[PATCH v7 07/17] mm: Close race in generic_access_phys

From: Daniel Vetter <daniel.vetter@ffwll.ch>
To: DRI Development <dri-devel@lists.freedesktop.org>,
	LKML <linux-kernel@vger.kernel.org>
Cc: kvm@vger.kernel.org, linux-mm@kvack.org,
	linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org, linux-media@vger.kernel.org,
	"Daniel Vetter" <daniel.vetter@ffwll.ch>,
	"Daniel Vetter" <daniel.vetter@intel.com>,
	"Jason Gunthorpe" <jgg@ziepe.ca>,
	"Dan Williams" <dan.j.williams@intel.com>,
	"Kees Cook" <keescook@chromium.org>,
	"Benjamin Herrensmidt" <benh@kernel.crashing.org>,
	"Dave Airlie" <airlied@linux.ie>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"John Hubbard" <jhubbard@nvidia.com>,
	"Jérôme Glisse" <jglisse@redhat.com>, "Jan Kara" <jack@suse.cz>,
	"Chris Wilson" <chris@chris-wilson.co.uk>
Subject: [PATCH v7 07/17] mm: Close race in generic_access_phys
Date: Fri, 27 Nov 2020 17:41:21 +0100	[thread overview]
Message-ID: <20201127164131.2244124-8-daniel.vetter@ffwll.ch> (raw)
In-Reply-To: <20201127164131.2244124-1-daniel.vetter@ffwll.ch>

Way back it was a reasonable assumptions that iomem mappings never
change the pfn range they point at. But this has changed:

- gpu drivers dynamically manage their memory nowadays, invalidating
  ptes with unmap_mapping_range when buffers get moved

- contiguous dma allocations have moved from dedicated carvetouts to
  cma regions. This means if we miss the unmap the pfn might contain
  pagecache or anon memory (well anything allocated with GFP_MOVEABLE)

- even /dev/mem now invalidates mappings when the kernel requests that
  iomem region when CONFIG_IO_STRICT_DEVMEM is set, see 3234ac664a87
  ("/dev/mem: Revoke mappings when a driver claims the region")

Accessing pfns obtained from ptes without holding all the locks is
therefore no longer a good idea. Fix this.

Since ioremap might need to manipulate pagetables too we need to drop
the pt lock and have a retry loop if we raced.

While at it, also add kerneldoc and improve the comment for the
vma_ops->access function. It's for accessing, not for moving the
memory from iomem to system memory, as the old comment seemed to
suggest.

References: 28b2ee20c7cb ("access_process_vm device memory infrastructure")
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Benjamin Herrensmidt <benh@kernel.crashing.org>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: linux-mm@kvack.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-samsung-soc@vger.kernel.org
Cc: linux-media@vger.kernel.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
--
v2: Fix inversion in the retry check (John).

v4: While at it, use offset_in_page (Chris Wilson)
---
 include/linux/mm.h |  3 ++-
 mm/memory.c        | 46 +++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index c4cc8ea1402c..b50cbb33d0f3 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -574,7 +574,8 @@ struct vm_operations_struct {
 	vm_fault_t (*pfn_mkwrite)(struct vm_fault *vmf);
 
 	/* called by access_process_vm when get_user_pages() fails, typically
-	 * for use by special VMAs that can switch between memory and hardware
+	 * for use by special VMAs. See also generic_access_phys() for a generic
+	 * implementation useful for any iomem mapping.
 	 */
 	int (*access)(struct vm_area_struct *vma, unsigned long addr,
 		      void *buf, int len, int write);
diff --git a/mm/memory.c b/mm/memory.c
index c48f8df6e502..ac32039ce941 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -4847,28 +4847,68 @@ int follow_phys(struct vm_area_struct *vma,
 	return ret;
 }
 
+/**
+ * generic_access_phys - generic implementation for iomem mmap access
+ * @vma: the vma to access
+ * @addr: userspace addres, not relative offset within @vma
+ * @buf: buffer to read/write
+ * @len: length of transfer
+ * @write: set to FOLL_WRITE when writing, otherwise reading
+ *
+ * This is a generic implementation for &vm_operations_struct.access for an
+ * iomem mapping. This callback is used by access_process_vm() when the @vma is
+ * not page based.
+ */
 int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
 			void *buf, int len, int write)
 {
 	resource_size_t phys_addr;
 	unsigned long prot = 0;
 	void __iomem *maddr;
-	int offset = addr & (PAGE_SIZE-1);
+	pte_t *ptep, pte;
+	spinlock_t *ptl;
+	int offset = offset_in_page(addr);
+	int ret = -EINVAL;
+
+	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
+		return -EINVAL;
+
+retry:
+	if (follow_pte(vma->vm_mm, addr, &ptep, &ptl))
+		return -EINVAL;
+	pte = *ptep;
+	pte_unmap_unlock(ptep, ptl);
 
-	if (follow_phys(vma, addr, write, &prot, &phys_addr))
+	prot = pgprot_val(pte_pgprot(pte));
+	phys_addr = (resource_size_t)pte_pfn(pte) << PAGE_SHIFT;
+
+	if ((write & FOLL_WRITE) && !pte_write(pte))
 		return -EINVAL;
 
 	maddr = ioremap_prot(phys_addr, PAGE_ALIGN(len + offset), prot);
 	if (!maddr)
 		return -ENOMEM;
 
+	if (follow_pte(vma->vm_mm, addr, &ptep, &ptl))
+		goto out_unmap;
+
+	if (!pte_same(pte, *ptep)) {
+		pte_unmap_unlock(ptep, ptl);
+		iounmap(maddr);
+
+		goto retry;
+	}
+
 	if (write)
 		memcpy_toio(maddr + offset, buf, len);
 	else
 		memcpy_fromio(buf, maddr + offset, len);
+	ret = len;
+	pte_unmap_unlock(ptep, ptl);
+out_unmap:
 	iounmap(maddr);
 
-	return len;
+	return ret;
 }
 EXPORT_SYMBOL_GPL(generic_access_phys);
 #endif
-- 
2.29.2

