On 8/5/26 14:58, Christian König wrote:
On 7/31/26 18:43, Paolo Bonzini wrote:
This ensures that fixup_user_fault() users see a writable PTE when
they request one. The flip side is that vmw_bo_vm_fault() now has
to record by hand the write fault, because .pfn_mkwrite() is not
invoked.
Prefaulting works as before because only the first entry comes out
writable, while the following ones still end up executing
the .pfn_mkwrite() callback.
Please split that patch for TTM/VMWGFX. The TTM part looks
reasonable, but VMGFX is a completely different beast.
Note that the TTM change alone would break vmwgfx without the other
part. This is not obvious, and it's why I placed them together given
the TTM part is just one line of code, but if you prefer I can split
them (the TTM change can go second).
Let me know if "looks reasonable" counts as "Acked-by" for that part or not.
By the way, see also
https://lists.freedesktop.org/archives/dri-devel/2026-August/586696.html
- it touches the same code, and the mistake was noticed by sashiko when
reviewing this one.
Paolo
