While testing [1] and [2] with a SRIOV enabled dGPU, it was noticed that migrating a BO to System RAM before exporting it as a dmabuf results in considerable performance degradation while running in a Guest VM. For example, running a simple 3D app such as weston-simple-egl would yield ~50 FPS instead of ~59 FPS, assuming a mode of 1920x1080@60. One fix to this problem is to not migrate the BO and keep it in VRAM during export. However, given that the GPU running in PF mode on the Host cannot effectively access the PCI BAR addresses backing the imported dmabuf BO, they need to be translated into LMEM addresses (DPA) to enable this use-case to work properly. With this patch series applied, it would become possible to display (via Qemu GTK UI) Guest VM compositor's framebuffer (created in its LMEM) on the Host without having to make any copies of it or a roundtrip to System RAM. And, weston-simple-egl can now achieve ~59 FPS while running with Gnome Wayland in the Guest VM. [1] https://lore.kernel.org/dri-devel/20240813035509.3360760-1-vivek.kasireddy@xxxxxxxxx/ [2] https://lore.kernel.org/dri-devel/20240624065552.1572580-1-vivek.kasireddy@xxxxxxxxx/ Patchset overview: Patch 1: PCI driver patch to not block P2P DMA between VF and PF Patch 2: Prevent BO migration to System RAM while running in VM Patch 3: Helper function to get VF's starting address in LMEM Patch 4-5: Create new SG for VRAM based dmabuf BO to hold translated LMEM addresses This series is tested using the following method: - Run Qemu with the following relevant options: qemu-system-x86_64 -m 4096m .... -device vfio-pci,host=0000:03:00.1 -device virtio-vga,max_outputs=1,blob=true,xres=1920,yres=1080 -display gtk,gl=on -object memory-backend-memfd,id=mem1,size=4096M -machine memory-backend=mem1 ... - Run Gnome Wayland with the following options in the Guest VM: # cat /usr/lib/udev/rules.d/61-mutter-primary-gpu.rules ENV{DEVNAME}=="/dev/dri/card1", TAG+="mutter-device-preferred-primary", TAG+="mutter-device-disable-kms-modifiers" # XDG_SESSION_TYPE=wayland dbus-run-session -- /usr/bin/gnome-shell --wayland --no-x11 & Cc: Michal Wajdeczko <michal.wajdeczko@xxxxxxxxx> Cc: Michał Winiarski <michal.winiarski@xxxxxxxxx> Cc: Simona Vetter <simona@xxxxxxxx> Cc: Matthew Auld <matthew.auld@xxxxxxxxx> Cc: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx> Cc: Dongwon Kim <dongwon.kim@xxxxxxxxx> Vivek Kasireddy (5): PCI/P2PDMA: Don't enforce ACS check for functions of same device drm/xe/dmabuf: Don't migrate BO to System RAM while running in VF mode drm/xe/pf: Add a helper function to get a VF's starting address in LMEM drm/xe/bo: Create a new sg for dmabuf BOs that are associated with a VF drm/xe/pt: Add an additional check for dmabuf BOs while updating PTEs drivers/gpu/drm/xe/xe_bo.c | 108 ++++++++++++++++++++- drivers/gpu/drm/xe/xe_bo_types.h | 6 ++ drivers/gpu/drm/xe/xe_dma_buf.c | 9 +- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 23 +++++ drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + drivers/gpu/drm/xe/xe_pt.c | 5 +- drivers/pci/p2pdma.c | 12 ++- include/linux/pci.h | 1 + 8 files changed, 158 insertions(+), 7 deletions(-) -- 2.45.1
