All of lore.kernel.org
 help / color / mirror / Atom feed
diff for duplicates of <571A60EB.5030603@linux.vnet.ibm.com>

diff --git a/a/1.1.hdr b/a/1.1.hdr
deleted file mode 100644
index 1067781..0000000
--- a/a/1.1.hdr
+++ /dev/null
@@ -1,2 +0,0 @@
-Content-Type: text/plain; charset=windows-1252; format=flowed
-Content-Transfer-Encoding: 7bit
diff --git a/a/1.2.hdr b/a/1.2.hdr
deleted file mode 100644
index a140fea..0000000
--- a/a/1.2.hdr
+++ /dev/null
@@ -1,2 +0,0 @@
-Content-Type: text/html; charset=windows-1252
-Content-Transfer-Encoding: quoted-printable
diff --git a/a/1.1.txt b/N1/1.txt
similarity index 100%
rename from a/1.1.txt
rename to N1/1.txt
diff --git a/a/1.2.bin b/N1/2.bin
similarity index 100%
rename from a/1.2.bin
rename to N1/2.bin
diff --git a/a/2.hdr b/N1/2.hdr
index 5216513..c1a262e 100644
--- a/a/2.hdr
+++ b/N1/2.hdr
@@ -1,4 +1,2 @@
-Content-Type: text/plain; charset="utf-8"
-MIME-Version: 1.0
-Content-Transfer-Encoding: base64
-Content-Disposition: inline
+Content-Type: text/html; charset=windows-1252
+Content-Transfer-Encoding: 8bit
diff --git a/a/2.txt b/a/2.txt
deleted file mode 100644
index d258c5f..0000000
--- a/a/2.txt
+++ /dev/null
@@ -1,4 +0,0 @@
-_______________________________________________
-Linuxppc-dev mailing list
-Linuxppc-dev@lists.ozlabs.org
-https://lists.ozlabs.org/listinfo/linuxppc-dev
diff --git a/a/content_digest b/N1/content_digest
index 28971ed..4a347c9 100644
--- a/a/content_digest
+++ b/N1/content_digest
@@ -11,7 +11,7 @@
   linux-kernel@vger.kernel.org
  " linuxppc-dev@lists.ozlabs.org\0"
  "Cc\0ego@linux.vnet.ibm.com\0"
- "\02:1.1\0"
+ "\01:1\0"
  "b\0"
  "Hi Stewart,\n"
  "\n"
@@ -207,260 +207,254 @@
  "> in ms?\n"
  "\n"
  Yes its 2000 ms.
- "\02:1.2\0"
- "b\0"
- "<html>\r\n"
- "  <head>\r\n"
- "    <meta content=\"text/html; charset=windows-1252\"\r\n"
- "      http-equiv=\"Content-Type\">\r\n"
- "  </head>\r\n"
- "  <body text=\"#000000\" bgcolor=\"#FFFFFF\">\r\n"
- "    <pre>Hi Stewart,\r\n"
- "\r\n"
- "</pre>\r\n"
- "    On 04/20/2016 03:41 AM, Stewart Smith wrote:<br>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <pre wrap=\"\">Akshay Adiga <a class=\"moz-txt-link-rfc2396E\" href=\"mailto:akshay.adiga@linux.vnet.ibm.com\">&lt;akshay.adiga@linux.vnet.ibm.com&gt;</a> writes:\r\n"
- "</pre>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">Iozone results show fairly consistent performance boost.\r\n"
- "YCSB on redis shows improved Max latencies in most cases.\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">What about power consumption?\r\n"
- "\r\n"
- "</pre>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">Iozone write/rewite test were made with filesizes 200704Kb and 401408Kb\r\n"
- "with different record sizes . The following table shows IOoperations/sec\r\n"
- "with and without patch.\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">Iozone Results ( in op/sec) ( mean over 3 iterations )\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">What's the variance between runs?\r\n"
- "</pre>\r\n"
- "    </blockquote>\r\n"
- "    <pre>Re-Ran Iozone test\r\n"
- "\r\n"
- "w/o : without patch,  w : with patch , stdev : standard deviation , avg ; average \r\n"
- "\r\n"
- "Iozone Results for ReWrite\r\n"
- "+----------+--------+-----------+------------+-----------+-----------+---------+\r\n"
- "| filesize | reclen |  w/o(avg) | w/o(stdev) |   w(avg)  |  w(stdev) | change% |\r\n"
- "+----------+--------+-----------+------------+-----------+-----------+---------+\r\n"
- "|  200704  |   1    |  795070.4 |  5813.51   |  805127.8 |  16872.59 |  1.264  |\r\n"
- "|  200704  |   2    | 1448973.8 |  23058.79  | 1472098.8 |  18062.73 |  1.595  |\r\n"
- "|  200704  |   4    |  2413444  |  85988.09  | 2562535.8 |  48649.35 |  6.177  |\r\n"
- "|  200704  |   8    |  3827453  |  87710.52  | 3846888.2 |  86438.51 |  0.507  |\r\n"
- "|  200704  |   16   | 5276096.8 |  73208.19  | 5425961.6 | 170774.75 |  2.840  |\r\n"
- "|  200704  |   32   | 6742930.6 |  22789.45  | 6848904.4 | 257768.84 |  1.571  |\r\n"
- "|  200704  |   64   | 7059479.2 | 300725.26  |  7373635  | 285106.90 |  4.450  |\r\n"
- "|  200704  |  128   | 7097647.2 | 408171.71  |  7716500  | 266139.68 |  8.719  |\r\n"
- "|  200704  |  256   |  6710810  | 314594.13  | 7661752.6 | 454049.27 |  14.170 |\r\n"
- "|  200704  |  512   | 7034675.4 | 516152.97  | 7378583.2 | 613617.57 |  4.888  |\r\n"
- "|  200704  |  1024  | 6265317.2 | 446101.38  | 7540629.6 | 294865.20 |  20.355 |\r\n"
- "|  401408  |   1    |  802233.2 |  4263.92   |   817507  |  17727.09 |  1.903  |\r\n"
- "|  401408  |   2    | 1461892.8 |  53678.12  |  1482872  |  45670.30 |  1.435  |\r\n"
- "|  401408  |   4    | 2629686.8 |  24365.33  | 2673196.2 |  41576.78 |  1.654  |\r\n"
- "|  401408  |   8    | 4156353.8 |  70636.85  | 4149330.4 |  56521.84 |  -0.168 |\r\n"
- "|  401408  |   16   |  5895437  |  63762.43  | 5924167.4 | 396311.75 |  0.487  |\r\n"
- "|  401408  |   32   | 7330826.6 | 167080.53  | 7785889.2 | 245434.99 |  6.207  |\r\n"
- "|  401408  |   64   | 8298555.2 | 328890.89  | 8482416.8 | 249698.02 |  2.215  |\r\n"
- "|  401408  |  128   | 8241108.6 | 490560.96  |  8686478  | 224816.21 |  5.404  |\r\n"
- "|  401408  |  256   | 8038080.6 | 327704.66  | 8372327.4 | 210978.18 |  4.158  |\r\n"
- "|  401408  |  512   | 8229523.4 | 371701.73  | 8654695.2 | 296715.07 |  5.166  |\r\n"
- "+----------+--------+-----------+------------+-----------+-----------+---------+\r\n"
- "\r\n"
- "Iozone results for Write \r\n"
- "+----------+--------+-----------+------------+-----------+------------+---------+\r\n"
- "| filesize | reclen |  w/o(avg) | w/o(stdev) |   w(avg)  |  w(stdev)  | change% |\r\n"
- "+----------+--------+-----------+------------+-----------+------------+---------+\r\n"
- "|  200704  |   1    |   575825  |  7,876.69  |  569388.4 |  6,699.59  |  -1.12  |\r\n"
- "|  200704  |   2    | 1061229.4 |  7,589.50  | 1045193.2 | 19,785.85  |  -1.51  |\r\n"
- "|  200704  |   4    |  1808329  | 13,040.67  | 1798138.4 | 50,367.19  |  -0.56  |\r\n"
- "|  200704  |   8    | 2822953.4 | 19,948.89  | 2830305.6 | 21,202.77  |   0.26  |\r\n"
- "|  200704  |   16   |  3976987  | 62,201.72  | 3909063.8 | 268,640.51 |  -1.71  |\r\n"
- "|  200704  |   32   | 4959358.2 | 112,052.99 |  4760303  | 330,343.73 |  -4.01  |\r\n"
- "|  200704  |   64   | 5452454.6 | 628,078.72 | 5692265.6 | 190,562.91 |   4.40  |\r\n"
- "|  200704  |  128   | 5645246.8 | 10,455.85  | 5653330.2 | 18,153.76  |   0.14  |\r\n"
- "|  200704  |  256   | 5855897.2 | 184,854.25 |  5402069  | 538,523.04 |  -7.75  |\r\n"
- "|  200704  |  512   |  5515904  | 326,198.86 | 5639976.4 |  8,480.46  |   2.25  |\r\n"
- "|  200704  |  1024  | 5471718.2 | 415,179.15 | 5399414.6 | 686,124.50 |  -1.32  |\r\n"
- "|  401408  |   1    |  584786.6 |  1,256.59  |  587237.2 |  6,552.55  |   0.42  |\r\n"
- "|  401408  |   2    | 1047018.8 | 26,567.72  | 1040926.8 | 16,495.93  |  -0.58  |\r\n"
- "|  401408  |   4    | 1815465.8 | 16,426.92  | 1773652.6 | 38,169.02  |  -2.30  |\r\n"
- "|  401408  |   8    |  2814285  | 27,374.53  |  2756608  | 96,689.13  |  -2.05  |\r\n"
- "|  401408  |   16   |  3931646  | 129,648.79 | 3805793.4 | 141,368.40 |  -3.20  |\r\n"
- "|  401408  |   32   | 4875353.4 | 146,203.70 |  4884084  | 265,484.01 |   0.18  |\r\n"
- "|  401408  |   64   | 5479805.8 | 349,995.36 | 5565292.2 | 20,645.45  |   1.56  |\r\n"
- "|  401408  |  128   |  5598486  | 195,680.23 |  5645125  | 62,017.38  |   0.83  |\r\n"
- "|  401408  |  256   |  5803148  | 328,683.02 |  5657215  | 20,579.28  |  -2.51  |\r\n"
- "|  401408  |  512   | 5565091.4 | 166,123.57 | 5725974.4 | 169,506.29 |   2.89  |\r\n"
- "+----------+--------+-----------+------------+-----------+------------+---------+\r\n"
- "\r\n"
- "</pre>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <pre wrap=\"\">\r\n"
- "</pre>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">Tested with YCSB workload (50% update + 50% read) over redis for 1 million\r\n"
- "records and 1 million operation. Each test was carried out with target\r\n"
- "operations per second and persistence disabled.\r\n"
- "\r\n"
- "Max-latency (in us)( mean over 5 iterations )\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">What's the variance between runs?\r\n"
- "\r\n"
- "std dev? 95th percentile?\r\n"
- "\r\n"
- "</pre>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">---------------------------------------------------------------\r\n"
- "op/s    Operation       with patch      without patch   %change\r\n"
- "---------------------------------------------------------------\r\n"
- "15000   Read            61480.6         50261.4         22.32\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">This seems fairly significant regression. Any idea why at 15K op/s\r\n"
- "there's such a regression?\r\n"
- "</pre>\r\n"
- "    </blockquote>\r\n"
- "    <pre>Just Re-Ran the test for power numbers. \r\n"
- "Results for YCSB+Redis test.\r\n"
- "P95 : 95 Percentile \r\n"
- "P99 : 99 Percentile\r\n"
- "\r\n"
- "Power numbers are taken for one run of YCSB+redis test which has 50% Read + 50% Update.\r\n"
- "Maximum Latency has clearly gone down for all cases will less than 5% increase in power.\r\n"
- "\r\n"
- "\r\n"
- "+------------+----------+--------+------------+---------+---------+----------------+\r\n"
- "|   Op/sec   | Testcase | AvgLat |   MaxLat   |   P95   |   P99   |     Power      |\r\n"
- "+------------+----------+--------+------------+---------+---------+----------------+\r\n"
- "|   15000    |   Read   |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  |  51.8  |  127903.0  |   55.8  |  145.2  |     602.7      |\r\n"
- "| w/o patch  |  StdDev  | 5.692  | 105355.497 |  11.232 |   2.04  |      5.11      |\r\n"
- "| with patch | Average  | 53.28  |  30834.2   |   72.2  |  151.2  |     629.01     |\r\n"
- "| with patch |  StdDev  | 2.348  |  8928.323  |  15.74  |  3.544  |      3.25      |\r\n"
- "|     -      | <b>Change%  |  2.86  |   -75.89   |  29.39  |   4.13  | 4.36535589846</b>  |\r\n"
- "|   25000    |   Read   |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 53.78  |  123743.0  |   85.4  |  152.2  |     617.95     |\r\n"
- "| w/o patch  |  StdDev  | 4.593  |  80224.53  |  5.886  |   4.49  |      1.32      |\r\n"
- "| with patch | Average  | 49.65  |  84101.4   |   84.2  |  154.4  |     651.64     |\r\n"
- "| with patch |  StdDev  | 1.658  | 72656.042  |  4.261  |  2.332  |      8.76      |\r\n"
- "|     -      |<b> Change%  | -7.68  |   -32.04   |  -1.41  |   1.45  |  5.4518974027 </b> |\r\n"
- "|   35000    |   Read   |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 56.07  |  57391.0   |   93.0  |  147.6  |     636.39     |\r\n"
- "| w/o patch  |  StdDev  | 1.391  | 34494.839  |  1.789  |  2.871  |      2.92      |\r\n"
- "| with patch | Average  | 56.46  |  39634.2   |   95.0  |  149.2  |     653.44     |\r\n"
- "| with patch |  StdDev  | 3.174  |  6089.848  |  3.347  |   3.37  |      4.4       |\r\n"
- "|     -      |<b> Change%  |  0.69  |   -30.94   |   2.15  |   1.08  |  2.6791747199 </b> |\r\n"
- "|   40000    |   Read   |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  |  58.6  |  80427.8   |   97.2  |  147.4  |     636.85     |\r\n"
- "| w/o patch  |  StdDev  | 1.105  | 59327.584  |  0.748  |  2.498  |      1.51      |\r\n"
- "| with patch | Average  | 58.76  |  45291.8   |   97.2  |  149.0  |     656.12     |\r\n"
- "| with patch |  StdDev  | 1.675  | 10486.954  |  2.482  |  3.406  |      6.97      |\r\n"
- "|     -      | <b>Change%  |  0.27  |   -43.69   |   0.0   |   1.09  |  3.0258302583</b>  |\r\n"
- "|   45000    |   Read   |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 69.02  |  120027.8  |  102.6  |  149.6  |     640.68     |\r\n"
- "| w/o patch  |  StdDev  |  0.74  | 96288.811  |  1.855  |  1.497  |      7.65      |\r\n"
- "| with patch | Average  | 69.65  |  98024.6   |  102.0  |  147.8  |     653.09     |\r\n"
- "| with patch |  StdDev  |  1.14  | 78041.439  |   2.28  |  1.939  |      3.91      |\r\n"
- "|     -      <b>| Change%  |  0.92  |   -18.33   |  -0.58  |   -1.2  | 1.93700443279</b>  |\r\n"
