diff for duplicates of <20110729234857.GA30941@fieldses.org> diff --git a/a/1.txt b/N1/1.txt index c382263..5fc79aa 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -1,29 +1,25 @@ On Sat, Jul 30, 2011 at 09:30:25AM +1000, NeilBrown wrote: -> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote= -: ->=20 -> >=20 -> >=20 +> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote: +> +> > +> > > > Sent from my iPhone -> >=20 -> > On 30/07/2011, at 2:53, "J. Bruce Fields" <bfields@fieldses.org> wr= -ote: -> >=20 +> > +> > On 30/07/2011, at 2:53, "J. Bruce Fields" <bfields@fieldses.org> wrote: +> > > > > On Fri, Jul 29, 2011 at 12:48:36PM -0400, bfields wrote: > > >> On Fri, Jul 29, 2011 at 11:30:05PM +1000, Greg Banks wrote: > > >>> > > >>> > > >>> Sent from my iPhone > > >>> -> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> = -=20 +> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> > > >>> wrote: > > >>> -> > >>>> Le vendredi 29 juillet 2011 =C3=A0 21:58 +1000, Greg Banks a =C3= -=A9crit : +> > >>>> Le vendredi 29 juillet 2011 à 21:58 +1000, Greg Banks a écrit : > > >>>> > > >>>>> -> > >>>>> Sure, and a whole lot of the callsites are ("..._%d", cpu), =20 +> > >>>>> Sure, and a whole lot of the callsites are ("..._%d", cpu), > > >>>>> hence the > > >>>>> unfortune :( > > >>>> @@ -37,8 +33,7 @@ ote: > > >>>> If SVC_POOL_PERNODE is selected, we could name them : > > >>>> nfsd_n0 -> nfsd_nN > > >>>> -> > >>>> That would help to check with "ps aux" which cpu/nodes are und= -er +> > >>>> That would help to check with "ps aux" which cpu/nodes are under > > >>>> stress. > > >>>> > > >>>> @@ -47,39 +42,31 @@ er > > >> > > >> Yup, patch welcomed.--b. > > > -> > > (Annoying fact: some initscripts stop nfsd using a rough equivale= -nt of -> > > "killall nfsd". So the name of the threads is arguably ABI. I t= -hink -> > > those initscripts are nuts and deserve what they get, but that ma= -y be +> > > (Annoying fact: some initscripts stop nfsd using a rough equivalent of +> > > "killall nfsd". So the name of the threads is arguably ABI. I think +> > > those initscripts are nuts and deserve what they get, but that may be > > > because I'm forgetting the reason they do that.) > > > -> >=20 -> > We had this discussion in May-June 2008; it's because the nfsds wer= -e =20 +> > +> > We had this discussion in May-June 2008; it's because the nfsds were > > once many many years ago userspace threads. ->=20 -> Even when they became kernel threads, 'kill' was the only way to kill= - them - +> +> Even when they became kernel threads, 'kill' was the only way to kill them - > at first. ->=20 -> >=20 -> > The "killall nfsd" semantics in those scripts are awful and lead to= - =20 -> > problems shutting down when there are lots of threads. It would =20 -> > probably be an improvement to provide a better shutdown mechanism a= -nd =20 +> +> > +> > The "killall nfsd" semantics in those scripts are awful and lead to +> > problems shutting down when there are lots of threads. It would +> > probably be an improvement to provide a better shutdown mechanism and > > force distros to use it. ->=20 +> > rpc.nfsd 0 ->=20 +> > will stop all nfsd threads. Follow with ->=20 +> > exportfs -f ->=20 -> and you should be done. I'm not 100% sure about the nfsv4 thread tho= -ugh - +> +> and you should be done. I'm not 100% sure about the nfsv4 thread though - > would need to check. Should be fine. diff --git a/a/content_digest b/N1/content_digest index e594e3c..155a819 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -8,7 +8,6 @@ "ref\020110729165345.GM23194@fieldses.org\0" "ref\0F562C6F4-3FC2-4952-A903-8E87E9EF2D9D@fastmail.fm\0" "ref\020110730093025.716f3f50@notabene.brown\0" - "ref\020110730093025.716f3f50-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org\0" "From\0J. Bruce Fields <bfields@fieldses.org>\0" "Subject\0Re: [PATCH] sunrpc: use better NUMA affinities\0" "Date\0Fri, 29 Jul 2011 19:48:57 -0400\0" @@ -23,31 +22,27 @@ "\00:1\0" "b\0" "On Sat, Jul 30, 2011 at 09:30:25AM +1000, NeilBrown wrote:\n" - "> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote=\n" - ":\n" - ">=20\n" - "> >=20\n" - "> >=20\n" + "> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote:\n" + "> \n" + "> > \n" + "> > \n" "> > Sent from my iPhone\n" - "> >=20\n" - "> > On 30/07/2011, at 2:53, \"J. Bruce Fields\" <bfields@fieldses.org> wr=\n" - "ote:\n" - "> >=20\n" + "> > \n" + "> > On 30/07/2011, at 2:53, \"J. Bruce Fields\" <bfields@fieldses.org> wrote:\n" + "> > \n" "> > > On Fri, Jul 29, 2011 at 12:48:36PM -0400, bfields wrote:\n" "> > >> On Fri, Jul 29, 2011 at 11:30:05PM +1000, Greg Banks wrote:\n" "> > >>>\n" "> > >>>\n" "> > >>> Sent from my iPhone\n" "> > >>>\n" - "> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> =\n" - "=20\n" + "> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> \n" "> > >>> wrote:\n" "> > >>>\n" - "> > >>>> Le vendredi 29 juillet 2011 =C3=A0 21:58 +1000, Greg Banks a =C3=\n" - "=A9crit :\n" + "> > >>>> Le vendredi 29 juillet 2011 \303\240 21:58 +1000, Greg Banks a \303\251crit :\n" "> > >>>>\n" "> > >>>>>\n" - "> > >>>>> Sure, and a whole lot of the callsites are (\"..._%d\", cpu), =20\n" + "> > >>>>> Sure, and a whole lot of the callsites are (\"..._%d\", cpu), \n" "> > >>>>> hence the\n" "> > >>>>> unfortune :(\n" "> > >>>>\n" @@ -61,8 +56,7 @@ "> > >>>> If SVC_POOL_PERNODE is selected, we could name them :\n" "> > >>>> nfsd_n0 -> nfsd_nN\n" "> > >>>>\n" - "> > >>>> That would help to check with \"ps aux\" which cpu/nodes are und=\n" - "er\n" + "> > >>>> That would help to check with \"ps aux\" which cpu/nodes are under\n" "> > >>>> stress.\n" "> > >>>>\n" "> > >>>>\n" @@ -71,43 +65,35 @@ "> > >>\n" "> > >> Yup, patch welcomed.--b.\n" "> > >\n" - "> > > (Annoying fact: some initscripts stop nfsd using a rough equivale=\n" - "nt of\n" - "> > > \"killall nfsd\". So the name of the threads is arguably ABI. I t=\n" - "hink\n" - "> > > those initscripts are nuts and deserve what they get, but that ma=\n" - "y be\n" + "> > > (Annoying fact: some initscripts stop nfsd using a rough equivalent of\n" + "> > > \"killall nfsd\". So the name of the threads is arguably ABI. I think\n" + "> > > those initscripts are nuts and deserve what they get, but that may be\n" "> > > because I'm forgetting the reason they do that.)\n" "> > >\n" - "> >=20\n" - "> > We had this discussion in May-June 2008; it's because the nfsds wer=\n" - "e =20\n" + "> > \n" + "> > We had this discussion in May-June 2008; it's because the nfsds were \n" "> > once many many years ago userspace threads.\n" - ">=20\n" - "> Even when they became kernel threads, 'kill' was the only way to kill=\n" - " them -\n" + "> \n" + "> Even when they became kernel threads, 'kill' was the only way to kill them -\n" "> at first.\n" - ">=20\n" - "> >=20\n" - "> > The \"killall nfsd\" semantics in those scripts are awful and lead to=\n" - " =20\n" - "> > problems shutting down when there are lots of threads. It would =20\n" - "> > probably be an improvement to provide a better shutdown mechanism a=\n" - "nd =20\n" + "> \n" + "> > \n" + "> > The \"killall nfsd\" semantics in those scripts are awful and lead to \n" + "> > problems shutting down when there are lots of threads. It would \n" + "> > probably be an improvement to provide a better shutdown mechanism and \n" "> > force distros to use it.\n" - ">=20\n" + "> \n" "> rpc.nfsd 0\n" - ">=20\n" + "> \n" "> will stop all nfsd threads. Follow with\n" - ">=20\n" + "> \n" "> exportfs -f\n" - ">=20\n" - "> and you should be done. I'm not 100% sure about the nfsv4 thread tho=\n" - "ugh -\n" + "> \n" + "> and you should be done. I'm not 100% sure about the nfsv4 thread though -\n" "> would need to check.\n" "\n" "Should be fine.\n" "\n" --b. -48c8925fa8023a25a8e575e9218d29a2702edf6fc5a08cb16e92927e9a1816d4 +9412124e0838173486990738657f0054becb7ff13cf0f7d481aa061482cd2f11
diff --git a/a/1.txt b/N2/1.txt index c382263..194feb9 100644 --- a/a/1.txt +++ b/N2/1.txt @@ -1,29 +1,25 @@ On Sat, Jul 30, 2011 at 09:30:25AM +1000, NeilBrown wrote: -> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote= -: ->=20 -> >=20 -> >=20 +> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb-97jfqw80gc6171pxa8y+qA@public.gmane.org> wrote: +> +> > +> > > > Sent from my iPhone -> >=20 -> > On 30/07/2011, at 2:53, "J. Bruce Fields" <bfields@fieldses.org> wr= -ote: -> >=20 +> > +> > On 30/07/2011, at 2:53, "J. Bruce Fields" <bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org> wrote: +> > > > > On Fri, Jul 29, 2011 at 12:48:36PM -0400, bfields wrote: > > >> On Fri, Jul 29, 2011 at 11:30:05PM +1000, Greg Banks wrote: > > >>> > > >>> > > >>> Sent from my iPhone > > >>> -> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> = -=20 +> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> > > >>> wrote: > > >>> -> > >>>> Le vendredi 29 juillet 2011 =C3=A0 21:58 +1000, Greg Banks a =C3= -=A9crit : +> > >>>> Le vendredi 29 juillet 2011 à 21:58 +1000, Greg Banks a écrit : > > >>>> > > >>>>> -> > >>>>> Sure, and a whole lot of the callsites are ("..._%d", cpu), =20 +> > >>>>> Sure, and a whole lot of the callsites are ("..._%d", cpu), > > >>>>> hence the > > >>>>> unfortune :( > > >>>> @@ -37,8 +33,7 @@ ote: > > >>>> If SVC_POOL_PERNODE is selected, we could name them : > > >>>> nfsd_n0 -> nfsd_nN > > >>>> -> > >>>> That would help to check with "ps aux" which cpu/nodes are und= -er +> > >>>> That would help to check with "ps aux" which cpu/nodes are under > > >>>> stress. > > >>>> > > >>>> @@ -47,41 +42,37 @@ er > > >> > > >> Yup, patch welcomed.--b. > > > -> > > (Annoying fact: some initscripts stop nfsd using a rough equivale= -nt of -> > > "killall nfsd". So the name of the threads is arguably ABI. I t= -hink -> > > those initscripts are nuts and deserve what they get, but that ma= -y be +> > > (Annoying fact: some initscripts stop nfsd using a rough equivalent of +> > > "killall nfsd". So the name of the threads is arguably ABI. I think +> > > those initscripts are nuts and deserve what they get, but that may be > > > because I'm forgetting the reason they do that.) > > > -> >=20 -> > We had this discussion in May-June 2008; it's because the nfsds wer= -e =20 +> > +> > We had this discussion in May-June 2008; it's because the nfsds were > > once many many years ago userspace threads. ->=20 -> Even when they became kernel threads, 'kill' was the only way to kill= - them - +> +> Even when they became kernel threads, 'kill' was the only way to kill them - > at first. ->=20 -> >=20 -> > The "killall nfsd" semantics in those scripts are awful and lead to= - =20 -> > problems shutting down when there are lots of threads. It would =20 -> > probably be an improvement to provide a better shutdown mechanism a= -nd =20 +> +> > +> > The "killall nfsd" semantics in those scripts are awful and lead to +> > problems shutting down when there are lots of threads. It would +> > probably be an improvement to provide a better shutdown mechanism and > > force distros to use it. ->=20 +> > rpc.nfsd 0 ->=20 +> > will stop all nfsd threads. Follow with ->=20 +> > exportfs -f ->=20 -> and you should be done. I'm not 100% sure about the nfsv4 thread tho= -ugh - +> +> and you should be done. I'm not 100% sure about the nfsv4 thread though - > would need to check. Should be fine. --b. +-- +To unsubscribe from this list: send the line "unsubscribe linux-nfs" in +the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org +More majordomo info at http://vger.kernel.org/majordomo-info.html diff --git a/a/content_digest b/N2/content_digest index e594e3c..0e25700 100644 --- a/a/content_digest +++ b/N2/content_digest @@ -9,45 +9,41 @@ "ref\0F562C6F4-3FC2-4952-A903-8E87E9EF2D9D@fastmail.fm\0" "ref\020110730093025.716f3f50@notabene.brown\0" "ref\020110730093025.716f3f50-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org\0" - "From\0J. Bruce Fields <bfields@fieldses.org>\0" + "From\0J. Bruce Fields <bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>\0" "Subject\0Re: [PATCH] sunrpc: use better NUMA affinities\0" "Date\0Fri, 29 Jul 2011 19:48:57 -0400\0" - "To\0NeilBrown <neilb@suse.de>\0" - "Cc\0Greg Banks <gnb@fastmail.fm>" - Eric Dumazet <eric.dumazet@gmail.com> - Christoph Hellwig <hch@infradead.org> - linux-nfs@vger.kernel.org <linux-nfs@vger.kernel.org> - David Miller <davem@davemloft.net> - linux-kernel <linux-kernel@vger.kernel.org> - " netdev <netdev@vger.kernel.org>\0" + "To\0NeilBrown <neilb-l3A5Bk7waGM@public.gmane.org>\0" + "Cc\0Greg Banks <gnb-97jfqw80gc6171pxa8y+qA@public.gmane.org>" + Eric Dumazet <eric.dumazet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> + Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> + linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> + David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org> + linux-kernel <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> + " netdev <netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>\0" "\00:1\0" "b\0" "On Sat, Jul 30, 2011 at 09:30:25AM +1000, NeilBrown wrote:\n" - "> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb@fastmail.fm> wrote=\n" - ":\n" - ">=20\n" - "> >=20\n" - "> >=20\n" + "> On Sat, 30 Jul 2011 06:34:44 +1000 Greg Banks <gnb-97jfqw80gc6171pxa8y+qA@public.gmane.org> wrote:\n" + "> \n" + "> > \n" + "> > \n" "> > Sent from my iPhone\n" - "> >=20\n" - "> > On 30/07/2011, at 2:53, \"J. Bruce Fields\" <bfields@fieldses.org> wr=\n" - "ote:\n" - "> >=20\n" + "> > \n" + "> > On 30/07/2011, at 2:53, \"J. Bruce Fields\" <bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org> wrote:\n" + "> > \n" "> > > On Fri, Jul 29, 2011 at 12:48:36PM -0400, bfields wrote:\n" "> > >> On Fri, Jul 29, 2011 at 11:30:05PM +1000, Greg Banks wrote:\n" "> > >>>\n" "> > >>>\n" "> > >>> Sent from my iPhone\n" "> > >>>\n" - "> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet@gmail.com> =\n" - "=20\n" + "> > >>> On 29/07/2011, at 22:11, Eric Dumazet <eric.dumazet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> \n" "> > >>> wrote:\n" "> > >>>\n" - "> > >>>> Le vendredi 29 juillet 2011 =C3=A0 21:58 +1000, Greg Banks a =C3=\n" - "=A9crit :\n" + "> > >>>> Le vendredi 29 juillet 2011 \303\240 21:58 +1000, Greg Banks a \303\251crit :\n" "> > >>>>\n" "> > >>>>>\n" - "> > >>>>> Sure, and a whole lot of the callsites are (\"..._%d\", cpu), =20\n" + "> > >>>>> Sure, and a whole lot of the callsites are (\"..._%d\", cpu), \n" "> > >>>>> hence the\n" "> > >>>>> unfortune :(\n" "> > >>>>\n" @@ -61,8 +57,7 @@ "> > >>>> If SVC_POOL_PERNODE is selected, we could name them :\n" "> > >>>> nfsd_n0 -> nfsd_nN\n" "> > >>>>\n" - "> > >>>> That would help to check with \"ps aux\" which cpu/nodes are und=\n" - "er\n" + "> > >>>> That would help to check with \"ps aux\" which cpu/nodes are under\n" "> > >>>> stress.\n" "> > >>>>\n" "> > >>>>\n" @@ -71,43 +66,39 @@ "> > >>\n" "> > >> Yup, patch welcomed.--b.\n" "> > >\n" - "> > > (Annoying fact: some initscripts stop nfsd using a rough equivale=\n" - "nt of\n" - "> > > \"killall nfsd\". So the name of the threads is arguably ABI. I t=\n" - "hink\n" - "> > > those initscripts are nuts and deserve what they get, but that ma=\n" - "y be\n" + "> > > (Annoying fact: some initscripts stop nfsd using a rough equivalent of\n" + "> > > \"killall nfsd\". So the name of the threads is arguably ABI. I think\n" + "> > > those initscripts are nuts and deserve what they get, but that may be\n" "> > > because I'm forgetting the reason they do that.)\n" "> > >\n" - "> >=20\n" - "> > We had this discussion in May-June 2008; it's because the nfsds wer=\n" - "e =20\n" + "> > \n" + "> > We had this discussion in May-June 2008; it's because the nfsds were \n" "> > once many many years ago userspace threads.\n" - ">=20\n" - "> Even when they became kernel threads, 'kill' was the only way to kill=\n" - " them -\n" + "> \n" + "> Even when they became kernel threads, 'kill' was the only way to kill them -\n" "> at first.\n" - ">=20\n" - "> >=20\n" - "> > The \"killall nfsd\" semantics in those scripts are awful and lead to=\n" - " =20\n" - "> > problems shutting down when there are lots of threads. It would =20\n" - "> > probably be an improvement to provide a better shutdown mechanism a=\n" - "nd =20\n" + "> \n" + "> > \n" + "> > The \"killall nfsd\" semantics in those scripts are awful and lead to \n" + "> > problems shutting down when there are lots of threads. It would \n" + "> > probably be an improvement to provide a better shutdown mechanism and \n" "> > force distros to use it.\n" - ">=20\n" + "> \n" "> rpc.nfsd 0\n" - ">=20\n" + "> \n" "> will stop all nfsd threads. Follow with\n" - ">=20\n" + "> \n" "> exportfs -f\n" - ">=20\n" - "> and you should be done. I'm not 100% sure about the nfsv4 thread tho=\n" - "ugh -\n" + "> \n" + "> and you should be done. I'm not 100% sure about the nfsv4 thread though -\n" "> would need to check.\n" "\n" "Should be fine.\n" "\n" - --b. + "--b.\n" + "--\n" + "To unsubscribe from this list: send the line \"unsubscribe linux-nfs\" in\n" + "the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org\n" + More majordomo info at http://vger.kernel.org/majordomo-info.html -48c8925fa8023a25a8e575e9218d29a2702edf6fc5a08cb16e92927e9a1816d4 +9aeaf79f82677f8601177b34141a307c95c4412e956aac5236eabb3cb43360b3
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.