Thanks Roland and Mark. I had found the post on LWN (and a rgrep user_regset in the kernel gives much fun when so many architectures except arm appear). In terms of "good story", I meant to convince company people who already have hard times reaching time to market for their SoC (I guess I am one of these guys ;-) ). For us, acceptable/not too bad (or at least "existing") code may forbid to investigate "better" solution, as we have so many topics in bad shape. And we do not leverage all features.
But frankly speaking, I have seen that improving/maturing much in last years. This is for example why I have been able to spend time on systemtap and other tools/methodologies/frameworks. And the optimizations/debug techniques we could propose to dev teams thanks to this have justified afterwards the time spent. So as it makes sense, I will try to revive this topic. Regards fred Frederic Turgis OMAP Platform Business Unit - OMAP System Engineering - Platform Enablement Texas Instruments France SA, 821 Avenue Jack Kilby, 06270 Villeneuve Loubet. 036 420 040 R.C.S Antibes. Capital de EUR 753.920 -----Original Message----- From: Mark Wielaard [mailto:m...@redhat.com] Sent: Saturday, April 23, 2011 11:13 AM To: Turgis, Frederic Cc: Frank Ch. Eigler; utrace-devel@redhat.com; Roland McGrath Subject: Re: utrace support on ARM > I will see if I can interest people in TI and Linaro. > I will need a good story... ;-) This describes why you want user_regset: http://thread.gmane.org/gmane.linux.kernel/616284 This describes why you want tracehooks: http://thread.gmane.org/gmane.linux.kernel/707950 Both those patch sets have been accepted in mainline and most architectures have implemented the necessary arch specific support already. Read the whole threads for some feedback from other arch maintainers that might be useful when adopting it for ARM too. Cheers, Mark
