[PATCH 4.9 08/11] iommu/amd: Fix sleeping in atomic in increase_address_space()

[gregkh]

From: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

From: Andrey Ryabinin <arbn@xxxxxxxxxxxxxxx>

commit 140456f994195b568ecd7fc2287a34eadffef3ca upstream.

increase_address_space() calls get_zeroed_page(gfp) under spin_lock with
disabled interrupts. gfp flags passed to increase_address_space() may allow
sleeping, so it comes to this:

 BUG: sleeping function called from invalid context at mm/page_alloc.c:4342
 in_atomic(): 1, irqs_disabled(): 1, pid: 21555, name: epdcbbf1qnhbsd8

 Call Trace:
  dump_stack+0x66/0x8b
  ___might_sleep+0xec/0x110
  __alloc_pages_nodemask+0x104/0x300
  get_zeroed_page+0x15/0x40
  iommu_map_page+0xdd/0x3e0
  amd_iommu_map+0x50/0x70
  iommu_map+0x106/0x220
  vfio_iommu_type1_ioctl+0x76e/0x950 [vfio_iommu_type1]
  do_vfs_ioctl+0xa3/0x6f0
  ksys_ioctl+0x66/0x70
  __x64_sys_ioctl+0x16/0x20
  do_syscall_64+0x4e/0x100
  entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fix this by moving get_zeroed_page() out of spin_lock/unlock section.

Fixes: 754265bcab ("iommu/amd: Fix race in increase_address_space()")
Signed-off-by: Andrey Ryabinin <arbn@xxxxxxxxxxxxxxx>
Acked-by: Will Deacon <will@xxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20210217143004.19165-1-arbn@xxxxxxxxxxxxxxx
Signed-off-by: Joerg Roedel <jroedel@xxxxxxx>
Signed-off-by: Andrey Ryabinin <arbn@xxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 drivers/iommu/amd_iommu.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -1331,24 +1331,26 @@ static void increase_address_space(struc
 	unsigned long flags;
 	u64 *pte;
 
+	pte = (void *)get_zeroed_page(gfp);
+	if (!pte)
+		goto out;
+
 	spin_lock_irqsave(&domain->lock, flags);
 
 	if (WARN_ON_ONCE(domain->mode == PAGE_MODE_6_LEVEL))
 		/* address space already 64 bit large */
 		goto out;
 
-	pte = (void *)get_zeroed_page(gfp);
-	if (!pte)
-		goto out;
-
 	*pte             = PM_LEVEL_PDE(domain->mode,
 					virt_to_phys(domain->pt_root));
 	domain->pt_root  = pte;
 	domain->mode    += 1;
 	domain->updated  = true;
+	pte              = NULL;
 
 out:
 	spin_unlock_irqrestore(&domain->lock, flags);
+	free_page((unsigned long)pte);
 
 	return;
 }