[v2,03/46] mm: Add kmap_local_folio()

Commit Message

Matthew Wilcox June 22, 2021, 12:15 p.m. UTC

This allows us to map a portion of a folio.  Callers can only expect
to access up to the next page boundary.

Signed-off-by: Matthew Wilcox (Oracle) <[e-mail]>
---
 include/linux/highmem-internal.h | 11 +++++++++
 include/linux/highmem.h          | 38 ++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

Comments

Christoph Hellwig June 23, 2021, 7:58 a.m. UTC | #1

On Tue, Jun 22, 2021 at 01:15:08PM +0100, Matthew Wilcox (Oracle) wrote:
> This allows us to map a portion of a folio.  Callers can only expect
> to access up to the next page boundary.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <[e-mail]>

This looks sensible to me:

Reviewed-by: Christoph Hellwig <[e-mail]>

While we're at it:  flushing incoherent mappings for use with kmap
has been a complete mess.  I've been trying to fix this with
kunmap_local_dirty, but that could use another review, as well as a
folio version probably give that one prime use case of folios is to have
them mapped into userspace.

Patch

diff --git a/include/linux/highmem-internal.h b/include/linux/highmem-internal.h
index 7902c7d8b55f..d5d6f930ae1d 100644
--- a/include/linux/highmem-internal.h
+++ b/include/linux/highmem-internal.h
@@ -73,6 +73,12 @@  static inline void *kmap_local_page(struct page *page)
 	return __kmap_local_page_prot(page, kmap_prot);
 }
 
+static inline void *kmap_local_folio(struct folio *folio, size_t offset)
+{
+	struct page *page = folio_page(folio, offset / PAGE_SIZE);
+	return __kmap_local_page_prot(page, kmap_prot) + offset % PAGE_SIZE;
+}
+
 static inline void *kmap_local_page_prot(struct page *page, pgprot_t prot)
 {
 	return __kmap_local_page_prot(page, prot);
@@ -160,6 +166,11 @@  static inline void *kmap_local_page(struct page *page)
 	return page_address(page);
 }
 
+static inline void *kmap_local_folio(struct folio *folio, size_t offset)
+{
+	return page_address(&folio->page) + offset;
+}
+
 static inline void *kmap_local_page_prot(struct page *page, pgprot_t prot)
 {
 	return kmap_local_page(page);
diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 832b49b50c7b..c7c664e0315b 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -96,6 +96,44 @@  static inline void kmap_flush_unused(void);
  */
 static inline void *kmap_local_page(struct page *page);
 
+/**
+ * kmap_local_folio - Map a page in this folio for temporary usage
+ * @folio:	The folio to be mapped.
+ * @offset:	The byte offset within the folio.
+ *
+ * Returns: The virtual address of the mapping
+ *
+ * Can be invoked from any context.
+ *
+ * Requires careful handling when nesting multiple mappings because the map
+ * management is stack based. The unmap has to be in the reverse order of
+ * the map operation:
+ *
+ * addr1 = kmap_local_folio(page1, offset1);
+ * addr2 = kmap_local_folio(page2, offset2);
+ * ...
+ * kunmap_local(addr2);
+ * kunmap_local(addr1);
+ *
+ * Unmapping addr1 before addr2 is invalid and causes malfunction.
+ *
+ * Contrary to kmap() mappings the mapping is only valid in the context of
+ * the caller and cannot be handed to other contexts.
+ *
+ * On CONFIG_HIGHMEM=n kernels and for low memory pages this returns the
+ * virtual address of the direct mapping. Only real highmem pages are
+ * temporarily mapped.
+ *
+ * While it is significantly faster than kmap() for the higmem case it
+ * comes with restrictions about the pointer validity. Only use when really
+ * necessary.
+ *
+ * On HIGHMEM enabled systems mapping a highmem page has the side effect of
+ * disabling migration in order to keep the virtual address stable across
+ * preemption. No caller of kmap_local_folio() can rely on this side effect.
+ */
+static inline void *kmap_local_folio(struct folio *folio, size_t offset);
+
 /**
  * kmap_atomic - Atomically map a page for temporary usage - Deprecated!
  * @page:	Pointer to the page to be mapped