Date: Mon, 19 Mar 2012 21:06:36 +0100 From: Davide Italiano <davide.italiano@gmail.com> To: Gleb Smirnoff <glebius@freebsd.org>, George Neville-Neil <gnn@freebsd.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r233045 - in head/sys: conf kern Message-ID: <CACYV=-HnKmkcw-p2i9o_VyKGkp6V5yzO0B61uCBOAebg=VHd-A@mail.gmail.com> In-Reply-To: <CACYV=-EmbDNep1rPwF6jfHB2M8JuhLDFvrsvZoBMuVB7kOC1bA@mail.gmail.com> References: <201203162032.q2GKWBRV033715@svn.freebsd.org> <20120319194431.GD30704@FreeBSD.org> <CACYV=-EmbDNep1rPwF6jfHB2M8JuhLDFvrsvZoBMuVB7kOC1bA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 19, 2012 at 8:54 PM, Davide Italiano <davide.italiano@gmail.com> wrote: > 2012/3/19 Gleb Smirnoff <glebius@freebsd.org>: >> =A0Davide, >> >> On Fri, Mar 16, 2012 at 08:32:11PM +0000, Davide Italiano wrote: >> D> Author: davide >> D> Date: Fri Mar 16 20:32:11 2012 >> D> New Revision: 233045 >> D> URL: http://svn.freebsd.org/changeset/base/233045 >> D> >> D> Log: >> D> =A0 Add rudimentary profiling of the hash table used in the in the um= tx code to >> D> =A0 hold active lock queues. >> D> >> D> =A0 Reviewed by: =A0 =A0 =A0 attilio >> D> =A0 Approved by: =A0 =A0 =A0 davidxu, gnn (mentor) >> D> =A0 MFC after: 3 weeks >> D> >> D> Modified: >> D> =A0 head/sys/conf/NOTES >> D> =A0 head/sys/conf/options >> D> =A0 head/sys/kern/kern_umtx.c >> >> ... >> >> D> =A0static void >> D> =A0umtxq_sysinit(void *arg __unused) >> D> =A0{ >> D> @@ -232,8 +265,15 @@ umtxq_sysinit(void *arg __unused) >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0TAILQ_INIT(&umtxq_chains[i= ][j].uc_pi_list); >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0umtxq_chains[i][j].uc_busy= =3D 0; >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0umtxq_chains[i][j].uc_wait= ers =3D 0; >> D> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0#ifdef UMTX_PROFILING >> D> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0umtxq_chains[i][j].length = =3D 0; >> D> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0umtxq_chains[i][j].max_lengt= h =3D 0; >> D> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0#endif >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0} >> D> =A0 =A0 =A0} >> D> + =A0 =A0#ifdef UMTX_PROFILING >> D> + =A0 =A0umtx_init_profiling(); >> D> + =A0 =A0#endif >> D> =A0 =A0 =A0mtx_init(&umtx_lock, "umtx lock", NULL, MTX_SPIN); >> D> =A0 =A0 =A0EVENTHANDLER_REGISTER(process_exec, umtx_exec_hook, NULL, >> D> =A0 =A0 =A0 =A0 =A0EVENTHANDLER_PRI_ANY); >> D> @@ -384,6 +424,14 @@ umtxq_insert_queue(struct umtx_q *uq, in >> D> >> D> =A0 =A0 =A0TAILQ_INSERT_TAIL(&uh->head, uq, uq_link); >> D> =A0 =A0 =A0uh->length++; >> D> + =A0 =A0#ifdef UMTX_PROFILING >> D> + =A0 =A0uc->length++; >> D> + =A0 =A0if (uc->length > uc->max_length) { >> D> + =A0 =A0 =A0 =A0 =A0 =A0uc->max_length =3D uc->length; >> D> + =A0 =A0 =A0 =A0 =A0 =A0if (uc->max_length > max_length) >> D> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0max_length =3D uc->max_lengt= h; >> D> + =A0 =A0} >> D> + =A0 =A0#endif >> D> =A0 =A0 =A0uq->uq_flags |=3D UQF_UMTXQ; >> D> =A0 =A0 =A0uq->uq_cur_queue =3D uh; >> D> =A0 =A0 =A0return; >> D> @@ -401,6 +449,9 @@ umtxq_remove_queue(struct umtx_q *uq, in >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0uh =3D uq->uq_cur_queue; >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0TAILQ_REMOVE(&uh->head, uq, uq_link); >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0uh->length--; >> D> + =A0 =A0 =A0 =A0 =A0 =A0#ifdef UMTX_PROFILING >> D> + =A0 =A0 =A0 =A0 =A0 =A0uc->length--; >> D> + =A0 =A0 =A0 =A0 =A0 =A0#endif >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0uq->uq_flags &=3D ~UQF_UMTXQ; >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0if (TAILQ_EMPTY(&uh->head)) { >> D> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0KASSERT(uh->length =3D=3D = 0, >> >> These indented ifdefs look like a major violation of style used througho= ut >> the FreeBSD kernel code. Can you please keep with common style? >> >> -- >> Totus tuus, Glebius. > > Heh, > sorry, also Juli Mallet noticed this, I'm writing a fix for this and > after I'll have approval from my mentor I'll commit. > > Davide This should fix: http://people.freebsd.org/~davide/umtx_stylefix.diff Can you plase give it a closer look? George, if everythin' is ok, can I have also your approval?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CACYV=-HnKmkcw-p2i9o_VyKGkp6V5yzO0B61uCBOAebg=VHd-A>