Re: [RFC PATCH 00/28] Introduce GPU SVM and Xe SVM implementation

2024-09-25 Thread Simona Vetter
On Tue, Sep 24, 2024 at 07:36:21PM +, Matthew Brost wrote: > On Tue, Sep 24, 2024 at 11:16:01AM +0200, Simona Vetter wrote: > > On Tue, Aug 27, 2024 at 07:48:33PM -0700, Matthew Brost wrote: > > > Continuation of SVM work by Oak Zeng [1][2] based on community feedback. > > > Introduces GPU SVM

Re: [RFC PATCH 00/28] Introduce GPU SVM and Xe SVM implementation

2024-09-24 Thread Matthew Brost
On Tue, Sep 24, 2024 at 11:16:01AM +0200, Simona Vetter wrote: > On Tue, Aug 27, 2024 at 07:48:33PM -0700, Matthew Brost wrote: > > Continuation of SVM work by Oak Zeng [1][2] based on community feedback. > > Introduces GPU SVM layer and new Xe uAPI. Supports GPU page faults for > > system allocati

Re: [RFC PATCH 00/28] Introduce GPU SVM and Xe SVM implementation

2024-09-24 Thread Simona Vetter
On Tue, Aug 27, 2024 at 07:48:33PM -0700, Matthew Brost wrote: > Continuation of SVM work by Oak Zeng [1][2] based on community feedback. > Introduces GPU SVM layer and new Xe uAPI. Supports GPU page faults for > system allocations (e.g., malloc), runtime allocations (e.g., binding a > BO), migrati

[RFC PATCH 00/28] Introduce GPU SVM and Xe SVM implementation

2024-08-27 Thread Matthew Brost
Continuation of SVM work by Oak Zeng [1][2] based on community feedback. Introduces GPU SVM layer and new Xe uAPI. Supports GPU page faults for system allocations (e.g., malloc), runtime allocations (e.g., binding a BO), migration to and from VRAM, and unified eviction (BO and SVM VRAM allocations