[PATCH v5 1/7] udmabuf: pre-fault when first page fault
Kasireddy, Vivek
vivek.kasireddy at intel.com
Fri Sep 6 08:12:08 UTC 2024
Hi Huan,
> Subject: [PATCH v5 1/7] udmabuf: pre-fault when first page fault
>
> The current udmabuf mmap uses a page fault to populate the vma.
>
> However, the current udmabuf has already obtained and pinned the folio
> upon completion of the creation.This means that the physical memory has
> already been acquired, rather than being accessed dynamically.
>
> As a result, the page fault has lost its purpose as a demanding
> page. Due to the fact that page fault requires trapping into kernel mode
> and filling in when accessing the corresponding virtual address in mmap,
> when creating a large size udmabuf, this represents a considerable
> overhead.
>
> This patch first the pfn into page table, and then pre-fault each pfn
> into vma, when first access. Should know, if anything wrong when
> pre-fault, will not report it's error, else, report when task access it
> at the first time.
>
> Suggested-by: Vivek Kasireddy <vivek.kasireddy at intel.com>
> Signed-off-by: Huan Yang <link at vivo.com>
> ---
> drivers/dma-buf/udmabuf.c | 35 +++++++++++++++++++++++++++++++++--
> 1 file changed, 33 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c
> index 047c3cd2ceff..0a8c231a36e1 100644
> --- a/drivers/dma-buf/udmabuf.c
> +++ b/drivers/dma-buf/udmabuf.c
> @@ -43,7 +43,8 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault
> *vmf)
> struct vm_area_struct *vma = vmf->vma;
> struct udmabuf *ubuf = vma->vm_private_data;
> pgoff_t pgoff = vmf->pgoff;
> - unsigned long pfn;
> + unsigned long addr, end, pfn;
> + vm_fault_t ret;
>
> if (pgoff >= ubuf->pagecount)
> return VM_FAULT_SIGBUS;
> @@ -51,7 +52,37 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault
> *vmf)
> pfn = folio_pfn(ubuf->folios[pgoff]);
> pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT;
>
> - return vmf_insert_pfn(vma, vmf->address, pfn);
> + ret = vmf_insert_pfn(vma, vmf->address, pfn);
> + if (ret & VM_FAULT_ERROR)
> + return ret;
> +
> + /* pre fault */
> + pgoff = vma->vm_pgoff;
> + end = vma->vm_end;
Nit: use vma->vm_end directly in the loop below, as end is used only once.
> + addr = vma->vm_start;
> +
> + for (; addr < end; pgoff++, addr += PAGE_SIZE) {
> + if (addr == vmf->address)
> + continue;
> +
> + if (WARN_ON(pgoff >= ubuf->pagecount))
> + break;
> +
> + pfn = folio_pfn(ubuf->folios[pgoff]);
> +
Nit: no need for a blank line here.
> + pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT;
> +
> + /**
> + * If something wrong, due to this vm fault success,
> + * do not report in here, report only when true access
> + * this addr.
> + * So, don't update ret here, just break.
Please rewrite the above comments as:
* If the below vmf_insert_pfn() fails, we do not return an error here
* during this pre-fault step. However, an error will be returned if the
* failure occurs when the addr is truly accessed.
With that,
Acked-by: Vivek Kasireddy <vivek.kasireddy at intel.com>
> + */
> + if (vmf_insert_pfn(vma, addr, pfn) & VM_FAULT_ERROR)
> + break;
> + }
> +
> + return ret;
> }
>
> static const struct vm_operations_struct udmabuf_vm_ops = {
> --
> 2.45.2
More information about the dri-devel
mailing list