<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body>
    <p>Hi Christian and Lang, <br>
    </p>
    <p><br>
    </p>
    <p>I can't apply it cleanly because of
      <a class="moz-txt-link-freetext" href="https://patchwork.freedesktop.org/patch/436305/">https://patchwork.freedesktop.org/patch/436305/</a>. This series needs
      a rebase.</p>
    <p><br>
    </p>
    <p>Regards,</p>
    <p>Nirmoy<br>
    </p>
    <div class="moz-cite-prefix">On 6/7/2021 4:42 PM, Das, Nirmoy wrote:<br>
    </div>
    <blockquote type="cite" cite="mid:73ffb605-3d7c-3493-38e3-a0dac7914ccd@amd.com">
      
      <p><br>
      </p>
      <div class="moz-cite-prefix">On 6/7/2021 3:01 PM, Christian König
        wrote:<br>
      </div>
      <blockquote type="cite" cite="mid:1a18dab0-45ee-3406-05f6-fe3ae3b2d136@gmail.com"> Hi
        Nirmoy,<br>
        <br>
        can you give that patch set a testing round and an review?<br>
      </blockquote>
      <p><br>
      </p>
      <p>Sure! I will get back asap.<br>
      </p>
      <blockquote type="cite" cite="mid:1a18dab0-45ee-3406-05f6-fe3ae3b2d136@gmail.com"> <br>
        The memory stress test with low VRAM on APUs should be
        sufficient and you should have that setup still around.<br>
        <br>
        Thanks in advance,<br>
        Christian.<br>
        <br>
        <div class="moz-cite-prefix">Am 07.06.21 um 14:58 schrieb Yu,
          Lang:<br>
        </div>
        <blockquote type="cite" cite="mid:DM6PR12MB4250A90CF8058DD28AB1B04FFB389@DM6PR12MB4250.namprd12.prod.outlook.com">
          <p style="font-family:Arial;font-size:10pt;color:#0000FF;margin:15pt;" align="Left"> [AMD Official Use Only]<br>
          </p>
          <br>
          <div>
            <div>
              <div>
                <div dir="ltr" data-ogsc="" style="">
                  <div dir="ltr" data-ogsc="" style="">Thanks,please let
                    Nirmoy do it.</div>
                  <div dir="ltr" data-ogsc="" style=""><br>
                  </div>
                  <div dir="ltr" data-ogsc="" style="">Regards,</div>
                  <div dir="ltr" data-ogsc="" style="">Lang</div>
                </div>
              </div>
              <div><br>
              </div>
              <div class="ms-outlook-ios-signature">获取 <a href="https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Faka.ms%2Fo0ukef&data=04%7C01%7Cnirmoy.das%40amd.com%7Cd789a01a337743638ed308d929c28fa0%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637586737878220366%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=TkIx1d6hvZtrP6iQihfUlrB6oBP9QZGfNa08anYm4X4%3D&reserved=0" originalsrc="https://aka.ms/o0ukef" shash="gszyxGAXNrgArUnrhQQeivk7AuAAyNE6bTPhVIW3B2JI8UTjQe4Hi8Lk6Kr99OWLAN4GK15/4cvTc8W9JXG5kNVhxnFhCIGbj5eG+r5yOSVEosc8ou1+9NuZVFTMkZbYaYCOwHVZpbmEz58h22Tv++5lAulLk1VBhk/dFqFyNlg=" moz-do-not-send="true">Outlook for iOS</a></div>
            </div>
            <hr style="display:inline-block;width:98%" tabindex="-1">
            <div id="divRplyFwdMsg" dir="ltr"><font style="font-size:11pt" face="Calibri, sans-serif" color="#000000"><b>发件人:</b> Christian König <a class="moz-txt-link-rfc2396E" href="mailto:ckoenig.leichtzumerken@gmail.com" moz-do-not-send="true"><ckoenig.leichtzumerken@gmail.com></a><br>
                <b>发送时间:</b> Monday, June 7, 2021 8:41:49 PM<br>
                <b>收件人:</b> Thomas Hellström (Intel) <a class="moz-txt-link-rfc2396E" href="mailto:thomas_os@shipmail.org" moz-do-not-send="true"><thomas_os@shipmail.org></a>;
                Yu, Lang <a class="moz-txt-link-rfc2396E" href="mailto:Lang.Yu@amd.com" moz-do-not-send="true"><Lang.Yu@amd.com></a><br>
                <b>抄送:</b> <a class="moz-txt-link-abbreviated" href="mailto:dri-devel@lists.freedesktop.org" moz-do-not-send="true">dri-devel@lists.freedesktop.org</a>
                <a class="moz-txt-link-rfc2396E" href="mailto:dri-devel@lists.freedesktop.org" moz-do-not-send="true"><dri-devel@lists.freedesktop.org></a><br>
                <b>主题:</b> Re: [PATCH 1/4] drm/ttm: add
                TTM_PL_FLAG_TEMPORARY flag v3</font>
              <div> </div>
            </div>
            <div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
                  <div class="PlainText">Thanks!<br>
                    <br>
                    Lang can anybody from your team give as an
                    reviewed-by/test-by on the <br>
                    amdgpu patches? I just want another pair of eyes
                    looking over it.<br>
                    <br>
                    If nobody else has time I will ask Nirmoy for a
                    quick testing round.<br>
                    <br>
                    Thanks,<br>
                    Christian.<br>
                    <br>
                    Am 07.06.21 um 14:38 schrieb Thomas Hellström
                    (Intel):<br>
                    > Sure. LGTM,<br>
                    ><br>
                    > Reviewed-by: Thomas Hellström <a class="moz-txt-link-rfc2396E" href="mailto:thomas.hellstrom@linux.intel.com" moz-do-not-send="true"><thomas.hellstrom@linux.intel.com></a><br>
                    ><br>
                    ><br>
                    > On 6/7/21 2:36 PM, Christian König wrote:<br>
                    >> Thomas any comments on this?<br>
                    >><br>
                    >> Is the purpose of this now clear enough?<br>
                    >><br>
                    >> Thanks,<br>
                    >> Christian.<br>
                    >><br>
                    >> Am 01.06.21 um 14:25 schrieb Christian
                    König:<br>
                    >>> From: Lang Yu <a class="moz-txt-link-rfc2396E" href="mailto:Lang.Yu@amd.com" moz-do-not-send="true"><Lang.Yu@amd.com></a><br>
                    >>><br>
                    >>> Sometimes drivers need to use bounce
                    buffers to evict BOs. While <br>
                    >>> those reside<br>
                    >>> in some domain they are not necessarily
                    suitable for CS.<br>
                    >>><br>
                    >>> Add a flag so that drivers can note
                    that a bounce buffers needs to be<br>
                    >>> reallocated during validation.<br>
                    >>><br>
                    >>> v2: add detailed comments<br>
                    >>> v3 (chk): merge commits and rework
                    commit message<br>
                    >>><br>
                    >>> Suggested-by: Christian König <a class="moz-txt-link-rfc2396E" href="mailto:christian.koenig@amd.com" moz-do-not-send="true"><christian.koenig@amd.com></a><br>
                    >>> Signed-off-by: Lang Yu <a class="moz-txt-link-rfc2396E" href="mailto:Lang.Yu@amd.com" moz-do-not-send="true"><Lang.Yu@amd.com></a><br>
                    >>> Signed-off-by: Christian König <a class="moz-txt-link-rfc2396E" href="mailto:christian.koenig@amd.com" moz-do-not-send="true"><christian.koenig@amd.com></a><br>
                    >>> ---<br>
                    >>>   drivers/gpu/drm/ttm/ttm_bo.c    | 3
                    +++<br>
                    >>>   include/drm/ttm/ttm_placement.h | 7
                    +++++--<br>
                    >>>   2 files changed, 8 insertions(+), 2
                    deletions(-)<br>
                    >>><br>
                    >>> diff --git
                    a/drivers/gpu/drm/ttm/ttm_bo.c <br>
                    >>> b/drivers/gpu/drm/ttm/ttm_bo.c<br>
                    >>> index 51a94fd63bd7..6b393502198e 100644<br>
                    >>> --- a/drivers/gpu/drm/ttm/ttm_bo.c<br>
                    >>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c<br>
                    >>> @@ -929,6 +929,9 @@ static bool
                    ttm_bo_places_compat(const struct <br>
                    >>> ttm_place *places,<br>
                    >>>   {<br>
                    >>>       unsigned i;<br>
                    >>>   +    if (mem->placement &
                    TTM_PL_FLAG_TEMPORARY)<br>
                    >>> +        return false;<br>
                    >>> +<br>
                    >>>       for (i = 0; i < num_placement;
                    i++) {<br>
                    >>>           const struct ttm_place *heap
                    = &places[i];<br>
                    >>>   diff --git
                    a/include/drm/ttm/ttm_placement.h <br>
                    >>> b/include/drm/ttm/ttm_placement.h<br>
                    >>> index aa6ba4d0cf78..8995c9e4ec1b 100644<br>
                    >>> --- a/include/drm/ttm/ttm_placement.h<br>
                    >>> +++ b/include/drm/ttm/ttm_placement.h<br>
                    >>> @@ -47,8 +47,11 @@<br>
                    >>>    * top of the memory area, instead of
                    the bottom.<br>
                    >>>    */<br>
                    >>>   -#define TTM_PL_FLAG_CONTIGUOUS  (1
                    << 19)<br>
                    >>> -#define TTM_PL_FLAG_TOPDOWN     (1
                    << 22)<br>
                    >>> +#define TTM_PL_FLAG_CONTIGUOUS  (1
                    << 0)<br>
                    >>> +#define TTM_PL_FLAG_TOPDOWN     (1
                    << 1)<br>
                    >>> +<br>
                    >>> +/* For multihop handling */<br>
                    >>> +#define TTM_PL_FLAG_TEMPORARY   (1
                    << 2)<br>
                    >>>     /**<br>
                    >>>    * struct ttm_place<br>
                    <br>
                  </div>
                </span></font></div>
          </div>
        </blockquote>
        <br>
      </blockquote>
    </blockquote>
  </body>
</html>