From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754079Ab1LIPrz (ORCPT ); Fri, 9 Dec 2011 10:47:55 -0500 Received: from tac.ki.iif.hu ([193.6.222.43]:48161 "EHLO tac.ki.iif.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751036Ab1LIPrx (ORCPT ); Fri, 9 Dec 2011 10:47:53 -0500 From: Ferenc Wagner To: Jeremy Fitzhardinge Cc: Konrad Rzeszutek Wilk , Zhenzhong Duan , linux-x86_64@vger.kernel.org, linux-kernel@vger.kernel.org, xen-devel@lists.xensource.com Subject: Re: [PATCH] cpu idle ticks show twice in xen pvm guest References: <20111010155319.GA29140@phenom.oracle.com> <4E934EC9.5030909@goop.org> Date: Fri, 09 Dec 2011 16:47:39 +0100 In-Reply-To: <4E934EC9.5030909@goop.org> (Jeremy Fitzhardinge's message of "Mon, 10 Oct 2011 13:00:09 -0700") Message-ID: <87vcppbvc4.fsf@tac.ki.iif.hu> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.2 (gnu/linux) MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-=-= Jeremy Fitzhardinge writes: >> On Wed, Oct 05, 2011 at 10:11:58PM -0700, Zhenzhong Duan wrote: >> >>> Run below test on xen pvm. >>> # x=$(cat /proc/stat | grep cpu0 | awk '{print $5}') && sleep 60 \ >>> && y=$(cat /proc/stat | grep cpu0 | awk '{print $5}') \ >>> && echo -e "X:$x\nY:$y\nIDLE:" $(echo "scale=3; ($y-$x)/6000*100" | bc) >>> >>> @ X:58562301 >>> @ Y:58574282 >>> @ IDLE: 199.600 >>> >>> Normal idle percent should be around 100%. >>> xen_timer_interrupt called account_idle_ticks to account hypervisor stolen idle ticks >>> but these ticks will be accounted again when idle ticks restarted. >>> >>> Signed-off-by: Zhenzhong Duan >>> Signed-off-by: Joe Jin > > Does this affect the accounting of stolen ticks? If it does, that's not > necessarily a showstopper for this patch, but we'll need to do some more > thinking about it. Certainly, accurate accounting for idleness is > important. Please see also http://thread.gmane.org/gmane.linux.kernel/734441, where I found that the counter doubling isn't always present under 2.6.26. However, after going to 2.6.32 (Debian lenny-backports kernel, 4th of April on the graph below) that instability seems to disappear. Please note that the following graph shows halved idle and iowait percentages. --=-=-= Content-Type: image/png Content-Disposition: inline; filename=cpu.png Content-Transfer-Encoding: base64 Content-Description: Modified Munin graph iVBORw0KGgoAAAANSUhEUgAAAfUAAAEGCAYAAAB8TgymAAAABmJLR0QA/wD/AP+gvaeTAAAgAElE QVR4nO2dfXQUVZr/vx3QySAtCZ0AkbwRQUZRRCJxlbwAkcgZd4POHpjfWUXEEUSjZ1FXB5goIhGY UXxDE10xwrLrmQ2ODmF0WRhJJ+PLKEFllOwQzAudQDDkpaERCIH07490Nf1WXVXdVXWr6z6fc3Iq XV1V3/s8VdW3nvvcutfy9ddfu0EQBEEQRExz4sQJxLEuBEEQBEEQ0XHixAk0NzdjqLBi/PjxLMtD EARBEESEbNu2DQAoUicIgiAIs0CVOkEQBEGYBKrUCYIgCMIkDJXehCAIgiAIFuzcuTPk+jlz5oRc T5E6QRAEQRiUOXPmYM6cOZg1a5b3czgoUicIgiAIA3PmzBkcOHAAQ4YMgdPpxNCh4lU3VeoEQRAE YVBaWlrQ3NyMsWPHIjMzE3v37kVGRobo9lSpEwRBEIRB6enpwc0334xhw4YBAGbPnh12e6rUCYIg CMKgHD9+HMePHw9aTx3lCIIgCCLGoI5yBEEQBGEiqKMcQRAEQZgAbjvKtdTVAQDG5eczLglBEARB qAOXHeXO9/Xhz2VlAIBFO3Zg6E9+wrhEBEEQBBE9SjvKmaJS3/v22zh55Ij3/5sfeohxiQiCIAgi eiwWC4qKimCxWAAAbrcbu3btEt0+5nu/nzxyBF9u2uT9/OWmTd4KniAIgiBiGavVira2NgwMDGBg YABtbW2wWq2i2+taqVutVu+fL21tbSgsLITNZkNhYSHa29vDrvfl4+eew/m+Pu/n8319+Pi557Q1 hCAIgiB04LrrrkNHRwd2796N3bt3o6OjA5MnTxbdXtdK3eVyweVyBa1fuXIlcnNzceTIEeTm5qK0 tDTsel/uLC/H4wcO4PT/+394/MABPH7gAO4sL1dctjO7dys3SCVI25x6RtAmm0nbrLostfXUtVqt uOmmm3Dbbbfhtttuw0033YRPPvlEdHtD5NTr6uqwYcMGxMfHo6SkBDk5OWHXa4Hlsss0OzZpG0Ob J1tZ67LU5tFmltpks7EwRE7d6XQiKSkJRUVFsNlscDqdYdf7cqGnB67KSgz09gL19Rg+fHhEy0vO nYtq/2iWAwcOMNHlzW4e/czT+WWtS/7mx99SuvF9fXBVVuJcQ0PU9ePOnTuD/sJh+frrr90AMH78 +KjF5WK1Wv2a4TMyMlBfX4/k5GR0dnYiJycHra2toutDsWbNGvz2t78NLWjxLN3q2kFExz6Lxe9z tucE7YMl1OaiCPvF2vkNtF8MMb9kBxgs129S+ynVi/R4co8rd79AxMonVU6p76X8nO0OOI7nPAeu F0PudREtQnmC7kOZ5STCc+rUKc2OvXPnzqBX2rZt2wbAIM3veXl5KC8vx/Lly1FRUYGCgoKw67Xg VF0dhjMauIZX7YmptRjeHqwt9WMdKUb1s1x7pbYL9f2p1LqQPpZz3Ej0fNcHakdqp9LrQdCNtPxS 34fb71RqXVAQIbf83so/wutfzrn2w1POID3PerkPiZempuJce7voQ4JAtA8LoR4+wt1XagUNgUjp skbXSD1UN3yXywWHw4GFCxdi//79mDJlCrZs2YK0tDTR9aGgSD0GUTsgibXzq09ARgRgqRqMaNzz 52krJFyPAZWk3BaPWEOtlpHAytf7MCByvwTpBvhddWSeJi0j9VAwidRD9XwHgPT0dNTU1MherwVG jeJMra00sohWj0M/6+3jSLTVrmQDdYXjy9UX+yyFe/68sDZH2iIiF1bnWquWEW/lHqYF41RqHbLb RdIdAdureZ2Fup99WwbUTGF0dXXh4MGD3oeE4cOHY+LEiUhKSgq5PZOcuhZQpB6DUKTORjagkgr8 kRP78Yt2P7EfU7nHVYrSSlktvOWPMmIU84vW50HzFowoibScUn6T2l6gfv58ABItA271IvWPP/4Y 1113nbcS7+rqwrfffovCwkK/7QyVUzcCPEZxzLUpUtdeV4aPxX685Ea4YuuLW1tRnZmp2nHlIqar B8WtrRcrHYSO/AXkPjxJrReY+7vnZfk70ocose0Ef0s9RCl9WJPa3lK1DcWtrdj+5BN+38t9mIv0 OsuGe7DvhGcstGj7QsjBYrF4h4mVgip1Dyw7PXCrrXNTIY9+VtIMrTasKlZBV297fbXDIVXJKuXi /tLacsqjdDu551mpnXIi8urMTCbneXh7vk/5tG0inDx5MhobG/HVV18Nag8fbpwR5VhhqdomeeJP eaZuZQG32qn6apvNz1LXtaVqG+b+7nnVdeVSLPL6qVl1A7X3WSy6vZ4WqK0nWuuGu85Dacv5vY+2 PHreV8nJyZg+fbp3RLnp06cjOTlZdHuK1D3wGMWx1ra++AOAbbJzW9Hm+szq53D+YRUth9LWK6Iy is0Xm2PZ2M2bv1nphvKzkHpRA6Ud5WI+Uu/v74fD4UBXVxcAoLm5OeRy0tmzYb9v2LEj7PdaLk/V 1THRZW13cWvrxfMyqRmWqm249j+2hvzsu13g0m+/WPNzCHuULAW/TDp7Nsh/ALDo8GHR7bVe+p5f HnSBi/4OvF4Dz4+Z7BZ0tbZPjr+NotvY2AiHw4GOjg5Ey/79+zFhwgTMnj0bs2fPxoQJE7B//37R 7bno/W7x9Ap0zzN2r07e8J4Xma8SSfbajbXzG2XLrJJXrZRsT6gD+Z1f3PPmqdr7ffLkyd7I/Pjx 42F7v8d8pK4WZsu3xoK2kA/TOgcmwKOfjZJf5kE3UFuv6zqUtp4Yxd9m1RU6yu3atQu7du3CoUOH wnaUo5y6B7PmW42srXc+jEc/857vJG3z6rLU1lM3OTk5bMe4QChS98BjFMdaW+nTbrSRT6z6OdBu JX6gKIq0zarLUltP3VCztIWbqY0idQ88RnGstaN92lVawfPoZ4qiSNusuiy19dYNnJEtHBSpe4jV KC6WtcWeduW8fx1JxG5UP2uZe6UoirTNqstSm6XNUlCk7oHHKI61NuXUtYfViFuCNk+6vGqTzcbC EJF6dXU1Jk2ahJEjR2LWrFloaWkBALS1taGwsBA2mw2FhYVob2/XrAxGjeLMrK33026s+zmSiJ6i KNI2qy5LbT11lTS9Awap1B955BG89tprOHbsGH79619jxYoVAICVK1ciNzcXR44cQW5uLkpLSzUr A49RHGttitS1h6Io0jarLkttitRl4DsDzWeffQYAqKurQ0lJCeLj41FSUgK73a6ZfqxHcbGoTZG6 9lAURdpm1WWpbeScuiEq9RdffBFLly5FamoqPvnkE5w8eRIA4HQ6kZSUhKKiIthsNjidzqB9L/T0 wFVZiYHeXpzyVPqByzs8zfli35+y2zE8Pz/s91ouMTDARJe13XHuwbGxhfMT7TLm/Jxq9y9/qj3k Mhq/VGdmquZfpUu1z6/Rdcnf/PhbSvd8dzdclZU419AAvTHcMLF1dXVYvHgxDh48iIyMDNTX1yM5 ORmdnZ3IyclBq8gTkqxhYoVhRt3B2/A41zZr7bnPh57/OVKkhok1mp+F61JAahjcSGA9tzgLbR5t ZqlNNgej5jCxcjHcMLFutxuNjY349a9/jXmeH+e8vDyUl5ejr68PFRUVKCgo0Eyfx3wra23KqWsP 5TtJ26y6LLX10v38889x9OhRuN0hIlERDFGpW61WXH755ZgzZw7y8/Px1FNPAQDWrl0Lu92OlJQU 1NbWoqysTLMy8JhvZa1NOXXtoXwnaZtVl6W2XrrXXHMNjh8/jtraWnz//ffo6+uT3McQ76m7XK6Q 69PT01FTU6NLGXiM4lhrU6SuPRRFkbZZdVlq66U7YsQIXH/99ejr60NbWxs+//xzJCYmIjMzEyNG jAi5jyEidSPAYxTHWpside2hKIq0zarLUltv3Z/85CcYP3488vPzkZycjIYwHfAMEakbAR6jONba FKmHR42R4CiKIm2z6rLUZqUbFxeHK664AldccYX4NjqWx9DwGMWx1qZIXXsoiiJts+qy1Dbye+pc RepC5ONG8KtDsRbFmUGbIvXQqDlWO0VRpG1WXZbaNKJcDMBjFMdamyJ17aEoirTNqstS28iResxX 6v39/XA4HOjq6gIANDc3h1xOOnvWbxn4fWdqatj9tVwOz89nosva7urMzKDzEu0y1vzsW35L1TZc +x9bVfVH05gxqh5PyVKL82tkXYD8zYu/pXQbGxvhcDjQ0dEBvTHciHKRImdEOYFQI48ZbbQxHrRp RDntp0Sl0b5I26y6LLVpRLkYIFbyrWbSppy69lC+k7TNqstSm3VOfefOnaLfUaXugcd8K2ttyqlr D+U7Sdusuiy1jZxT56r3ezh4jOJYa1Okrj0URZG2WXVZauupGy4qDwVF6h54jOJYa1Okrj0URZG2 WXVZauupO2fOnKC/cFCl7oHHKI61NkXq2kNRFGmbVZelNuucejgMUalXV1dj8uTJSEhIwOTJk7Fj xw4AQFtbGwoLC2Gz2VBYWIj29nbNysBjFMdamyJ17aEoirTNqstSm3VOPVy0bohX2lJTU7F582bk 5eWhtrYW9913H9rb27FgwQJkZWVhxYoVWLduHQ4fPozNmzeHPEa0r7QR+qP2K12xdl71eKWNIAj9 UfOVtq6uLhw8eNB7vOHDh2PixIlISkry285Qr7Slp6cjLi4OFosFQ4YMQaanaaOurg4lJSWIj49H SUkJ7Ha7ZmXgMYpjrU2RuvZQFEXaZtVlqa2n7v79+zFhwgTMnj0bs2fPxoQJE7B//37R7Q1RqW/c uBELFy6EzWbDvffei1dffRUA4HQ6kZSUhKKiIthsNjidzqB9L/T0wFVZiYHeXpzyVPqByztaWvyW obYbnp8vur/WSwwMMNFlbXec2w0g+PxEuow1P6tld7hldWamLjqhlmqfX6Prkr/58beU7vnubrgq K3EuzBSpSrBYLN4/yW2N0Pw+depUrF+/HgUFBbDb7fjNb36D+vp6ZGRkoL6+HsnJyejs7EROTg5a RZ6QaES52NOmEeVoRDkz6fKqTTYHo2bz+/Hjx9HY2OjX/H7VVVchOTnZbzuh+d0Q76n/8MMPAOBt fj927BgAIC8vD+Xl5Vi+fDkqKipQUFCgWRl47BnNWpt6v2sP9UwmbbPqstTWUzc5OTmoAg+HIZrf X331VTz66KNITk7GsmXLsHHjRgDA2rVrYbfbkZKSgtraWpSVlWlWBh7zray1KaeuPZTvJG2z6rLU 1lN3586dIf/EMETzuxpQ7/fYg3q/U+93gjAjaja/79y5E3PmzPEufdf5Yqje70aAxyiOtTZF6tpD URRpm1WXpbaeunFxcXC5XLBYLDh58iROnTqFoUPFM+eGyKkbAR7zray1KaeuPZTvJG2z6rLU1lM3 NTUVf/3rX3H11Vfjq6++Qn9/PyZOnCi6PUXqHniM4lhrU6SuPRRFkbZZdVlq66l7zTXXYPbs2UhP T8eMGTO8/4tBlboHHqM41toUqWsPRVGkbVZdltqsZ2mj+dRlwGMUx1qbInXtoSiKtM2qy1Jbb125 Pd8BE1Tq/f39cDgc6OrqAgA0NzeHXE46e9ZvGfh9Z2pq2P21XA7Pz2eiy9ru6szMoPMS7TLW/Ky2 /YHLpjFjdNEJtdTi/BpZFyB/8+JvKd3GxkY4HA50dHRADZRMvUqvtHkw2mhjPGjTiHI0opyZdHnV JpuDUfOVNrnQK20B8JhvZa1NOXXtoXwnaZtVl6U2zaceA/CYb2WtTTl17aF8J2mbVZelNuv51MNB lboHHqM41toUqWsPRVGkbVZdltoUqccAPEZxrLUpUtceiqJI26y6LLUpUo8BeIziWGur/bS7z2LB vjDzDfPoZ4qiSNusuiy1KVKPAXiM4lhrq/20mw03suEW/Z5HP1MURdpm1WWpbeRI3RBjv1utVr/P I0eOxOHDh9HW1oZ7770X33zzDaZMmYItW7Yg1fO+r9rwGMWx1qacuvZQFEXaZtVlqa2nbldXFw4e POh9RW748OGYOHEikpKSQm5viEjd5XJ5/5YsWYL77rsPALBy5Urk5ubiyJEjyM3NRWlpqWZl4DGK Y61NOXXtoSiKtM2qy1JbT939+/djwoQJmD17NmbPno0JEyZg//79otsbIlIXOHHiBKqqqvDll18C AOrq6rBhwwbEx8ejpKQEOTk5mmnzGMWx1qZIXXsoiiJts+qy1NZb12KxwBKmv5AvhojUBd555x3c dtttSElJAQA4nU4kJSWhqKgINpsNTqczaJ8LPT1wVVZioLcXp+x2AAha3tHS4rcMtd2pujrR/bVe dr3xBhNd1nY/63l4Czw/kS5PpYbXM5qf1bI73LK4tVUXnVBLtc+v0XXJ3/z4W0r3fHc3XJWVONfQ gGiZPHkyGhsbsWvXLuzatQuHDh3C5MmTRbc3zDCx58+fx3XXXYf/+q//wtSpUwEAGRkZqK+vR3Jy Mjo7O5GTk4NWkWaPaIeJJfRH7WFS3fM951W8r5yh0GOYWIIg9IeGiQWwfft2pKameit0AMjLy0N5 eTn6+vpQUVGBgoICzfR5zLey1qacuvZQvpO0zarLUltP3cAZ2qRmajNMpf7aa6+hpKTEb93atWth t9uRkpKC2tpalJWVaabPY76VtTbl1LWH8p2kbVZdltp66wozs8mZpc0wlXpNTQ3uuOMOv3Xp6emo qalBT08P9uzZg7S0NM30eYziWGtTpK49FEWRtll1WWrTe+oxAI9RHGtttZ92LVWDOSU3QveZ4NHP FEWRtll1WWrTiHIxAI9RHGttitS1h6Io0jarLkttPXV9m97lQJW6Bx6jONbalFPXHoqiSNusuiy1 KVKPAXiM4lhr8xqpS008oyYURZG2WXVZarPu/R4OU+fUvT+cVVWS2/IYxbHW5jVSvzjpjPbvqVMU Rdpm1WWpraduqGb3mHilLVL6+/vhcDjQ1dUFAGhubvYus+FG4qQmAMCks2f9lr7bAUDDjh0h1+ux PFVXx0SXtd3Fra1B50WtpeH9PKlZE7sDl4sOH9ZFJ9RSy/NrRF2A/M2Lv6V0Gxsb4XA40NHRAb0x zIhy0RJyRDlPoC70ihagEeWMgVYjqhn+/IpclwRBmAMaUc4AGCXfyoW2ZfCP15y6nlC+k7TNqstS W++cupx1AqbOqSuJhIySb+VJm9ecup5QvpO0zarLUltvXanOcb5QpO6BxyiOtTZF6tpDURRpm1WX pbbeusLwsHKGiTV1Tl0sZ2v4nKvZ0TinbPjzSzl1gjA1lFM3ADxGcay1KVLXHoqiSNusuiy1aex3 Cfr6+rBq1SpUVVXh+PHjAACXy4W2tjbce++9+OabbzBlyhRs2bIFqampmpSBx3wra23KqWsP5TtJ 26y6LLX11BXLp4s1wxsiUl+zZg3+8pe/4H/+539w8uRJuFwuAMDKlSuRm5uLI0eOIDc3F6WlpZqV gccojrU2ReraQ1EUaZtVl6W23mO/z5kzB+PGjcOIESMwa9assHl1Q+TUr7rqKlRWViI3N9dvfUZG Bvbu3YtRo0ahs7MTOTk5aBVxJuXUYwet3k8XMPz5pZw6QZgaNXPqZ86cwYEDB3Du3DlceeWVaGtr w7XXXov4+Hi/7QyVU+/s7MTu3bsxevRoXH311fjjH/8IAHA6nUhKSkJRURFsNhucTmfQvhd6euCq