Re: mapping user space buffer to kernel address space

[Linus Torvalds]

On Wed, 18 Oct 2000, Jeff Garzik wrote:
> 
> Well coolio.  Would somebody be up for sanity checking my audio mmap
> code (attached)?  It doesn't look too hard at all to get the audio
> drivers doing the correct thing.

Looks reasonable - although your "max_size" checks are wrong at mmap time.
You should check for something like

        if (size > max_size)
                goto out;
        if (offset > max_size - size)
                goto out;

instead of your current (offset >= max_size) check. As it stands, I think
you can mmap past the end by just having size and offset < max_size, but
the _sum_ being bigger..

> Since this code works in my local tests, my two concerns at this point
> are correct vma->vm_pgoff treatment, and correct vm_flags/vm_file usage.

You should drop the 

        vma->vm_file = file;

as the VM layer will have done that already.

Also, the VM_LOCKED | VM_SHM thing is wrong, because it will cause
mis-calulcation of the mm->vm_locked fields, and it's also unnecessary.
Sure, it will cause the VM layer to not swap, but the no-op "swapping"
shouldn't much hurt anyway, as it's easy enough to re-populate the page
again.

So either just drop it, or make sure that the locked page count doesn't
get corrupted.

                Linus

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/