[PATCH v2 1/5] mm/hmm: HMM API to enable P2P DMA for device private pages

[Yonatan Maman <ymaman@nvidia.com>]

From: Yonatan Maman <Ymaman@Nvidia.com>

hmm_range_fault() by default triggered a page fault on device private
when HMM_PFN_REQ_FAULT flag was set. pages, migrating them to RAM. In some
cases, such as with RDMA devices, the migration overhead between the
device (e.g., GPU) and the CPU, and vice-versa, significantly degrades
performance. Thus, enabling Peer-to-Peer (P2P) DMA access for device
private page might be crucial for minimizing data transfer overhead.

Introduced an API to support P2P DMA for device private pages,includes:
 - Leveraging the struct pagemap_ops for P2P Page Callbacks. This callback
   involves mapping the page for P2P DMA and returning the corresponding
   PCI_P2P page.

 - Utilizing hmm_range_fault for initializing P2P DMA. The API
   also adds the HMM_PFN_REQ_TRY_P2P flag option for the
   hmm_range_fault caller to initialize P2P. If set, hmm_range_fault
   attempts initializing the P2P connection first, if the owner device
   supports P2P, using p2p_page. In case of failure or lack of support,
   hmm_range_fault will continue with the regular flow of migrating the
   page to RAM.

This change does not affect previous use-cases of hmm_range_fault,
because both the caller and the page owner must explicitly request and
support it to initialize P2P connection.

Signed-off-by: Yonatan Maman <Ymaman@Nvidia.com>
Signed-off-by: Gal Shalom <GalShalom@Nvidia.com>
---
 include/linux/hmm.h      |  2 ++
 include/linux/memremap.h |  8 ++++++
 mm/hmm.c                 | 57 +++++++++++++++++++++++++++++++---------
 3 files changed, 55 insertions(+), 12 deletions(-)

diff --git a/include/linux/hmm.h b/include/linux/hmm.h
index db75ffc949a7..988c98c0edcc 100644
--- a/include/linux/hmm.h
+++ b/include/linux/hmm.h
@@ -27,6 +27,7 @@ struct mmu_interval_notifier;
  * HMM_PFN_P2PDMA_BUS - Bus mapped P2P transfer
  * HMM_PFN_DMA_MAPPED - Flag preserved on input-to-output transformation
  *                      to mark that page is already DMA mapped
+ * HMM_PFN_ALLOW_P2P - Allow returning PCI P2PDMA page
  *
  * On input:
  * 0                 - Return the current state of the page, do not fault it.
@@ -47,6 +48,7 @@ enum hmm_pfn_flags {
 	HMM_PFN_DMA_MAPPED = 1UL << (BITS_PER_LONG - 4),
 	HMM_PFN_P2PDMA     = 1UL << (BITS_PER_LONG - 5),
 	HMM_PFN_P2PDMA_BUS = 1UL << (BITS_PER_LONG - 6),
+	HMM_PFN_ALLOW_P2P = 1UL << (BITS_PER_LONG - 7),
 
 	HMM_PFN_ORDER_SHIFT = (BITS_PER_LONG - 11),
 
diff --git a/include/linux/memremap.h b/include/linux/memremap.h
index 4aa151914eab..79becc37df00 100644
--- a/include/linux/memremap.h
+++ b/include/linux/memremap.h
@@ -89,6 +89,14 @@ struct dev_pagemap_ops {
 	 */
 	vm_fault_t (*migrate_to_ram)(struct vm_fault *vmf);
 
+	/*
+	 * Used for private (un-addressable) device memory only. Return a
+	 * corresponding PFN for a page that can be mapped to device
+	 * (e.g using dma_map_page)
+	 */
+	int (*get_dma_pfn_for_device)(struct page *private_page,
+				      unsigned long *dma_pfn);
+
 	/*
 	 * Handle the memory failure happens on a range of pfns.  Notify the
 	 * processes who are using these pfns, and try to recover the data on
diff --git a/mm/hmm.c b/mm/hmm.c
index feac86196a65..089e522b346b 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -232,6 +232,49 @@ static inline unsigned long pte_to_hmm_pfn_flags(struct hmm_range *range,
 	return pte_write(pte) ? (HMM_PFN_VALID | HMM_PFN_WRITE) : HMM_PFN_VALID;
 }
 
+static bool hmm_handle_device_private(struct hmm_range *range,
+				      unsigned long pfn_req_flags,
+				      swp_entry_t entry,
+				      unsigned long *hmm_pfn)
+{
+	struct page *page = pfn_swap_entry_to_page(entry);
+	struct dev_pagemap *pgmap = page_pgmap(page);
+	int ret;
+
+	pfn_req_flags &= range->pfn_flags_mask;
+	pfn_req_flags |= range->default_flags;
+
+	/*
+	 * Don't fault in device private pages owned by the caller,
+	 * just report the PFN.
+	 */
+	if (pgmap->owner == range->dev_private_owner) {
+		*hmm_pfn = swp_offset_pfn(entry);
+		goto found;
+	}
+
+	/*
+	 * P2P for supported pages, and according to caller request
+	 * translate the private page to the match P2P page if it fails
+	 * continue with the regular flow
+	 */
+	if (pfn_req_flags & HMM_PFN_ALLOW_P2P &&
+	    pgmap->ops->get_dma_pfn_for_device) {
+		ret = pgmap->ops->get_dma_pfn_for_device(page, hmm_pfn);
+		if (!ret)
+			goto found;
+
+	}
+
+	return false;
+
+found:
+	*hmm_pfn |= HMM_PFN_VALID;
+	if (is_writable_device_private_entry(entry))
+		*hmm_pfn |= HMM_PFN_WRITE;
+	return true;
+}
+
 static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr,
 			      unsigned long end, pmd_t *pmdp, pte_t *ptep,
 			      unsigned long *hmm_pfn)
@@ -255,19 +298,9 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr,
 	if (!pte_present(pte)) {
 		swp_entry_t entry = pte_to_swp_entry(pte);
 
-		/*
-		 * Don't fault in device private pages owned by the caller,
-		 * just report the PFN.
-		 */
 		if (is_device_private_entry(entry) &&
-		    page_pgmap(pfn_swap_entry_to_page(entry))->owner ==
-		    range->dev_private_owner) {
-			cpu_flags = HMM_PFN_VALID;
-			if (is_writable_device_private_entry(entry))
-				cpu_flags |= HMM_PFN_WRITE;
-			new_pfn_flags = swp_offset_pfn(entry) | cpu_flags;
-			goto out;
-		}
+		    hmm_handle_device_private(range, pfn_req_flags, entry, hmm_pfn))
+			return 0;
 
 		required_fault =
 			hmm_pte_need_fault(hmm_vma_walk, pfn_req_flags, 0);
-- 
2.34.1