<html>
    <head>
      <base href="https://bugs.freedesktop.org/" />
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - page allocation failure: order:5, mode:0x240c0c0"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=93458#c6">Comment # 6</a>
              on <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - page allocation failure: order:5, mode:0x240c0c0"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=93458">bug 93458</a>
              from <span class="vcard"><a class="email" href="mailto:freedesktop@mattwhitlock.name" title="Matt Whitlock <freedesktop@mattwhitlock.name>"> <span class="fn">Matt Whitlock</span></a>
</span></b>
        <pre>(In reply to Matt Whitlock from <a href="show_bug.cgi?id=93458#c5">comment #5</a>)
<span class="quote">> I would venture a guess that the memory allocated for a GPU FIFO buffer
> needs to be physically continuous, as it will be used for DMA</span >

I take it back. Looking into the source code, the allocation is happening in
nvkm_ramht_new(…), which is allocating a struct nvkm_ramht containing a
variably-sized array of struct nvkm_ramht_data elements. As these elements, in
turn, contain pointers, it seems very unlikely that they'd be DMA'd, so
presumably the struct nvkm_ramht doesn't need to be physically contiguous and
could safely be allocated by vzalloc(…).</pre>
        </div>
      </p>
      <hr>
      <span>You are receiving this mail because:</span>
      
      <ul>
          <li>You are the assignee for the bug.</li>
      </ul>
    </body>
</html>