<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:DengXian;
panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:"\@DengXian";
panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
p.MsoPlainText, li.MsoPlainText, div.MsoPlainText
{mso-style-priority:99;
mso-style-link:"Plain Text Char";
margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
span.PlainTextChar
{mso-style-name:"Plain Text Char";
mso-style-priority:99;
mso-style-link:"Plain Text";
font-family:"Calibri",sans-serif;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="#0563C1" vlink="#954F72">
<div class="WordSection1">
<p class="MsoPlainText">Hi Christian,<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">Now vm_size was set to <span style="background:yellow;mso-highlight:yellow">
0x4 0000</span> GB by below commit:<o:p></o:p></p>
<p class="MsoPlainText">1bf621c42137926ac249af761c0190a9258aa0db drm/amdgpu: Remove unnecessary VM size calculations<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">So that max_pfn would be <span style="background:yellow;mso-highlight:yellow">
0x10 0000 0000.<o:p></o:p></span></p>
<p class="MsoPlainText">amdgpu_csa_vaddr would make max_pfn << 12 to get 0x1 0000 0000 0000, and then minus AMDGPU_VA_RESERVED_SIZE, to get
<span style="background:yellow;mso-highlight:yellow">0xFFFF FFF0 0000</span><o:p></o:p></p>
<p class="MsoPlainText">unfortunately this number was between AMDGPU_GMC_HOLE_START and AMDGPU_GMC_HOLE_END, so that amdgpu_gmc_sign_extend was called to make it
<span style="background:yellow;mso-highlight:yellow">0xFFFF FFFF FFF0 0000</span><o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">in amdgpu_driver_open_kms, extended csa_addr cannot be passed into amdgpu_map_static_csa directly, it would be above the limit of max_pfn.<o:p></o:p></p>
<p class="MsoPlainText">So that csa_addr was restricted by AMDGPU_GMC_HOLE_MASK to make it possible for amdgpu_vm_alloc_pts.<o:p></o:p></p>
<p class="MsoPlainText">But this restriction by AMDGPU_GMC_HOLE_MASK would make the address fall back into AMDGPU_GMC_HOLE again, which causing GPU reset.<o:p></o:p></p>
<p class="MsoPlainText">We just put amdgpu_csa_vaddr back to AMDGPU_GMC_HOLE_START, to avoid the address touching AMDGPU_GMC_HOLE.<o:p></o:p></p>
<p class="MsoPlainText">By the way, if max_pfn was shift much to the left, it would always get zero, with or without min(*,*).<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">BR,<o:p></o:p></p>
<p class="MsoPlainText">Wentao<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">-----Original Message-----<br>
From: Koenig, Christian <Christian.Koenig@amd.com> <br>
Sent: Tuesday, January 15, 2019 4:02 PM<br>
To: Liu, Monk <Monk.Liu@amd.com>; Lou, Wentao <Wentao.Lou@amd.com>; amd-gfx@lists.freedesktop.org; Zhu, Rex <Rex.Zhu@amd.com><br>
Subject: Re: [PATCH] drm/amdgpu: csa_vaddr should not larger than AMDGPU_GMC_HOLE_START</p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">Am 15.01.19 um 07:19 schrieb Liu, Monk:<o:p></o:p></p>
<p class="MsoPlainText">> The max_pfn is now 1'0000'0000'0000'0000 (bytes) which is above 48 bit now, and it with AMDGPU_GMC_HOLE_MASK make it to zero ....<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> And in code "amdgpu_driver_open_kms()" I saw @Zhu, Rex write the code as :<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> "csa_addr = amdgpu_csa_vadr(adev) & AMDGPU_GMC_HOLE_MASK", I think this is wrong since you intentionally place the csa above GMC hole, right ?<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">The fix is just completely incorrect since min(adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT, AMDGPU_GMC_HOLE_START) still gives you 0 when we shift max_pfn to much to the left.<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">The correct solution is to substract the reserved size first and then shift. E.g.:<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">addr = (max_pfn - (AMDGPU_VA_RESERVED_SIZE >> AMDGPU_PAGE_SHIFT)) << AMDGPU_PAGE_SHIFT;<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">Regards,<o:p></o:p></p>
<p class="MsoPlainText">Christian.<o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> Looks like we should modify this place<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> /Monk<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> -----Original Message-----<o:p></o:p></p>
<p class="MsoPlainText">> From: amd-gfx <<a href="mailto:amd-gfx-bounces@lists.freedesktop.org"><span style="text-decoration:none">amd-gfx-bounces@lists.freedesktop.org</span></a>> On Behalf Of
<o:p></o:p></p>
<p class="MsoPlainText">> Christian K?nig<o:p></o:p></p>
<p class="MsoPlainText">> Sent: Monday, January 14, 2019 9:05 PM<o:p></o:p></p>
<p class="MsoPlainText">> To: Lou, Wentao <<a href="mailto:Wentao.Lou@amd.com"><span style="text-decoration:none">Wentao.Lou@amd.com</span></a>>;
<a href="mailto:amd-gfx@lists.freedesktop.org"><span style="text-decoration:none">amd-gfx@lists.freedesktop.org</span></a><o:p></o:p></p>
<p class="MsoPlainText">> Subject: Re: [PATCH] drm/amdgpu: csa_vaddr should not larger than
<o:p></o:p></p>
<p class="MsoPlainText">> AMDGPU_GMC_HOLE_START<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> Am 14.01.19 um 09:40 schrieb wentalou:<o:p></o:p></p>
<p class="MsoPlainText">>> After removing unnecessary VM size calculations, vm_manager.max_pfn
<o:p></o:p></p>
<p class="MsoPlainText">>> would reach 0x10,0000,0000 max_pfn << AMDGPU_GPU_PAGE_SHIFT exceeding
<o:p></o:p></p>
<p class="MsoPlainText">>> AMDGPU_GMC_HOLE_START would caused GPU reset.<o:p></o:p></p>
<p class="MsoPlainText">>><o:p> </o:p></p>
<p class="MsoPlainText">>> Change-Id: I47ad0be2b0bd9fb7490c4e1d7bb7bdacf71132cb<o:p></o:p></p>
<p class="MsoPlainText">>> Signed-off-by: wentalou <<a href="mailto:Wentao.Lou@amd.com"><span style="text-decoration:none">Wentao.Lou@amd.com</span></a>><o:p></o:p></p>
<p class="MsoPlainText">> NAK, that is incorrect. We intentionally place the csa above the GMC hole.<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">> Regards,<o:p></o:p></p>
<p class="MsoPlainText">> Christian.<o:p></o:p></p>
<p class="MsoPlainText">><o:p> </o:p></p>
<p class="MsoPlainText">>> ---<o:p></o:p></p>
<p class="MsoPlainText">>> drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 3 ++-<o:p></o:p></p>
<p class="MsoPlainText">>> 1 file changed, 2 insertions(+), 1 deletion(-)<o:p></o:p></p>
<p class="MsoPlainText">>><o:p> </o:p></p>
<p class="MsoPlainText">>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c<o:p></o:p></p>
<p class="MsoPlainText">>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c<o:p></o:p></p>
<p class="MsoPlainText">>> index 7e22be7..dd3bd01 100644<o:p></o:p></p>
<p class="MsoPlainText">>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c<o:p></o:p></p>
<p class="MsoPlainText">>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c<o:p></o:p></p>
<p class="MsoPlainText">>> @@ -26,7 +26,8 @@<o:p></o:p></p>
<p class="MsoPlainText">>> <o:p></o:p></p>
<p class="MsoPlainText">>> uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev)<o:p></o:p></p>
<p class="MsoPlainText">>> {<o:p></o:p></p>
<p class="MsoPlainText">>> - uint64_t addr = adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT;<o:p></o:p></p>
<p class="MsoPlainText">>> + uint64_t addr = min(adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT,<o:p></o:p></p>
<p class="MsoPlainText">>> + AMDGPU_GMC_HOLE_START);<o:p></o:p></p>
<p class="MsoPlainText">>> <o:p></o:p></p>
<p class="MsoPlainText">>> addr -= AMDGPU_VA_RESERVED_SIZE;<o:p></o:p></p>
<p class="MsoPlainText">>> addr = amdgpu_gmc_sign_extend(addr);<o:p></o:p></p>
<p class="MsoPlainText">> _______________________________________________<o:p></o:p></p>
<p class="MsoPlainText">> amd-gfx mailing list<o:p></o:p></p>
<p class="MsoPlainText">> <a href="mailto:amd-gfx@lists.freedesktop.org"><span style="text-decoration:none">amd-gfx@lists.freedesktop.org</span></a><o:p></o:p></p>
<p class="MsoPlainText">> <a href="https://lists.freedesktop.org/mailman/listinfo/amd-gfx">
<span style="text-decoration:none">https://lists.freedesktop.org/mailman/listinfo/amd-gfx</span></a><o:p></o:p></p>
<p class="MsoPlainText"><o:p> </o:p></p>
</div>
</body>
</html>