[PATCH] vfio/type1: Fix task tracking for QEMU vCPU hotplug

[Alex Williamson <alex.williamson@redhat.com>]

MAP_DMA ioctls might be called from various threads within a process,
for example when using QEMU, the vCPU threads are often generating
these calls and we therefore take a reference to that vCPU task.
However, QEMU also supports vCPU hotplug on some machines and the task
that called MAP_DMA may have exited by the time UNMAP_DMA is called,
resulting in the mm_struct pointer being NULL and thus a failure to
match against the existing mapping.

To resolve this, we instead take a reference to the thread
group_leader, which has the same mm_struct and resource limits, but
is less likely exit, at least in the QEMU case.  A difficulty here is
guaranteeing that the capabilities of the group_leader match that of
the calling thread, which we resolve by tracking CAP_IPC_LOCK at the
time of calling rather than at an indeterminate time in the future.
Potentially this also results in better efficiency as this is now
recorded once per MAP_DMA ioctl.

Reported-by: Xu Yandong <xuyandong2@huawei.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
---
 drivers/vfio/vfio_iommu_type1.c |   73 +++++++++++++++++++++++++--------------
 1 file changed, 47 insertions(+), 26 deletions(-)

diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index 5c212bf29640..1dca77f37fd0 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -83,6 +83,7 @@ struct vfio_dma {
 	size_t			size;		/* Map size (bytes) */
 	int			prot;		/* IOMMU_READ/WRITE */
 	bool			iommu_mapped;
+	bool			lock_cap;	/* capable(CAP_IPC_LOCK) */
 	struct task_struct	*task;
 	struct rb_root		pfn_list;	/* Ex-user pinned pfn list */
 };
@@ -253,29 +254,25 @@ static int vfio_iova_put_vfio_pfn(struct vfio_dma *dma, struct vfio_pfn *vpfn)
 	return ret;
 }
 
-static int vfio_lock_acct(struct task_struct *task, long npage, bool *lock_cap)
+static int vfio_lock_acct(struct vfio_dma *dma, long npage, bool async)
 {
 	struct mm_struct *mm;
-	bool is_current;
 	int ret;
 
 	if (!npage)
 		return 0;
 
-	is_current = (task->mm == current->mm);
-
-	mm = is_current ? task->mm : get_task_mm(task);
+	mm = async ? get_task_mm(dma->task) : dma->task->mm;
 	if (!mm)
 		return -ESRCH; /* process exited */
 
 	ret = down_write_killable(&mm->mmap_sem);
 	if (!ret) {
 		if (npage > 0) {
-			if (lock_cap ? !*lock_cap :
-			    !has_capability(task, CAP_IPC_LOCK)) {
+			if (!dma->lock_cap) {
 				unsigned long limit;
 
-				limit = task_rlimit(task,
+				limit = task_rlimit(dma->task,
 						RLIMIT_MEMLOCK) >> PAGE_SHIFT;
 
 				if (mm->locked_vm + npage > limit)
@@ -289,7 +286,7 @@ static int vfio_lock_acct(struct task_struct *task, long npage, bool *lock_cap)
 		up_write(&mm->mmap_sem);
 	}
 
-	if (!is_current)
+	if (async)
 		mmput(mm);
 
 	return ret;
@@ -400,7 +397,7 @@ static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr,
  */
 static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
 				  long npage, unsigned long *pfn_base,
-				  bool lock_cap, unsigned long limit)
+				  unsigned long limit)
 {
 	unsigned long pfn = 0;
 	long ret, pinned = 0, lock_acct = 0;
@@ -431,7 +428,7 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
 	 * pages are already counted against the user.
 	 */
 	if (!vfio_find_vpfn(dma, iova)) {
-		if (!lock_cap && current->mm->locked_vm + 1 > limit) {
+		if (!dma->lock_cap && current->mm->locked_vm + 1 > limit) {
 			put_pfn(*pfn_base, dma->prot);
 			pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n", __func__,
 					limit << PAGE_SHIFT);
@@ -456,7 +453,7 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
 		}
 
 		if (!vfio_find_vpfn(dma, iova)) {
-			if (!lock_cap &&
+			if (!dma->lock_cap &&
 			    current->mm->locked_vm + lock_acct + 1 > limit) {
 				put_pfn(pfn, dma->prot);
 				pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n",
@@ -469,7 +466,7 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
 	}
 
 out:
-	ret = vfio_lock_acct(current, lock_acct, &lock_cap);
+	ret = vfio_lock_acct(dma, lock_acct, false);
 
 unpin_out:
 	if (ret) {
@@ -498,7 +495,7 @@ static long vfio_unpin_pages_remote(struct vfio_dma *dma, dma_addr_t iova,
 	}
 
 	if (do_accounting)
-		vfio_lock_acct(dma->task, locked - unlocked, NULL);
+		vfio_lock_acct(dma, locked - unlocked, true);
 
 	return unlocked;
 }
@@ -515,7 +512,7 @@ static int vfio_pin_page_external(struct vfio_dma *dma, unsigned long vaddr,
 