- "|   15000    |  Update  |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 48.144 |  86847.0   |   52.4  |  189.2  |     602.7      |\r\n"
- "| w/o patch  |  StdDev  | 5.971  | 41580.919  |  16.427 |  8.376  |      5.11      |\r\n"
- "| with patch | Average  | 47.964 |  31106.2   |   58.4  |  182.2  |     629.01     |\r\n"
- "| with patch |  StdDev  | 3.003  |  4906.179  |  7.088  |  6.177  |      3.25      |\r\n"
- "|     -      | <b>Change%  | -0.37  |   -64.18   |  11.45  |   -3.7  | -3.69978858351</b> |\r\n"
- "|   25000    |  Update  |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 51.856 |  102808.6  |   87.0  |  182.4  |     617.95     |\r\n"
- "| w/o patch  |  StdDev  | 5.721  | 79308.823  |  4.899  |  7.965  |      1.32      |\r\n"
- "| with patch | Average  | 46.07  |  74623.0   |   86.2  |  183.0  |     651.64     |\r\n"
- "| with patch |  StdDev  | 1.779  | 77511.229  |  4.069  |  7.014  |      8.76      |\r\n"
- "|     -      | <b>Change%  | -11.16 |   -27.42   |  -0.92  |   0.33  | 0.328947368421</b> |\r\n"
- "|   35000    |  Update  |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 54.142 |  51074.2   |   93.6  |  181.8  |     636.39     |\r\n"
- "| w/o patch  |  StdDev  | 1.671  | 36877.588  |  1.497  |  8.035  |      2.92      |\r\n"
- "| with patch | Average  | 54.034 |  44731.8   |   94.4  |  184.4  |     653.44     |\r\n"
- "| with patch |  StdDev  | 3.363  |  13400.4   |   1.02  |  7.172  |      4.4       |\r\n"
- "|     -      |<b> Change%  |  -0.2  |   -12.42   |   0.85  |   1.43  |  1.4301430143</b>  |\r\n"
- "|   40000    |  Update  |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 57.528 |  71672.6   |   98.4  |  184.8  |     636.85     |\r\n"
- "| w/o patch  |  StdDev  | 1.111  | 63103.862  |  1.744  |  9.282  |      1.51      |\r\n"
- "| with patch | Average  | 57.738 |  32101.4   |   98.0  |  186.4  |     656.12     |\r\n"
- "| with patch |  StdDev  | 1.294  |  4481.801  |  1.673  |   7.71  |      6.97      |\r\n"
- "|     -      | <b>Change%  |  0.37  |   -55.21   |  -0.41  |   0.87  | 0.865800865801 </b>|\r\n"
- "|   45000    |  Update  |   -    |     -      |    -    |    -    |       -        |\r\n"
- "| w/o patch  | Average  | 69.97  |  117183.0  |  105.4  |  182.4  |     640.68     |\r\n"
- "| w/o patch  |  StdDev  | 0.925  | 99836.076  |   1.2   |  9.091  |      7.65      |\r\n"
- "| with patch | Average  | 70.508 |  104175.0  |  103.2  |  185.4  |     653.09     |\r\n"
- "| with patch |  StdDev  | 1.463  |  74438.13  |   1.47  |  7.915  |      3.91      |\r\n"
- "|     -      |<b> Change%  |  0.77  |   -11.1    |  -2.09  |   1.64  | 1.64473684211  </b>|\r\n"
- "+------------+----------+--------+------------+---------+---------+----------------+\r\n"
- "</pre>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <pre wrap=\"\">\r\n"
- "</pre>\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">--- a/drivers/cpufreq/powernv-cpufreq.c\r\n"
- "+++ b/drivers/cpufreq/powernv-cpufreq.c\r\n"
- "@@ -36,12 +36,56 @@\r\n"
- " #include &lt;asm/reg.h&gt;\r\n"
- " #include &lt;asm/smp.h&gt; /* Required for cpu_sibling_mask() in UP configs */\r\n"
- " #include &lt;asm/opal.h&gt;\r\n"
- "+#include &lt;linux/timer.h&gt;\r\n"
- " \r\n"
- " #define POWERNV_MAX_PSTATES\t256\r\n"
- " #define PMSR_PSAFE_ENABLE\t(1UL &lt;&lt; 30)\r\n"
- " #define PMSR_SPR_EM_DISABLE\t(1UL &lt;&lt; 31)\r\n"
- " #define PMSR_MAX(x)\t\t((x &gt;&gt; 32) &amp; 0xFF)\r\n"
- " \r\n"
- "+#define MAX_RAMP_DOWN_TIME\t\t\t\t5120\r\n"
- "+/*\r\n"
- "+ * On an idle system we want the global pstate to ramp-down from max value to\r\n"
- "+ * min over a span of ~5 secs. Also we want it to initially ramp-down slowly and\r\n"
- "+ * then ramp-down rapidly later on.\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">Where does 5 seconds come from?\r\n"
- "\r\n"
- "Why 5 and not 10, or not 2? Is there some time period inherit in\r\n"
- "hardware or software that this is computed from?</pre>\r\n"
- "    </blockquote>\r\n"
- "    <pre> As global pstates are per-chip and there are max 12 cores, so if the system is really\r\n"
- "\240idle, considering 5 seconds for each cores, it should take 60 seconds for the chip to\r\n"
- "\240go to pmin.</pre>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <blockquote type=\"cite\">\r\n"
- "        <pre wrap=\"\">+/* Interval after which the timer is queued to bring down global pstate */\r\n"
- "+#define GPSTATE_TIMER_INTERVAL\t\t\t\t2000\r\n"
- "</pre>\r\n"
- "      </blockquote>\r\n"
- "      <pre wrap=\"\">in ms?</pre>\r\n"
- "    </blockquote>\r\n"
- "    <pre>Yes its 2000 ms.\r\n"
- "</pre>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <pre wrap=\"\">\r\n"
- "</pre>\r\n"
- "    </blockquote>\r\n"
- "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\r\n"
- "      <pre wrap=\"\">\r\n"
- "</pre>\r\n"
- "    </blockquote>\r\n"
- "    <br>\r\n"
- "  </body>\r\n"
- "</html>\r\n"
  "\01:2\0"
  "b\0"
- "_______________________________________________\n"
- "Linuxppc-dev mailing list\n"
- "Linuxppc-dev@lists.ozlabs.org\n"
- https://lists.ozlabs.org/listinfo/linuxppc-dev
+ "<html>\n"
+ "  <head>\n"
+ "    <meta content=\"text/html; charset=windows-1252\"\n"
+ "      http-equiv=\"Content-Type\">\n"
+ "  </head>\n"
+ "  <body text=\"#000000\" bgcolor=\"#FFFFFF\">\n"
+ "    <pre>Hi Stewart,\n"
+ "\n"
+ "</pre>\n"
+ "    On 04/20/2016 03:41 AM, Stewart Smith wrote:<br>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <pre wrap=\"\">Akshay Adiga <a class=\"moz-txt-link-rfc2396E\" href=\"mailto:akshay.adiga@linux.vnet.ibm.com\">&lt;akshay.adiga@linux.vnet.ibm.com&gt;</a> writes:\n"
+ "</pre>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">Iozone results show fairly consistent performance boost.\n"
+ "YCSB on redis shows improved Max latencies in most cases.\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">What about power consumption?\n"
+ "\n"
+ "</pre>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">Iozone write/rewite test were made with filesizes 200704Kb and 401408Kb\n"
+ "with different record sizes . The following table shows IOoperations/sec\n"
+ "with and without patch.\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">Iozone Results ( in op/sec) ( mean over 3 iterations )\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">What's the variance between runs?\n"
+ "</pre>\n"
+ "    </blockquote>\n"
+ "    <pre>Re-Ran Iozone test\n"
+ "\n"
+ "w/o : without patch,  w : with patch , stdev : standard deviation , avg ; average \n"
+ "\n"
+ "Iozone Results for ReWrite\n"
+ "+----------+--------+-----------+------------+-----------+-----------+---------+\n"
+ "| filesize | reclen |  w/o(avg) | w/o(stdev) |   w(avg)  |  w(stdev) | change% |\n"
+ "+----------+--------+-----------+------------+-----------+-----------+---------+\n"
+ "|  200704  |   1    |  795070.4 |  5813.51   |  805127.8 |  16872.59 |  1.264  |\n"
+ "|  200704  |   2    | 1448973.8 |  23058.79  | 1472098.8 |  18062.73 |  1.595  |\n"
+ "|  200704  |   4    |  2413444  |  85988.09  | 2562535.8 |  48649.35 |  6.177  |\n"
+ "|  200704  |   8    |  3827453  |  87710.52  | 3846888.2 |  86438.51 |  0.507  |\n"
+ "|  200704  |   16   | 5276096.8 |  73208.19  | 5425961.6 | 170774.75 |  2.840  |\n"
+ "|  200704  |   32   | 6742930.6 |  22789.45  | 6848904.4 | 257768.84 |  1.571  |\n"
+ "|  200704  |   64   | 7059479.2 | 300725.26  |  7373635  | 285106.90 |  4.450  |\n"
+ "|  200704  |  128   | 7097647.2 | 408171.71  |  7716500  | 266139.68 |  8.719  |\n"
+ "|  200704  |  256   |  6710810  | 314594.13  | 7661752.6 | 454049.27 |  14.170 |\n"
+ "|  200704  |  512   | 7034675.4 | 516152.97  | 7378583.2 | 613617.57 |  4.888  |\n"
+ "|  200704  |  1024  | 6265317.2 | 446101.38  | 7540629.6 | 294865.20 |  20.355 |\n"
+ "|  401408  |   1    |  802233.2 |  4263.92   |   817507  |  17727.09 |  1.903  |\n"
+ "|  401408  |   2    | 1461892.8 |  53678.12  |  1482872  |  45670.30 |  1.435  |\n"
+ "|  401408  |   4    | 2629686.8 |  24365.33  | 2673196.2 |  41576.78 |  1.654  |\n"
+ "|  401408  |   8    | 4156353.8 |  70636.85  | 4149330.4 |  56521.84 |  -0.168 |\n"
