From: Serge Petrenko <sergepetrenko@tarantool.org> To: Alexander Turenko <alexander.turenko@tarantool.org> Cc: tarantool-patches@dev.tarantool.org, Vladislav Shpilevoy <v.shpilevoy@tarantool.org> Subject: Re: [Tarantool-patches] [PATCH v2 1/2] fiber: reset clock stats on fiber.top_enable() Date: Fri, 15 Nov 2019 21:23:42 +0300 [thread overview] Message-ID: <794D2738-DF93-410A-9F5D-D0CC8908CCC3@tarantool.org> (raw) In-Reply-To: <20191115161138.qy3u42snfkourbh7@tkn_work_nb> [-- Attachment #1: Type: text/plain, Size: 5395 bytes --] Hi! Thank you for your answer! Fixed codestyle and answered your question inline. -- Serge Petrenko sergepetrenko@tarantool.org <mailto:sergepetrenko@tarantool.org> > 15 нояб. 2019 г., в 19:11, Alexander Turenko <alexander.turenko@tarantool.org> написал(а): > > I have no any knowledge around fiber.top(), just asked a question > because of curiosity. > > WBR, Alexander Turenko. > > On Fri, Nov 15, 2019 at 05:58:58PM +0300, Serge Petrenko wrote: >> We didn't refresh last remembered clock on fiber.top_enable() >> This means that the fiber issuing fiber.top_enable() would get a huge >> chunk of cpu time on its first yield. Fix this. >> Also reset clock_delta and cpu_miss_count. >> >> If fiber.top() is issued on the same ev loop iteration as >> fiber.top_enable(), clock_delta_last is 0 for both cord and all the >> fibers, so report "instant" and "average" stats per last iteration as 0 >> instead of NaN. >> >> Follow-up #2694 >> --- >> src/lib/core/fiber.c | 21 +++++++++++++++++++++ >> src/lua/fiber.c | 12 ++++++++++-- >> 2 files changed, 31 insertions(+), 2 deletions(-) >> >> diff --git a/src/lib/core/fiber.c b/src/lib/core/fiber.c >> index aebaba7f0..258c094f5 100644 >> --- a/src/lib/core/fiber.c >> +++ b/src/lib/core/fiber.c >> @@ -1203,9 +1203,30 @@ fiber_top_enable() >> ev_check_start(cord()->loop, &cord()->check_event); >> fiber_top_enabled = true; >> >> + /* >> + * Reset cord and fiber clock stats in order to >> + * count from zero even on reenable. >> + */ >> cord()->clock_acc = 0; >> cord()->cpu_miss_count_last = 0; >> cord()->clock_delta_last = 0; >> + cord()->clock_delta = 0; >> + >> + struct fiber * fiber; > > Code style: struct fiber *fiber; Fixed > >> + rlist_foreach_entry(fiber, &cord()->alive, link) { >> + fiber->clock_acc = 0; >> + fiber->clock_delta_last = 0; >> + fiber->clock_delta = 0; >> + fiber->cputime = 0; >> + } >> + >> + cord()->sched.clock_acc = 0; >> + cord()->sched.clock_delta_last = 0; >> + cord()->sched.clock_delta = 0; >> + cord()->sched.cputime = 0; >> + >> + cord()->clock_last = __rdtscp(&cord()->cpu_id_last); >> + cord()->cpu_miss_count = 0; > > Should not we doing such clean up in fiber_top_disable() rather then > here? It doesn’t matter IMO. Well, yes, you don’t need a cleanup when you call fiber.top_enable() for the first time, but, other than that, I see no difference. > >> struct timespec ts; >> if (clock_gettime(CLOCK_THREAD_CPUTIME_ID, &ts) != 0) { >> say_debug("clock_gettime(): failed to get this" >> diff --git a/src/lua/fiber.c b/src/lua/fiber.c >> index 8b3b22e55..647505643 100644 >> --- a/src/lua/fiber.c >> +++ b/src/lua/fiber.c >> @@ -335,11 +335,19 @@ lbox_fiber_top_entry(struct fiber *f, void *cb_ctx) >> lua_newtable(L); >> >> lua_pushliteral(L, "average"); >> - lua_pushnumber(L, f->clock_acc / (double)cord()->clock_acc * 100); >> + if (cord()->clock_acc) >> + lua_pushnumber(L, f->clock_acc / (double)cord()->clock_acc * 100); >> + else >> + lua_pushnumber(L, 0); >> lua_settable(L, -3); >> + >> lua_pushliteral(L, "instant"); >> - lua_pushnumber(L, f->clock_delta_last / (double)cord()->clock_delta_last * 100); >> + if (cord()->clock_delta_last) >> + lua_pushnumber(L, f->clock_delta_last / (double)cord()->clock_delta_last * 100); >> + else >> + lua_pushnumber(L, 0); >> lua_settable(L, -3); > > Code style: lines over 80 chars. Fixed > >> + >> lua_pushliteral(L, "time"); >> lua_pushnumber(L, f->cputime / (double) FIBER_TIME_RES); >> lua_settable(L, -3); >> -- >> 2.21.0 (Apple Git-122) diff --git a/src/lib/core/fiber.c b/src/lib/core/fiber.c index 258c094f5..c6aab18b2 100644 --- a/src/lib/core/fiber.c +++ b/src/lib/core/fiber.c @@ -1212,7 +1212,7 @@ fiber_top_enable() cord()->clock_delta_last = 0; cord()->clock_delta = 0; - struct fiber * fiber; + struct fiber *fiber; rlist_foreach_entry(fiber, &cord()->alive, link) { fiber->clock_acc = 0; fiber->clock_delta_last = 0; diff --git a/src/lua/fiber.c b/src/lua/fiber.c index 647505643..9268b0a19 100644 --- a/src/lua/fiber.c +++ b/src/lua/fiber.c @@ -335,17 +335,21 @@ lbox_fiber_top_entry(struct fiber *f, void *cb_ctx) lua_newtable(L); lua_pushliteral(L, "average"); - if (cord()->clock_acc) - lua_pushnumber(L, f->clock_acc / (double)cord()->clock_acc * 100); - else + if (cord()->clock_acc) { + lua_pushnumber(L, f->clock_acc / + (double)cord()->clock_acc * 100); + } else { lua_pushnumber(L, 0); + } lua_settable(L, -3); lua_pushliteral(L, "instant"); - if (cord()->clock_delta_last) - lua_pushnumber(L, f->clock_delta_last / (double)cord()->clock_delta_last * 100); - else + if (cord()->clock_delta_last) { + lua_pushnumber(L, f->clock_delta_last / + (double)cord()->clock_delta_last * 100); + } else { lua_pushnumber(L, 0); + } lua_settable(L, -3); lua_pushliteral(L, "time"); [-- Attachment #2: Type: text/html, Size: 15315 bytes --]
next prev parent reply other threads:[~2019-11-15 18:23 UTC|newest] Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-11-15 14:58 [Tarantool-patches] [PATCH v2 0/2] fiber.top(): minor fixup Serge Petrenko 2019-11-15 14:58 ` [Tarantool-patches] [PATCH v2 1/2] fiber: reset clock stats on fiber.top_enable() Serge Petrenko 2019-11-15 16:11 ` Alexander Turenko 2019-11-15 18:23 ` Serge Petrenko [this message] 2019-11-15 21:39 ` Vladislav Shpilevoy 2019-11-18 16:11 ` Serge Petrenko 2019-11-15 14:58 ` [Tarantool-patches] [PATCH v2 2/2] app/fiber: wait till a full event loop iteration ends Serge Petrenko 2019-11-15 15:35 ` Serge Petrenko 2019-11-15 16:27 ` Alexander Turenko
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=794D2738-DF93-410A-9F5D-D0CC8908CCC3@tarantool.org \ --to=sergepetrenko@tarantool.org \ --cc=alexander.turenko@tarantool.org \ --cc=tarantool-patches@dev.tarantool.org \ --cc=v.shpilevoy@tarantool.org \ --subject='Re: [Tarantool-patches] [PATCH v2 1/2] fiber: reset clock stats on fiber.top_enable()' \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox