From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Brown Subject: Re: RFC: android logger feedback request Date: Wed, 21 Dec 2011 16:59:56 -0800 Message-ID: <20111222005956.GA4862@codeaurora.org> References: <4EF264C3.6000104@am.sony.com> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <4EF264C3.6000104@am.sony.com> Sender: linux-embedded-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Tim Bird Cc: linux-embedded , linux kernel , Arnd Bergmann , john stultz , Greg KH , Brian Swetland On Wed, Dec 21, 2011 at 02:59:15PM -0800, Tim Bird wrote: > In Android, this system uses a fixed set of device nodes with > well-known names: /dev/log/main, /dev/log/events, /dev/log/radio > and /dev/log/system. These names seem very specific to Android's use case. Would we want the mechanism to be more general, or configurable, so that another embedded-type system would be able to have their own log types. But, the biggest question I have is to understand why this is a kernel driver. In essence, it is just shuttling data from various processes to something that eventually needs to read that data. In other words, it is doing a subset of what syslog does. If the concern is about a userspace program crashing, wouldn't the userspace tool that reads this data also be able to crash? The driver clearly has no kernel API, since it exports no symbols. I could see more of an argument for it if certain kernel things were able to write to the log, but then we are also duplicating other functionality. David -- Sent by an employee of the Qualcomm Innovation Center, Inc. The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.