Date: Thu, 19 May 2016 13:57:53 -0700 From: Alfred Perlstein <bright@mu.org> To: Konstantin Belousov <kostikbel@gmail.com> Cc: geom@freebsd.org, arch@freebsd.org Subject: Re: Removing Giant asserts from geom Message-ID: <53397f3f-1056-ceb7-ce3a-5269ac1d29e2@mu.org> In-Reply-To: <20160519191247.GQ89104@kib.kiev.ua> References: <20160519105634.GO89104@kib.kiev.ua> <573DEA73.4080408@mu.org> <20160519191247.GQ89104@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On 5/19/16 12:12 PM, Konstantin Belousov wrote: > On Thu, May 19, 2016 at 09:31:47AM -0700, Alfred Perlstein wrote: >> It seems like it should be the opposite, the DROP_GIANTs should be >> turned into mtx_assert(&Giant, MA_NOTOWNED) as giant is removed from the >> tree. >> >> Meaning Giant should be pushed further back until it is eliminated. >> Doing as this patch proposes hides that we still have callers holding >> Giant which is not good. > Did you read the third paragraph of my email ? OK, and why is thread0 needing Giant for so long? > FWIW, the assumed model of the kernel locking which must be in somebody > mind when talking about 'pushing back Giant' is not true for last 5-6 > years for our kernel in general, and for the VFS in particular. OK, makes sense, still would prefer to have assertions that don't allow mistakes to creep in. FreeBSD's assertions on locking and VFS make it much easier to develop under. -Alfred
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?53397f3f-1056-ceb7-ce3a-5269ac1d29e2>