rMRAby9O2e0A4F3e0dISchm43Sm7Hafq6kKu12PZ9cYbTHRZ2S2ch2e//NLvs1pLo/v5zt/+Dpaq barbHWpZ3Nqqi06opVbn16i65G9+/C2le767G67KSpxraEC0fPbZZxg5ciRuvvlmjB49GpMmTcJ3 330nur0hIvW0tDRs2rQJM2fOxKeffop7770Xhw8fRkZGBurr65GcnEyRuongPVLX2n6CINiiZqT+ 448/4rLLLvNbd+7cOVx66aV+6wwVqd9yyy1+ny2eiVjy8vJQXl6Ovr4+VFRUoKCgQLMy8JhvZa1N OXXtoXwnaZtVl6W2nrp/+ctfgmZp27Nnj+j2hqjU169fj/Xr12P06NF4+OGH8eqrrwIA1q5dC7vd jpSUFNTW1qKsrEwVvVBTX/LYM5q1NvV+1x7qmUzaZtVlqa23rtAxTs7gM4ao1MeNG4eamhr09vbi wIEDKC4uBgCkp6ejpqYGPT092LNnD9LS0lTRy4Z7cPpLzxjkAJ9RHGttrZ52xeYrV91Wn+tH6nuK 1M2vy6s22WwsDPGeuhFgEUkJFU+22627toAZI/WL85X7E62tYudLznmkSN38urxqk83GwhCRuhEI F0kJkZ9YBCi2Xmo7ocUgUFvu8eTqhMOMkboYUrZK+c3bwiO1PiCC32ex4FuVWpmUQlEUaZtVl6W2 3u+p+y4D/w/EnJG68INaJX+XcJGUWOQX9L2g6/kYWEHIjSCl9KRQsr8ZI3WBwAhazFbvdiJ+k/pe imy4gfaIdo0aiqJI26y6LLVpRLkYQFHEKjOXKkRwYhGewLdpaWEjxMAIUqrlQAly7I5ax+OPwONo /bQb6HcxW6XOT+D3ov4Ic12cSqWcutl1edUmm7VF6Bw3a9Ys7+dwmDJSlztSl7BdvWW+33rvD7hO qe7r2ttCrheLEGVHlAEtB6EYnp8vmRNW3BIhHCaggrtY7kG/6/a06ynHQQy+EinVh0G2PxQwvJ1y 6mbX5VWbbNYeYVS5IUOGwOl0YuhQ8ao75iP1/v5+OBwOdHV1AQCam5sBAJPOnhVdWqq24dr/2Or9 nA03flpQjWy4kTipafA4k5qxz2LBrmuv9X4OuWwWWR9uaQGar734+VRqXUi9wPKI6nmOF3J73/08 2/nqNOzY4b9fc0D5xOxrvqgnu5w+y0lnz6K4tTXseYpmGUp3YmqtuJ0+23ntEfk+kmVDwQ4/v2hl d+By0eHDuur5LrU8v0bUBcjfvPhbSrexsREOhwMdHR2IlpaWFnz22WcYPnw4brjhBuzduxfp6emi 2xtiRDk18B1RTumIXe758kYgEyJ7Yft9iC7XqhTV9QIjapEIWylS5dRrzHOx86qWHwOvB6nvBd0b qxR09iAIIuZQc0S5ffv24eqrr8awYcPCbmeoEeVUQyrXHYZwOU9L1Ta/ikj4LJWLFdsv8LPcfKuU nlICe2WrkaMH5JdT997vHj+r7Uep8yxoq60rB8p3krZZdVlq66mbnZ0tWaH7YsqculIuRlSBn5VF 8JF+L5VvjbY8YvsF9srWKnIVW691XipQV20/Ky+HqoeVBeU7Sdusuiy1jfyeuqkq9WiadYtbW4P2 16qZOPC4c3/3vCoXidKHC/f8eTiVWhdU2cl9KFDqn8Dti1tbdb05Am2N9PyK7RfueCwjClY/QKy0 ebSZpTbZbCxMValHQyQnSK1KX662mJ5UZBmunEp6Zasdyep1U0Ta0iFsr8Z55jGiIJv50CabjQVV 6h5i+WkzmohTjna0EbkYevtcaYuImi01FKmbX5dXbbLZWJiq9/vvbryRdTG4QM0IliAIwmyo2ftd Lobq/W61Wv3+BNra2lBYWAibzYbCwkK0t2s31iaPPTgj1Q7Vu1sv7VjSM4I22UzaZtVlqW3kWdoM EalbrVa4XK6g9QsWLEBWVhZWrFiBdevW4fDhw9i8eXPIY1CkThAEQRgB7iN1AEhNTcXo0aPxz//8 z3A4HACAuro6lJSUID4+HiUlJbDb7Zrp8/i0yZs2T7b66S6fN/jHQpsBPJ5nltpks7EwRKXucrnQ 3t6OAwcO4Oqrr8aiRYsAAE6nE0lJSSgqKoLNZoPT6Qza90JPD1yVlRjo7cUdLS0AENGyOjMzqv2j WcZ5xhhnoc+T3Tz6uTozE3ck2JnYzcrfPJ5nlnbz6G8p3fPd3XBVVuJcQwP0xhDN776cPXsWV1xx BXp6epCRkYH6+nokJyejs7MTOTk5aBV5Qoq2+Z3HHpy8afNkq5/uG08Mflivb8dGHnsm86hNNgdD ze8eTp06hY0bN2Ly5MkAgLy8PJSXl6Ovrw8VFRUoKCjQTJvHdy150+bJVta6LLV5tJmlNtlsLAxR qQu93jMzM2G32/H2228DANauXQu73Y6UlBTU1tairKxMszLwmBfiTZsnW1nrstTm0WaW2mSzsTBc 83ukUO93ghjEPX9wkHmLMBuc0ElO5+Z3guAVan43ADw+bfKmzYutliw3LFlu3XUD4TGK4lGbbDYW VKl74DEvxJs2T7ay1mWpzaPNLLXJZmNBlboHHp82edPmyVbWuiy1ebSZpTbZbCyoUvfA49Mmb9o8 2cpal6U2jzaz1CabjQVV6h54fNrkTZsnW1nrstTm0WaW2mSzsYj5Sr2/vx8OhwNdXV0AgElnz0a0 bBozJqr9o1lWZ2Yy0eXNbq78PK7ZX9fzmQd/c3WeDWA3j/6W0m1sbITD4UBHRwf0hl5p88DjqEi8 aXNjq88rbDSiHGmbVZelNo0oFwPwmBfiTZsnW1nrstTm0WaW2mSzsaBK3QOPeSHetHmylbUuS20e bWapTTYbC6rUPfD4tMmbNk+2stZlqc2jzSy1yWZjQZW6Bx6fNnnT5slW1rostXm0maU22WwsqFL3 wOPTJm/aPNnKWpelNo82s9Qmm42FoSr1DRs2wGq1ej+3tbWhsLAQNpsNhYWFaG9v10ybx6dN3rR5 spW1LkttHm1mqW04m5fPu/gGiJ66BsEwlfrf/vY3vP76637rVq5cidzcXBw5cgS5ubkoLS3VTJ/H p03etHmylbUuS20ebWapTTYbC0NU6n19fVi8eDHWrVvnt76urg4lJSWIj49HSUkJ7Ha7ZmUw3NMm ace8nhG0i1tbkdV0JbKarmSizQIezzNLbbLZWBiiUn/22Wfxs5/9DL/85S/91judTiQlJaGoqAg2 mw1OpzNo3ws9PXBVVmKgtxd3tLQAQETL6szMqPaPZhnndjPR5c1ubvycYPfTvdWewMRuVv7m5jwb xG7D+dvn+tdV12d5vrsbrspKnGtogN4YYkS5ESNGYGBgwG+dy+VCRkYG6ipDyboAACAASURBVOvr kZycjM7OTuTk5KBV5AmJRpQjbaPpMdMOGFHuuyfeAAA0b1uvj74Ho472Rdrm0BXV9rn+ddX1gfsR 5U6cOAGXywWXywUA3mVeXh7Ky8vR19eHiooKFBQUaFYGHvNCvGmbxVb3/Plwz5+vu65SeMx38qhN NhsLQ1TqYqxduxZ2ux0pKSmora1FWVmZZlo85oV40zaLrZYsNyxZbt11lcJjvpNHbbLZWAxlXYBA hCgdANLT01FTU6OLLo9Pm7xp82Kr0Cmuedt6VGdmIks3ZX94jKJ41CabjYWhI3U94fFpkzdtnmwF BpvpXY89pruuAI9RFI/aZLOxoErdA49Pm7xp82QrMNhMb720TXddAR6jKB61yWZjQZW6Bx6fNnnT 5slWr25CHRNdgM8oikdtstlYUKXugcenTd60ebLVq+vMZ6IL8BlF8ahNNhsLqtQ98Pi0yZu26W0N MeY1ReqkbVZdltoUqWtIf38/HA4Hurq6AACTzp6NaNk0ZkxU+0ezrM7MZKLLm92m8vO45rDrxzWP G9RNTPX7zIO/TXWeY8BuQ/pb7P7Q6Tw3NjbC4XCgo6MDemOIEeXUgEaUI22j6amtLQw4431HPXDE LE+U7n2l7comFCfU4bt9iwY/04hypG0iXVFtzkeUM9x76qzgMS/Em3as2yp3wBk/XWc+vadO2qbU Zantq+t92K6qYlKWQGK++V0teMwL8abNk61eXcqpk7ZJdVlq++oqGeFRD6hS98Dj0yZv2jzZ6tWl 3u+kbVJdltrU+z0G4PFpkzdtnmz16lKkTtom1WWpTb3fYwAenzZ50+bJVq8uReqkbVJdltoUqccA PD5t8qbNk61eXYrUSdukuiy1KVKXoLq6Gjk5ObDZbMjPz8enn34KAGhra0NhYSFsNhsKCwvR3t6u XRk4fNrkTZsnW726FKmTtkl1WWpTpC7B+++/j61bt+Lo0aN44okncM899wAAVq5cidzcXBw5cgS5 ubkoLS3VrAw8Pm3yps2TrV5ditRJ26S6YtpZTVd6x2rQU9coGOI99c2bNwMAzpw5g/PnzyMhIQEA UFdXhw0bNiA+Ph4lJSXIycnRrAw8Pm3yps2TrV5dek+dtE2qy0rbaO+lB2KISB0ArFYrRo0ahWXL luHtt98GADidTiQlJaGoqAg2mw1OpzNovws9PXBVVmKgtxd3tLQAQETL4tbWqPaPZvnsl18y0eXN 7pj3c4Ldfyny/a32wYfiJW9n4/53sr2fzX5+Wevydj+x1hXzt9bX+51TazB3am3wfemz3fnubrgq K3GuoQF6Y6hhYk+fPo2PPvoIzz//PL744gtkZGSgvr4eycnJ6OzsRE5ODlpFmj2iHSaWIAxPwGQt UsPEBqL3MLEEwYKsecsBaHi9Bw5DG2JYWpbDxBoiUn/wwQfR1taGIUOGYMiQId5B8PPy8lBeXo6+ vj5UVFSgoKBAszIYLS9E2rGvp5W2kpzhrLoE1XSVQjlePrS5tJlhXxUpDFGp5+Xl4ec//znGjBmD 9evXY9OmTQCAtWvXwm63IyUlBbW1tSgrK9OsDLzlhXjU5slWgT35wSkrvaAcLx/amuuGmFJYDW33 /Pne/LhSWL5VIoUhKvV/+Zd/wbfffove3l588cUXKCoqAgCkp6ejpqYGPT092LNnD9LS0jQrA49P m7xpm9bWMD96FKmTtll1o9WOZsx2itRjAB6fsHnT5slWAYrUSdusupprh2shoEjd+MTq0yZpG1dP b+1QuXaK1NlpR9O8G622nmilK8d/kWhLHVf0e59KniL1GMC0T5ukzUzPCNoUqbPT1ntKTrP5W47/ ItGWOq4sXZ9IXY/BbpRAlboHozzdk7Z59DTXDtM8KECROmmbVVct7UhaVChSjwGM8nRP2ubRM4I2 ReqkrZuujIfMqAhxfDVsjqRFhXLqGtLf3w+Hw4Guri4AwKSzZyNaLjp8OKr9o1kWt7Yy0eXNblP4 eVwzAGBc8zi/z4HrheXcHdf6b8+Bvw11noXzY2K7/XRl2OuePx9Nq1fLO37A9R14/EWHD4c8nuT1 Huq4y+dh0pvZYfWEz4uu3RFWr7GxEQ6Hwzvmip4YakS5aKAR5QizETTGdMCIcc1XNvltL5XX425E uRAjfXGlrzdy7VW6XSCBI7kFrBdGlGuyDFbA3ihcbD8pJPbz3o8+9xf3I8oZgVjPC5G28fSi1Q5s FoykQw7l1EnbrLpA+Ny2lh0VjZxTN8QsbUaAx1wYb9o82SpAOXXSNoyuRIQcyexnrHLblFOPAXh8 wuZNO1ZsVfP9ZorUSdvousL1HklkzSpipkg9BuDxCZs37VixVc0mQ4rUIyeiyDFGrjEj6QZe716/ y7gP/CJmoSXAk1NXjYAWhqymK/EdAAT0aTEKFKl74PEJmzdtnmwVoEg9cqQix6AWleXzULx+tCra kRDr/hZQErGziphZ3ldSGKJS//3vf4/rr78eI0eORG5uLurqBk9UW1sbCgsLYbPZUFhYiPb2ds3K wOMTNm/aPNkqQJG6Coi8fx2q8mGZazWNv5VoM/I3y/tKCkNU6h9++CHeffddHD16FA8//DDuuece AMDKlSuRm5uLI0eOIDc3F6WlpZqVgccojjdtnmwV4CJSD6h0jdobW3Ntk0TqirQpUg/CEDn1rVu3 ev/Py8vDyZMn0d/fj7q6OmzYsAHx8fEoKSlBTk6OZmXgMYrjTTtmbZXIFYZ7zc1UkbrM95vZR45s 3kunSF0CFUe7o0hdJt3d3bjrrruwdOlSXHLJJXA6nUhKSkJRURFsNhuczmBHXujpgauyEgO9vbij pQUAIloWt7ZGtX80y2e//JKJLm92G93Prsceg3v+fNyRYB9cH7C81Z7gtxRb77ucVZdwcftYP7+C PyTWe3XFtleqF3ic9cnA8nkX1/t8Lk6oky5vrPhb6jpV6u9Afwb6UebS9z4pTqgTvU+kjrPk7Wxk NV2pWF+4r4L0fOw9390NV2UlzjU0QG8MM6LcN998gwULFuAXv/gFVq1ahbi4OGRkZKC+vh7Jycno 7OxETk4OWkWaemhEOSLmkYgkAiNyYUQ5uQPSxPyIciIjiAnr3c0WAMEj8EU8opuEniiB25ltRDkx u6Ts1XJceB/ERlwM/F5AbDup4wdCI8r5sHXrVixbtgybNm3C6tWrERc3WKy8vDyUl5ejr68PFRUV KCgo0KwMPOZbedPmyVaBmMypRzgxiNBxTW9dX3jKqQu9/7XWDTeSohr+Fjt+4HrfzyHvK60ntJGJ IXLqDz30EADg1ltv9a47duwY1q5di4ULF+KVV17BlClTsGXLFs3KELP5VtI2rJ4RtM2QU1fy3nIk ukqPH1abo5y611+Z4e2N1L9yWqC06P0uR5dy6hK4XK6gv8suuwzp6emoqalBT08P9uzZg7S0NM3K wGMUx5u22WyVMxa8ESN1pSPmKR1pTKmv1Rwj3MyRuth5k9LVagz2rKYrcf872ZL3gdw5E5TMrUC9 32MAHqM43rR5slXAUJG6p2ny4g98dBGtWASo1ljkkWxf7cwHVB7QTC5qX2OBI+qJVcxi51ntvgSh cuXhrm+lkx8pgSL1GMBsURxps9czgrYRI3W1EIsAhd7RLPCN1COZVS8qbUYju7G8r1hd3xSpG4xQ YzrzGMXxps2TrQKGitQDkBpbPdJcrPXSNiArRGSncS901mOCa9aHQaKFguV95Xt9a/UAFeq4FKkb DOEJ1DdHxGMUx5u22nqSuWGf3rChtNWcjU0M34hCDz1fos21RpqLFctr6+1v1ZHoXa3XmPeBCC0j QS0TIuVVswWDIvVguIzUBXxzezxGcabVFsnpqa2nJDfsqx0cgWoXQfpGFHro+aLY3yq9DiTWI1pv f2uFWAsHq3vZ2zKC6CpqqffLQ8EqYqZIXUP6+/vhcDjQ1dUFAJh09qy85bhmv+Wiw4eV7a/isri1 lYmu6exePg+T3swe/CycX639PK4Z7vnz0bR6tej3gX62ZLlxbWFT6P19rsuspitRuPtWv/Xjmscp Ws7dca3f51B+0WoZ0t8B911QecS+V7AsTqgL8pciPc91FMr/UkvB30H6Kl5v3uvH93oP4e+w16XC 38eo/R1wPftuF7g+1HaBy0B/y7kf5Bw30vsqsFyNjY1wOBzo6OiA3hhmRLloUTSinNhIUToTyXzN XBBp71mtz2tguaT0pEbekiAwcom2yVKv3LLodS1mt9yR2hQSSeQXan+BwOOInZ+gz8JIYxLXtWK/ CUiN4CahJxA0Ep9CIh3JTel1rdb9oDZCudzNFsDt5ntEOSPAIrcc9QhYKsCTtl56oXK3sdz7PdL3 ypnNGqbxu+KKRxqTSbTvc0f6fn7Uujq/m6+Wv6MhnK4W7+Qrgeucui/h8lGiT7QB30cacWv9fqme 2kqIWDvCSD7a3thyCbyplR5XzQgkqtyfxHvlUnYxmzWM4ZzmRn7bQDPdCP2txnVOOfVgTBmpR9LL NdxTrtQTrXe91Ni/It8HaUc5hrCSJ2+K1C+i1chXliw35k6tjXj/aHoLaxnJhHqLxJfi9aNlXcdq 90qPNHJUo1d2RP4OvN8jvP9l308qj1HOchQ9I0bqrDFlpB5JL1dNnnIDZ4+SOyKTjsS0tsIcZdQt A1HAakxwPSIKsftNbgSn9oOU0shRbkWu2pjgKo+4plrfnAivcx5bRkLpRtuHQy1MGalHgpKoUcn7 yYB0BGj2aDnIXx7/SGlHG8EF+l1UT6XIJVx5WUUzsiIKjWaX0stmIcIW/uTqajHim6oRnMzzYvQ+ DFqOrEeRejCmjNS9KHgiVhLFqf2+6/Ynnxw8rtwnbbH5oyNAj0j94shUyrTV9rPWs4aFe3Azc6Qu htoRnNx5sL/btwhZqirLR5G/VX6QEvsdUXMWulAEnmc9I1YjRepGwRCRutVq9f750tbWhsLCQths NhQWFqK9vV2zMvg95eo8L+7cqbXhbziR8kjmgGXYIevp3nMcpS0Ueo2AJYXa8z5HknuXEz0yjxy1 yrVqdD+J+SvQZj3HYJ9Vl6D7mO8CYr8javcVCbSPcurGwhCRusvlAoCgSn3lypXIzc3Fhx9+iHXr 1qG0tBSbN2+OWEfsiVXrJ1kpxCIaxeWK4IczmhaKaP0mRBbe40u1OERYMQjlcz8ps2VDg1mmTBWp i5yHwOsh2l7Rkb6PLGazHhGkmVpGpBD8+R0QsmVEjwcbitSDMUSkLkZdXR1KSkoQHx+PkpIS2O32 qI4Xrvd6NL2TlRIY8Yo96cp9wpadew4RMRW3tgatl3u8aCMAIbJQo8VBiZ7s46oYYRoxp652r/Og PgxR2hxpxCsVRak1/7Zcbb3G3Nf7/XwBllErRerBGLpSdzqdSEpKQlFREWw2G5zO4KejCz09cFVW YqC3F3e0tAAA7kiw+y/F1vssq535wevXJwPL54nvJ+O4oZZ3Tq2BJcvt/RyHAQAXJ0ZQqhd4PKn9 vDotLajOzAzaznu8SP0p0y+C3dH6U7FepDpRLKud+ZK6t9oTVF/uyXd6Py95e3Do06DzLPe6l7of pM6vzGW0dscNymriT6mlr7+FpeT9qdJSq+tbuG5ixd+sde9IsON8dzdclZU419AAvTHUMLFWq9Xb FA8AGRkZqK+vR3JyMjo7O5GTk4NWkbyo3zCxYsN3hsvvJtQpb75SaXhL2dpqD6e5fhuKW1tR/cYT 6h5f5nGC7I5UX6meRsOSSmov/yGsrhbNlbPqEoKaCpssg+NTa51uEruu1RouVIxQNutFKG29XnES /C31Cq1c5J4Po/mbtW7zlU1wN4HZMLGGyKmLkZeXh/LycixfvhwVFRUoKCiI7EAyfrwjyUeplYuX q61F7l+L3u9yy6laDlBm5ezV07Ey99NeLr2d2oT64dGr74hYr2itYZnv1FJbqk+A4G+9KnMBs/rb iLpyMETzu2/Pd9//165dC7vdjpSUFNTW1qKsrEyzMkSSj1KrV6lcbS1GPNOiB7rccgbZrfFbByx7 6Roxp641PNqsh7ZYbpv8bX5dORiq+T0a1qxZg9+991vWxSCIiNErkmU14pXRZtXSC7X8Lfc9fbV1 CGWwbn43RKRuBHiM4njT5slWAYrUY19bbm98VmPem83fRtaVA0XqBMEIVhGRXpG62vPBxypy/S2W M5fym1rzxRPqQJG6QeAxiuNNmydbBSiKMr622PzsSiNoVmPex5q/Y1lXDhSpEwQjzBqpUwToj5S/ o/WX2PGppYQNFKkbBB6jON60ebJVQI2IQquR3bTCqJFjYASudsR8/zvZIY+n9Vj0RvU3K13WD08x X6n39/fD4XCgq6sLADBpXHNEy6bE1Kj2j2ZZ7cxnosub3Ub087jmcZouW1ITQ66XU+6spitRuPtW yf18t/O1a0++U3P7Qi1Z6crxt1a6k8Y1+9mt1/VlVH+z1m1sbITD4UBHRwf0hprfPUQ0opxKkLY6 SA3Ocf872diT7xTtiKS0WVpqZDTfz7PqErBp0b6Q+2mJnBHO5HbQUtqBy4ijfbHQ1rr5Wzh+4PWt 58x0RvK3EXSb0MSs+Z3LSl3uMJVS66WOJxex3JfcXJnWaKUnVZmoVcnK1ZPaL/B7pcOdGiW3GWk5 lFbqBMErvpX6ByUlmLViBUakpmqqGRPDxEZKJJVscUIdvpM4TuB6qUpXjMAfVeGpL3A/pZWUVGUS