<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta name="Generator" content="Microsoft Word 15 (filtered medium)"> <style></style> </head> <body lang="EN-US" link="blue" vlink="purple" style="word-wrap:break-word"> <div class="WordSection1"> <o:p> </o:p> <o:p> </o:p> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in"> From: Zhang, Jianxun <jianxun.zhang@intel.com> Sent: Monday, March 3, 2025 12:23 PM To: Cavitt, Jonathan <jonathan.cavitt@intel.com>; intel-xe@lists.freedesktop.org Cc: Gupta, saurabhg <saurabhg.gupta@intel.com>; Zuo, Alex <alex.zuo@intel.com>; joonas.lahtinen@linux.intel.com; Brost, Matthew <matthew.brost@intel.com>; dri-devel@lists.freedesktop.org Subject: Re: [PATCH v3 6/6] drm/xe/xe_vm: Implement xe_vm_get_property_ioctl<o:p></o:p> </div> </div> <div class="MsoNormal" align="center" style="text-align:center"> <hr size="2" width="98%" align="center"> </div> <div id="divRplyFwdMsg"> From: Cavitt, Jonathan <<a href="mailto:jonathan.cavitt@intel.com">jonathan.cavitt@intel.com</a>> Sent: Friday, February 28, 2025 10:21 AM To: <a href="mailto:intel-xe@lists.freedesktop.org">intel-xe@lists.freedesktop.org</a> <<a href="mailto:intel-xe@lists.freedesktop.org">intel-xe@lists.freedesktop.org</a>> Cc: Gupta, saurabhg <<a href="mailto:saurabhg.gupta@intel.com">saurabhg.gupta@intel.com</a>>; Zuo, Alex <<a href="mailto:alex.zuo@intel.com">alex.zuo@intel.com</a>>; Cavitt, Jonathan <<a href="mailto:jonathan.cavitt@intel.com">jonathan.cavitt@intel.com</a>>; <a href="mailto:joonas.lahtinen@linux.intel.com">joonas.lahtinen@linux.intel.com</a> <<a href="mailto:joonas.lahtinen@linux.intel.com">joonas.lahtinen@linux.intel.com</a>>; Brost, Matthew <<a href="mailto:matthew.brost@intel.com">matthew.brost@intel.com</a>>; Zhang, Jianxun <<a href="mailto:jianxun.zhang@intel.com">jianxun.zhang@intel.com</a>>; <a href="mailto:dri-devel@lists.freedesktop.org">dri-devel@lists.freedesktop.org</a> <<a href="mailto:dri-devel@lists.freedesktop.org">dri-devel@lists.freedesktop.org</a>> Subject: [PATCH v3 6/6] drm/xe/xe_vm: Implement xe_vm_get_property_ioctl <o:p></o:p> <div> <o:p></o:p> </div> </div> <div> Add support for userspace to request a list of observed failed pagefaults from a specified VM. v2: - Only allow querying of failed pagefaults (Matt Brost) Signed-off-by: Jonathan Cavitt <<a href="mailto:jonathan.cavitt@intel.com">jonathan.cavitt@intel.com</a>> Suggested-by: Matthew Brost <<a href="mailto:matthew.brost@intel.com">matthew.brost@intel.com</a>> --- drivers/gpu/drm/xe/xe_device.c | 3 ++ drivers/gpu/drm/xe/xe_vm.c | 79 ++++++++++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_vm.h | 2 + 3 files changed, 84 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 9454b51f7ad8..43accae152ff 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -193,6 +193,9 @@ static const struct drm_ioctl_desc xe_ioctls[] = { DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(XE_VM_GET_PROPERTY, xe_vm_get_property_ioctl, + DRM_RENDER_ALLOW), + }; static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg) diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 6211b971bbbd..00d2c62ccf53 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -3234,6 +3234,85 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) return err; } +static u32 xe_vm_get_property_size(struct xe_vm *vm, u32 property) +{ + u32 size = 0; + + switch (property) { + case DRM_XE_VM_GET_PROPERTY_FAULTS: + spin_lock(&vm->pfs.lock); + size = vm->pfs.len * sizeof(struct drm_xe_pf);<o:p></o:p> </div> <div> + spin_unlock(&vm->pfs.lock); + return size; + default: + return -EINVAL; + } +} + +static int fill_property_pfs(struct xe_vm *vm, + struct drm_xe_vm_get_property *args, + u32 size) +{ + struct drm_xe_pf __user *usr_ptr = u64_to_user_ptr(args->ptr); + struct drm_xe_pf *fault_list; + struct drm_xe_pf *fault; + struct xe_vm_pf_entry *entry; + int i = 0; + + if (copy_from_user(&fault_list, usr_ptr, size)) + return -EFAULT;<o:p></o:p> <o:p> </o:p> </div> <div> > Why is copying from user space memory needed in this query ioctl?