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;
Date:   Sun, 07 Jun 2020 18:53:32 -0400
From:   "Alex Xu (Hello71)" <alex_y_xu@yahoo.ca>
Subject: Re: 82fef0ad811f "x86/mm: unencrypted non-blocking DMA allocations
 use coherent pools" was Re: next-0519 on thinkpad x60: sound related? window
 manager crash
To:     David Rientjes <rientjes@google.com>
Cc:     alsa-devel@alsa-project.org, bp@alien8.de, hch@infradead.org,
        hch@lst.de, hpa@zytor.com, linux-kernel@vger.kernel.org,
        mingo@redhat.com, perex@perex.cz, tglx@linutronix.de,
        tiwai@suse.com, x86@kernel.org, Pavel Machek <pavel@ucw.cz>
References: <20200520111136.GA3802@amd>
        <1591545088.74ii116nf2.none@localhost> <20200607163803.GA10303@duo.ucw.cz>
        <alpine.DEB.2.22.394.2006071209470.84952@chino.kir.corp.google.com>
In-Reply-To: <alpine.DEB.2.22.394.2006071209470.84952@chino.kir.corp.google.com>
MIME-Version: 1.0
Message-Id: <1591570155.2tn9o40h95.none@localhost>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Content-Length: 5386
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Excerpts from David Rientjes's message of June 7, 2020 3:41 pm:
> On Sun, 7 Jun 2020, Pavel Machek wrote:
>=20
>> > I have a similar issue, caused between aaa2faab4ed8 and b170290c2836.
>> >=20
>> > [   20.263098] BUG: unable to handle page fault for address: ffffb2b58=
2cc2000
>> > [   20.263104] #PF: supervisor write access in kernel mode
>> > [   20.263105] #PF: error_code(0x000b) - reserved bit violation
>> > [   20.263107] PGD 3fd03b067 P4D 3fd03b067 PUD 3fd03c067 PMD 3f8822067=
 PTE 8000273942ab2163
>> > [   20.263113] Oops: 000b [#1] PREEMPT SMP
>> > [   20.263117] CPU: 3 PID: 691 Comm: mpv Not tainted 5.7.0-11262-gb170=
290c2836 #1
>> > [   20.263119] Hardware name: To Be Filled By O.E.M. To Be Filled By O=
.E.M./B450 Pro4, BIOS P4.10 03/05/2020
>> > [   20.263125] RIP: 0010:__memset+0x24/0x30
>> > [   20.263128] Code: cc cc cc cc cc cc 0f 1f 44 00 00 49 89 f9 48 89 d=
1 83 e2 07 48 c1 e9 03 40 0f b6 f6 48 b8 01 01 01 01 01 01 01 01 48 0f af c=
6 <f3> 48 ab 89 d1 f3 aa 4c 89 c8 c3 90 49 89 f9 40 88 f0 48 89 d1 f3
>> > [   20.263131] RSP: 0018:ffffb2b583d07e10 EFLAGS: 00010216
>> > [   20.263133] RAX: 0000000000000000 RBX: ffff8b8000102c00 RCX: 000000=
0000004000
>> > [   20.263134] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffb2=
b582cc2000
>> > [   20.263136] RBP: ffff8b8000101000 R08: 0000000000000000 R09: ffffb2=
b582cc2000
>> > [   20.263137] R10: 0000000000005356 R11: ffff8b8000102c18 R12: 000000=
0000000000
>> > [   20.263139] R13: 0000000000000000 R14: ffff8b8039944200 R15: ffffff=
ff9794daa0
>> > [   20.263141] FS:  00007f41aa4b4200(0000) GS:ffff8b803ecc0000(0000) k=
nlGS:0000000000000000
>> > [   20.263143] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> > [   20.263144] CR2: ffffb2b582cc2000 CR3: 00000003b6731000 CR4: 000000=
00003406e0
>> > [   20.263146] Call Trace:
>> > [   20.263151]  ? snd_pcm_hw_params+0x3f3/0x47a
>> > [   20.263154]  ? snd_pcm_common_ioctl+0xf2/0xf73
>> > [   20.263158]  ? snd_pcm_ioctl+0x1e/0x29
>> > [   20.263161]  ? ksys_ioctl+0x77/0x91
>> > [   20.263163]  ? __x64_sys_ioctl+0x11/0x14
>> > [   20.263166]  ? do_syscall_64+0x3d/0xf5
>> > [   20.263170]  ? entry_SYSCALL_64_after_hwframe+0x44/0xa9
>> > [   20.263173] Modules linked in: uvcvideo videobuf2_vmalloc videobuf2=
_memops videobuf2_v4l2 videodev snd_usb_audio videobuf2_common snd_hwdep sn=
d_usbmidi_lib input_leds snd_rawmidi led_class
>> > [   20.263182] CR2: ffffb2b582cc2000
>> > [   20.263184] ---[ end trace c6b47a774b91f0a0 ]---
>> > [   20.263187] RIP: 0010:__memset+0x24/0x30
>> > [   20.263190] Code: cc cc cc cc cc cc 0f 1f 44 00 00 49 89 f9 48 89 d=
1 83 e2 07 48 c1 e9 03 40 0f b6 f6 48 b8 01 01 01 01 01 01 01 01 48 0f af c=
6 <f3> 48 ab 89 d1 f3 aa 4c 89 c8 c3 90 49 89 f9 40 88 f0 48 89 d1 f3
>> > [   20.263192] RSP: 0018:ffffb2b583d07e10 EFLAGS: 00010216
>> > [   20.263193] RAX: 0000000000000000 RBX: ffff8b8000102c00 RCX: 000000=
0000004000
>> > [   20.263195] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffb2=
b582cc2000
>> > [   20.263196] RBP: ffff8b8000101000 R08: 0000000000000000 R09: ffffb2=
b582cc2000
>> > [   20.263197] R10: 0000000000005356 R11: ffff8b8000102c18 R12: 000000=
0000000000
>> > [   20.263199] R13: 0000000000000000 R14: ffff8b8039944200 R15: ffffff=
ff9794daa0
>> > [   20.263201] FS:  00007f41aa4b4200(0000) GS:ffff8b803ecc0000(0000) k=
nlGS:0000000000000000
>> > [   20.263202] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> > [   20.263204] CR2: ffffb2b582cc2000 CR3: 00000003b6731000 CR4: 000000=
00003406e0
>> >=20
>> > I bisected this to 82fef0ad811f "x86/mm: unencrypted non-blocking DMA=20
>> > allocations use coherent pools". Reverting 1ee18de92927 resolves the=20
>> > issue.
>> >=20
>> > Looks like Thinkpad X60 doesn't have VT-d, but could still be DMA=20
>> > related.
>>=20
>> Note that newer -next releases seem to behave okay for me. The commit
>> pointed out by siection is really simple:
>>=20
>> AFAIK you could verify it is responsible by turning off
>> CONFIG_AMD_MEM_ENCRYPT on latest kernel...
>>=20
>> Best regards,
>> 								Pavel
>>=20
>> index 1d6104ea8af0..2bf2222819d3 100644
>> --- a/arch/x86/Kconfig
>> +++ b/arch/x86/Kconfig
>> @@ -1520,6 +1520,7 @@ config X86_CPA_STATISTICS
>>  config AMD_MEM_ENCRYPT
>>         bool "AMD Secure Memory Encryption (SME) support"
>>         depends on X86_64 && CPU_SUP_AMD
>> +       select DMA_COHERENT_POOL
>>         select DYNAMIC_PHYSICAL_MASK
>>         select ARCH_USE_MEMREMAP_PROT
>>         select ARCH_HAS_FORCE_DMA_UNENCRYPTED
>=20
> Thanks for the report!
>=20
> Besides CONFIG_AMD_MEM_ENCRYPT, do you have CONFIG_DMA_DIRECT_REMAP=20
> enabled?  If so, it may be caused by the virtual address passed to the=20
> set_memory_{decrypted,encrypted}() functions.
>=20
> And I assume you are enabling SME by using mem_encrypt=3Don on the kernel=
=20
> command line or CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT is enabled.
>=20
> We likely need an atomic pool for devices that support DMA to addresses i=
n=20
> sme_me_mask as well.  I can test this tomorrow, but wanted to get it out=20
> early to see if it helps?

This patch doesn't seem to help. I have the same problem (kernel page=20
fault, __memset, snd_pcm_hw_params...).

I don't have CONFIG_DMA_DIRECT_REMAP enabled, and AFAICT it doesn't seem=20
to be selectable currently on x86, unless there are some patches=20
floating around for that.