Date: Tue, 31 May 2011 23:26:32 +0530 From: "Jayachandran C." <c.jayachandran@gmail.com> To: Adrian Chadd <adrian@freebsd.org> Cc: freebsd-mips@freebsd.org Subject: Re: hwpmc and mips? Message-ID: <BANLkTi=D0UQF=OX4j1TD%2B0YJuvSZ9PURjA@mail.gmail.com> In-Reply-To: <BANLkTinZnjaCQ=qEd7gP%2B9zF-ebB95Gv-A@mail.gmail.com> References: <BANLkTiktv1XoBNPqcTB7KwVh6CoK1Df5Tw@mail.gmail.com> <BANLkTinZnjaCQ=qEd7gP%2B9zF-ebB95Gv-A@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, May 31, 2011 at 3:38 PM, Adrian Chadd <adrian@freebsd.org> wrote: > On 31 May 2011 18:00, Adrian Chadd <adrian@freebsd.org> wrote: > >> I'm likely going to look at this sometime after July (and I'm sure >> it's just a case of hooking into the right IRQ handler) but if someone >> beat me to it *hint*, I'd very much appreciate it. > > To prove I'm not completely slack, I went spelunking through the > source to see how hwpmc hooks into things. > > There's no HWPMC_HOOKS used anywhere in sys/mips/* or sys/mips/*/*, > save config files. > > I thus have a feeling the hwpmc glue for sampling mode just isn't > there for any MIPS core at the present time. > > Would someone more versed in the ways of MIPS stack frame magic and > trap handling care to take a stab at it? We have some work in progress on hwpmc on XLR, there is a hang on SMP which we need to fix before merging it. The counters are very different from the mips24k, and unfortunately the counters generated the same interrupt as the count/compare timer. On mips, getting the stack trace is going to be difficult - I'm not sure if this can be done efficiently in the pmc interrupt handler at all. JC.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BANLkTi=D0UQF=OX4j1TD%2B0YJuvSZ9PURjA>