Commit 40ddacf2 authored by Leon Romanovsky's avatar Leon Romanovsky Committed by Jason Gunthorpe

RDMA/umem: Don't hold mmap_sem for too long

DMA mapping is time consuming operation and doesn't need to be performed
with mmap_sem semaphore is held.

The semaphore only needs to be held for accounting and get_user_pages
related activities.
Signed-off-by: default avatarHuy Nguyen <huyn@mellanox.com>
Signed-off-by: default avatarLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: default avatarJason Gunthorpe <jgg@mellanox.com>
parent 6869e000
...@@ -84,7 +84,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, ...@@ -84,7 +84,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
struct ib_umem *umem; struct ib_umem *umem;
struct page **page_list; struct page **page_list;
struct vm_area_struct **vma_list; struct vm_area_struct **vma_list;
unsigned long locked;
unsigned long lock_limit; unsigned long lock_limit;
unsigned long cur_base; unsigned long cur_base;
unsigned long npages; unsigned long npages;
...@@ -149,15 +148,16 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, ...@@ -149,15 +148,16 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
npages = ib_umem_num_pages(umem); npages = ib_umem_num_pages(umem);
down_write(&current->mm->mmap_sem);
locked = npages + current->mm->pinned_vm;
lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) { down_write(&current->mm->mmap_sem);
current->mm->pinned_vm += npages;
if ((current->mm->pinned_vm > lock_limit) && !capable(CAP_IPC_LOCK)) {
up_write(&current->mm->mmap_sem);
ret = -ENOMEM; ret = -ENOMEM;
goto out; goto out;
} }
up_write(&current->mm->mmap_sem);
cur_base = addr & PAGE_MASK; cur_base = addr & PAGE_MASK;
...@@ -176,14 +176,16 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, ...@@ -176,14 +176,16 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
need_release = 1; need_release = 1;
sg_list_start = umem->sg_head.sgl; sg_list_start = umem->sg_head.sgl;
down_read(&current->mm->mmap_sem);
while (npages) { while (npages) {
ret = get_user_pages_longterm(cur_base, ret = get_user_pages_longterm(cur_base,
min_t(unsigned long, npages, min_t(unsigned long, npages,
PAGE_SIZE / sizeof (struct page *)), PAGE_SIZE / sizeof (struct page *)),
gup_flags, page_list, vma_list); gup_flags, page_list, vma_list);
if (ret < 0) {
if (ret < 0) up_read(&current->mm->mmap_sem);
goto out; goto out;
}
umem->npages += ret; umem->npages += ret;
cur_base += ret * PAGE_SIZE; cur_base += ret * PAGE_SIZE;
...@@ -199,6 +201,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, ...@@ -199,6 +201,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
/* preparing for next loop */ /* preparing for next loop */
sg_list_start = sg; sg_list_start = sg;
} }
up_read(&current->mm->mmap_sem);
umem->nmap = ib_dma_map_sg_attrs(context->device, umem->nmap = ib_dma_map_sg_attrs(context->device,
umem->sg_head.sgl, umem->sg_head.sgl,
...@@ -215,13 +218,14 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, ...@@ -215,13 +218,14 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
out: out:
if (ret < 0) { if (ret < 0) {
down_write(&current->mm->mmap_sem);
current->mm->pinned_vm -= ib_umem_num_pages(umem);
up_write(&current->mm->mmap_sem);
if (need_release) if (need_release)
__ib_umem_release(context->device, umem, 0); __ib_umem_release(context->device, umem, 0);
kfree(umem); kfree(umem);
} else }
current->mm->pinned_vm = locked;
up_write(&current->mm->mmap_sem);
if (vma_list) if (vma_list)
free_page((unsigned long) vma_list); free_page((unsigned long) vma_list);
free_page((unsigned long) page_list); free_page((unsigned long) page_list);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment