New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
vk: Enable use of a passthrough DMA layer if supported #9613
Commits on Jan 23, 2021
-
Configuration menu - View commit details
-
Copy full SHA for 644aeaa - Browse repository at this point
Copy the full SHA 644aeaaView commit details -
vk/dma: Disable memory inheritance
- It is not possible to emulate passthrough memory cleanly, and we don't need to A stupid race condition appears when trying to synchronize DMA blocks with memory inheritance. Since the usage pattern is to acquire a range and then load or write+flush, this new data is going to be.. overwritten by the commandbuffer execution sequence later. Acquiring a scratch buffer to hold CPU content during the transition is not worth the effort.. as the data will be destroyed anyway during the transfer process immediately afterwards. Fixes data corruption when moving data around using the emulated DMA passthrough
Configuration menu - View commit details
-
Copy full SHA for d851ee3 - Browse repository at this point
Copy the full SHA d851ee3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 8d67a1a - Browse repository at this point
Copy the full SHA 8d67a1aView commit details -
Configuration menu - View commit details
-
Copy full SHA for 0476260 - Browse repository at this point
Copy the full SHA 0476260View commit details
Commits on Jan 24, 2021
-
win32: Be a bit more optimistic with allocated ranges; we get contigu…
…ous ranges more often than not
Configuration menu - View commit details
-
Copy full SHA for ba90322 - Browse repository at this point
Copy the full SHA ba90322View commit details -
Configuration menu - View commit details
-
Copy full SHA for aa4cfd5 - Browse repository at this point
Copy the full SHA aa4cfd5View commit details -
Some spec violations fixes Make the option dynamic
Configuration menu - View commit details
-
Copy full SHA for f56cee2 - Browse repository at this point
Copy the full SHA f56cee2View commit details -
Configuration menu - View commit details
-
Copy full SHA for 0d05801 - Browse repository at this point
Copy the full SHA 0d05801View commit details -
Configuration menu - View commit details
-
Copy full SHA for ba56be9 - Browse repository at this point
Copy the full SHA ba56be9View commit details -
Configuration menu - View commit details
-
Copy full SHA for fdfc7bc - Browse repository at this point
Copy the full SHA fdfc7bcView commit details -
vk/dma: Fix multiple logical bugs
- Fix range chaining. - Add validation checks that no overlaps exist.
Configuration menu - View commit details
-
Copy full SHA for d0ffc33 - Browse repository at this point
Copy the full SHA d0ffc33View commit details -
gl: Avoid type shenanigans used to fake byteswap for select formats
- Just use the reversed type instead. The new uploader backend combines swizzle+swap so there is no need for tricks anymore
Configuration menu - View commit details
-
Copy full SHA for 3223786 - Browse repository at this point
Copy the full SHA 3223786View commit details -
Configuration menu - View commit details
-
Copy full SHA for 38eefdb - Browse repository at this point
Copy the full SHA 38eefdbView commit details -
vk/dma: Disable pasthrough DMA for NVIDIA+windows
- The driver seems to cache page mapping as long as allocation has not been removed - This is undesirable as we cannot stop the emulator to remove stale allocations every time a page is unmapped
Configuration menu - View commit details
-
Copy full SHA for 8eb9382 - Browse repository at this point
Copy the full SHA 8eb9382View commit details