On Wed, 27 Sep 2017 13:01:51 +0200 Richard Genoud <richard.gen...@gmail.com> wrote:
> 2017-09-27 12:15 GMT+02:00 Richard Genoud <richard.gen...@gmail.com>: > > 2017-09-27 12:04 GMT+02:00 Boris Brezillon > > <boris.brezil...@free-electrons.com>: > >> On Wed, 27 Sep 2017 11:05:57 +0200 > >> Richard Genoud <richard.gen...@gmail.com> wrote: > >> > >>> Hi Boris, Nicolas ! > >>> > >>> Since commit f88fc122cc34 ("mtd: nand: Cleanup/rework the atmel_nand > >>> driver") > >>> strange things happen when nand-ecc-strength = <4>; (previously > >>> atmel,pmecc-cap). > >>> > >>> I first saw that a NULL pointer dereference happened when "udevadm > >>> trigger" was launched. > >>> With strace, I nailed it down to : > >>> > >>> sh-4.3# echo change > /sys/devices/virtual/bdi/mtd-1/uevent > >>> [ 86.696275] Unable to handle kernel NULL pointer dereference at > >>> virtual address 00000001 > >>> [ 86.704285] pgd = c717c000 > >>> [ 86.707072] [00000001] *pgd=c06d9a7000000000, *pte=00000000, > >>> *ppte=00000000 > >>> [ 86.713979] Internal error: Oops: 17 [#3] ARM > >>> [ 86.718306] CPU: 0 PID: 1 Comm: sh Tainted: G D W > >>> 4.11.0-rc1-00056-gf88fc122cc34-dirty #75 > >>> [ 86.727443] Hardware name: Atmel AT91SAM9 > >>> [ 86.731424] task: c7880b60 task.stack: c7884000 > >>> [ 86.735926] PC is at strlen+0x14/0x2c > >>> [ 86.739556] LR is at kobject_get_path+0x34/0xac > >>> [ 86.744046] pc : [<c023bc08>] lr : [<c0235020>] psr: 20000013 > >>> [ 86.744046] sp : c7885dc0 ip : c7885dd0 fp : c7885dcc > >>> [ 86.755439] r10: 00000002 r9 : 00000000 r8 : c7885f78 > >>> [ 86.760627] r7 : 014000c0 r6 : c7ab2308 r5 : 00000001 r4 : c7ab2308 > >>> [ 86.767106] r3 : 00000001 r2 : 00000001 r1 : 014000c0 r0 : 00000001 > >>> [ 86.773588] Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM > >>> Segment none > >>> [ 86.780672] Control: 0005317f Table: 2717c000 DAC: 00000051 > >>> [ 86.786372] Process sh (pid: 1, stack limit = 0xc7884190) > >>> [ 86.791730] Stack: (0xc7885dc0 to 0xc7886000) > >>> [ 86.796075] 5dc0: c7885df4 c7885dd0 c0235020 c023bc04 c0728bf8 > >>> c79f1000 c7ab2308 c78c2b00 > >>> [ 86.804195] 5de0: c04f4610 c7885f78 c7885e44 c7885df8 c0236244 > >>> c0234ffc c00a53b4 00000074 > >>> [ 86.812315] 5e00: 00107000 c7885ea8 c7885e64 c05d604b c717b420 > >>> c05b4aa8 0000081f 00000007 > >>> [ 86.820438] 5e20: c7ab2300 c7199ea0 c79baae0 c7885f78 c7199eb0 > >>> 00000007 c7885e54 c7885e48 > >>> [ 86.828559] 5e40: c0236640 c0236188 c7885e74 c7885e58 c02a5834 > >>> c023663c c7885e9c 00000002 > >>> [ 86.836681] 5e60: c7bf1f50 c79baae0 c7885e84 c7885e78 c02a37b8 > >>> c02a5800 c7885e9c c7885e88 > >>> [ 86.844801] 5e80: c0128fc8 c02a37a0 00000000 00000000 c7885ed4 > >>> c7885ea0 c01281e4 c0128f8c > >>> [ 86.852922] 5ea0: 00000000 00000000 c7880b60 c01280b8 00106cf8 > >>> c7215c20 c7885f78 00000007 > >>> [ 86.861045] 5ec0: c7884000 00106cf8 c7885f44 c7885ed8 c00caec0 > >>> c01280c8 0000081f 00107d00 > >>> [ 86.869167] 5ee0: c06d0f7c c7885fb0 00053177 00001180 00000178 > >>> c7885fac c7885f04 c00091e4 > >>> [ 86.877288] 5f00: c001128c c000e088 00000158 c00cb114 000012bc > >>> 00000000 bec504d0 b6e8bbec > >>> [ 86.885409] 