On Wed, 14 Dec 2022 11:22:48 -0800
Steve Sistare <steven.sistare@xxxxxxxxxx> wrote:
> When a vfio container is preserved across exec using the VFIO_UPDATE_VADDR
> interfaces, locked_vm of the new mm becomes 0. If the user later unmaps a
> dma mapping, locked_vm underflows to a large unsigned value, and a
> subsequent dma map request fails with ENOMEM in __account_locked_vm.
>
> To avoid underflow, do not decrement locked_vm during unmap if the
> dma's mm has changed. To restore the correct locked_vm count, when
> VFIO_DMA_MAP_FLAG_VADDR is used and the dma's mm has changed, add
> the mapping's pinned page count to the new mm->locked_vm, subject
> to the rlimit. Now that mediated devices are excluded when using
> VFIO_UPDATE_VADDR, the amount of pinned memory equals the size of
> the mapping.
>
> Fixes: c3cbab24db38 ("vfio/type1: implement interfaces to update vaddr")
>
> Signed-off-by: Steve Sistare <steven.sistare@xxxxxxxxxx>
> ---
> drivers/vfio/vfio_iommu_type1.c | 49 +++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 49 insertions(+)
>
> diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
> index b04f485..e719c13 100644
> --- a/drivers/vfio/vfio_iommu_type1.c
> +++ b/drivers/vfio/vfio_iommu_type1.c
> @@ -100,6 +100,7 @@ struct vfio_dma {
> struct task_struct *task;
> struct rb_root pfn_list; /* Ex-user pinned pfn list */
> unsigned long *bitmap;
> + struct mm_struct *mm;
> };
>
> struct vfio_batch {
> @@ -424,6 +425,12 @@ static int vfio_lock_acct(struct vfio_dma *dma, long npage, bool async)
> if (!mm)
> return -ESRCH; /* process exited */
>
> + /* Avoid locked_vm underflow */
> + if (dma->mm != mm && npage < 0) {
> + mmput(mm);
> + return 0;
> + }
How about initialize ret = 0 and jump to the existing mmput() with a
goto, so there's no assumptions about whether we need the mmput() or
not.
> +
> ret = mmap_write_lock_killable(mm);
> if (!ret) {
> ret = __account_locked_vm(mm, abs(npage), npage > 0, dma->task,
> @@ -1180,6 +1187,7 @@ static void vfio_remove_dma(struct vfio_iommu *iommu, struct vfio_dma *dma)
> vfio_unmap_unpin(iommu, dma, true);
> vfio_unlink_dma(iommu, dma);
> put_task_struct(dma->task);
> + mmdrop(dma->mm);
> vfio_dma_bitmap_free(dma);
> if (dma->vaddr_invalid) {
> iommu->vaddr_invalid_count--;
> @@ -1578,6 +1586,42 @@ static bool vfio_iommu_iova_dma_valid(struct vfio_iommu *iommu,
> return list_empty(iova);
> }
>
> +static int vfio_change_dma_owner(struct vfio_dma *dma)
> +{
> + int ret = 0;
> + struct mm_struct *mm = get_task_mm(dma->task);
> +
> + if (dma->mm != mm) {
> + long npage = dma->size >> PAGE_SHIFT;
> + bool new_lock_cap = capable(CAP_IPC_LOCK);
> + struct task_struct *new_task = current->group_leader;
> +
> + ret = mmap_write_lock_killable(new_task->mm);
> + if (ret)
> + goto out;
> +
> + ret = __account_locked_vm(new_task->mm, npage, true,
> + new_task, new_lock_cap);
> + mmap_write_unlock(new_task->mm);
> + if (ret)
> + goto out;
> +
> + if (dma->task != new_task) {
> + vfio_lock_acct(dma, -npage, 0);
> + put_task_struct(dma->task);
> + dma->task = get_task_struct(new_task);
> + }
IIUC, we're essentially open coding vfio_lock_acct() in the previous
section so that we can be sure we've accounted the new task before we
credit the previous task. However, I was under the impression the task
remained the same, but the mm changes, which is how we end up with the
underflow described in the commit log. What circumstances cause a task
change? Thanks,
Alex
> + mmdrop(dma->mm);
> + dma->mm = new_task->mm;
> + mmgrab(dma->mm);
> + dma->lock_cap = new_lock_cap;
> + }
> +out:
> + if (mm)
> + mmput(mm);
> + return ret;
> +}
> +
> static int vfio_dma_do_map(struct vfio_iommu *iommu,
> struct vfio_iommu_type1_dma_map *map)
> {
> @@ -1627,6 +1671,9 @@ static int vfio_dma_do_map(struct vfio_iommu *iommu,
> dma->size != size) {
> ret = -EINVAL;
> } else {
> + ret = vfio_change_dma_owner(dma);
> + if (ret)
> + goto out_unlock;
> dma->vaddr = vaddr;
> dma->vaddr_invalid = false;
> iommu->vaddr_invalid_count--;
> @@ -1687,6 +1734,8 @@ static int vfio_dma_do_map(struct vfio_iommu *iommu,
> get_task_struct(current->group_leader);
> dma->task = current->group_leader;
> dma->lock_cap = capable(CAP_IPC_LOCK);
> + dma->mm = dma->task->mm;
> + mmgrab(dma->mm);
>
> dma->pfn_list = RB_ROOT;
>
