SNB, IVB/HSW/BDW) now pass probe.
  Gen8 Broadwell/Cherryview uses DDI display engine (same as Gen9) —
  expected to work with current register paths.
  Gen4-Gen7 (I965G through Haswell) use FDI display engine which differs
  from DDI. They will probe successfully but display init uses DDI_BUF_CTL
  registers that don't exist on FDI hardware. Full FDI support is documented
  as future work.

info.rs: +56 entries covering all pre-Gen9 generations
  Gen4: 18 IDs (I965G, G33, Q33/Q35, GM965, G45, GM45, Pineview)
  Gen5: 2 IDs (Ironlake desktop/mobile)
  Gen6: 7 IDs (Sandy Bridge GT1/GT2 desktop/mobile)
  Gen7: 11 IDs (Ivy Bridge, Haswell ULT/ULX GT1/GT2/GT3)
  Gen8: 18 IDs (Broadwell ULT/ULX GT1/GT2/GT3, Cherryview GT1/GT2)
  Total: 56 new entries → info.rs now has 157 device IDs
 2026-06-01 22:53:37 +03:00
vasilito a31299a52e intel: gaps 5-6 — HuC firmware, VBT LFP/eDP/DTD parsing 
huc.rs: NEW — HuC firmware upload and authentication
  DMA-based firmware staging with GGTT mapping
  WOPCM upload + GuC interrupt notification
  Authentication polling with timeout fallback (non-fatal)
  Wired in mod.rs alongside GuC loading

vbt.rs: expanded parsing for laptop panel support
  eDP block 27: link rate, lane count, T3/T12 timing
  LFP backlight block 43: PWM frequency, min brightness
  Generic DTD block 42: panel native mode timing
  PanelTiming struct with pixel clock, h/v active/blank/sync
  VbtInfo now carries edp_link_rate, lfp_backlight, panel_native_dtd
 2026-06-01 22:45:12 +03:00
vasilito 8182391e21 intel: gaps 1-4 — device IDs, MOCS tables, GT interrupts, workarounds 
info.rs: +38 device IDs (ADL-S 8, ADL-N 5, RPL-S 8, RKL 6, CML 4, JSL 5, ICL 4)
  Coverage: 63 → 101 IDs (~18% → ~29% of Linux 7.1 i915)

mocs.rs: NEW — per-generation MOCS table initialization
  Gen9: LNCFCMOCS registers (64 entries with UC/WB cacheability)
  Gen12+: GEN12_GLOBAL_MOCS registers (64 entries with UC/WT/WB)
  Fixes all GPU memory accesses defaulting to uncacheable

gt.rs: GT interrupt registers + handler
  GEN8_GT_IER/IIR/IMR: render user, context switch, GuC interrupts
  enable_gt_interrupts/disable_gt_interrupts/handle_gt_interrupt
  Wired into driver init and IRQ processing loop

mod.rs: MOCS init after CDCLK, GT interrupt enable after GT init,
  GT interrupt handling in process_irq
 2026-06-01 22:41:16 +03:00
vasilito 2ae3eb9d02 intel: P0 fixes — wire ATOMIC ioctl, fix SYNCOBJ caps, update plan 
scheme.rs:
- DRM_IOCTL_MODE_ATOMIC: actually call driver.atomic_commit() instead of returning
  empty. This was dead code — the Intel driver's atomic_commit was fully implemented
  but unreachable from userspace. Single-line fix unblocks KWin/Wayland.
- DRM_CAP_SYNCOBJ: 0 → 1. Syncobjs were fully implemented but advertised as unavailable.
- DRM_CAP_SYNCOBJ_TIMELINE: 0 → 1. Timeline-based syncobj manager exists.
- DRM_CAP_ATOMIC_ASYNC_PAGE_FLIP: 0 → 1. Async page flip support advertised.

INTEL-DRIVER-FULL-IMPLEMENTATION-PLAN.md:
- Cross-reference analysis from 3 background agents vs Linux 7.1 i915
- P0 gaps: MOCS tables absent, HuC/GSC firmware missing, render state needed
- P1 gaps: GT interrupts, VBT LFP/eDP/DTD parsing, missing device IDs
- Updated priority with effort estimates per gap
 2026-06-01 22:27:25 +03:00
vasilito af465a4a30 intel: fix EDID read — remove unused mutable binding, clean imports 2026-06-01 22:10:20 +03:00
vasilito cc95ab1e25 fix: bootloader builds with rustup nightly + --cfg aes_force_soft 
Direct build command proven working:
  RUST_TARGET_PATH=targets RUSTFLAGS='--cfg aes_force_soft' \
  rustup run nightly cargo -Zunstable-options rustc \
    --target x86_64-unknown-uefi -Z build-std=core,alloc \
    --bin bootloader --release -- --emit link=bootloader.efi

UEFI target file fixed: removed -sse restriction causing LLVM
'Do not know how to split' error in aes crate.

Updated rust-toolchain.toml to 'nightly' (latest).
 2026-06-01 22:06:48 +03:00
vasilito a2b5557e94 intel: VRAM size reporting — total_vram_bytes, free_vram_bytes 2026-06-01 21:59:39 +03:00
vasilito dca17ce146 intel: EU/subslice runtime detection via GT_SLICE_INFO and EU_DISABLE fuses 2026-06-01 21:58:50 +03:00
vasilito 25b9bd60c6 intel: scaler/rotation, display power gating, GPU statistics 
- cursor.rs: plane scaler (PS_CTRL/PS_WIN/PS_SIZE) with nearest filter,
  rotation property (0/90/180/270) via PLANE_ROT_CTL registers
- display_power.rs: gate_ddi_wells + gate_aux_wells per active port count
- gt.rs: GpuStats struct and gpu_stats() method for utilization reporting
 2026-06-01 21:55:45 +03:00
vasilito 53cab7e445 intel: Phase 7+8 — debug/observability and GuC submission 
- hangcheck.rs: GpuErrorState struct with full register snapshot
- guc.rs: CTB channels (H2G/G2H) with GGTT descriptors, ADS setup
- mod.rs: wire CTB init and ADS setup after GuC firmware upload
 2026-06-01 21:45:52 +03:00
vasilito 6e55e00c53 fix: bootloader recipe TARGET handling, add -Zunstable-options to mk files 
- Fix bootloader recipe: pass correct TARGET on make command line
  instead of hardcoding x86_64-unknown-uefi (breaks BIOS build)
- Add cargo -Zunstable-options to x86_64-unknown-uefi.mk and
  x86-unknown-none.mk for custom target support
- Add x86_64-unknown-uefi.json target file

Remaining: redoxer toolchain cargo/rust-src version mismatch
prevents build-std compilation. Needs 'make prefix' to rebuild
toolchain with matching versions.
 2026-06-01 21:44:55 +03:00
vasilito a36954d85c intel: Phase 6 — platform enablement 
- info.rs: GMD_ID runtime detection (MTL+), stepping field, EU counts
- mod.rs: read PCI revision + GMD_ID during init, override device info
- gt.rs: per-stepping workarounds (DG2 commmon slice, MTL/ARL row chicken,
  BMG G21 A0-specific L3 prefetch disable)
 2026-06-01 21:40:54 +03:00
vasilito d042a26375 fix: bootloader target file + -Zunstable-options + TARGET override 
Bootloader needs x86_64-unknown-uefi (UEFI target) but redoxer
sets TARGET=x86_64-unknown-redox (OS target). Added:
- x86_64-unknown-uefi.json custom target file
- cargo -Zunstable-options in Makefile for custom target support
- TARGET= override in recipe (may still be overridden by redoxer)
 2026-06-01 21:37:20 +03:00
vasilito 48397c6419 intel: Phase 5 — runtime power management 
- RPS interactive governor: fast ramp-up on activity, slow ramp-down on idle
- Runtime PM with wakeref counting and RC6 transitions
- Forcewake automatically taken on first wakeref, released on last
- Frequency tracking with min/max/target per-GT state
 2026-06-01 21:29:43 +03:00
vasilito e01a4b2dcf intel: Phase 4 — atomic modeset and color pipeline 
- kms/atomic.rs: AtomicState, atomic_check with mode+bandwidth validation
- driver.rs: atomic_commit default method on GpuDriver trait
- mod.rs: IntelDriver atomic_commit with atomic_check → set_crtc dispatch
- gamma.rs: degamma LUT (sRGB linearize), CSC identity, CTM identity
 2026-06-01 21:25:18 +03:00
vasilito 8b0cabaa47 fix: m4 builds successfully - complete gnulib cross-compilation recipe 
Root cause chain discovered and fixed:
1. GCC built-in stddef.h shadowed by relibc's _STDDEF_H guard
   → fix_types.h with guarded typedefs for 15+ sys types
2. gnulib configure bakes raw typedefs into GL_CFLAG_GNULIB_WARNINGS
   → strip them from Makefiles after configure
3. __fseterr/__freadahead don't exist in relibc
   → compile C stubs and inject into link via Makefile patch

The recipe pattern is documented and reusable for other gnulib packages
(ninja-build, diffutils, etc.).

Also: bootloader recipe needs RUSTFLAGS=-Zunstable-options for
custom target support after redoxer toolchain restore.
 2026-06-01 21:22:04 +03:00
vasilito 28436dc604 intel: Phase 3 — wire execlist port and PDP registers 
- Create ExeclistPort during driver init with context control registers
- Store execlist_port in IntelDriver for submission routing
- Wire PDP0_LDW/UDW register writes in cs_submit before ring batch
 2026-06-01 21:15:37 +03:00