Date: Fri, 26 Nov 2004 23:31:59 +0300 From: Gleb Smirnoff <glebius@freebsd.org> To: Scott Long <scottl@freebsd.org> Cc: cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/netgraph netgraph.h ng_base.c ng_source.c src/sys/netgraph/atm/sscop ng_sscop_cust.h src/sys/netgraph/atm/uni ng_uni_cust.h src/sys/netgraph/bluetooth/hci ng_hci_main.c ng_hci_misc.c ng_hci_var.h src/sys/netgraph/bluetooth/l2cap ... Message-ID: <20041126203159.GB87167@cell.sick.ru> In-Reply-To: <41A74A92.3090605@freebsd.org> References: <200411261029.iAQATJHG003436@repoman.freebsd.org> <41A74A92.3090605@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Nov 26, 2004 at 08:24:02AM -0700, Scott Long wrote: S> One of the stated goals for 5-STABLE was that we would retain API and S> ABI stability except when absolutely neccessary when fixing bugs or S> security holes. I know that we didn't explicitily name netgraph in S> the list of 'stable' subsystems, but it was assumed to be part of S> 'networking'. Can you please explain why this ABI change is required S> for 5-STABLE? We are going to test netgraph(4) for mpsafeness soon. Without this change it is impossible to turn netgraph ISR to MPSAFE. We are not going to enable MPSAFE netgraph ISR in RELENG_5, but we want to bring RELENG_5 to a state, when a small a 1-line patch enables MPSAFE ISR. In this case users running RELENG_5 will have an easy opportunity to test. I'm afraid that CURRENT userbase does not run netgraph widely, and thus we will not find resuorces for thourough testing. I understand all problems caused by API change, but we need this to go on. P.S. And only a small number of nodes affected. -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041126203159.GB87167>