@@ -95,11 +95,11 @@ struct vfio_dma {
size_t size; /* Map size (bytes) */
int prot; /* IOMMU_READ/WRITE */
bool iommu_mapped;
- bool lock_cap; /* capable(CAP_IPC_LOCK) */
bool vaddr_invalid;
struct task_struct *task;
struct rb_root pfn_list; /* Ex-user pinned pfn list */
unsigned long *bitmap;
+ struct vm_account vm_account;
};
struct vfio_batch {
@@ -412,31 +412,6 @@ static int vfio_iova_put_vfio_pfn(struct vfio_dma *dma, struct vfio_pfn *vpfn)
return ret;
}
-static int vfio_lock_acct(struct vfio_dma *dma, long npage, bool async)
-{
- struct mm_struct *mm;
- int ret;
-
- if (!npage)
- return 0;
-
- mm = async ? get_task_mm(dma->task) : dma->task->mm;
- if (!mm)
- return -ESRCH; /* process exited */
-
- ret = mmap_write_lock_killable(mm);
- if (!ret) {
- ret = __account_locked_vm(mm, abs(npage), npage > 0, dma->task,
- dma->lock_cap);
- mmap_write_unlock(mm);
- }
-
- if (async)
- mmput(mm);
-
- return ret;
-}
-
/*
* Some mappings aren't backed by a struct page, for example an mmap'd
* MMIO range for our own or another device. These use a different
@@ -715,16 +690,8 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
* externally pinned pages are already counted against
* the user.
*/
- if (!rsvd && !vfio_find_vpfn(dma, iova)) {
- if (!dma->lock_cap &&
- mm->locked_vm + lock_acct + 1 > limit) {
- pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n",
- __func__, limit << PAGE_SHIFT);
- ret = -ENOMEM;
- goto unpin_out;
- }
+ if (!rsvd && !vfio_find_vpfn(dma, iova))
lock_acct++;
- }
pinned++;
npage--;
@@ -744,7 +711,11 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
}
out:
- ret = vfio_lock_acct(dma, lock_acct, false);
+ if (vm_account_pinned(&dma->vm_account, lock_acct)) {
+ ret = -ENOMEM;
+ lock_acct = 0;
+ pr_warn("%s: RLIMIT_MEMLOCK exceeded\n", __func__);
+ }
unpin_out:
if (batch->size == 1 && !batch->offset) {
@@ -759,6 +730,7 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
put_pfn(pfn, dma->prot);
}
vfio_batch_unpin(batch, dma);
+ vm_unaccount_pinned(&dma->vm_account, lock_acct);
return ret;
}
@@ -782,7 +754,7 @@ static long vfio_unpin_pages_remote(struct vfio_dma *dma, dma_addr_t iova,
}
if (do_accounting)
- vfio_lock_acct(dma, locked - unlocked, true);
+ vm_unaccount_pinned(&dma->vm_account, locked - unlocked);
return unlocked;
}
@@ -805,7 +777,7 @@ static int vfio_pin_page_external(struct vfio_dma *dma, unsigned long vaddr,
ret = 0;
if (do_accounting && !is_invalid_reserved_pfn(*pfn_base)) {
- ret = vfio_lock_acct(dma, 1, true);
+ ret = vm_account_pinned(&dma->vm_account, 1);
if (ret) {
put_pfn(*pfn_base, dma->prot);
if (ret == -ENOMEM)
@@ -833,7 +805,7 @@ 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, -unlocked, true);
+ vm_unaccount_pinned(&dma->vm_account, unlocked);
return unlocked;
}
@@ -921,7 +893,7 @@ static int vfio_iommu_type1_pin_pages(void *iommu_data,
ret = vfio_add_to_pfn_list(dma, iova, phys_pfn);
if (ret) {
if (put_pfn(phys_pfn, dma->prot) && do_accounting)
- vfio_lock_acct(dma, -1, true);
+ vm_unaccount_pinned(&dma->vm_account, 1);
goto pin_unwind;
}
@@ -1162,7 +1134,7 @@ static long vfio_unmap_unpin(struct vfio_iommu *iommu, struct vfio_dma *dma,
}
if (do_accounting) {
- vfio_lock_acct(dma, -unlocked, true);
+ vm_unaccount_pinned(&dma->vm_account, unlocked);
return 0;
}
return unlocked;
@@ -1674,7 +1646,8 @@ 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);
+ vm_account_init(&dma->vm_account, dma->task, NULL, VM_ACCOUNT_TASK |
+ (capable(CAP_IPC_LOCK) ? VM_ACCOUNT_BYPASS : 0));
dma->pfn_list = RB_ROOT;
@@ -2398,7 +2371,7 @@ static void vfio_iommu_unmap_unpin_reaccount(struct vfio_iommu *iommu)
if (!is_invalid_reserved_pfn(vpfn->pfn))
locked++;
}
- vfio_lock_acct(dma, locked - unlocked, true);
+ vm_unaccount_pinned(&dma->vm_account, locked - unlocked);
}
}
This switches the charge to pinned_vm to be consistent with other drivers that pin pages with FOLL_LONGTERM. It also allows the use of the vm_account helper struct which makes a future change to implement cgroup accounting of pinned pages easier to implement as that requires a reference to the cgroup to be maintained. Signed-off-by: Alistair Popple <apopple@nvidia.com> Cc: Alex Williamson <alex.williamson@redhat.com> Cc: Cornelia Huck <cohuck@redhat.com> Cc: kvm@vger.kernel.org Cc: linux-kernel@vger.kernel.org --- drivers/vfio/vfio_iommu_type1.c | 59 +++++++++------------------------- 1 file changed, 16 insertions(+), 43 deletions(-)