5f20: c7215c20 c7215c20 00000000 00000007 00106cf8 > >>> c7885f78 c7885f74 c7885f48 > >>> [ 86.893531] 5f40: c00cb160 c00cae94 c00e6e04 c00e6568 00000000 > >>> 00000000 c7215c20 c7215c20 > >>> [ 86.901652] 5f60: 00000007 00106cf8 c7885fa4 c7885f78 c00cb2dc > >>> c00cb0b0 00000000 00000000 > >>> [ 86.909773] 5f80: 00000007 00106cf8 b6e8dd50 00000004 c000a544 > >>> 00000000 00000000 c7885fa8 > >>> [ 86.917895] 5fa0: c000a3a0 c00cb2a0 00000007 00106cf8 00000001 > >>> 00106cf8 00000007 00000000 > >>> [ 86.926015] 5fc0: 00000007 00106cf8 b6e8dd50 00000004 00000007 > >>> 00000004 00000000 000e9124 > >>> [ 86.934139] 5fe0: 00000000 bec50a3c b6db63d0 b6e107ac 60000010 > >>> 00000001 ffffffff ffffffff > >>> [ 86.942277] [<c023bc08>] (strlen) from [<c0235020>] > >>> (kobject_get_path+0x34/0xac) > >>> [ 86.949620] [<c0235020>] (kobject_get_path) from [<c0236244>] > >>> (kobject_uevent_env+0xcc/0x4b4) > >>> [ 86.958083] [<c0236244>] (kobject_uevent_env) from [<c0236640>] > >>> (kobject_uevent+0x14/0x18) > >>> [ 86.966287] [<c0236640>] (kobject_uevent) from [<c02a5834>] > >>> (uevent_store+0x44/0x64) > >>> [ 86.973987] [<c02a5834>] (uevent_store) from [<c02a37b8>] > >>> (dev_attr_store+0x28/0x34) > >>> [ 86.981672] [<c02a37b8>] (dev_attr_store) from [<c0128fc8>] > >>> (sysfs_kf_write+0x4c/0x58) > >>> [ 86.989525] [<c0128fc8>] (sysfs_kf_write) from [<c01281e4>] > >>> (kernfs_fop_write+0x12c/0x1c4) > >>> [ 86.997737] [<c01281e4>] (kernfs_fop_write) from [<c00caec0>] > >>> (__vfs_write+0x3c/0x11c) > >>> [ 87.005596] [<c00caec0>] (__vfs_write) from [<c00cb160>] > >>> (vfs_write+0xc0/0x164) > >>> [ 87.012855] [<c00cb160>] (vfs_write) from [<c00cb2dc>] > >>> (SyS_write+0x4c/0x8c) > >>> [ 87.019854] [<c00cb2dc>] (SyS_write) from [<c000a3a0>] > >>> (ret_fast_syscall+0x0/0x38) > >>> [ 87.027364] Code: e92dd800 e24cb004 e1a03000 e1a02003 (e5d21000) > >>> [ 87.033544] ---[ end trace 29af93c3c072b1f4 ]--- > >>> [ 87.039277] Kernel panic - not syncing: Attempted to kill init! > >>> exitcode=0x0000000b > >>> > >>> This is fun because it really doesn't seem to have anything to do with > >>> atmel-nand... > >>> > >>> I first found that on my custom board, built around an at91sam9g35-cm, > >>> but I managed to trigger it > >>> on an at91sam9g35-ek board, with a 4.13.3 kernel. > >>> > >>> NB: I couldn't trigger this with ecc-strength = 2 > >>> > >>> So, here is my configuration: > >>> - at91sam9g35-ek board with the image > >>> ftp://www.at91.com/pub/demo/linux4sam_5.6/linux4sam-poky-at91sam9x5ek-5.6.zip > >>> I flashed this image a first time as is, and then I flashed only the rfs > >>> with eccType 0xc0902405. > >>> > >>> - Kernel 4.13.3 with the quick'n dirty patch: > >>> --- a/arch/arm/boot/dts/at91sam9x5cm.dtsi > >>> +++ b/arch/arm/boot/dts/at91sam9x5cm.dtsi > >>> @@ -56,7 +56,7 @@ > >>> cs-gpios = <&pioD 4 > >>> GPIO_ACTIVE_HIGH>; > >>> nand-bus-width = <8>; > >>> nand-ecc-mode = "hw"; > >>> - nand-ecc-strength = <2>; > >>> + nand-ecc-strength = <4>; > >>> nand-ecc-step-size = <512>; > >>> nand-on-flash-bbt; > >>> label = "atmel_nand"; > >>> > >>> A minimal defconfig (attached) > >>> > >>> To trigger the