 	ret = vaddr_get_pfn(mm, vaddr, dma->prot, pfn_base);
 	if (!ret && do_accounting && !is_invalid_reserved_pfn(*pfn_base)) {
-		ret = vfio_lock_acct(dma->task, 1, NULL);
+		ret = vfio_lock_acct(dma, 1, true);
 		if (ret) {
 			put_pfn(*pfn_base, dma->prot);
 			if (ret == -ENOMEM)
@@ -542,7 +539,7 @@ static int vfio_unpin_page_external(struct vfio_dma *dma, dma_addr_t iova,
 	unlocked = vfio_iova_put_vfio_pfn(dma, vpfn);
 
 	if (do_accounting)
-		vfio_lock_acct(dma->task, -unlocked, NULL);
+		vfio_lock_acct(dma, -unlocked, true);
 
 	return unlocked;
 }
@@ -834,7 +831,7 @@ static long vfio_unmap_unpin(struct vfio_iommu *iommu, struct vfio_dma *dma,
 		unlocked += vfio_sync_unpin(dma, domain, &unmapped_region_list);
 
 	if (do_accounting) {
-		vfio_lock_acct(dma->task, -unlocked, NULL);
+		vfio_lock_acct(dma, -unlocked, true);
 		return 0;
 	}
 	return unlocked;
@@ -1049,14 +1046,12 @@ static int vfio_pin_map_dma(struct vfio_iommu *iommu, struct vfio_dma *dma,
 	size_t size = map_size;
 	long npage;
 	unsigned long pfn, limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
-	bool lock_cap = capable(CAP_IPC_LOCK);
 	int ret = 0;
 
 	while (size) {
 		/* Pin a contiguous chunk of memory */
 		npage = vfio_pin_pages_remote(dma, vaddr + dma->size,
-					      size >> PAGE_SHIFT, &pfn,
-					      lock_cap, limit);
+					      size >> PAGE_SHIFT, &pfn, limit);
 		if (npage <= 0) {
 			WARN_ON(!npage);
 			ret = (int)npage;
@@ -1131,8 +1126,36 @@ static int vfio_dma_do_map(struct vfio_iommu *iommu,
 	dma->iova = iova;
 	dma->vaddr = vaddr;
 	dma->prot = prot;
-	get_task_struct(current);
-	dma->task = current;
+
+	/*
+	 * We need to be able to both add to a task's locked memory and test
+	 * against the locked memory limit and we need to be able to do both
+	 * outside of this call path as pinning can be asynchronous via the
+	 * external interfaces for mdev devices.  RLIMIT_MEMLOCK requires a
+	 * task_struct and VM locked pages requires an mm_struct, however
+	 * holding an indefinite mm reference is not recommended, therefore we
+	 * only hold a reference to a task.  We could hold a reference to
+	 * current, however QEMU uses this call path through vCPU threads,
+	 * which can be killed resulting in a NULL mm and failure in the unmap
+	 * path when called via a different thread.  Avoid this problem by
+	 * using the group_leader as threads within the same group require
+	 * both CLONE_THREAD and CLONE_VM and will therefore use the same
+	 * mm_struct.
+	 *
+	 * Previously we also used the task for testing CAP_IPC_LOCK at the
+	 * time of pinning and accounting, however has_capability() makes use
+	 * of real_cred, a copy-on-write field, so we can't guarantee that it
+	 * matches group_leader, or in fact that it might not change by the
+	 * time it's evaluated.  If a process were to call MAP_DMA with
+	 * CAP_IPC_LOCK but later drop it, it doesn't make sense that they
+	 * possibly see different results for an iommu_mapped vfio_dma vs
+	 * externally mapped.  Therefore track CAP_IPC_LOCK in vfio_dma at the
+	 * time of calling MAP_DMA.
+	 */
+	get_task_struct(current->group_leader);
+	dma->task = current->group_leader;
+	dma->lock_cap = capable(CAP_IPC_LOCK);
+
 	dma->pfn_list = RB_ROOT;
 
 	/* Insert zero-sized and grow as we map chunks of it */
@@ -1167,7 +1190,6 @@ static int vfio_iommu_replay(struct vfio_iommu *iommu,
 	struct vfio_domain *d;
 	struct rb_node *n;
 	unsigned long limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
-	bool lock_cap = capable(CAP_IPC_LOCK);
 	int ret;
 
 	/* Arbitrarily pick the first domain in the list for lookups */
@@ -1214,8 +1236,7 @@ static int vfio_iommu_replay(struct vfio_iommu *iommu,
 
 				npage = vfio_pin_pages_remote(dma, vaddr,
 							      n >> PAGE_SHIFT,
-							      &pfn, lock_cap,
-							      limit);
+							      &pfn, limit);
 				if (npage <= 0) {
 					WARN_ON(!npage);
 					ret = (int)npage;
@@ -1492,7 +1513,7 @@ static void vfio_iommu_unmap_unpin_reaccount(struct vfio_iommu *iommu)
 			if (!is_invalid_reserved_pfn(vpfn->pfn))
 				locked++;
 		}
-		vfio_lock_acct(dma->task, locked - unlocked, NULL);
+		vfio_lock_acct(dma, locked - unlocked, true);
 	}
 }