Re: [PATCH v3 0/2] Align mmap address for DAX pmd mappings

[Toshi Kani]

On 4/21/2016 8:22 PM, Matthew Wilcox wrote:
> On Thu, Apr 21, 2016 at 07:43:39PM -0400, Toshi Kani wrote:
> > On 4/21/2016 4:21 PM, Mike Kravetz wrote:
> > > Might want to keep the future possibility of PUD_SIZE THP in mind?
> > Yes, this is why the func name does not say 'pmd'. It can be extended to
> > support
> > PUD_SIZE in future.
> Sure ... but what does that look like?  I think it should look a little
> like this:

Yes, I had something similar in mind, too.  Do you want me to use this 
version without the call with PUD_SIZE?

> unsigned long __thp_get_unmapped_area(struct file *filp, unsigned long len,
>                          loff_t off, unsigned long flags, unsigned long size);
> {
>          unsigned long addr;
>          loff_t off_end = off + len;
>          loff_t off_align = round_up(off, size);
>          unsigned long len_size;
>
>          if ((off_end <= off_align) || ((off_end - off_align) < size))
>                  return NULL;
>
>          len_size = len + size;
>          if ((len_size < len) || (off + len_size) < off)
>                  return NULL;
>
>          addr = current->mm->get_unmapped_area(filp, NULL, len_size,
>                                                  off >> PAGE_SHIFT, flags);
>          if (IS_ERR_VALUE(addr))
>                  return NULL;
>   
>          addr += (off - addr) & (size - 1);
>          return addr;
> }
>
> unsigned long thp_get_unmapped_area(struct file *filp, unsigned long addr,
>                  unsigned long len, unsigned long pgoff, unsigned long flags)
> {
>          loff_t off = (loff_t)pgoff << PAGE_SHIFT;
>
>          if (addr)
>                  goto out;
>          if (IS_DAX(filp->f_mapping->host) && !IS_ENABLED(CONFIG_FS_DAX_PMD))
>                  goto out;
>          /* Kirill, please fill in the right condition here for THP pagecache */
>
>          addr = __thp_get_unmapped_area(filp, len, off, flags, PUD_SIZE);
>          if (addr)
>                  return addr;
>          addr = __thp_get_unmapped_area(filp, len, off, flags, PMD_SIZE);
>          if (addr)
>                  return addr;
>
>   out:
>          return current->mm->get_unmapped_area(filp, addr, len, pgoff, flags);
> }
>
> By the way, I added an extra check here, when we add len and size
> (PMD_SIZE in the original), we need to make sure that doesn't wrap.
> NB: I'm not even compiling these suggestions, just throwing them out
> here as ideas to be criticised.

Yes, I agree with the extra check.  Thanks for pointing this out.

> Also, len_size is a stupid name, but I can't think of a better one.

How about len_pad?

Thanks,
-Toshi