<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <meta name="Generator" content="Microsoft Exchange Server"> <style></style> </head> <body> <div> <div dir="auto"> <div dir="auto"> <div dir="auto">Hi guys, <div dir="auto"><br> </div> <div dir="auto">thanks for pointing this out Nirmoy.</div> <div dir="auto"><br> </div> <div dir="auto">Yeah, could be that I forgot to commit the patch. Currently I don't know at which end of the chaos I should start to clean up.</div> <div dir="auto"><br> </div> <div dir="auto">Christian.</div> </div> <div><br> <div class="x_elided-text">Am 25.03.2020 12:09 schrieb "Das, Nirmoy" <Nirmoy.Das@amd.com>:<br type="attribution"> <blockquote style="margin:0 0 0 0.8ex; border-left:1px #ccc solid; padding-left:1ex"> <div><font size="2"><span style="font-size:11pt"> <div>Hi Xinhui,<br> <br> <br> Can you please check if you can reproduce the crash with <br> <a href="https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html">https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html</a><br> <br> Christian fix it earlier, I think he forgot to push it.<br> <br> <br> Regards,<br> <br> Nirmoy<br> <br> On 3/25/20 12:07 PM, xinhui pan wrote:<br> > gpu recover will call sdma suspend/resume. In this period, ring will be<br> > disabled. So the vm_pte_scheds(sdma.instance[X].ring.sched)->ready will<br> > be false.<br> ><br> > If we submit any jobs in this ring-disabled period. We fail to pick up<br> > a rq for vm entity and entity->rq will set to NULL.<br> > amdgpu_vm_sdma_commit did not check the entity->rq, so fix it. Otherwise<br> > hit panic.<br> ><br> > Cc: Christian König <christian.koenig@amd.com><br> > Cc: Alex Deucher <alexander.deucher@amd.com><br> > Cc: Felix Kuehling <Felix.Kuehling@amd.com><br> > Signed-off-by: xinhui pan <xinhui.pan@amd.com><br> > ---<br> > drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 2 ++<br> > 1 file changed, 2 insertions(+)<br> ><br> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > index cf96c335b258..d30d103e48a2 100644<br> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > @@ -95,6 +95,8 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,<br> > int r;<br> > <br> > entity = p->direct ? &p->vm->direct : &p->vm->delayed;<br> > + if (!entity->rq)<br> > + return -ENOENT;<br> > ring = container_of(entity->rq->sched, struct amdgpu_ring, sched);<br> > <br> > WARN_ON(ib->length_dw == 0);<br> </div> </span></font></div> </blockquote> </div> <br> </div> </div> <div><br> <div class="x_elided-text">Am 25.03.2020 12:09 schrieb "Das, Nirmoy" <Nirmoy.Das@amd.com>:<br type="attribution"> <blockquote style="margin:0 0 0 0.8ex; border-left:1px #ccc solid; padding-left:1ex"> <div><font size="2"><span style="font-size:11pt"> <div>Hi Xinhui,<br> <br> <br> Can you please check if you can reproduce the crash with <br> <a href="https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html">https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html</a><br> <br> Christian fix it earlier, I think he forgot to push it.<br> <br> <br> Regards,<br> <br> Nirmoy<br> <br> On 3/25/20 12:07 PM, xinhui pan wrote:<br> > gpu recover will call sdma suspend/resume. In this period, ring will be<br> > disabled. So the vm_pte_scheds(sdma.instance[X].ring.sched)->ready will<br> > be false.<br> ><br> > If we submit any jobs in this ring-disabled period. We fail to pick up<br> > a rq for vm entity and entity->rq will set to NULL.<br> > amdgpu_vm_sdma_commit did not check the entity->rq, so fix it. Otherwise<br> > hit panic.<br> ><br> > Cc: Christian König <christian.koenig@amd.com><br> > Cc: Alex Deucher <alexander.deucher@amd.com><br> > Cc: Felix Kuehling <Felix.Kuehling@amd.com><br> > Signed-off-by: xinhui pan <xinhui.pan@amd.com><br> > ---<br> > drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 2 ++<br> > 1 file changed, 2 insertions(+)<br> ><br> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > index cf96c335b258..d30d103e48a2 100644<br> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > @@ -95,6 +95,8 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,<br> > int r;<br> > <br> > entity = p->direct ? &p->vm->direct : &p->vm->delayed;<br> > + if (!entity->rq)<br> > + return -ENOENT;<br> > ring = container_of(entity->rq->sched, struct amdgpu_ring, sched);<br> > <br> > WARN_ON(ib->length_dw == 0);<br> </div> </span></font></div> </blockquote> </div> <br> </div> </div> <div class="x_gmail_extra"><br> <div class="x_gmail_quote">Am 25.03.2020 12:09 schrieb "Das, Nirmoy" <Nirmoy.Das@amd.com>:<br type="attribution"> </div> </div> </div> <font size="2"><span style="font-size:11pt;"> <div class="PlainText">Hi Xinhui,<br> <br> <br> Can you please check if you can reproduce the crash with <br> <a href="https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html">https://lists.freedesktop.org/archives/amd-gfx/2020-February/046414.html</a><br> <br> Christian fix it earlier, I think he forgot to push it.<br> <br> <br> Regards,<br> <br> Nirmoy<br> <br> On 3/25/20 12:07 PM, xinhui pan wrote:<br> > gpu recover will call sdma suspend/resume. In this period, ring will be<br> > disabled. So the vm_pte_scheds(sdma.instance[X].ring.sched)->ready will<br> > be false.<br> ><br> > If we submit any jobs in this ring-disabled period. We fail to pick up<br> > a rq for vm entity and entity->rq will set to NULL.<br> > amdgpu_vm_sdma_commit did not check the entity->rq, so fix it. Otherwise<br> > hit panic.<br> ><br> > Cc: Christian König <christian.koenig@amd.com><br> > Cc: Alex Deucher <alexander.deucher@amd.com><br> > Cc: Felix Kuehling <Felix.Kuehling@amd.com><br> > Signed-off-by: xinhui pan <xinhui.pan@amd.com><br> > ---<br> > drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 2 ++<br> > 1 file changed, 2 insertions(+)<br> ><br> > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > index cf96c335b258..d30d103e48a2 100644<br> > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c<br> > @@ -95,6 +95,8 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,<br> > int r;<br> > <br> > entity = p->direct ? &p->vm->direct : &p->vm->delayed;<br> > + if (!entity->rq)<br> > + return -ENOENT;<br> > ring = container_of(entity->rq->sched, struct amdgpu_ring, sched);<br> > <br> > WARN_ON(ib->length_dw == 0);<br> </div> </span></font> </body> </html>