EDAC of cache on A53

[leimaohui]

I backport the patches of EDAC come from https://lkml.org/lkml/2018/3/14/1203  and https://lkml.org/lkml/2018/3/14/1204 to Nxp linux-4.14, and tested on ls1043a. kernel panic occurs. The following is part of log when kernel panic occurs. I wonder these patches are not applicable for 4.14. Is there any one have suggestions? 

    ---------------------------------------------

    [    2.982670] IR MCE Keyboard/mouse protocol handler initialized

    [    2.982672] IR XMP protocol handler initialized

    [    2.984352] imx2-wdt 2ad0000.wdog: timeout 60 sec (nowayout=0)

    [    2.987466] scsi_eh_0[1210]: PC Alignment exception: pc=0000000005f5e0ff sp=ffff000008003ef0

    [    2.987469] Internal error: Oops - SP/PC alignment exception: 8a000000 [#1] PREEMPT SMP

    [    2.987471] Modules linked in:

    [    2.987477] CPU: 0 PID: 1210 Comm: scsi_eh_0 Not tainted 4.14.16-dirty #3

    [    2.987478] Hardware name: LS1043A RDB Board (DT)

    [    2.987480] task: ffff8000749da2c0 task.stack: ffff00000b710000

    [    2.987483] PC is at 0x5f5e0ff

    [    2.987485] LR is at 0x5f5e0ff

    [    2.987487] pc : [<0000000005f5e0ff>] lr : [<0000000005f5e0ff>] pstate: 600001c5

    [    2.987488] sp : ffff000008003ef0

    [    2.987489] x29: ffff000008003f00 x28: ffff8000749da2c0

    [    2.987492] x27: ffff0000092d9000 x26: ffff000008004000

    [    2.987495] x25: ffff000008000000 x24: ffff000009452e38

    [    2.987498] x23: ffff00000801f000 x22: 0000000000000001

    [    2.987501] x21: 00000000000001c0 x20: 0000000000000000

    [    2.987504] x19: ffff000009520500 x18: 0000000000000006

    [    2.987506] x17: 0000000000000007 x16: 0000000000000001

    [    2.987509] x15: ffff00000947339d x14: 34783020736b7269

    [    2.987512] x13: 7571203030317830 x12: a5a5a5a5a5a5a5a5

    [    2.987515] x11: 5a5a5a5a5a5a5a5a x10: a5a5a5a5a5a5a5a5

    [    2.987518] x9 : 5a5a5a5a5a5a5a5a x8 : a5a5a5a5a5a5a5a5

    [    2.987521] x7 : 5a5a5a5a5a5a5a5a x6 : a5a5a5a5a5a5a5a5

    [    2.987523] x5 : 5a5a5a5a5a5a5a5a x4 : 0000000000000000

    [    2.987526] x3 : 0000000000000000 x2 : 0000000000000000

    [    2.987529] x1 : 0000000000000000 x0 : ffffffffffffffff

    [    2.987532] Process scsi_eh_0 (pid: 1210, stack limit = 0xffff00000b710000)

    [    2.987534] Call trace:

    [    2.987537] Exception stack(0xffff000008003db0 to 0xffff000008003ef0)

    [    2.987540] 3da0:                                   ffffffffffffffff 0000000000000000

    [    2.987542] 3dc0: 0000000000000000 0000000000000000 0000000000000000 5a5a5a5a5a5a5a5a

    [    2.987545] 3de0: a5a5a5a5a5a5a5a5 5a5a5a5a5a5a5a5a a5a5a5a5a5a5a5a5 5a5a5a5a5a5a5a5a

    [    2.987548] 3e00: a5a5a5a5a5a5a5a5 5a5a5a5a5a5a5a5a a5a5a5a5a5a5a5a5 7571203030317830

    [    2.987551] 3e20: 34783020736b7269 ffff00000947339d 0000000000000001 0000000000000007

    [    2.987553] 3e40: 0000000000000006 ffff000009520500 0000000000000000 00000000000001c0

    [    2.987556] 3e60: 0000000000000001 ffff00000801f000 ffff000009452e38 ffff000008000000

    [    2.987559] 3e80: ffff000008004000 ffff0000092d9000 ffff8000749da2c0 ffff000008003f00

    [    2.987562] 3ea0: 0000000005f5e0ff ffff000008003ef0 0000000005f5e0ff 00000000600001c5

    [    2.987565] 3ec0: ffff000008003ee0 ffff00000813e178 0001000000000000 ffff0000080cdfd8