iuKEOny3b5Hkdmrp+RJot1qVsNh2UnpSumLHlVovaDeLfqsdoSKKSCtjpfsZNYoibXPostRWotts t8Px178i51e/wrT77sPQ+HhNy2bKSF3rCSPMilp+In8TBMEzvpH6hkmTvOtHpKVh1ooVyIp0HpMw xHzze2NjI959913v56+++gpTp06N+HjnW1owdNw4NYpG2gbV5slW1rostXm0maU22Rweq6eyBYAR qamDlfqMGaqXKeYr9UDWrFmDp556KuL9z372GeJvuUXFEpG20bR5spW1LkttHm1mqU02h2fDpEkY Gh+Paffdh5xf/Uqz5nd6Tz2A/kOHSNvk2jzZylqXpTaPNrPUJpvDk1VQgHv/+EfcUlKieT4dAIYs Xbr0GQAYOXKk5mJaUldXF/l86wCGpqYibtgwFUtE2kbT5slW1rostXm0maU22Ryeq2+/HfEjRmhc IqChoQGAiZrfGxoacM0117AuBkEQBEHojuma36lCJ6xWK+siECbDarV6/wgiFoipSr21tRUzZszA yJEjMWPGDDgcDsl91LoZrVYrXnnlFdWPK1f78ssvx/jx4/Fv//ZvOHPmjG7agj6LHzZfXT30rVYr Nm7cCAB49dVXdbe3oaEBVqvV24ymB6xtNsK1FQ6XywWXy6WqNovzDACff/45ZsyYAZvNhsmTJ+O/ //u/ddHl9XebFTFVqa9YsQK33347Ojo6MGfOHKxYsUJX/S1btug+9J/AiRMnUFtbi56eHjz99NO6 amvxw6ZUW68ybNmyBW63G1u2bNFcK5Ddu3cjLi4Ou3fv1lWXpc2sri2W1zSr87xgwQIsXrwY7e3t +NOf/oSamhpddHn+3WZBTFXqn3zyCX71q1/hJz/5CRYvXoxPPvkEALB//37k5uYiISHB70lM+F+t SOCuu+7C66+/7rfuiy++wLRp05CYmIhp06bhiy++AABMnz7de9Ps2bMHubm5UWlbLBaMHTsWa9as wR//+EcAg+/q33rrrRg5ciSmTJmCTz/9FIC4P9RE8GliYiJuueUWr61WqxUVFRVIT09HWlqaJhWF mN0A8PTTT2PUqFHIzs7Gvn37FB975MiReOGFF5CUlORdJ2ar8N3TTz+N0aNHY/r06VHZ9ec//xl3 3303/vznP/sdv7S0FMnJyUE2qaWtxOaZM2diz549AIA//OEPmD17dsS6YgRes773sZbXlpiu2oid 51DaX331FW688UYkJydj1apVUZXpsssug8PhwLfffovk5GS88cYbAMTvp3DXnhJ4/t1mQUxV6idP nkRCwuCUd4mJiThx4gQAYOnSpViwYAF++OEHv6dv4X+1nsqXLl2Kd99916sLACUlJXjkkUdw9OhR lJSU4OGHHwYw+FS8detWAMDWrVuxYMGCqPUBYMyYMd4Z6e6//34sWbIEHR0deOGFF1BSUuItZyh/ qIng066uLrz44otYsmSJ97uhQ4fi4MGDqKysxKpVq6LSCdX8LmY3AGRlZeHw4cNYtmwZHnroIcV6 ixcvRllZGRYvXuxdF85WAEhOTkZLSwumTZsWoZXA6dOnsXfvXjz11FPYu3cvTp8+7f1u/PjxcDgc WLZsmZ+tamkrsbmkpMR7Xb/yyit49NFHI9aNBDWvLRaEO8+hKCkpwYMPPojDhw8jKysrKu33338f P/zwAx5//HFMnDgRVVVVAMLfT+GuPbnQ77a+xFTv97S0NPztb39DYmIiuru7MXXqVBw+fBiJiYk4 cuQIhoV4xcBqtapyYQjHeemll+ByufD888/D5XIhISEBHR0d+OlPf4rTp09j7Nix6O3tRW9vL667 7jp8+umnmD59Or777jvvhR2pNgAcOXIEM2bMwKFDh5CYmIjz5897t7NYLDh58mRYf0TKhQsXYLPZ 4HQOjnf81ltv4eWXX0Z7ezsGBga82larFb29vRg6dGhQ2ZUitq+Y3VarFZ2dnfjpT3+KM2fOYOzY sejp6YlYT/gsZquwTU9PDy655JKIbBTYuXMn5s2b5/28bds2zJkzJ6xNamgrtfn8+fPIzs7G6tWr UVZWhr1798JisURuOIKvLd8ynT9/HomJiXC5XKpeW0p01dASCHeeQ2knJibi6NGj3vM/atQoVcpx 4MAB3H777WhtbdXsfhLg9Xdbb2Ky9/stt9yCt99+G319fdi0aZO3yXHixInYunUrzp07F7TPpZde KqtjhlweeOAB7xMuMPgwtG3bNpw5cwZVVVXeh6PExETceuutuPvuuzF79uyoLwy3242jR4/imWee QXFxMQDg+uuvx7Zt23D8+HG4XC5vRRPOH0opLS3FyZMn8fHHHyMtLc27fuXKlXj55Zdx7NgxvPfe e3C73d7vhB9drRCzG4D3XLz33nu46qqrVNELZyuAqCt0YDDP+uyzz8LlcuHZZ5/1y7f62hT48K2G dijEbB46dCjuvfdePPDAA3jkkUeiqtDFri2bzYaPP/4YZ8+exebNm/32UePaikR3zJgx+Oabb6LW FjvPYtoTJ07E73//e5w5cwZ/+MMfotJevHgxWlpa0NfXh4aGBly4cAGA/Psp0sCP599tFsRUpb5u 3Trs2LEDKSkp+Oijj7B27VoAQEVFBbZs2YLk5OSgHMySJUuQnZ2tWn5s2LBhfs2Ur732Gl599VWk pKTgtddew2uvveb97p577sE333yjShPOiBEjkJeXh+HDh2PNmjUABqPljRs3Ii0tza95Opw/lHLF FVfgmmuuwQMPPODX3Pn4449j0aJFmDBhAv7v//4vKg2liNkNDOYH09PT8dJLLwXl0SJFD1t3796N oqIiAEBRUZFfpe5rk+/1pSXhbF64cCHi4+Mxf/78qDTErq1nnnkGCxcuxFVXXeWteNQkEt2VK1fi tttui/p+EjvPYtqvv/46Nm7ciPT0dBw8eBBxcZH/ZM+cORNz587FmDFjsH79elRUVACQfz9Feu3x /LvNgphqficI3lCrGVItBgYGsGnTJrS0tGDdunWsi8MNAwMD+Oijj7Bq1aqIO6wpxWjXHhEeU8+n ThCENowYMQKTJ0/GBx98wLoo3GC1WmGxWJCZmala6xNhXqhSJwgDY7RIyWjl4QGWY0QQsUdM5dQJ giAIghCHKnWCIAiCMAlUqRMEQRCESaBKnSAIgiBMAlXqBEEQBGESour9LmdgAOpByZef9LDVSO/P qmWvHJuEbXw1jeKHcOjpIzUx0nUWjlj1b7hySKFlOd955x2sWLECP/74oyIdo/gv6lfastzikww0 W5qjPXxEGMW5fmS5xb9rjm7sbKMRxlKoYanhzq3P8JNByBx1TYlNvuORR8OXX36J5cuXe2dd++ij j/Diiy/i66+/xrBhw1BQUIDnnnsOGRkZUekAgDvMVWGReVXofd7F9OT8vujpWwBw/5f4d5a75B3D SPdVfZjvbtRYe+3atdi1axcmT57st17qvBvFf9T8ThCcsnHjRjzyyCPez2+88QaeeOIJOBwOHDhw ADfccEPUQ8HyCvk2dvnhhx+CKvRYgmmlvn37dtx4441ITEz0G3O4q6sL6enp3mkJ7XY78vLyJPcT +1/g2LFjmDt3Lmw2W9A83FarFQUFBVi0aBGuu+46PPjgg5raroRw8zyL+QIIb69wHLXmAlcLq9WK N998E+np6cjIyPCbxCJwClZfTp48iaVLl4Ycv1rKDywJZ9P333+P6dOnY9SoUYqmGZVjr8PhwP79 +72TAwFAdXU1brvtNlx22WW4/PLLsXTpUrS0tERmmIqE89GhQ4eQn5+PkSNHIj8/H4cOHUJ3dzfG jRsXNK3p6dOnkZWVhe7ubtH54sPpSf2+CMSSb4Hw/hW+D/U7cejQIe/1uXr1as3mn1dKuN/EUNeL gLCt2+1WVK8YzX9MK/UHH3wQGzdu9M4OJDRfJCUl4eabb/b+oG/evBmLFi2S3E/sf4Fly5bhrrvu wtGjR4PmDQaAF198Ee+99x42b96MP/3pT5rZrSZivgCk7RVoamqKaj5utXE6nfj73/+Ol156CU8/ /bR3fbj5lR999FFYLBZ8/fXXEfuBBeFseuyxx/CLX/wCbW1tiqbRlWNveXk5lixZgiFDhogeZ/Xq 1bjzzjtl62qF1HkvLi5GR0cHbr/9djz66KOw2WzIz88PGsr2gw8+QH5+Pmw2m+h88eH0pH5fBGLJ t4D8ecsDfycef/xx3HnnnWhra0N8fLyWRVREuN/EUNeLgJy6JJSvm5wt6gAABOhJREFUjOa/qCZ0 sVqtkjn1cMb+4z/+I7q7uzFjxgxkZ2fj5z//uffHa8eOHXj55ZdRVVWFG264AQ0NDRg+fLjkfkK5 QumOHj3a7+k9cF7sEydOYMSIEThx4gQSEhL8piCMBqvVKplTD+cnsTmvgfC+CGevcJyOjg6vX9XA arVK5tSlbP3hhx8wbNgwXLhwAYmJiUHnIdT5HTNmDA4ePIgRI0YEHVPKD9FgtVolc+qRdlIaNWoU mpqaYLVacfLkSYwdO1b0OvBFyl6Xy4Vp06ahvr4+5Lk/c+YMHnnkEfz973/Hhx9+GNKnShi8JsLn 1NXy0fjx49HZ2Ymamho899xzePfddzFhwgQ0Nzdj/vz5KC0txcyZM/G///u/+M1vfoPvv/8eFy5c CHlNiPk3XG5Vb98K5ZHKqUfTUU7sd0Lq+owUq9UqmVOPtF4Ru14C9ZWed6n99PCfIeZT/+CDD7B6 9WqMHDkSlZWVmDt3rve7OXPmoKmpCaWlpfinf/onP4eE2y8ccXFxfk9vgTexMK1hXFxc0JzZLLFY LN7pGAObFMP5QspeAKpW6Goh3IBDhgxR5TzI8UOsEso/UvZu3rwZd955Z8hz39jYiJkzZ2Lo0KHY tWuXKpWOlohdHzNmzMCxY8dQXl6OgYEBvP766+js7MSMGTMAAA899BBWrVqFzs5OdHR0qHa/m8m3 vhjxd0KMcL+JrH7X9fQf00r9kksuQVFRER577DGsXLnSb97mSy65BPPnz8d//ud/Bs1rG24/YLBS CDXv9cyZM1FWVoYff/xRG4M0IjU1Fdu3b8fp06eDZmkK54tYtTcSbr/9dqxYsQLd3d1B38WqH266 6SZs2rQJ586dw5tvvhn0fUpKCvbu3Ru0Ppy9Fy5cwFtvvRWyz8i7776LuXPn4sknn8Qbb7yhqMmf FdOmTcNbb72Fvr4+vPnmm97mTYvFgoULF+KVV17B/Pnz8dJLL2HhwoWwWAZ72p89exZJSUk4d+4c 1q9fr0hT7PfFbL6VIicnB2+99RbOnTuHf//3f2ddHC/hfhPFrhc5iJ33SNHKf0wrdaFzgc1mw7/+ 67+ioqLC7/vCwkKMGzcO//AP/6Bov5KSEhQUFAR1PHjllVfw/fffY9y4cWE7NhiNZ555BsuWLcPP fvYzJCQk+H0Xzhexaq8Y4TqsvPTSSxgYGMD1118f9J2R/RDOphdeeAFVVVUYO3YsTp06FbTv008/ jeLiYkXX+fbt2zFlyhSkp6cHHe+BBx5Ae3s7Fi5c6Nf5h/XDkNR5f//99zFmzBhs374dL774ove7 u+++G5deeinWr1+P+Ph43HXXxXe71q1bh1/+8pe4+uqrkZqaKlsPEP99iUXfAvI7AAayYcMGbNu2 DWPHjg1qQWRJuN/EcNeLFGLn3Wj+izqnLkWkOYLz58/j/vvvxzXXXIMnn3wyomMYBS39ZDR4shWI PXtnzZqF3/72t7p2jIw1H0UKC98CxvGvVM5ZyXGkMMP1Eki0/hNy6lFV6lpitVoxdepUVFdXx1Tu iSAIgkfUqtR5Ra1KPeoR5bSCLg6CIIjYgX6zo0Mt/9GIcgRBEARhEqhSJwiCIAiTQJU6QRAEQZgE qtQJgiAIwiRQpU4QBEEQJsHy9ddfu0+cOIHmZjZznxMEQRAEoQ5xVKETBEEQhDn4/4AYmMnejagV AAAAAElFTkSuQmCC --=-=-= (I haven't collected steal values, so the numbers don't sum up to 100%.) I'd be grateful if this discrepancy could be cleared up eventually! It's heartening to see some progress after more than three years. :) Actually, as Munin doesn't half the idle and iowait values, but truncates the (then overflowing) graph at 100%, I was rather surprised to see iowait completely disappear after the kernel upgrade, and concluded that it was somehow converted into buggy-looping in blkfront. Now I see this isn't the case, but the steadily increasing system CPU usage between reboots is still a mystery. I'll start a separate thread for that, just wanted to provide some motivation for this topic. -- Thanks, Feri. --=-=-=--