Date: Sun, 27 Feb 2005 08:40:16 +0800 From: David Xu <davidxu@freebsd.org> To: Olivier Houchard <cognet@freebsd.org> Cc: cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/arm/include machdep.h pmap.h sysarch.h src/sys/arm/arm cpufunc_asm.S genassym.c machdep.c pmap.c swtch.S src/sys/arm/xscale std.xscale src/sys/arm/xscale/i80321 std.i80321 Message-ID: <422116F0.7030604@freebsd.org> In-Reply-To: <200502261859.j1QIx1fL008419@repoman.freebsd.org> References: <200502261859.j1QIx1fL008419@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
I am thinking why you can finish the task in such short time. :=) Should sysarch(ARM_SET_TP,...) write the thread pointer to ARM_TP_ADDRESS now not just remembers it or did I miss something ? David Xu Olivier Houchard wrote: >cognet 2005-02-26 18:59:01 UTC > > FreeBSD src repository > > Modified files: > sys/arm/include sysarch.h pmap.h machdep.h > sys/arm/arm cpufunc_asm.S genassym.c machdep.c pmap.c > swtch.S > sys/arm/xscale/i80321 std.i80321 > Added files: > sys/arm/xscale std.xscale > Log: > Instead of using sysarch() to store-retrieve the tp, add a magic address, > ARM_TP_ADDRESS, where the tp will be stored. On CPUs that support it, a cache > line will be allocated and locked for this address, so that it will never go > to RAM. On CPUs that does not, a page is allocated for it (it will be a bit > slower, and is wrong for SMP, but should be fine for UP). > The tp is still stored in the mdthread struct, and at each context switch, > ARM_TP_ADDRESS gets updated. > > Suggested by: davidxu > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?422116F0.7030604>