Commit 140456f9 authored by Andrey Ryabinin's avatar Andrey Ryabinin Committed by Joerg Roedel

iommu/amd: Fix sleeping in atomic in increase_address_space()

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: 754265bc ("iommu/amd: Fix race in increase_address_space()")
Signed-off-by: default avatarAndrey Ryabinin <arbn@yandex-team.com>
Acked-by: default avatarWill Deacon <will@kernel.org>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20210217143004.19165-1-arbn@yandex-team.comSigned-off-by: default avatarJoerg Roedel <jroedel@suse.de>
parent 45e606f2
......@@ -182,6 +182,10 @@ static bool increase_address_space(struct protection_domain *domain,
bool ret = true;
u64 *pte;
pte = (void *)get_zeroed_page(gfp);
if (!pte)
return false;
spin_lock_irqsave(&domain->lock, flags);
if (address <= PM_LEVEL_SIZE(domain->iop.mode))
......@@ -191,10 +195,6 @@ static bool increase_address_space(struct protection_domain *domain,
if (WARN_ON_ONCE(domain->iop.mode == PAGE_MODE_6_LEVEL))
goto out;
pte = (void *)get_zeroed_page(gfp);
if (!pte)
goto out;
*pte = PM_LEVEL_PDE(domain->iop.mode, iommu_virt_to_phys(domain->iop.root));
domain->iop.root = pte;
......@@ -208,10 +208,12 @@ static bool increase_address_space(struct protection_domain *domain,
*/
amd_iommu_domain_set_pgtable(domain, pte, domain->iop.mode);
pte = NULL;
ret = true;
out:
spin_unlock_irqrestore(&domain->lock, flags);
free_page((unsigned long)pte);
return ret;
}
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment