Fixes for 4.14

Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/queue-4.14/mm-fix-exec-activate_mm-vs-tlb-shootdown-and-lazy-tl.patch b/queue-4.14/mm-fix-exec-activate_mm-vs-tlb-shootdown-and-lazy-tl.patch
new file mode 100644 (file)
index 0000000..e96eb93
--- /dev/null
+++ b/queue-4.14/mm-fix-exec-activate_mm-vs-tlb-shootdown-and-lazy-tl.patch
@@ -0,0 +1,115 @@
+From 6743ee4f371a583700264efca0504a5b2b4437cd Mon Sep 17 00:00:00 2001
+From: Sasha Levin <sashal@kernel.org>
+Date: Mon, 14 Sep 2020 14:52:16 +1000
+Subject: mm: fix exec activate_mm vs TLB shootdown and lazy tlb switching race
+
+From: Nicholas Piggin <npiggin@gmail.com>
+
+commit d53c3dfb23c45f7d4f910c3a3ca84bf0a99c6143 upstream.
+
+Reading and modifying current->mm and current->active_mm and switching
+mm should be done with irqs off, to prevent races seeing an intermediate
+state.
+
+This is similar to commit 38cf307c1f20 ("mm: fix kthread_use_mm() vs TLB
+invalidate"). At exec-time when the new mm is activated, the old one
+should usually be single-threaded and no longer used, unless something
+else is holding an mm_users reference (which may be possible).
+
+Absent other mm_users, there is also a race with preemption and lazy tlb
+switching. Consider the kernel_execve case where the current thread is
+using a lazy tlb active mm:
+
+  call_usermodehelper()
+    kernel_execve()
+      old_mm = current->mm;
+      active_mm = current->active_mm;
+      *** preempt *** -------------------->  schedule()
+                                               prev->active_mm = NULL;
+                                               mmdrop(prev active_mm);
+                                             ...
+                      <--------------------  schedule()
+      current->mm = mm;
+      current->active_mm = mm;
+      if (!old_mm)
+          mmdrop(active_mm);
+
+If we switch back to the kernel thread from a different mm, there is a
+double free of the old active_mm, and a missing free of the new one.
+
+Closing this race only requires interrupts to be disabled while ->mm
+and ->active_mm are being switched, but the TLB problem requires also
+holding interrupts off over activate_mm. Unfortunately not all archs
+can do that yet, e.g., arm defers the switch if irqs are disabled and
+expects finish_arch_post_lock_switch() to be called to complete the
+flush; um takes a blocking lock in activate_mm().
+
+So as a first step, disable interrupts across the mm/active_mm updates
+to close the lazy tlb preempt race, and provide an arch option to
+extend that to activate_mm which allows architectures doing IPI based
+TLB shootdowns to close the second race.
+
+This is a bit ugly, but in the interest of fixing the bug and backporting
+before all architectures are converted this is a compromise.
+
+Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
+Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
+[mpe: Manual backport to 4.19 due to membarrier_exec_mmap(mm) changes]
+Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
+Link: https://lore.kernel.org/r/20200914045219.3736466-2-npiggin@gmail.com
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+---
+ arch/Kconfig |  7 +++++++
+ fs/exec.c    | 15 ++++++++++++++-
+ 2 files changed, 21 insertions(+), 1 deletion(-)
+
+diff --git a/arch/Kconfig b/arch/Kconfig
+index 77b3e21c4844e..95567f6832752 100644
+--- a/arch/Kconfig
++++ b/arch/Kconfig
+@@ -339,6 +339,13 @@ config HAVE_RCU_TABLE_FREE
+ config HAVE_RCU_TABLE_INVALIDATE
+       bool
+ 
++config ARCH_WANT_IRQS_OFF_ACTIVATE_MM
++      bool
++      help
++        Temporary select until all architectures can be converted to have
++        irqs disabled over activate_mm. Architectures that do IPI based TLB
++        shootdowns should enable this.
++
+ config ARCH_HAVE_NMI_SAFE_CMPXCHG
+       bool
+ 
+diff --git a/fs/exec.c b/fs/exec.c
+index 92fd21f0b0ff7..e8d1e67059775 100644
+--- a/fs/exec.c
++++ b/fs/exec.c
+@@ -1024,10 +1024,23 @@ static int exec_mmap(struct mm_struct *mm)
+               }
+       }
+       task_lock(tsk);
++
++      local_irq_disable();
+       active_mm = tsk->active_mm;
+-      tsk->mm = mm;
+       tsk->active_mm = mm;
++      tsk->mm = mm;
++      /*
++       * This prevents preemption while active_mm is being loaded and
++       * it and mm are being updated, which could cause problems for
++       * lazy tlb mm refcounting when these are updated by context
++       * switches. Not all architectures can handle irqs off over
++       * activate_mm yet.
++       */
++      if (!IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM))
++              local_irq_enable();
+       activate_mm(active_mm, mm);
++      if (IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM))
++              local_irq_enable();
+       tsk->mm->vmacache_seqnum = 0;
+       vmacache_flush(tsk);
+       task_unlock(tsk);
+-- 
+2.27.0
+

diff --git a/queue-4.14/series b/queue-4.14/series
index 43a0a5650047faf60273331b30a537cefe625de8..6a3ab7292af4d1699e234e7212ee751a8d712acf 100644 (file)
--- a/queue-4.14/series
+++ b/queue-4.14/series
@@ -82,3 +82,4 @@ perf-core-fix-race-in-the-perf_mmap_close-function.patch
 revert-kernel-reboot.c-convert-simple_strtoul-to-kstrtoint.patch
 reboot-fix-overflow-parsing-reboot-cpu-number.patch
 convert-trailing-spaces-and-periods-in-path-components.patch
+mm-fix-exec-activate_mm-vs-tlb-shootdown-and-lazy-tl.patch