Special VM details¶

The special VM is configured with a page structure where we can dynamicallymap BOs which need to be copied and cleared, dynamically map other VM’s pagetable BOs for updates, and identity map the entire device’s VRAM with 1 GBpages.

Currently the page structure consists of 32 physical pages with 16 beingreserved for BO mapping during copies and clear, 1 reserved for kernel binds,several pages are needed to setup the identity mappings (exact number basedon how many bits of address space the device has), and the rest are reserveduser bind operations.

TODO: Diagram of layout

Bind jobs¶

A bind job consist of two batches and runs either on the migrate engine(kernel binds) or the bind engine passed in (user binds). In both cases theVM of the engine is the migrate VM.

The first batch is used to update the migration VM page structure to point tothe bind VM page table BOs which need to be updated. A physical page isrequired for this. If it is a user bind, the page is allocated from pool ofpages reserved user bind operations with drm_suballoc managing this pool. Ifit is a kernel bind, the page reserved for kernel binds is used.

The first batch is only required for devices without VRAM as when the devicehas VRAM the bind VM page table BOs are in VRAM and the identity mapping canbe used.

The second batch is used to program page table updated in the bind VM. Whynot just one batch? Well the TLBs need to be invalidated between these twobatches and that only can be done from the ring.

When the bind job complete, the page allocated is returned the pool of pagesreserved for user bind operations if a user bind. No need do this for kernelbinds as the reserved kernel page is serially used by each job.

Copy / clear jobs¶

A copy or clear job consist of two batches and runs on the migrate engine.

Like binds, the first batch is used update the migration VM page structure.In copy jobs, we need to map the source and destination of the BO into pagethe structure. In clear jobs, we just need to add 1 mapping of BO into thepage structure. We use the 16 reserved pages in migration VM for mappings,this gives us a maximum copy size of 16 MB and maximum clear size of 32 MB.

The second batch is used do either do the copy or clear. Again similar tobinds, two batches are required as the TLBs need to be invalidated from thering between the batches.

More than one job will be generated if the BO is larger than maximum copy /clear size.

Future work¶

Update copy and clear code to use identity mapped VRAM.

Can we rework the use of the pages async binds to use all the entries in eachpage?

Using large pages for sysmem mappings.

Is it possible to identity map the sysmem? We should explore this.