Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210318083236.43578-1-daniel@qtec.com> <MN2PR12MB37755DAAB03FC0065E98EF1183699@MN2PR12MB3775.namprd12.prod.outlook.com>
 <CADnq5_MMTBXt50sqJvYmXWn0uBALz2fNvpSrCfjSuMa99VLa=w@mail.gmail.com> <375f0915-83b3-c729-b95f-939d828d24d0@amd.com>
In-Reply-To: <375f0915-83b3-c729-b95f-939d828d24d0@amd.com>
From:   Daniel Gomez <daniel@qtec.com>
Date:   Mon, 22 Mar 2021 10:38:50 +0100
Message-ID: <CAH1Ww+TNMpk-LYYaM=SG9XGK4wcGY3+w8rJzihUw=4EpzVV9EQ@mail.gmail.com>
Subject: Re: [PATCH] drm/radeon/ttm: Fix memory leak userptr pages
To:     Felix Kuehling <felix.kuehling@amd.com>
Cc:     Alex Deucher <alexdeucher@gmail.com>,
        "Koenig, Christian" <Christian.Koenig@amd.com>,
        David Airlie <airlied@linux.ie>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "amd-gfx@lists.freedesktop.org" <amd-gfx@lists.freedesktop.org>,
        "dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org>,
        "Deucher, Alexander" <Alexander.Deucher@amd.com>,
        "dagmcr@gmail.com" <dagmcr@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, 19 Mar 2021 at 21:29, Felix Kuehling <felix.kuehling@amd.com> wrote=
:
>
> This caused a regression in kfdtest in a large-buffer stress test after
> memory allocation for user pages fails:

I'm sorry to hear that. BTW, I guess you meant amdgpu leak patch and
not this one.
Just some background for the mem leak patch if helps to understand this:
The leak was introduce here:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?=
id=3D0b988ca1c7c4c73983b4ea96ef7c2af2263c87eb
where the bound status was introduced for all drm drivers including
radeon and amdgpu. So this patch just reverts the logic to the
original code but keeping the bound status. In my case, the binding
code allocates the user pages memory and returns without bounding (at
amdgpu_gtt_mgr_has_gart_addr). So,
when the unbinding happens, the memory needs to be cleared to prevent the l=
eak.

>
> [17359.536303] amdgpu: init_user_pages: Failed to get user pages: -16
> [17359.543746] BUG: kernel NULL pointer dereference, address: 00000000000=
00000
> [17359.551494] #PF: supervisor read access in kernel mode
> [17359.557375] #PF: error_code(0x0000) - not-present page
> [17359.563247] PGD 0 P4D 0
> [17359.566514] Oops: 0000 [#1] SMP PTI
> [17359.570728] CPU: 8 PID: 5944 Comm: kfdtest Not tainted 5.11.0-kfd-fkue=
hlin #193
> [17359.578760] Hardware name: ASUS All Series/X99-E WS/USB 3.1, BIOS 3201=
 06/17/2016
> [17359.586971] RIP: 0010:amdgpu_ttm_backend_unbind+0x52/0x110 [amdgpu]
> [17359.594075] Code: 48 39 c6 74 1b 8b 53 0c 48 8d bd 80 a1 ff ff e8 24 6=
2 00 00 85 c0 0f 85 ab 00 00 00 c6 43 54 00 5b 5d c3 48 8b 46 10 8b 4e 50 <=
48> 8b 30 48 85 f6 74 ba 8b 50 0c 48 8b bf 80 a1 ff ff 83 e1 01 45
> [17359.614340] RSP: 0018:ffffa4764971fc98 EFLAGS: 00010206
> [17359.620315] RAX: 0000000000000000 RBX: ffff950e8d4edf00 RCX: 000000000=
0000000
> [17359.628204] RDX: 0000000000000000 RSI: ffff950e8d4edf00 RDI: ffff950ea=
dec5e80
> [17359.636084] RBP: ffff950eadec5e80 R08: 0000000000000000 R09: 000000000=
0000000
> [17359.643958] R10: 0000000000000246 R11: 0000000000000001 R12: ffff950c0=
3377800
> [17359.651833] R13: ffff950eadec5e80 R14: ffff950c03377858 R15: 000000000=
0000000
> [17359.659701] FS:  00007febb20cb740(0000) GS:ffff950ebfc00000(0000) knlG=
S:0000000000000000
> [17359.668528] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [17359.675012] CR2: 0000000000000000 CR3: 00000006d700e005 CR4: 000000000=
01706e0
> [17359.682883] Call Trace:
> [17359.686063]  amdgpu_ttm_backend_destroy+0x12/0x70 [amdgpu]
> [17359.692349]  ttm_bo_cleanup_memtype_use+0x37/0x60 [ttm]
> [17359.698307]  ttm_bo_release+0x278/0x5e0 [ttm]
> [17359.703385]  amdgpu_bo_unref+0x1a/0x30 [amdgpu]
> [17359.708701]  amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu+0x7e5/0x910 [amdg=
pu]
> [17359.716307]  kfd_ioctl_alloc_memory_of_gpu+0x11a/0x220 [amdgpu]
> [17359.723036]  kfd_ioctl+0x223/0x400 [amdgpu]
> [17359.728017]  ? kfd_dev_is_large_bar+0x90/0x90 [amdgpu]
> [17359.734152]  __x64_sys_ioctl+0x8b/0xd0
> [17359.738796]  do_syscall_64+0x2d/0x40
> [17359.743259]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> [17359.749205] RIP: 0033:0x7febb083b6d7
> [17359.753681] Code: b3 66 90 48 8b 05 b1 47 2d 00 64 c7 00 26 00 00 00 4=
8 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <=
48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 47 2d 00 f7 d8 64 89 01 48
> [17359.774340] RSP: 002b:00007ffdb5522cd8 EFLAGS: 00000202 ORIG_RAX: 0000=
000000000010
> [17359.782668] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007febb=
083b6d7
> [17359.790566] RDX: 00007ffdb5522d60 RSI: 00000000c0284b16 RDI: 000000000=
0000003
> [17359.798459] RBP: 00007ffdb5522d10 R08: 00007ffdb5522dd0 R09: 00000000c=
4000004
> [17359.806352] R10: 0000000000000000 R11: 0000000000000202 R12: 000055941=
6e4e2aa
> [17359.814251] R13: 0000000000000000 R14: 0000000000000021 R15: 000000000=
0000000
> [17359.822140] Modules linked in: ip6table_filter ip6_tables iptable_filt=
er amdgpu x86_pkg_temp_thermal drm_ttm_helper ttm iommu_v2 gpu_sched ip_tab=
les x_tables
> [17359.837776] CR2: 0000000000000000
> [17359.841888] ---[ end trace a6f27d64475b28c8 ]---
> [17359.847318] RIP: 0010:amdgpu_ttm_backend_unbind+0x52/0x110 [amdgpu]
> [17359.854479] Code: 48 39 c6 74 1b 8b 53 0c 48 8d bd 80 a1 ff ff e8 24 6=
2 00 00 85 c0 0f 85 ab 00 00 00 c6 43 54 00 5b 5d c3 48 8b 46 10 8b 4e 50 <=
48> 8b 30 48 85 f6 74 ba 8b 50 0c 48 8b bf 80 a1 ff ff 83 e1 01 45
> [17359.874929] RSP: 0018:ffffa4764971fc98 EFLAGS: 00010206
> [17359.881014] RAX: 0000000000000000 RBX: ffff950e8d4edf00 RCX: 000000000=
0000000
> [17359.889007] RDX: 0000000000000000 RSI: ffff950e8d4edf00 RDI: ffff950ea=
dec5e80
> [17359.897008] RBP: ffff950eadec5e80 R08: 0000000000000000 R09: 000000000=
0000000
> [17359.905020] R10: 0000000000000246 R11: 0000000000000001 R12: ffff950c0=
3377800
> [17359.913034] R13: ffff950eadec5e80 R14: ffff950c03377858 R15: 000000000=
0000000
> [17359.921050] FS:  00007febb20cb740(0000) GS:ffff950ebfc00000(0000) knlG=
S:0000000000000000
> [17359.930047] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [17359.936674] CR2: 0000000000000000 CR3: 00000006d700e005 CR4: 000000000=
01706e0

From what I understand, the init_user_pages fails (returns EBUSY) and
the code goes to allocate_init_user_pages_failed where the unbind and
the userptr clear occurs.
Can we prevent this if we save the bounding status + userptr alloc? so
the function amdgpu_ttm_backend_unbind returns without trying to clear
the userptr memory?

Something like:

amdgpu_ttm_backend_bind:
    if (gtt->userptr) {
        r =3D amdgpu_ttm_tt_pin_userptr(bdev, ttm);
        if (r) ...
       gtt->sg_table =3D true;
   }

amdgpu_ttm_backend_unbind:
if (gtt->sg_table) {
        if (gtt->user_ptr) ...
}

If you agree, I'll send a v2 patch. Otherwise, maybe we could return
within amdgpu_ttm_tt_unpin_userptr if memory hasn't been allocated.

Any other ideas?

Regards,
Daniel

>
> Reverting this patch fixes the problem for me.
>
> Regards,
>    Felix
>
> On 2021-03-18 10:57 p.m., Alex Deucher wrote:
> > Applied.  Thanks!
> >
> > Alex
> >
> > On Thu, Mar 18, 2021 at 5:00 AM Koenig, Christian
> > <Christian.Koenig@amd.com> wrote:
> >> Reviewed-by: Christian K=C3=B6nig <christian.koenig@amd.com>
> >> ________________________________
> >> Von: Daniel Gomez <daniel@qtec.com>
> >> Gesendet: Donnerstag, 18. M=C3=A4rz 2021 09:32
> >> Cc: dagmcr@gmail.com <dagmcr@gmail.com>; Daniel Gomez <daniel@qtec.com=
>; Deucher, Alexander <Alexander.Deucher@amd.com>; Koenig, Christian <Chris=
tian.Koenig@amd.com>; David Airlie <airlied@linux.ie>; Daniel Vetter <danie=
l@ffwll.ch>; amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>;=
 dri-devel@lists.freedesktop.org <dri-devel@lists.freedesktop.org>; linux-k=
ernel@vger.kernel.org <linux-kernel@vger.kernel.org>
> >> Betreff: [PATCH] drm/radeon/ttm: Fix memory leak userptr pages
> >>
> >> If userptr pages have been pinned but not bounded,
> >> they remain uncleared.
> >>
> >> Signed-off-by: Daniel Gomez <daniel@qtec.com>
> >> ---
> >>   drivers/gpu/drm/radeon/radeon_ttm.c | 5 +++--
> >>   1 file changed, 3 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/rad=
eon/radeon_ttm.c
> >> index e8c66d10478f..bbcc6264d48f 100644
> >> --- a/drivers/gpu/drm/radeon/radeon_ttm.c
> >> +++ b/drivers/gpu/drm/radeon/radeon_ttm.c
> >> @@ -485,13 +485,14 @@ static void radeon_ttm_backend_unbind(struct ttm=
_bo_device *bdev, struct ttm_tt
> >>           struct radeon_ttm_tt *gtt =3D (void *)ttm;
> >>           struct radeon_device *rdev =3D radeon_get_rdev(bdev);
> >>
> >> +       if (gtt->userptr)
> >> +               radeon_ttm_tt_unpin_userptr(bdev, ttm);
> >> +
> >>           if (!gtt->bound)
> >>                   return;
> >>
> >>           radeon_gart_unbind(rdev, gtt->offset, ttm->num_pages);
> >>
> >> -       if (gtt->userptr)
> >> -               radeon_ttm_tt_unpin_userptr(bdev, ttm);
> >>           gtt->bound =3D false;
> >>   }
> >>
> >> --
> >> 2.30.2
> >>
> >> _______________________________________________
> >> dri-devel mailing list
> >> dri-devel@lists.freedesktop.org
> >> https://lists.freedesktop.org/mailman/listinfo/dri-devel
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel