Date: Mon, 12 Dec 2005 00:58:25 -0500 (EST) From: Daniel Eischen <deischen@freebsd.org> To: Jason Evans <jasone@canonware.com> Cc: Johan Bucht <bucht@acc.umu.se>, current@freebsd.org Subject: Re: New libc malloc patch Message-ID: <Pine.GSO.4.43.0512120055020.22334-100000@sea.ntplx.net> In-Reply-To: <9FAD2B4B-C167-42D7-A8E7-BE03F4C07543@canonware.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 11 Dec 2005, Jason Evans wrote: > On Dec 11, 2005, at 5:48 PM, Johan Bucht wrote: > > > * Locking primitive > > The biggest issue and as David Xu pointed out is probably the locking > > primitives. The SPINLOCK use has a limit in the threading library and > > makes is hard to have a lot of mutexes. I ended up using a wrapper > > around the umtx_lock function to get recursive mutexes and it would > > probably be better to extend the umtx functions to handle recursion. > > This would probably also be appreciated by other malloc > > implementations. > > Might be interesting to implement some of the ideas from the Linux > > futex > > implementation to help umtx. > > I have been contemplating creating a separate spinlock API that > doesn't require the threads library to track the spinlocks across > fork. This would (if I understand correctly) remove the current > static spinlock limitations. What about using pthread_atfork()? > As for supporting recursive spinlocks, I doubt that the overhead > would be acceptable in general. If I could get rid of the need for > the one recursive lock in malloc.c, I certainly would. =) Why do we need a recursive mutex? Can you not restructure the code so that it is not needed? -- DE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.4.43.0512120055020.22334-100000>