<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <meta content="text/html; charset=us-ascii"> <style type="text/css" style="display:none">  </style> </head> <body dir="ltr"> <div dir="auto">That's irrelevant in this case. CPU mapping's are destroyed automatically when the BO is freed. <div dir="auto"><br> </div> <div dir="auto">Giving the CPU pointer to the free function is only for extra checking that we can't access it anymore.</div> <div dir="auto"><br> </div> <div dir="auto">Christian.</div> </div> <div class="gmail_extra"><br> <div class="gmail_quote">Am 28.08.2019 13:05 schrieb "Yin, Tianci (Rico)" <Tianci.Yin@amd.com>:<br type="attribution"> </div> </div> <div> <blockquote style="border-color:rgb(200,200,200); border-left:3px solid rgb(200,200,200); padding-left:1ex; margin-left:0.8ex; color:rgb(102,102,102)"> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <font size="2"><span style="font-size:11pt">Might be better to use a local variable for the CPU pointer instead of <br> </span></font></div> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <font size="2"><span style="font-size:11pt">duplicating the calling parameters.</span></font></div> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <br> </div> </blockquote> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> Local variable can't be used in cleaner.<br> </div> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <br> </div> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <br> </div> <div dir="ltr" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0); text-align:left"> <br> </div> <div id="appendonsend"></div> <hr tabindex="-1" style="display:inline-block; width:98%"> <div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>From:</b> Christian König <ckoenig.leichtzumerken@gmail.com><br> <b>Sent:</b> Wednesday, August 28, 2019 17:57<br> <b>To:</b> Yin, Tianci (Rico) <Tianci.Yin@amd.com>; amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org><br> <b>Cc:</b> Xu, Feifei <Feifei.Xu@amd.com>; Ma, Le <Le.Ma@amd.com>; Xiao, Jack <Jack.Xiao@amd.com>; Zhang, Hawking <Hawking.Zhang@amd.com><br> <b>Subject:</b> Re: [PATCH] drm/amdgpu/psp: keep TMR in visible vram region for SRIOV</font> <div> </div> </div> <div class="BodyFragment"><font size="2"><span style="font-size:11pt"> <div class="PlainText">Am 28.08.19 um 11:25 schrieb Tianci Yin:<br> > From: "Tianci.Yin" <tianci.yin@amd.com><br> ><br> > Fix compute ring test failure in sriov scenario.<br> ><br> > Signed-off-by: Tianci.Yin <tianci.yin@amd.com><br> > ---<br> > drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 14 ++++++++++----<br> > drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 1 +<br> > 2 files changed, 11 insertions(+), 4 deletions(-)<br> ><br> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c<br> > index 9f7cc5b..92c68c9 100644<br> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c<br> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c<br> > @@ -261,9 +261,15 @@ static int psp_tmr_init(struct psp_context *psp)<br> > }<br> > }<br> > <br> > - ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_SIZE,<br> > - AMDGPU_GEM_DOMAIN_VRAM,<br> > - &psp->tmr_bo, &psp->tmr_mc_addr, NULL);<br> > + psp->tmr_buf = NULL;<br> > + if (!amdgpu_sriov_vf(psp->adev))<br> > + ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_SIZE,<br> > + AMDGPU_GEM_DOMAIN_VRAM,<br> > + &psp->tmr_bo, &psp->tmr_mc_addr, NULL);<br> > + else<br> > + ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_SIZE,<br> > + AMDGPU_GEM_DOMAIN_VRAM,<br> > + &psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);<br> <br> Might be better to use a local variable for the CPU pointer instead of <br> duplicating the calling parameters.<br> <br> BTW: How did you solved the stolen_vga_memory problem I pointed out in <br> the original patch set?<br> <br> Christian.<br> <br> > <br> > return ret;<br> > }<br> > @@ -1216,7 +1222,7 @@ static int psp_hw_fini(void *handle)<br> > <br> > psp_ring_destroy(psp, PSP_RING_TYPE__KM);<br> > <br> > - amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, NULL);<br> > + amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);<br> > amdgpu_bo_free_kernel(&psp->fw_pri_bo,<br> > &psp->fw_pri_mc_addr, &psp->fw_pri_buf);<br> > amdgpu_bo_free_kernel(&psp->fence_buf_bo,<br> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h<br> > index bc0947f..b73d4aa 100644<br> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h<br> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h<br> > @@ -171,6 +171,7 @@ struct psp_context<br> > /* tmr buffer */<br> > struct amdgpu_bo *tmr_bo;<br> > uint64_t tmr_mc_addr;<br> > + void *tmr_buf;<br> > <br> > /* asd firmware and buffer */<br> > const struct firmware *asd_fw;<br> <br> </div> </span></font></div> </div> </body> </html>