linux-kernel - RE: [RFC PATCH v2 05/18] sched: add task flag for preempt IRQ tracking

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

Message-ID: <063D6719AE5E284EB5DD2968C1650D6D5F4B45CB@AcuExch.aculab.com>
Date:	Wed, 4 May 2016 15:16:31 +0000
From:	David Laight <David.Laight@...LAB.COM>
To:	'Andy Lutomirski' <luto@...capital.net>,
	Josh Poimboeuf <jpoimboe@...hat.com>
CC:	"linux-s390@...r.kernel.org" <linux-s390@...r.kernel.org>,
	Jiri Kosina <jikos@...nel.org>, Jessica Yu <jeyu@...hat.com>,
	Vojtech Pavlik <vojtech@...e.com>,
	Petr Mladek <pmladek@...e.com>,
	Peter Zijlstra <peterz@...radead.org>, X86 ML <x86@...nel.org>,
	Heiko Carstens <heiko.carstens@...ibm.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...hat.com>,
	"live-patching@...r.kernel.org" <live-patching@...r.kernel.org>,
	Jiri Slaby <jslaby@...e.cz>, Miroslav Benes <mbenes@...e.cz>,
	"linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>,
	"Chris J Arges" <chris.j.arges@...onical.com>
Subject: RE: [RFC PATCH v2 05/18] sched: add task flag for preempt IRQ
 tracking

From: Andy Lutomirski
> Sent: 02 May 2016 19:13
...
> I hope your plans include rewriting the current stack unwinder
> completely.  The thing in print_context_stack is (a)
> hard-to-understand and hard-to-modify crap and (b) is called in a loop
> from another file using totally ridiculous conventions.

I've seen a 'stack unwinder' that parsed the instruction stream
forwards looking for 'return' instructions.
I fixed it to add a few extra instructions needs to sort out the
exit path from hardware interrupts.

It only had to understand instructions that modified %sp and %bp
and remember which branch instructions and branch targets it had
used in order to find the correct exit path from a function.

Worked reasonably well without any debug info or guaranteed frame
pointers.
It did have to fall back on scanning the stack if it was inside
an infinite loop. Even on x86 it is reasonably possible to check
for 'call' instructions in this case.

	David