[Xen-devel] [PATCH v2] x86: Don't crash when mapping a page using EFI runtime page tables

[Ross Lagerwall <ross.lagerwall@xxxxxxxxxx>]

When an interrupt is received during an EFI runtime service call, Xen
may call map_domain_page() while using the EFI runtime page tables.
This fails because, although the EFI runtime page tables are a
copy of the idle domain's page tables, current points at a different
domain's vCPU.

To fix this, return NULL from mapcache_current_vcpu() when using the EFI
runtime page tables which is treated equivalently to running in an idle
vCPU.

This issue can be reproduced by repeatedly calling GetVariable() from
dom0 while using VT-d, since VT-d frequently maps a page from interrupt
context.

Example call trace:
[<ffff82d0801615dc>] __find_next_zero_bit+0x28/0x60
[<ffff82d08016a10e>] map_domain_page+0x4c6/0x4eb
[<ffff82d080156ae6>] map_vtd_domain_page+0xd/0xf
[<ffff82d08015533a>] msi_msg_read_remap_rte+0xe3/0x1d8
[<ffff82d08014e516>] iommu_read_msi_from_ire+0x31/0x34
[<ffff82d08016ff6c>] set_msi_affinity+0x134/0x17a
[<ffff82d0801737b5>] move_masked_irq+0x5c/0x98
[<ffff82d080173816>] move_native_irq+0x25/0x36
[<ffff82d08016ffcb>] ack_nonmaskable_msi_irq+0x19/0x20
[<ffff82d08016ffdb>] ack_maskable_msi_irq+0x9/0x37
[<ffff82d080173e8b>] do_IRQ+0x251/0x635
[<ffff82d080234502>] common_interrupt+0x62/0x70
[<00000000df7ed2be>] 00000000df7ed2be

Signed-off-by: Ross Lagerwall <ross.lagerwall@xxxxxxxxxx>
---

Changed in v2:
* Only read CR3 when checking if running on the runtime page tables
  if inside the efi runtime services lock and on the same CPU that
  took the lock.

 xen/arch/x86/domain_page.c | 13 +++++++++----
 xen/arch/x86/efi/stub.c    |  2 +-
 xen/common/efi/runtime.c   | 10 ++++++++--
 xen/include/xen/efi.h      |  2 +-
 4 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/xen/arch/x86/domain_page.c b/xen/arch/x86/domain_page.c
index 5f6f397..7954998 100644
--- a/xen/arch/x86/domain_page.c
+++ b/xen/arch/x86/domain_page.c
@@ -32,20 +32,25 @@ static inline struct vcpu *mapcache_current_vcpu(void)
         return NULL;
 
     /*
+     * When using efi runtime page tables, we have the equivalent of the idle
+     * domain's page tables but current may point at another domain's VCPU.
+     * Return NULL as though current is not properly set up yet.
+     */
+    if ( efi_enabled && efi_rs_using_pgtables() )
+        return NULL;
+
+    /*
      * If guest_table is NULL, and we are running a paravirtualised guest,
      * then it means we are running on the idle domain's page table and must
      * therefore use its mapcache.
      */
     if ( unlikely(pagetable_is_null(v->arch.guest_table)) && is_pv_vcpu(v) )
     {
-        unsigned long cr3;
-
         /* If we really are idling, perform lazy context switch now. */
         if ( (v = idle_vcpu[smp_processor_id()]) == current )
             sync_local_execstate();
         /* We must now be running on the idle page table. */
-        ASSERT((cr3 = read_cr3()) == __pa(idle_pg_table) ||
-               (efi_enabled && cr3 == efi_rs_page_table()));
+        ASSERT(read_cr3() == __pa(idle_pg_table));
     }
 
     return v;
diff --git a/xen/arch/x86/efi/stub.c b/xen/arch/x86/efi/stub.c
index 627009f..07c2bd0 100644
--- a/xen/arch/x86/efi/stub.c
+++ b/xen/arch/x86/efi/stub.c
@@ -12,7 +12,7 @@ void __init efi_init_memory(void) { }
 
 void efi_update_l4_pgtable(unsigned int l4idx, l4_pgentry_t l4e) { }
 
-paddr_t efi_rs_page_table(void)
+bool_t efi_rs_using_pgtables(void)
 {
     BUG();
     return 0;
diff --git a/xen/common/efi/runtime.c b/xen/common/efi/runtime.c
index 5ed8b01..ae87557 100644
--- a/xen/common/efi/runtime.c
+++ b/xen/common/efi/runtime.c
@@ -30,6 +30,7 @@ const CHAR16 *__read_mostly efi_fw_vendor;
 const EFI_RUNTIME_SERVICES *__read_mostly efi_rs;
 #ifndef CONFIG_ARM /* TODO - disabled until implemented on ARM */
 static DEFINE_SPINLOCK(efi_rs_lock);
+static unsigned int efi_rs_on_cpu = NR_CPUS;
 #endif
 
 UINTN __read_mostly efi_memmap_size;
@@ -66,6 +67,8 @@ unsigned long efi_rs_enter(void)
 
     spin_lock(&efi_rs_lock);
 
+    efi_rs_on_cpu = smp_processor_id();
+
     /* prevent fixup_page_fault() from doing anything */
     irq_enter();
 
@@ -100,13 +103,16 @@ void efi_rs_leave(unsigned long cr3)
         asm volatile ( "lgdt %0" : : "m" (gdt_desc) );
     }
     irq_exit();
+    efi_rs_on_cpu = NR_CPUS;
     spin_unlock(&efi_rs_lock);
     stts();
 }
 
-paddr_t efi_rs_page_table(void)
+bool_t efi_rs_using_pgtables(void)
 {
-    return efi_l4_pgtable ? virt_to_maddr(efi_l4_pgtable) : 0;
+    return efi_l4_pgtable &&
+           (smp_processor_id() == efi_rs_on_cpu) &&
+           (read_cr3() == virt_to_maddr(efi_l4_pgtable));
 }
 
 unsigned long efi_get_time(void)
diff --git a/xen/include/xen/efi.h b/xen/include/xen/efi.h
index 48de8e0..e74dad1 100644
--- a/xen/include/xen/efi.h
+++ b/xen/include/xen/efi.h
@@ -29,7 +29,7 @@ struct xenpf_efi_runtime_call;
 struct compat_pf_efi_runtime_call;
 
 void efi_init_memory(void);
-paddr_t efi_rs_page_table(void);
+bool_t efi_rs_using_pgtables(void);
 unsigned long efi_get_time(void);
 void efi_halt_system(void);
 void efi_reset_system(bool_t warm);
-- 
2.1.0


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel