Re: [PATCH 1/2] drm/radeon: fix invalid memory access in radeon_atrm_get_bios()

[Dave Airlie <airlied@gmail.com>]

On Tue, Jan 24, 2012 at 2:10 PM, Alex Deucher <alexdeucher@gmail.com> wrote:
> On Sun, Jan 22, 2012 at 9:43 AM, Igor Murzov
> <intergalactic.anonymous@gmail.com> wrote:
>> From 77c912ea1eca50a93a34d5be69f9dc96a8bef0d8 Mon Sep 17 00:00:00 2001
>> From: Igor Murzov <e-mail@date.by>
>> Date: Sun, 22 Jan 2012 19:02:27 +0400
>> Subject: [PATCH 1/2] drm/radeon: fix invalid memory access in radeon_atrm_get_bios()
>>
>> At a boot time I observed following bug:
>>
>>  BUG: unable to handle kernel paging request at ffff8800a4244000
>>  IP: [<ffffffff81275b5b>] memcpy+0xb/0x120
>>  PGD 1816063 PUD 1fe7d067 PMD 1ff9f067 PTE 80000000a4244160
>>  Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
>>  CPU 0
>>  Modules linked in: btusb bluetooth brcmsmac brcmutil crc8 cordic b43 radeon(+)
>>  mac80211 cfg80211 ttm ohci_hcd drm_kms_helper rfkill drm ssb agpgart mmc_core
>>  sp5100_tco video battery ac thermal processor rtc_cmos thermal_sys snd_hda_codec_hdmi
>>  joydev snd_hda_codec_conexant button bcma pcmcia snd_hda_intel snd_hda_codec
>>  snd_hwdep snd_pcm shpchp pcmcia_core k8temp snd_timer atl1c snd psmouse hwmon
>>  i2c_piix4 i2c_algo_bit soundcore evdev i2c_core ehci_hcd sg serio_raw snd_page_alloc
>>  loop btrfs
>>
>>  Pid: 1008, comm: modprobe Not tainted 3.3.0-rc1 #21 LENOVO 20046                           /AMD CRB
>>  RIP: 0010:[<ffffffff81275b5b>]  [<ffffffff81275b5b>] memcpy+0xb/0x120
>>  RSP: 0018:ffff8800aa72db00  EFLAGS: 00010246
>>  RAX: ffff8800a4150000 RBX: 0000000000001000 RCX: 0000000000000087
>>  RDX: 0000000000000000 RSI: ffff8800a4244000 RDI: ffff8800a4150bc8
>>  RBP: ffff8800aa72db78 R08: 0000000000000010 R09: ffffffff8174bbec
>>  R10: ffffffff812ee010 R11: 0000000000000001 R12: 0000000000001000
>>  R13: 0000000000010000 R14: ffff8800a4140000 R15: ffff8800aaba1800
>>  FS:  00007ff9a3bd4720(0000) GS:ffff8800afa00000(0000) knlGS:0000000000000000
>>  CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
>>  CR2: ffff8800a4244000 CR3: 00000000a9c18000 CR4: 00000000000006f0
>>  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>>  DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
>>  Process modprobe (pid: 1008, threadinfo ffff8800aa72c000, task ffff8800aa0e4000)
>>  Stack:
>>  ffffffffa04e7c7b 0000000000000001 0000000000010000 ffff8800aa72db28
>>  ffffffff00000001 0000000000001000 ffffffff8113cbef 0000000000000020
>>  ffff8800a4243420 ffff880000000002 ffff8800aa72db08 ffff8800a9d42000
>>  Call Trace:
>>  [<ffffffffa04e7c7b>] ? radeon_atrm_get_bios_chunk+0x8b/0xd0 [radeon]
>>  [<ffffffff8113cbef>] ? kmalloc_order_trace+0x3f/0xb0
>>  [<ffffffffa04a9298>] radeon_get_bios+0x68/0x2f0 [radeon]
>>  [<ffffffffa04c7a30>] rv770_init+0x40/0x280 [radeon]
>>  [<ffffffffa047d740>] radeon_device_init+0x560/0x600 [radeon]
>>  [<ffffffffa047ef4f>] radeon_driver_load_kms+0xaf/0x170 [radeon]
>>  [<ffffffffa043cdde>] drm_get_pci_dev+0x18e/0x2c0 [drm]
>>  [<ffffffffa04e7e95>] radeon_pci_probe+0xad/0xb5 [radeon]
>>  [<ffffffff81296c5f>] local_pci_probe+0x5f/0xd0
>>  [<ffffffff81297418>] pci_device_probe+0x88/0xb0
>>  [<ffffffff813417aa>] ? driver_sysfs_add+0x7a/0xb0
>>  [<ffffffff813418d8>] really_probe+0x68/0x180
>>  [<ffffffff81341be5>] driver_probe_device+0x45/0x70
>>  [<ffffffff81341cb3>] __driver_attach+0xa3/0xb0
>>  [<ffffffff81341c10>] ? driver_probe_device+0x70/0x70
>>  [<ffffffff813400ce>] bus_for_each_dev+0x5e/0x90
>>  [<ffffffff8134172e>] driver_attach+0x1e/0x20
>>  [<ffffffff81341298>] bus_add_driver+0xc8/0x280
>>  [<ffffffff813422c6>] driver_register+0x76/0x140
>>  [<ffffffff812976d6>] __pci_register_driver+0x66/0xe0
>>  [<ffffffffa043d021>] drm_pci_init+0x111/0x120 [drm]
>>  [<ffffffff8133c67a>] ? vga_switcheroo_register_handler+0x3a/0x60
>>  [<ffffffffa0229000>] ? 0xffffffffa0228fff
>>  [<ffffffffa02290ec>] radeon_init+0xec/0xee [radeon]
>>  [<ffffffff810002f2>] do_one_initcall+0x42/0x180
>>  [<ffffffff8109d8d2>] sys_init_module+0x92/0x1e0
>>  [<ffffffff815407a9>] system_call_fastpath+0x16/0x1b
>>  Code: 58 2a 43 50 88 43 4e 48 83 c4 08 5b c9 c3 66 90 e8 cb fd ff ff eb
>>  e6 90 90 90 90 90 90 90 90 90 48 89 f8 89 d1 c1 e9 03 83 e2 07 <f3> 48
>>  a5 89 d1 f3 a4 c3 20 48 83 ea 20 4c 8b 06 4c 8b 4e 08 4c
>>  RIP  [<ffffffff81275b5b>] memcpy+0xb/0x120
>>  RSP <ffff8800aa72db00>
>>  CR2: ffff8800a4244000
>>  ---[ end trace fcffa1599cf56382 ]---
>>
>> Call to acpi_evaluate_object() not always returns 4096 bytes chunks,
>> on my system it can return 2048 bytes chunk, so pass the length of
>> retrieved chunk to memcpy(), not the length of the recieving buffer.
>>
>> Signed-off-by: Igor Murzov <e-mail@date.by>

Hi Igor,

I'm not sure I understand, does your BIOS return 2K chunks always or
just for the last chunks?

since if it returns 2K always, won't your next patch break stuff?

we have a regression report against the second patch, just wondering
what it might be.

Dave.