From: peterz@infradead.org (Peter Zijlstra) Date: Fri, 18 Dec 2015 12:29:02 +0100 Subject: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable In-Reply-To: <5673DD60.7080302@linaro.org> References: <1449667265-17525-1-git-send-email-pmladek@suse.com> <1449667265-17525-5-git-send-email-pmladek@suse.com> <20151211124159.GB3729@pathway.suse.cz> <20151211145725.b0e81bb4bb18fcd72ef5f557@linux-foundation.org> <20151211232113.GZ8644@n2100.arm.linux.org.uk> <5673DD60.7080302@linaro.org> Message-ID: <20151218112902.GO6344@twins.programming.kicks-ass.net> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Fri, Dec 18, 2015 at 10:18:08AM +0000, Daniel Thompson wrote: > I'm not entirely sure that this is an improvement. What I do these days is delete everything in vprintk_emit() and simply call early_printk(). Kill the useless kmsg buffer crap and locking, just pound bytes to the UART registers without anything in between. The other semi usable solution is redirecting to trace_printk() and recovering the trace buffers from your kdump. But I've found that typically kdump doesn't work anymore if you properly wedge the machine. So this is very much a second rate solution. But this globally locked buffer, calling out to console drivers that do locking and even scheduling, is an unreliable unfixable trainwreck that I've given up on.