All of lore.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <lkp@intel.com>
To: Stefan Roesch <shr@devkernel.io>, kernel-team@fb.com
Cc: oe-kbuild-all@lists.linux.dev, shr@devkernel.io,
	akpm@linux-foundation.org, david@redhat.com, hannes@cmpxchg.org,
	riel@surriel.com, linux-kernel@vger.kernel.org,
	linux-mm@kvack.org
Subject: Re: [PATCH v3 1/4] mm/ksm: add ksm advisor
Date: Wed, 6 Dec 2023 23:17:02 +0800	[thread overview]
Message-ID: <202312062353.lBYSA7Eb-lkp@intel.com> (raw)
In-Reply-To: <20231204234906.1237478-2-shr@devkernel.io>

Hi Stefan,

kernel test robot noticed the following build errors:

[auto build test ERROR on 12d04a7bf0da67321229d2bc8b1a7074d65415a9]

url:    https://github.com/intel-lab-lkp/linux/commits/Stefan-Roesch/mm-ksm-add-ksm-advisor/20231205-075118
base:   12d04a7bf0da67321229d2bc8b1a7074d65415a9
patch link:    https://lore.kernel.org/r/20231204234906.1237478-2-shr%40devkernel.io
patch subject: [PATCH v3 1/4] mm/ksm: add ksm advisor
config: i386-randconfig-011-20231206 (https://download.01.org/0day-ci/archive/20231206/202312062353.lBYSA7Eb-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231206/202312062353.lBYSA7Eb-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202312062353.lBYSA7Eb-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: mm/ksm.o: in function `scan_time_advisor':
>> mm/ksm.c:423: undefined reference to `__divdi3'
>> ld: mm/ksm.c:435: undefined reference to `__divdi3'
   ld: mm/ksm.c:428: undefined reference to `__divdi3'


vim +423 mm/ksm.c

   383	
   384	/*
   385	 * The scan time advisor is based on the current scan rate and the target
   386	 * scan rate.
   387	 *
   388	 *      new_pages_to_scan = pages_to_scan * (scan_time / target_scan_time)
   389	 *
   390	 * To avoid pertubations it calculates a change factor of previous changes.
   391	 * A new change factor is calculated for each iteration and it uses an
   392	 * exponentially weighted moving average. The new pages_to_scan value is
   393	 * multiplied with that change factor:
   394	 *
   395	 *      new_pages_to_scan *= change facor
   396	 *
   397	 * In addition the new pages_to_scan value is capped by the max and min
   398	 * limits.
   399	 */
   400	static void scan_time_advisor(void)
   401	{
   402		unsigned int cpu_percent;
   403		unsigned long cpu_time;
   404		unsigned long cpu_time_diff;
   405		unsigned long cpu_time_diff_ms;
   406		unsigned long pages;
   407		unsigned long per_page_cost;
   408		unsigned long factor;
   409		unsigned long change;
   410		unsigned long last_scan_time;
   411		s64 scan_time;
   412	
   413		/* Convert scan time to seconds */
   414		scan_time = advisor_stop_scan();
   415		scan_time = div_s64(scan_time, MSEC_PER_SEC);
   416		scan_time = scan_time ? scan_time : 1;
   417	
   418		/* Calculate CPU consumption of ksmd background thread */
   419		cpu_time = task_sched_runtime(current);
   420		cpu_time_diff = cpu_time - advisor_ctx.cpu_time;
   421		cpu_time_diff_ms = cpu_time_diff / 1000 / 1000;
   422	
 > 423		cpu_percent = (cpu_time_diff_ms * 100) / (scan_time * 1000);
   424		cpu_percent = cpu_percent ? cpu_percent : 1;
   425		last_scan_time = prev_scan_time(&advisor_ctx, scan_time);
   426	
   427		/* Calculate scan time as percentage of target scan time */
   428		factor = ksm_advisor_target_scan_time * 100 / scan_time;
   429		factor = factor ? factor : 1;
   430	
   431		/*
   432		 * Calculate scan time as percentage of last scan time and use
   433		 * exponentially weighted average to smooth it
   434		 */
 > 435		change = scan_time * 100 / last_scan_time;
   436		change = change ? change : 1;
   437		change = ewma(advisor_ctx.change, change);
   438	
   439		/* Calculate new scan rate based on target scan rate. */
   440		pages = ksm_thread_pages_to_scan * 100 / factor;
   441		/* Update pages_to_scan by weighted change percentage. */
   442		pages = pages * change / 100;
   443	
   444		/* Cap new pages_to_scan value */
   445		per_page_cost = ksm_thread_pages_to_scan / cpu_percent;
   446		per_page_cost = per_page_cost ? per_page_cost : 1;
   447	
   448		pages = min(pages, per_page_cost * ksm_advisor_max_cpu);
   449		pages = max(pages, per_page_cost * ksm_advisor_min_cpu);
   450		pages = min(pages, ksm_advisor_max_pages);
   451	
   452		/* Update advisor context */
   453		advisor_ctx.change = change;
   454		advisor_ctx.scan_time = scan_time;
   455		advisor_ctx.cpu_time = cpu_time;
   456	
   457		ksm_thread_pages_to_scan = pages;
   458	}
   459	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

  reply	other threads:[~2023-12-06 15:17 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-04 23:49 [PATCH v3 0/4] mm/ksm: Add ksm advisor Stefan Roesch
2023-12-04 23:49 ` [PATCH v3 1/4] mm/ksm: add " Stefan Roesch
2023-12-06 15:17   ` kernel test robot [this message]
2023-12-12 13:44   ` David Hildenbrand
2023-12-12 18:32     ` Stefan Roesch
2023-12-04 23:49 ` [PATCH v3 2/4] mm/ksm: add sysfs knobs for advisor Stefan Roesch
2023-12-12 13:45   ` David Hildenbrand
2023-12-12 18:02     ` Stefan Roesch
2023-12-12 18:07       ` David Hildenbrand
2023-12-12 18:27         ` Stefan Roesch
2023-12-04 23:49 ` [PATCH v3 3/4] mm/ksm: add tracepoint for ksm advisor Stefan Roesch
2023-12-12 13:45   ` David Hildenbrand
2023-12-04 23:49 ` [PATCH v3 4/4] mm/ksm: document ksm advisor and its sysfs knobs Stefan Roesch
2023-12-12 13:46   ` David Hildenbrand
2023-12-05 20:46 ` [PATCH v3 0/4] mm/ksm: Add ksm advisor Andrew Morton
2023-12-07 23:18   ` Stefan Roesch

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=202312062353.lBYSA7Eb-lkp@intel.com \
    --to=lkp@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=hannes@cmpxchg.org \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=oe-kbuild-all@lists.linux.dev \
    --cc=riel@surriel.com \
    --cc=shr@devkernel.io \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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.