kernel panic on a 4.13.3 kernel: > >>> > >>> At uboot: > >>> setenv bootargs $bootargs init=/bin/sh > >>> tftpboot 0x22000000 zImage ; tftpboot 0x21000000 at91sam9g35ek.dtb > >>> bootz 0x22000000 - 0x21000000 > >>> > >>> mount -tsysfs none /sys > >>> mount -tproc none /proc > >>> > >>> echo "change" > /sys/devices/platform/leds/leds/pd21/uevent > >>> [ 21.130000] Unable to handle kernel NULL pointer dereference at > >>> virtual address 00000001 > >>> [ 21.140000] pgd = c7170000 > >>> [ 21.140000] [00000001] *pgd=c06f780000000000, *pte=60000013, > >>> *ppte=60000013 > >>> [ 21.150000] Internal error: Oops: 17 [#1] ARM > >>> [ 21.150000] CPU: 0 PID: 1 Comm: sh Not tainted 4.13.3-dirty #77 > >>> [ 21.150000] Hardware name: Atmel AT91SAM9 > >>> [ 21.150000] task: c787c800 task.stack: c78c2000 > >>> [ 21.150000] PC is at strlen+0x14/0x2c > >>> [ 21.150000] LR is at kobject_get_path+0x34/0xac > >>> [ 21.150000] pc : [<c04d172c>] lr : [<c04c7444>] psr: 20000013 > >>> [ 21.150000] sp : c78c3d90 ip : c78c3da0 fp : c78c3d9c > >>> [ 21.150000] r10: 00000002 r9 : c78c3e28 r8 : c7bdd408 > >>> [ 21.150000] r7 : 014000c0 r6 : c7bdd408 r5 : 00000001 r4 : c7bdd408 > >>> [ 21.150000] r3 : 00000001 r2 : 00000001 r1 : 014000c0 r0 : 00000001 > >>> [ 21.150000] Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM > >>> Segment none > >>> [ 21.150000] Control: 0005317f Table: 27170000 DAC: 00000051 > >>> [ 21.150000] Process sh (pid: 1, stack limit = 0xc78c2190) > >>> [ 21.150000] Stack: (0xc78c3d90 to 0xc78c4000) > >>> [ 21.150000] 3d80: c78c3dc4 > >>> c78c3da0 c04c7444 c04d1728 > >>> [ 21.150000] 3da0: c0747de8 c7995000 c7bdd408 c787fac0 c05087c4 > >>> c7bdd408 c78c3e14 c78c3dc8 > >>> [ 21.150000] 3dc0: c04c8630 c04c7420 c00b0ad8 c78c3e4c 0007c000 > >>> c78c3e28 c78c3e4c c05ee954 > >>> [ 21.150000] 3de0: 27e1718f c05fbcf7 00000000 00000000 00000007 > >>> 00000006 00000002 c7bdd408 > >>> [ 21.150000] 3e00: c05ee954 c713d0c0 c78c3e5c c78c3e18 c04c8b08 > >>> c04c8574 c00c5d58 c00c5600 > >>> [ 21.150000] 3e20: c78c3e2c c0521e9c c060f2f7 00000000 c00b6168 > >>> 00000007 c7bdd400 c7160320 > >>> [ 21.150000] 3e40: c713d0c0 c78c3f78 c7160330 00000007 c78c3e74 > >>> c78c3e60 c02a102c c04c8a44 > >>> [ 21.150000] 3e60: c7034dc0 c713d0c0 c78c3e84 c78c3e78 c029efb8 > >>> c02a1010 c78c3e9c c78c3e88 > >>> [ 21.150000] 3e80: c012b194 c029efa0 00000000 00000000 c78c3ed4 > >>> c78c3ea0 c012a3c8 c012b158 > >>> [ 21.150000] 3ea0: 00000000 00000000 0000000f c012a27c 00100f40 > >>> c7b82ea0 c78c3f78 00000007 > >>> [ 21.150000] 3ec0: c78c2000 00100f40 c78c3f44 c78c3ed8 c00cca24 > >>> c012a28c 00001000 00000000 > >>> [ 21.150000] 3ee0: 00001000 00000000 00000000 00000000 00000015 > >>> 07d4a7e8 00000015 0727b420 > >>> [ 21.150000] 3f00: 00000015 0727b420 0000148c c00ccc8c be9854d0 > >>> b6f74bec c78c3fa4 c78c3f28 > >>> [ 21.150000] 3f20: c7b82ea0 c7b82ea0 00000000 00000007 00100f40 > >>> c78c3f78 c78c3f74 c78c3f48 > >>> [ 21.150000] 3f40: c00cccd8 c00cc9f8 c00e8e48 c00e85ac 00000000 > >>> 00000000 c7b82ea0 c7b82ea0 > >>> [ 21.150000] 3f60: 00000007 00100f40 c78c3fa4 c78c3f78 c00cce54 > >>> c00ccc28 00000000 00000000 > >>> [ 