Re: [RFC v2 0/9] drm/xe/madvise: Add support for purgeable buffer objects

["Souza, Jose" <jose.souza@intel.com>]

On Mon, 2025-12-01 at 11:20 +0530, Arvind Yadav wrote:
> This patch series introduces comprehensive support for purgeable
> buffer objects
> in the Xe driver, enabling userspace to provide memory usage hints
> for better
> memory management under system pressure.
> 
> Overview:
> 
> Purgeable memory allows applications to mark buffer objects as "not
> currently
> needed" (DONTNEED), making them eligible for kernel reclamation
> during memory
> pressure. This helps prevent OOM conditions and enables more
> efficient GPU
> memory utilization for workloads with temporary or regeneratable data
> (caches,
> intermediate results, decoded frames, etc.).
> 
> Purgeable BO Lifecycle:
> 1. WILLNEED (default): BO actively needed, kernel preserves backing
> store
> 2. DONTNEED (user hint): BO contents discardable, eligible for
> purging
> 3. PURGED (kernel action): Backing store reclaimed during memory
> pressure
> 
> Key Design Principles:
>   - i915 compatibility: "Once purged, always purged" semantics -
> purged BOs
>      remain permanently invalid and must be destroyed/recreated
>   - Safety first: Only non-shared BOs can be marked DONTNEED to
> prevent
>     multi-process data corruption
>   - Multiple protection layers: Validation in madvise, VM bind, mmap,
> and
>     fault handlers
>   - Async TLB invalidation: Uses xe_bo_trigger_rebind() for non-
> blocking
>     GPU mapping invalidation
>   - Scratch PTE support: Fault-mode VMs use scratch pages for safe
> zero reads
>     on purged BO access
> 
> Error Handling:
>   - CPU access (mmap): Returns VM_FAULT_SIGBUS (SIGBUS signal to
> process)
>   - GPU access (non-scratch VM): Page fault fails with -EACCES, GPU
> context reset
>   - GPU access (scratch VM): Page fault succeeds, rebinds with
> scratch PTEs
>   - VM_BIND operations: MAP/PREFETCH rejected with -EINVAL
>   - Mmap offset ioctl: Rejected with -EINVAL for early error
> detection

uAPI and overal this lgtm, please let me know when you think this is
ready to be implemented on Mesa side.

> 
> v2 Changes:
>   - Reordered patches: Moved shared BO helper before main
> implementation for
>     proper dependency order
>   - Fixed reference counting in mmap offset validation (use
> drm_gem_object_put)
>   - Removed incorrect claims about madvise(WILLNEED) restoring purged
> BOs
>   - Fixed error code documentation inconsistencies
>   - Initialize purge_state_val fields to prevent kernel memory leaks
>   - Use xe_bo_trigger_rebind() for async TLB invalidation (Thomas
> Hellström)
>   - Add NULL rebind with scratch PTEs for fault mode (Thomas
> Hellström)
>   - Implement i915-compatible retained field logic (Thomas Hellström)
>   - Skip BO validation for purged BOs in page fault handler (crash
> fix)
>   - Add scratch VM check in page fault path (non-scratch VMs fail
> fault) 
>  
> 
> Arvind Yadav (7):
>   drm/xe/bo: Add purgeable bo state tracking and field madv to xe_bo
>   drm/xe/bo: Prevent purging of shared buffer objects
>   drm/xe/madvise: Implement purgeable buffer object support
>   drm/xe/bo: Handle CPU faults on purged buffer objects
>   drm/xe/bo: Prevent mmap of purged buffer objects
>   drm/xe/vm: Prevent binding of purged buffer objects
>   drm/xe: Add support for querying purgeable BO states
> 
> Himal Prasad Ghimiray (2):
>   drm/xe/uapi: Add UAPI support for purgeable buffer objects
>   drm/xe/uapi: Add UAPI for purgeable bo state to madvise query
> response
> 
>  drivers/gpu/drm/xe/xe_bo.c           | 92 ++++++++++++++++++++++++--
> --
>  drivers/gpu/drm/xe/xe_bo.h           | 57 +++++++++++++++++
>  drivers/gpu/drm/xe/xe_bo_types.h     |  3 +
>  drivers/gpu/drm/xe/xe_gt_pagefault.c | 19 ++++++
>  drivers/gpu/drm/xe/xe_pt.c           | 13 +++-
>  drivers/gpu/drm/xe/xe_vm.c           | 12 ++++
>  drivers/gpu/drm/xe/xe_vm_madvise.c   | 76 +++++++++++++++++++++++
>  include/uapi/drm/xe_drm.h            | 50 ++++++++++++++-
>  8 files changed, 308 insertions(+), 14 deletions(-)