lkml.org 
[lkml]   [2006]   [Apr]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [patch 1/5] mm: remap_vmalloc_range
    On Fri, Apr 21, 2006 at 12:29:38AM -0700, Andrew Morton wrote:
    >
    > When replacing calls to remap_pfn_rage() with calls to remap_valloc_range():
    >
    > - remap_pfn_range() sets VM_IO|VM_RESERVED|VM_PFNMAP on the user's vma.
    > remap_valloc_range() sets only VM_RESERVED.

    Yep, it doesn't use PFNMAPs (we can always user the underlying struct
    page), nor is it IO space. The only change that should be seen, as
    noted in patch 4/5, is that get_user_pages will work on all mappings
    now. I don't think there is a downside to this?

    >
    > - remap_pfn_range() has special handling for COWable user vma's, but
    > remap_valloc_range() does not.

    That's only for PFNMAPs. COW should continue to work fine.

    >
    > - are vma->vm_start and vma->vm_end always a multiple of PAGE_SIZE? (I
    > always forget). If not, remap_valloc_range() looks a tad buggy.

    I hope so.

    >
    >
    > pls explain.
    >
    >
    > - remap_valloc_range() can use ~PAGE_MASK, not PAGE_SIZE-1

    I initially did that when coding the function in mm/memory.c, but when
    adding all the vmalloc range checking I tried to stick with vmalloc
    convention.

    >
    > - remap_valloc_range() would lose a whole buncha typecasts if you use the
    > gcc pointer-arith-with-void* extension.

    Should I?
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2006-04-21 09:44    [W:4.060 / U:0.712 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site