Date: Sun, 25 Feb 2001 19:33:41 +0600 (ALMT) From: Boris Popov <bp@butya.kz> To: John Baldwin <jhb@FreeBSD.org> Cc: Bruce Evans <bde@zeta.org.au>, cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org Subject: Re: cvs commit: src/sys/i386/conf NOTES src/sys/conf options Message-ID: <Pine.BSF.4.21.0102251928210.1428-100000@lion.butya.kz> In-Reply-To: <XFMail.010225020933.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 25 Feb 2001, John Baldwin wrote: > > Now removing INVARIANT_SUPPORT is back on my todo list :-]. It enables > > a whole 28 lines of code (the _mtx_assert() function). It seems to > > I think Eivind's point was that you may want to just have support for > invariants in your kernel and then build a kernel module that you are debugging > with invariants. For example, if I wanted to debug a device driver that used > mutexes, I might build it with INVARIANTS so that the mtx_assert()'s do > something useful. However, I may not wish to turn on INVARIANTS in my base > kernel for some reason, and I can get by by building a kernel with just > INVARIANT_SUPPORT compiled in.. And this works until structures are the same for both cases. The most notable breakage was a kernel compiled with INVARIANTS and a KLD compiled without them. Any attempt to work with zone allocator leads to mysterious panics. Now it seems to be fixed (thanks DES :) -- Boris Popov http://www.butya.kz/~bp/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0102251928210.1428-100000>