Re: [PATCH] VM: vm_ops->fault takes explicit struct file

[Stefan Richter <stefanr@s5r6.in-berlin.de>]

Erez Zadok wrote:
> For convenience of review and bisectability, I present the changes as one
> patch here.  If desired, I can split it for each subsystem.

Hmm, a split patch won't be useful; it won't compile.

Besides, this change doesn't look too intrusive to me.  For example, 
changes like this...

[...]
> --- a/drivers/ieee1394/dma.c
> +++ b/drivers/ieee1394/dma.c
> @@ -232,7 +232,7 @@ void dma_region_sync_for_device(struct dma_region *dma, unsigned long offset,
>  #ifdef CONFIG_MMU
>  
>  static int dma_region_pagefault(struct vm_area_struct *vma,
> -				struct vm_fault *vmf)
> +				struct vm_fault *vmf, struct file *file)
>  {
>  	struct dma_region *dma = (struct dma_region *)vma->vm_private_data;
>  

...are unlikely to cause objections or merge rejects.  ACK from me for 
the ieee1394 part, FWIW.

PS:  It's an obscure and unimportant fact, but you are not required to 
Cc krh on changes in drivers/ieee1394; he is interested in 
drivers/firewire which is separate code.
-- 
Stefan Richter
-=====-==--- -=-- ==-=-
http://arcgraph.de/sr/