linux-kernel - Re: [PATCH 3/7] Revert "perf: Fix warning while reading ring buffer headers"

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] [day] [month] [year] [list]

Message-ID: <20100515033933.GC8150@nowhere>
Date:	Sat, 15 May 2010 05:39:35 +0200
From:	Frederic Weisbecker <fweisbec@...il.com>
To:	Ian Munsie <imunsie@....ibm.com>
Cc:	linux-kernel@...r.kernel.org,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Paul Mackerras <paulus@...ba.org>, Ingo Molnar <mingo@...e.hu>,
	Arnaldo Carvalho de Melo <acme@...hat.com>,
	Steven Rostedt <rostedt@...dmis.org>,
	Tom Zanussi <tzanussi@...il.com>,
	Xiao Guangrong <xiaoguangrong@...fujitsu.com>
Subject: Re: [PATCH 3/7] Revert "perf: Fix warning while reading ring
	buffer headers"

On Thu, May 13, 2010 at 04:03:48PM +1000, Ian Munsie wrote:
> From: Ian Munsie <imunsie@....ibm.com>
> 
> This reverts commit d00a47cce569a3e660a8c9de5d57af28d6a9f0f7.
>     "perf: Fix warning while reading ring buffer headers"
> 
> The reverted patch removed the processing of the header_page, skipping
> over it instead on the assumption that perf was not using any of the
> data from that header. The patch neglected to remove the header_page_
> variables which were initialised in the removed code, nor did it fix any
> code that was using those variables.
> 
> In particular, long_size was set based on one of those variables
> (header_page_size_size) to learn the size of a long from the kernel,
> which is necessary to correctly print out some of the trace information
> in some circumstances. For instance, the size of a long in a 64 bit
> kernel would differ from the size of a long in perf if it was compiled
> for a 32 bit userspace. Perf trace needs to know the size of a long in
> the kernel so that it can print out the correct value without
> truncation.



I don't understand.
In the format file we have the size of the fields beside their type name,
so why do we need this?

Thanks.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/