linux-kernel - [REGRESSION] Biscted to 'memblock, nobootmem: add memblock_virt_alloc

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <35c4753f-bcb5-46df-b273-e5b39db163f0@CO9EHSMHS018.ehs.local>
Date:	Tue, 28 Jan 2014 16:44:35 -0800
From:	Sören Brinkmann <soren.brinkmann@...inx.com>
To:	Michal Simek <michal.simek@...inx.com>,
	Yinghai Lu <yinghai@...nel.org>,
	Russell King <linux@....linux.org.uk>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	Santosh Shilimkar <santosh.shilimkar@...com>,
	Dave Hansen <dave.hansen@...el.com>,
	Ingo Molnar <mingo@...e.hu>, "H. Peter Anvin" <hpa@...or.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>
CC:	Sören Brinkmann <soren.brinkmann@...inx.com>,
	<linux-arm-kernel@...ts.infradead.org>,
	<linux-kernel@...r.kernel.org>
Subject: [REGRESSION] Biscted to 'memblock, nobootmem: add
 memblock_virt_alloc_low()'

Hi all,

I just found that the current Linux tree
(HEAD@ d891ea23d5203e5c47439b2a174f86a00b356a6c ) fails to boot on
Zynq (ARM).
The system runs into a panic relatively early (full boot log attached):
[    0.000000] Unable to handle kernel paging request at virtual address ffffffc0
[    0.000000] pgd = c0004000
[    0.000000] [ffffffc0] *pgd=2e7f6821, *pte=00000000, *ppte=00000000
[    0.000000] Internal error: Oops: 817 [#1] PREEMPT SMP ARM
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 3.13.0-xilinx-08988-gd891ea23d520 #22
[    0.000000] task: c0385ab0 ti: c037c000 task.ti: c037c000
[    0.000000] PC is at __memzero+0x4c/0x80
[    0.000000] LR is at 0x0
[    0.000000] pc : [<c01d9ccc>]    lr : [<00000000>]    psr: 000001d3
[    0.000000] sp : c037df0c  ip : 00000000  fp : 00000000
[    0.000000] r10: ffffffff  r9 : ffffffff  r8 : 3fffffc0
[    0.000000] r7 : ffffffff  r6 : ffffffc0  r5 : 00000040  r4 : 0000001c
[    0.000000] r3 : 00000000  r2 : 00000000  r1 : ffffffdc  r0 : ffffffc0
[    0.000000] Flags: nzcv  IRQs off  FIQs off  Mode SVC_32  ISA ARM  Segment kernel
[    0.000000] Control: 18c5387d  Table: 0000404a  DAC: 00000015
[    0.000000] Process swapper (pid: 0, stack limit = 0xc037c240)
[    0.000000] Stack: (0xc037df0c to 0xc037e000)
[    0.000000] df00:                            c0365cfc ffffffff 00000000 ffffffff 0000001c
[    0.000000] df20: ffffffff 00000000 00000000 413fc090 c038704c c0365f08 ffffffff c0370368
[    0.000000] df40: ffff1000 0002e7f5 00001000 00000007 c038dff0 c0370368 c039ebf4 c037dfd4
[    0.000000] df60: c0370368 c039ebf4 c08ef12c c038704c c038dff0 c035aabc ffffffff 18c5387d
[    0.000000] df80: ffffffff 0000406a 413fc090 c03843c0 00000000 c025eb74 c02f23a9 c037dfb4
[    0.000000] dfa0: c0008200 c037c000 00000001 c039e840 ffffffff 0000406a c03843c0 c0357854
[    0.000000] dfc0: 00000000 00000000 00000000 00000000 00000000 c0372918 18c5387d c0384410
[    0.000000] dfe0: c0372914 c038717c 0000406a 413fc090 00000000 00008074 00000000 00000000
[    0.000000] Code: e3110020 18a0500c 18a0500c e3110010 (18a0500c) 
[    0.000000] ---[ end trace 15c15b4afa9eff8e ]---
[    0.000000] Kernel panic - not syncing: Attempted to kill the idle task!

I bisected this issue down to commit:
ad6492b80f60a2139fa9bf8fd79b182fe5e3647c is the first bad commit
commit ad6492b80f60a2139fa9bf8fd79b182fe5e3647c
Author: Yinghai Lu <yinghai@...nel.org>
Date:   Mon Jan 27 17:06:49 2014 -0800

    memblock, nobootmem: add memblock_virt_alloc_low()
    
    The new memblock_virt APIs are used to replaced old bootmem API.
    
    We need to allocate page below 4G for swiotlb.
    
    That should fix regression on Andrew's system that is using swiotlb.
    
    Signed-off-by: Yinghai Lu <yinghai@...nel.org>
    Cc: Russell King <linux@....linux.org.uk>
    Cc: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
    Acked-by: Santosh Shilimkar <santosh.shilimkar@...com>
    Cc: Dave Hansen <dave.hansen@...el.com>
    Cc: Ingo Molnar <mingo@...e.hu>
    Cc: "H. Peter Anvin" <hpa@...or.com>
    Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
    Signed-off-by: Linus Torvalds <torvalds@...ux-foundation.org>

:040000 040000 75086461d0b57fca2e767b9be2e660a3fc34fc8b 2365645efe4f6ad75fcfaab3e59da2d9cf0968a2 M	arch
:040000 040000 106df01421a1d64347ee050c854e2540064a060b fab0ad7e09604c1ce795abdee894896619822838 M	include
:040000 040000 f67991f1b98d03d9791883a5b714bb1d8e6ae596 f6067483b5d7a5688bd873e75a3b4e89df104234 M	lib

Reverting that commit on top of HEAD results in a booting system again.
I also attached my kernel config.

Is this somehow related to this discussion regarding the non-DMAabble
memory (though that case doesn't seem to panic):
https://lkml.org/lkml/2014/1/27/212 ?

	Thanks,
	Sören


View attachment "boot.log" of type "text/plain" (5433 bytes)

View attachment "config" of type "text/plain" (33118 bytes)