    [    2.987567] 3ee0: ffff000008003f00 0000000005f5e0ff

    [    2.987569] [<0000000005f5e0ff>] 0x5f5e0ff

    [    2.987577] [<ffff00000814348c>] flush_smp_call_function_queue+0x9c/0x160

    [    2.987581] [<ffff0000081442b0>] generic_smp_call_function_single_interrupt+0x10/0x18

    [    2.987585] [<ffff000008090130>] handle_IPI+0x168/0x1d0

    [    2.987588] [<ffff000008081664>] gic_handle_irq+0x94/0xb0

    [    2.987590] Exception stack(0xffff00000b7135e0 to 0xffff00000b713720)

    [    2.987593] 35e0: 0000000000000020 ffff8000749da2c0 0000800072a8a000 ffffffffffffffff

    [    2.987596] 3600: 0000000000000000 0000000000000000 ffff0000094733e6 7361202c64657265

    [    2.987599] 3620: 622064656e676973 ffff000008656a58 00000000000c470f 2069636820643636

    [    2.987602] 3640: 0000000005f5e0ff 7571203030317830 34783020736b7269 ffff00000947339d

    [    2.987604] 3660: 0000000000000001 0000000000000007 0000000000000006 0000000000000000

    [    2.987607] 3680: ffff000009472000 ffff000009452e08 ffff000009474000 ffff000009472c08

    [    2.987610] 36a0: ffff000009452e38 0000000000000040 0000000000000000 ffff0000092d9000

    [    2.987613] 36c0: 0000000000000057 ffff00000b713720 ffff0000081165c0 ffff00000b713720

    [    2.987616] 36e0: ffff0000081165c4 0000000040000045 ffff000009315000 ffff000009452e38

    [    2.987619] 3700: ffffffffffffffff ffff0000094596c4 ffff00000b713720 ffff0000081165c4

    [    2.987621] [<ffff0000080830f0>] el1_irq+0xb0/0x124

    [    2.987626] [<ffff0000081165c4>] console_unlock+0x2b4/0x458

    [    2.987629] [<ffff0000081169c0>] vprintk_emit+0x258/0x2f8

    [    2.987632] [<ffff000008116be8>] vprintk_default+0x38/0x40

    [    2.987635] [<ffff0000081180c8>] vprintk_func+0xf8/0x198

    [    2.987637] [<ffff00000811781c>] printk+0x4c/0x54

    [    2.987642] [<ffff0000087004f0>] ata_link_printk+0x88/0x90

    [    2.987646] [<ffff000008705994>] ata_std_postreset+0xec/0xf0

    [    2.987649] [<ffff00000871c928>] ahci_postreset+0x30/0x80

    [    2.987654] [<ffff0000087102ac>] ata_eh_reset+0x534/0xc40

    [    2.987657] [<ffff000008710dbc>] ata_eh_recover+0x274/0x1360

    [    2.987660] [<ffff000008717bd8>] sata_pmp_error_handler+0x3c8/0x8b8

    [    2.987663] [<ffff00000871d810>] ahci_error_handler+0x20/0x70

    [    2.987666] [<ffff0000087123d0>] ata_scsi_port_error_handler+0x450/0x668

    [    2.987669] [<ffff000008712668>] ata_scsi_error+0x80/0xb8

    [    2.987674] [<ffff0000086d4a00>] scsi_error_handler+0xb0/0x340

    [    2.987678] [<ffff0000080e6a1c>] kthread+0xfc/0x128

    [    2.987682] [<ffff0000080853e4>] ret_from_fork+0x10/0x1c

    [    2.987687] Code: bad PC value

    [    2.987698] ---[ end trace cd30cde0f38c6e42 ]---

    [    2.987700] Kernel panic - not syncing: Fatal exception in interrupt

    [    2.987702] SMP: stopping secondary CPUs

    [    2.987705] Kernel Offset: disabled

    [    2.987707] CPU features: 0x002004

    [    2.987708] Memory Limit: none

    [    3.642790] ---[ end Kernel panic - not syncing: Fatal exception in interrupt

[leimaohui]

That's the base config for EDAC function, no matter which EDAC (memery, PCI2, cache)  and no matter for which processor, you have to enable this config. But that's not EDAC of cache for a53.

[Pavel]

Look at attached file. It is snapshot for EDAC driver enabling in NXP LSDK Linux BSP:

https://www.nxp.com/support/developer-resources/run-time-software/linux-software-and-development-too...

Use this BSP and driver on your board.

Have a great day,
Pavel Chubakov

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------