21.150000] 3f80: 00000007 00100f40 b6f76d50 00000004 c000a544 > >>> 00000000 00000000 c78c3fa8 > >>> [ 21.150000] 3fa0: c000a3a0 c00cce18 00000007 00100f40 00000001 > >>> 00100f40 00000007 00000000 > >>> [ 21.150000] 3fc0: 00000007 00100f40 b6f76d50 00000004 00000007 > >>> 00000004 00000000 000e9124 > >>> [ 21.150000] 3fe0: 00000000 be985a3c b6e9f3d0 b6ef97ac 60000010 > >>> 00000001 00000000 01010000 > >>> [ 21.150000] [<c04d172c>] (strlen) from [<c04c7444>] > >>> (kobject_get_path+0x34/0xac) > >>> [ 21.150000] [<c04c7444>] (kobject_get_path) from [<c04c8630>] > >>> (kobject_uevent_env+0xcc/0x4b8) > >>> [ 21.150000] [<c04c8630>] (kobject_uevent_env) from [<c04c8b08>] > >>> (kobject_synth_uevent+0xd4/0x33c) > >>> [ 21.150000] [<c04c8b08>] (kobject_synth_uevent) from [<c02a102c>] > >>> (uevent_store+0x2c/0x4c) > >>> [ 21.150000] [<c02a102c>] (uevent_store) from [<c029efb8>] > >>> (dev_attr_store+0x28/0x34) > >>> [ 21.150000] [<c029efb8>] (dev_attr_store) from [<c012b194>] > >>> (sysfs_kf_write+0x4c/0x58) > >>> [ 21.150000] [<c012b194>] (sysfs_kf_write) from [<c012a3c8>] > >>> (kernfs_fop_write+0x14c/0x1bc) > >>> [ 21.150000] [<c012a3c8>] (kernfs_fop_write) from [<c00cca24>] > >>> (__vfs_write+0x3c/0x130) > >>> [ 21.150000] [<c00cca24>] (__vfs_write) from [<c00cccd8>] > >>> (vfs_write+0xc0/0x164) > >>> [ 21.150000] [<c00cccd8>] (vfs_write) from [<c00cce54>] > >>> (SyS_write+0x4c/0x8c) > >>> [ 21.150000] [<c00cce54>] (SyS_write) from [<c000a3a0>] > >>> (ret_fast_syscall+0x0/0x38) > >>> [ 21.150000] Code: e92dd800 e24cb004 e1a03000 e1a02003 (e5d21000) > >>> [ 21.480000] ---[ end trace 3cc39b52c074a44c ]--- > >>> [ 21.490000] Kernel panic - not syncing: Attempted to kill init! > >>> exitcode=0x0000000b > >>> > >>> > >>> Or, we can just launch udev: > >>> udevd -d > >>> udevadm trigger > >>> (udevadm actually writes "changes" in uevent files). > >>> > >>> Now, a fun fact: If we add CONFIG_PM, there's no more kernel panic. > >>> (wtf?!) > >>> > >>> I also tried to boot with a nfs-root, echo changed > ".../uevent", no > >>> kernel panic. > >>> Then, ubiattach the mtd partition, echo changed > ".../uevent" => PANIC ! > >>> > >>> So, it really seems to be nand-related. > >>> > >>> Any idea ? > >>> > >> > >> Hm, it looks like a nasty buffer overflow. Can you try to double the > >> size here [1] (size *= 2) and see if that still happens? > > It seems that in [1], only the space for mu is allocated. > according to [2], dmu is (req->ecc.strength + 1)*sizeof(s32) octets more > and so is delta > So, IMHO, the allocation should be : > --- a/drivers/mtd/nand/atmel/pmecc.c > +++ b/drivers/mtd/nand/atmel/pmecc.c > @@ -363,7 +363,7 @@ atmel_pmecc_create_user(struct atmel_pmecc *pmecc, > size += (req->ecc.strength + 1) * sizeof(u16); > /* Reserve space for mu, dmu and delta. */ > size = ALIGN(size, sizeof(s32)); > - size += (req->ecc.strength + 1) * sizeof(s32); > + size += (req->ecc.strength + 1) * sizeof(s32) * 3; > > user = kzalloc(size, GFP_KERNEL); > if (!user) > > LGTM. Can you send a patch with the CC-stable+Fixes tags. I'll queue it for -rc3. Thanks a lot for reporting and fixing this bug. Boris