<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Hi Nirmoy,<br>
    <br>
    can you give that patch set a testing round and an review?<br>
    <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">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      <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://aka.ms/o0ukef" 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"><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"><thomas_os@shipmail.org></a>; Yu, Lang
            <a class="moz-txt-link-rfc2396E" href="mailto:Lang.Yu@amd.com"><Lang.Yu@amd.com></a><br>
            <b>抄送:</b> <a class="moz-txt-link-abbreviated" href="mailto:dri-devel@lists.freedesktop.org">dri-devel@lists.freedesktop.org</a>
            <a class="moz-txt-link-rfc2396E" href="mailto:dri-devel@lists.freedesktop.org"><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"><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"><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"><christian.koenig@amd.com></a><br>
                >>> Signed-off-by: Lang Yu
                <a class="moz-txt-link-rfc2396E" href="mailto:Lang.Yu@amd.com"><Lang.Yu@amd.com></a><br>
                >>> Signed-off-by: Christian König
                <a class="moz-txt-link-rfc2396E" href="mailto:christian.koenig@amd.com"><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>
  </body>
</html>