Commit Graph
19 Commits
Author SHA1 Message Date
Mac e77041b6f5 graphics: size the buffer and page-table constants explicitly
Names the upload/stream/download/device-local buffer sizes instead of repeating
literals, and makes the BDA page-table address width explicit.

Two of these are behaviour changes and are the ones to look at:

- DOWNLOAD_BUFFER_SIZE is raised from 32 MiB to 128 MiB. Marvel's Spider-Man:
  Miles Morales asks for a single 80 MiB download allocation, which 32 MiB
  cannot serve; the run dies with 'failed to map reusable download buffer,
  size=0x5000000'. This costs every title the larger reservation.
- CACHING_ADDRESS_BITS is 39, where the table previously covered 40 bits. That
  halves the BDA page table from 512 MiB to 256 MiB, at the cost of covering
  guest addresses only to 512 GiB. 38 bits was tried first and was too narrow -
  it produced an intermittent DMA barrier on an address just past 2^38.

The address-width reduction in particular is tuned against one title; if the
1 TiB window is load-bearing for something else, take the naming and leave the
value at 40.
2026-08-31 23:41:14 +05:30
nmzik e738535294 graphics: add RDNA2 FLAT BDA buffer cache and faults 2026-08-27 03:45:41 +02:00
nmzik 1779142805 renderer: invalidate buffers on memory unmap 2026-08-24 00:40:02 +02:00
nmzik 186621c6ff renderer: use LRU buffer cache collection 2026-08-24 00:17:02 +02:00
nmzik 11d4c0d2a2 renderer: synchronize bounded buffer downloads 2026-08-23 19:22:17 +02:00
nmzik aa69b5b184 renderer: rewrite buffer lookup around page index 2026-08-23 19:22:17 +02:00
nmzik 4d8d3b6d25 renderer: batch buffer cache uploads 2026-08-23 19:22:17 +02:00
nmzik 26c4168160 refactor(renderer): simplify buffer and texture cache ownership 2026-08-21 20:11:38 +02:00
nmzik 840d742230 renderer: align buffer synchronization flow 2026-08-18 03:53:45 +02:00
nmzik 698c9f802f renderer: remove global BufferCache locking 2026-08-08 08:02:35 +02:00
nmzik ace549ab21 reduce complexity of buffer overlap detection 2026-08-06 23:46:47 +02:00
nmzik 0cbae39e41 remove double locking. small perf boost 2026-08-06 23:23:28 +02:00
nmzik 51d0e02543 perf: sync only GPU-dirty readbacks 2026-08-04 02:26:50 +02:00
nmzik b877b4be9c graphics: perf - batch GPU page watcher updates at 4 MiB granularity 2026-08-02 08:39:50 +02:00
nmzik f830d6b2e4 renderer: remove legacy code left from refactoring 2026-07-31 23:13:48 +02:00
nmzik 2f5396c6a5 Rework guest memory tracking/virtual address space/direct and flexible memory (#135)
* Rework guest memory tracking

* add unknwon flag

* Fix macOS guest address-space reservation
2026-07-31 03:07:17 +02:00
nmzik 8a244677d7 fix(renderer): resolve delayed GPU page faults through buffer and texture caches 2026-07-29 18:47:19 +02:00
nmzik f6e01e5403 Optimize bulk memory invalidation (#124)
* Per page -> per range search (optimization)
2026-07-29 05:09:36 +02:00
nmzik a6b61d7aa0 refactor(graphics): split renderer and recompiler into focused modules 2026-07-29 00:30:24 +02:00