+ "|  401408  |   16   |  5895437  |  63762.43  | 5924167.4 | 396311.75 |  0.487  |\n"
+ "|  401408  |   32   | 7330826.6 | 167080.53  | 7785889.2 | 245434.99 |  6.207  |\n"
+ "|  401408  |   64   | 8298555.2 | 328890.89  | 8482416.8 | 249698.02 |  2.215  |\n"
+ "|  401408  |  128   | 8241108.6 | 490560.96  |  8686478  | 224816.21 |  5.404  |\n"
+ "|  401408  |  256   | 8038080.6 | 327704.66  | 8372327.4 | 210978.18 |  4.158  |\n"
+ "|  401408  |  512   | 8229523.4 | 371701.73  | 8654695.2 | 296715.07 |  5.166  |\n"
+ "+----------+--------+-----------+------------+-----------+-----------+---------+\n"
+ "\n"
+ "Iozone results for Write \n"
+ "+----------+--------+-----------+------------+-----------+------------+---------+\n"
+ "| filesize | reclen |  w/o(avg) | w/o(stdev) |   w(avg)  |  w(stdev)  | change% |\n"
+ "+----------+--------+-----------+------------+-----------+------------+---------+\n"
+ "|  200704  |   1    |   575825  |  7,876.69  |  569388.4 |  6,699.59  |  -1.12  |\n"
+ "|  200704  |   2    | 1061229.4 |  7,589.50  | 1045193.2 | 19,785.85  |  -1.51  |\n"
+ "|  200704  |   4    |  1808329  | 13,040.67  | 1798138.4 | 50,367.19  |  -0.56  |\n"
+ "|  200704  |   8    | 2822953.4 | 19,948.89  | 2830305.6 | 21,202.77  |   0.26  |\n"
+ "|  200704  |   16   |  3976987  | 62,201.72  | 3909063.8 | 268,640.51 |  -1.71  |\n"
+ "|  200704  |   32   | 4959358.2 | 112,052.99 |  4760303  | 330,343.73 |  -4.01  |\n"
+ "|  200704  |   64   | 5452454.6 | 628,078.72 | 5692265.6 | 190,562.91 |   4.40  |\n"
+ "|  200704  |  128   | 5645246.8 | 10,455.85  | 5653330.2 | 18,153.76  |   0.14  |\n"
+ "|  200704  |  256   | 5855897.2 | 184,854.25 |  5402069  | 538,523.04 |  -7.75  |\n"
+ "|  200704  |  512   |  5515904  | 326,198.86 | 5639976.4 |  8,480.46  |   2.25  |\n"
+ "|  200704  |  1024  | 5471718.2 | 415,179.15 | 5399414.6 | 686,124.50 |  -1.32  |\n"
+ "|  401408  |   1    |  584786.6 |  1,256.59  |  587237.2 |  6,552.55  |   0.42  |\n"
+ "|  401408  |   2    | 1047018.8 | 26,567.72  | 1040926.8 | 16,495.93  |  -0.58  |\n"
+ "|  401408  |   4    | 1815465.8 | 16,426.92  | 1773652.6 | 38,169.02  |  -2.30  |\n"
+ "|  401408  |   8    |  2814285  | 27,374.53  |  2756608  | 96,689.13  |  -2.05  |\n"
+ "|  401408  |   16   |  3931646  | 129,648.79 | 3805793.4 | 141,368.40 |  -3.20  |\n"
+ "|  401408  |   32   | 4875353.4 | 146,203.70 |  4884084  | 265,484.01 |   0.18  |\n"
+ "|  401408  |   64   | 5479805.8 | 349,995.36 | 5565292.2 | 20,645.45  |   1.56  |\n"
+ "|  401408  |  128   |  5598486  | 195,680.23 |  5645125  | 62,017.38  |   0.83  |\n"
+ "|  401408  |  256   |  5803148  | 328,683.02 |  5657215  | 20,579.28  |  -2.51  |\n"
+ "|  401408  |  512   | 5565091.4 | 166,123.57 | 5725974.4 | 169,506.29 |   2.89  |\n"
+ "+----------+--------+-----------+------------+-----------+------------+---------+\n"
+ "\n"
+ "</pre>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <pre wrap=\"\">\n"
+ "</pre>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">Tested with YCSB workload (50% update + 50% read) over redis for 1 million\n"
+ "records and 1 million operation. Each test was carried out with target\n"
+ "operations per second and persistence disabled.\n"
+ "\n"
+ "Max-latency (in us)( mean over 5 iterations )\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">What's the variance between runs?\n"
+ "\n"
+ "std dev? 95th percentile?\n"
+ "\n"
+ "</pre>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">---------------------------------------------------------------\n"
+ "op/s    Operation       with patch      without patch   %change\n"
+ "---------------------------------------------------------------\n"
+ "15000   Read            61480.6         50261.4         22.32\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">This seems fairly significant regression. Any idea why at 15K op/s\n"
+ "there's such a regression?\n"
+ "</pre>\n"
+ "    </blockquote>\n"
+ "    <pre>Just Re-Ran the test for power numbers. \n"
+ "Results for YCSB+Redis test.\n"
+ "P95 : 95 Percentile \n"
+ "P99 : 99 Percentile\n"
+ "\n"
+ "Power numbers are taken for one run of YCSB+redis test which has 50% Read + 50% Update.\n"
+ "Maximum Latency has clearly gone down for all cases will less than 5% increase in power.\n"
+ "\n"
+ "\n"
+ "+------------+----------+--------+------------+---------+---------+----------------+\n"
+ "|   Op/sec   | Testcase | AvgLat |   MaxLat   |   P95   |   P99   |     Power      |\n"
+ "+------------+----------+--------+------------+---------+---------+----------------+\n"
+ "|   15000    |   Read   |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  |  51.8  |  127903.0  |   55.8  |  145.2  |     602.7      |\n"
+ "| w/o patch  |  StdDev  | 5.692  | 105355.497 |  11.232 |   2.04  |      5.11      |\n"
+ "| with patch | Average  | 53.28  |  30834.2   |   72.2  |  151.2  |     629.01     |\n"
+ "| with patch |  StdDev  | 2.348  |  8928.323  |  15.74  |  3.544  |      3.25      |\n"
+ "|     -      | <b>Change%  |  2.86  |   -75.89   |  29.39  |   4.13  | 4.36535589846</b>  |\n"
+ "|   25000    |   Read   |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 53.78  |  123743.0  |   85.4  |  152.2  |     617.95     |\n"
+ "| w/o patch  |  StdDev  | 4.593  |  80224.53  |  5.886  |   4.49  |      1.32      |\n"
+ "| with patch | Average  | 49.65  |  84101.4   |   84.2  |  154.4  |     651.64     |\n"
+ "| with patch |  StdDev  | 1.658  | 72656.042  |  4.261  |  2.332  |      8.76      |\n"
+ "|     -      |<b> Change%  | -7.68  |   -32.04   |  -1.41  |   1.45  |  5.4518974027 </b> |\n"
+ "|   35000    |   Read   |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 56.07  |  57391.0   |   93.0  |  147.6  |     636.39     |\n"
+ "| w/o patch  |  StdDev  | 1.391  | 34494.839  |  1.789  |  2.871  |      2.92      |\n"
+ "| with patch | Average  | 56.46  |  39634.2   |   95.0  |  149.2  |     653.44     |\n"
+ "| with patch |  StdDev  | 3.174  |  6089.848  |  3.347  |   3.37  |      4.4       |\n"
+ "|     -      |<b> Change%  |  0.69  |   -30.94   |   2.15  |   1.08  |  2.6791747199 </b> |\n"
+ "|   40000    |   Read   |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  |  58.6  |  80427.8   |   97.2  |  147.4  |     636.85     |\n"
+ "| w/o patch  |  StdDev  | 1.105  | 59327.584  |  0.748  |  2.498  |      1.51      |\n"
+ "| with patch | Average  | 58.76  |  45291.8   |   97.2  |  149.0  |     656.12     |\n"
+ "| with patch |  StdDev  | 1.675  | 10486.954  |  2.482  |  3.406  |      6.97      |\n"
+ "|     -      | <b>Change%  |  0.27  |   -43.69   |   0.0   |   1.09  |  3.0258302583</b>  |\n"
+ "|   45000    |   Read   |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 69.02  |  120027.8  |  102.6  |  149.6  |     640.68     |\n"
+ "| w/o patch  |  StdDev  |  0.74  | 96288.811  |  1.855  |  1.497  |      7.65      |\n"
+ "| with patch | Average  | 69.65  |  98024.6   |  102.0  |  147.8  |     653.09     |\n"
+ "| with patch |  StdDev  |  1.14  | 78041.439  |   2.28  |  1.939  |      3.91      |\n"
+ "|     -      <b>| Change%  |  0.92  |   -18.33   |  -0.58  |   -1.2  | 1.93700443279</b>  |\n"
+ "|   15000    |  Update  |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 48.144 |  86847.0   |   52.4  |  189.2  |     602.7      |\n"
+ "| w/o patch  |  StdDev  | 5.971  | 41580.919  |  16.427 |  8.376  |      5.11      |\n"
+ "| with patch | Average  | 47.964 |  31106.2   |   58.4  |  182.2  |     629.01     |\n"
+ "| with patch |  StdDev  | 3.003  |  4906.179  |  7.088  |  6.177  |      3.25      |\n"
+ "|     -      | <b>Change%  | -0.37  |   -64.18   |  11.45  |   -3.7  | -3.69978858351</b> |\n"
+ "|   25000    |  Update  |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 51.856 |  102808.6  |   87.0  |  182.4  |     617.95     |\n"
+ "| w/o patch  |  StdDev  | 5.721  | 79308.823  |  4.899  |  7.965  |      1.32      |\n"
+ "| with patch | Average  | 46.07  |  74623.0   |   86.2  |  183.0  |     651.64     |\n"
+ "| with patch |  StdDev  | 1.779  | 77511.229  |  4.069  |  7.014  |      8.76      |\n"
+ "|     -      | <b>Change%  | -11.16 |   -27.42   |  -0.92  |   0.33  | 0.328947368421</b> |\n"
+ "|   35000    |  Update  |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 54.142 |  51074.2   |   93.6  |  181.8  |     636.39     |\n"
+ "| w/o patch  |  StdDev  | 1.671  | 36877.588  |  1.497  |  8.035  |      2.92      |\n"
+ "| with patch | Average  | 54.034 |  44731.8   |   94.4  |  184.4  |     653.44     |\n"
+ "| with patch |  StdDev  | 3.363  |  13400.4   |   1.02  |  7.172  |      4.4       |\n"
+ "|     -      |<b> Change%  |  -0.2  |   -12.42   |   0.85  |   1.43  |  1.4301430143</b>  |\n"
+ "|   40000    |  Update  |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 57.528 |  71672.6   |   98.4  |  184.8  |     636.85     |\n"
+ "| w/o patch  |  StdDev  | 1.111  | 63103.862  |  1.744  |  9.282  |      1.51      |\n"
+ "| with patch | Average  | 57.738 |  32101.4   |   98.0  |  186.4  |     656.12     |\n"
+ "| with patch |  StdDev  | 1.294  |  4481.801  |  1.673  |   7.71  |      6.97      |\n"
+ "|     -      | <b>Change%  |  0.37  |   -55.21   |  -0.41  |   0.87  | 0.865800865801 </b>|\n"
+ "|   45000    |  Update  |   -    |     -      |    -    |    -    |       -        |\n"
+ "| w/o patch  | Average  | 69.97  |  117183.0  |  105.4  |  182.4  |     640.68     |\n"
+ "| w/o patch  |  StdDev  | 0.925  | 99836.076  |   1.2   |  9.091  |      7.65      |\n"
+ "| with patch | Average  | 70.508 |  104175.0  |  103.2  |  185.4  |     653.09     |\n"
+ "| with patch |  StdDev  | 1.463  |  74438.13  |   1.47  |  7.915  |      3.91      |\n"
+ "|     -      |<b> Change%  |  0.77  |   -11.1    |  -2.09  |   1.64  | 1.64473684211  </b>|\n"
+ "+------------+----------+--------+------------+---------+---------+----------------+\n"
+ "</pre>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <pre wrap=\"\">\n"
+ "</pre>\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">--- a/drivers/cpufreq/powernv-cpufreq.c\n"
+ "+++ b/drivers/cpufreq/powernv-cpufreq.c\n"
+ "@@ -36,12 +36,56 @@\n"
+ " #include &lt;asm/reg.h&gt;\n"
+ " #include &lt;asm/smp.h&gt; /* Required for cpu_sibling_mask() in UP configs */\n"
+ " #include &lt;asm/opal.h&gt;\n"
+ "+#include &lt;linux/timer.h&gt;\n"
+ " \n"
+ " #define POWERNV_MAX_PSTATES\t256\n"
+ " #define PMSR_PSAFE_ENABLE\t(1UL &lt;&lt; 30)\n"
+ " #define PMSR_SPR_EM_DISABLE\t(1UL &lt;&lt; 31)\n"
+ " #define PMSR_MAX(x)\t\t((x &gt;&gt; 32) &amp; 0xFF)\n"
+ " \n"
+ "+#define MAX_RAMP_DOWN_TIME\t\t\t\t5120\n"
+ "+/*\n"
+ "+ * On an idle system we want the global pstate to ramp-down from max value to\n"
+ "+ * min over a span of ~5 secs. Also we want it to initially ramp-down slowly and\n"
+ "+ * then ramp-down rapidly later on.\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">Where does 5 seconds come from?\n"
+ "\n"
+ "Why 5 and not 10, or not 2? Is there some time period inherit in\n"
+ "hardware or software that this is computed from?</pre>\n"
+ "    </blockquote>\n"
+ "    <pre> As global pstates are per-chip and there are max 12 cores, so if the system is really\n"
+ "\240idle, considering 5 seconds for each cores, it should take 60 seconds for the chip to\n"
+ "\240go to pmin.</pre>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <blockquote type=\"cite\">\n"
+ "        <pre wrap=\"\">+/* Interval after which the timer is queued to bring down global pstate */\n"
+ "+#define GPSTATE_TIMER_INTERVAL\t\t\t\t2000\n"
+ "</pre>\n"
+ "      </blockquote>\n"
+ "      <pre wrap=\"\">in ms?</pre>\n"
+ "    </blockquote>\n"
+ "    <pre>Yes its 2000 ms.\n"
+ "</pre>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <pre wrap=\"\">\n"
+ "</pre>\n"
+ "    </blockquote>\n"
+ "    <blockquote cite=\"mid:87vb3dmep8.fsf@linux.vnet.ibm.com\" type=\"cite\">\n"
+ "      <pre wrap=\"\">\n"
+ "</pre>\n"
+ "    </blockquote>\n"
+ "    <br>\n"
+ "  </body>\n"
+ "</html>\n"
 
-11a2ef9f86b4b306bd59aa5a63e9b057b5b193066af1fec295870003dc86b7ee
+9b62af052eb867d3b613e34675647432f3e2cc76d315a948bf9608edc3824ffe

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.