mm/hmm: remove the customizable pfn format from hmm_range_fault

Presumably the intent here was that hmm_range_fault() could put the data
into some HW specific format and thus avoid some work. However, nothing
actually does that, and it isn't clear how anything actually could do that
as hmm_range_fault() provides CPU addresses which must be DMA mapped.

Perhaps there is some special HW that does not need DMA mapping, but we
don't have any examples of this, and the theoretical performance win of
avoiding an extra scan over the pfns array doesn't seem worth the
complexity. Plus pfns needs to be scanned anyhow to sort out any
DEVICE_PRIVATE pages.

This version replaces the uint64_t with an usigned long containing a pfn
and fixed flags. On input flags is filled with the HMM_PFN_REQ_* values,
on successful output it is filled with HMM_PFN_* values, describing the
state of the pages.

amdgpu is simple to convert, it doesn't use snapshot and doesn't use
per-page flags.

nouveau uses only 16 hmm_pte entries at most (ie fits in a few cache
lines), and it sweeps over its pfns array a couple of times anyhow. It
also has a nasty call chain before it reaches the dma map and hardware
suggesting performance isn't important:

   nouveau_svm_fault():
     args.i.m.method = NVIF_VMM_V0_PFNMAP
     nouveau_range_fault()
      nvif_object_ioctl()
       client->driver->ioctl()
	  struct nvif_driver nvif_driver_nvkm:
	    .ioctl = nvkm_client_ioctl
	   nvkm_ioctl()
	    nvkm_ioctl_path()
	      nvkm_ioctl_v0[type].func(..)
	      nvkm_ioctl_mthd()
	       nvkm_object_mthd()
		  struct nvkm_object_func nvkm_uvmm:
		    .mthd = nvkm_uvmm_mthd
		   nvkm_uvmm_mthd()
		    nvkm_uvmm_mthd_pfnmap()
		     nvkm_vmm_pfn_map()
		      nvkm_vmm_ptes_get_map()
		       func == gp100_vmm_pgt_pfn
			struct nvkm_vmm_desc_func gp100_vmm_desc_spt:
			  .pfn = gp100_vmm_pgt_pfn
			 nvkm_vmm_iter()
			  REF_PTES == func == gp100_vmm_pgt_pfn()
			    dma_map_page()

Link: https://lore.kernel.org/r/5-v2-b4e84f444c7d+24f57-hmm_no_flags_jgg@mellanox.com
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Ralph Campbell <rcampbell@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

1 files changed, 38 insertions, 63 deletions

diff --git a/include/linux/hmm.h b/include/linux/hmm.h
index 81c302c884c0..e912b9dc4633 100644
--- a/include/linux/hmm.h
+++ b/include/linux/hmm.h
@@ -19,45 +19,47 @@
 #include <linux/mmu_notifier.h>
 
 /*
- * hmm_pfn_flag_e - HMM flag enums
+ * On output:
+ * 0             - The page is faultable and a future call with 
+ *                 HMM_PFN_REQ_FAULT could succeed.
+ * HMM_PFN_VALID - the pfn field points to a valid PFN. This PFN is at
+ *                 least readable. If dev_private_owner is !NULL then this could
+ *                 point at a DEVICE_PRIVATE page.
+ * HMM_PFN_WRITE - if the page memory can be written to (requires HMM_PFN_VALID)
+ * HMM_PFN_ERROR - accessing the pfn is impossible and the device should
+ *                 fail. ie poisoned memory, special pages, no vma, etc
  *
- * Flags:
- * HMM_PFN_VALID: pfn is valid. It has, at least, read permission.
- * HMM_PFN_WRITE: CPU page table has write permission set
- *
- * The driver provides a flags array for mapping page protections to device
- * PTE bits. If the driver valid bit for an entry is bit 3,
- * i.e., (entry & (1 << 3)), then the driver must provide
- * an array in hmm_range.flags with hmm_range.flags[HMM_PFN_VALID] == 1 << 3.
- * Same logic apply to all flags. This is the same idea as vm_page_prot in vma
- * except that this is per device driver rather than per architecture.
+ * On input:
+ * 0                 - Return the current state of the page, do not fault it.
+ * HMM_PFN_REQ_FAULT - The output must have HMM_PFN_VALID or hmm_range_fault()
+ *                     will fail
+ * HMM_PFN_REQ_WRITE - The output must have HMM_PFN_WRITE or hmm_range_fault()
+ *                     will fail. Must be combined with HMM_PFN_REQ_FAULT.
  */
-enum hmm_pfn_flag_e {
-	HMM_PFN_VALID = 0,
-	HMM_PFN_WRITE,
-	HMM_PFN_FLAG_MAX
+enum hmm_pfn_flags {
+	/* Output flags */
+	HMM_PFN_VALID = 1UL << (BITS_PER_LONG - 1),
+	HMM_PFN_WRITE = 1UL << (BITS_PER_LONG - 2),
+	HMM_PFN_ERROR = 1UL << (BITS_PER_LONG - 3),
+
+	/* Input flags */
+	HMM_PFN_REQ_FAULT = HMM_PFN_VALID,
+	HMM_PFN_REQ_WRITE = HMM_PFN_WRITE,
+
+	HMM_PFN_FLAGS = HMM_PFN_VALID | HMM_PFN_WRITE | HMM_PFN_ERROR,
 };
 
 /*
- * hmm_pfn_value_e - HMM pfn special value
- *
- * Flags:
- * HMM_PFN_ERROR: corresponding CPU page table entry points to poisoned memory
- * HMM_PFN_NONE: corresponding CPU page table entry is pte_none()
+ * hmm_pfn_to_page() - return struct page pointed to by a device entry
  *
- * Driver provides values for none entry, error entry, and special entry.
- * Driver can alias (i.e., use same value) error and special, but
- * it should not alias none with error or special.
- *
- * HMM pfn value returned by hmm_vma_get_pfns() or hmm_vma_fault() will be:
- * hmm_range.values[HMM_PFN_ERROR] if CPU page table entry is poisonous,
- * hmm_range.values[HMM_PFN_NONE] if there is no CPU page table entry,
+ * This must be called under the caller 'user_lock' after a successful
+ * mmu_interval_read_begin(). The caller must have tested for HMM_PFN_VALID
+ * already.
  */
-enum hmm_pfn_value_e {
-	HMM_PFN_ERROR,
-	HMM_PFN_NONE,
-	HMM_PFN_VALUE_MAX
-};
+static inline struct page *hmm_pfn_to_page(unsigned long hmm_pfn)
+{
+	return pfn_to_page(hmm_pfn & ~HMM_PFN_FLAGS);
+}
 
 /*
  * struct hmm_range - track invalidation lock on virtual address range
@@ -66,12 +68,9 @@ enum hmm_pfn_value_e {
  * @notifier_seq: result of mmu_interval_read_begin()
  * @start: range virtual start address (inclusive)
  * @end: range virtual end address (exclusive)
- * @pfns: array of pfns (big enough for the range)
- * @flags: pfn flags to match device driver page table
- * @values: pfn value for some special case (none, special, error, ...)
+ * @hmm_pfns: array of pfns (big enough for the range)
  * @default_flags: default flags for the range (write, read, ... see hmm doc)
  * @pfn_flags_mask: allows to mask pfn flags so that only default_flags matter
- * @pfn_shift: pfn shift value (should be <= PAGE_SHIFT)
  * @dev_private_owner: owner of device private pages
  */
 struct hmm_range {
@@ -79,37 +78,13 @@ struct hmm_range {
 	unsigned long		notifier_seq;
 	unsigned long		start;
 	unsigned long		end;
-	uint64_t		*pfns;
-	const uint64_t		*flags;
-	const uint64_t		*values;
-	uint64_t		default_flags;
-	uint64_t		pfn_flags_mask;
-	uint8_t			pfn_shift;
+	unsigned long		*hmm_pfns;
+	unsigned long		default_flags;
+	unsigned long		pfn_flags_mask;
 	void			*dev_private_owner;
 };
 
 /*
- * hmm_device_entry_to_page() - return struct page pointed to by a device entry
- * @range: range use to decode device entry value
- * @entry: device entry value to get corresponding struct page from
- * Return: struct page pointer if entry is a valid, NULL otherwise
- *
- * If the device entry is valid (ie valid flag set) then return the struct page
- * matching the entry value. Otherwise return NULL.
- */
-static inline struct page *hmm_device_entry_to_page(const struct hmm_range *range,
-						    uint64_t entry)
-{
-	if (entry == range->values[HMM_PFN_NONE])
-		return NULL;
-	if (entry == range->values[HMM_PFN_ERROR])
-		return NULL;
-	if (!(entry & range->flags[HMM_PFN_VALID]))
-		return NULL;
-	return pfn_to_page(entry >> range->pfn_shift);
-}
-
-/*
  * Please see Documentation/vm/hmm.rst for how to use the range API.
  */
 int hmm_range_fault(struct hmm_range *range);