<o:p></o:p> </div> <div> > Does copy_from_user() automatically allocate kernel memory for fault_list?<o:p></o:p> <o:p> </o:p> Hrmm… No, you have a point. It probably isn’t necessary to copy the memory<o:p></o:p> from userspace if we don’t need to read anything from it. Though I’m fairly certain<o:p></o:p> it does automatically allocate kernel memory for the fault_list, as we do similar<o:p></o:p> things in xe_query.c without allocating memory for first pointer.<o:p></o:p> </div> <div> <o:p> </o:p> </div> <div> + + spin_lock(&vm->pfs.lock); + list_for_each_entry(entry, &vm->pfs.list, list) { + struct xe_pagefault *pf = entry->pf; + + fault = &fault_list[i++]; + fault->address = pf->page_addr; + fault->address_type = pf->address_type; + fault->address_precision = SZ_4K;<o:p></o:p> <o:p> </o:p> </div> <div> > If we can get the exact address, the precision should be 1, right?<o:p></o:p> > (<a href="https://registry.khronos.org/vulkan/specs/latest/man/html/VkDeviceFaultAddressInfoEXT.html">https://registry.khronos.org/vulkan/specs/latest/man/html/VkDeviceFaultAddressInfoEXT.html</a>)<o:p></o:p> <o:p> </o:p> I… don’t know, actually. That question would be better answered by Matthew Brost, I think.<o:p></o:p> I’ll change it to 1 for v3.<o:p></o:p> </div> <div> + } + spin_unlock(&vm->pfs.lock); + + if (copy_to_user(usr_ptr, &fault_list, size)) + return -EFAULT; + + return 0; +} + +int xe_vm_get_property_ioctl(struct drm_device *drm, void *data, + struct drm_file *file) +{ + struct xe_device *xe = to_xe_device(drm); + struct xe_file *xef = to_xe_file(file); + struct drm_xe_vm_get_property *args = data; + struct xe_vm *vm; + u32 size; + + if (XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1])) + return -EINVAL; + + vm = xe_vm_lookup(xef, args->vm_id); + if (XE_IOCTL_DBG(xe, !vm)) + return -ENOENT; + + size = xe_vm_get_property_size(vm, args->property); + if (size < 0) { + return size; + } else if (args->size != size) { + if (args->size) + return -EINVAL;<o:p></o:p> </div> <div> > If there is a change in the size of property between the first and the second calls, say, more faults added in this case, the 2nd call will fail unintendedly. <o:p></o:p> <o:p> </o:p> I suppose the alternative would be to ask userspace to allocate memory for the maximum<o:p></o:p> possible number of pagefaults, then return the number of pagefaults in “value” while saving<o:p></o:p> the pagefaults to “*ptr”.<o:p></o:p> <o:p> </o:p> Hmm… Give me a moment to see what that would look like…<o:p></o:p> -Jonathan Cavitt<o:p></o:p> <o:p> </o:p> </div> <div> + args->size = size;<o:p></o:p> </div> <div> + return 0; + } + + switch (args->property) { + case DRM_XE_VM_GET_PROPERTY_FAULTS: + return fill_property_pfs(vm, args, size); + default: + return -EINVAL; + } +} + /** * xe_vm_bind_kernel_bo - bind a kernel BO to a VM * @vm: VM to bind the BO to diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h index 4d94ab5c8ea4..bf6604465aa3 100644 --- a/drivers/gpu/drm/xe/xe_vm.h +++ b/drivers/gpu/drm/xe/xe_vm.h @@ -184,6 +184,8 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file); int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file); +int xe_vm_get_property_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); void xe_vm_close_and_put(struct xe_vm *vm); -- 2.43.0<o:p></o:p> </div> </div> </body> </html>