<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#ffffff" text="#000000">
    On 06/02/2011 05:41 AM, mani wrote:
    <blockquote
      cite="mid:BANLkTinbMfmF1m_vA6RiP5Y-dnOueyOKwg@mail.gmail.com"
      type="cite">Dear All,<br>
      <br>
      Any suggestions on the below point ? <br>
    </blockquote>
    <br>
    have you tried different benchmarks?<br>
    This can give you an idea of the impact. It seems that up to now,
    you've only executed sequential read and write access by one
    process.<br>
    <br>
    Personally I use the fio benchmark. For flash file systems, David
    Wagner from Free Electrons developed a benchmarksuite. Maybe it
    would be worse having a look on it.<br>
    <br>
    Best regards,<br>
    Matthias Brugger<br>
    <br>
    <blockquote
      cite="mid:BANLkTinbMfmF1m_vA6RiP5Y-dnOueyOKwg@mail.gmail.com"
      type="cite"><br>
      Thanks in advance.. <br>
      <br>
      <div class="gmail_quote">On Mon, May 30, 2011 at 10:53 AM, mani <span
          dir="ltr">&lt;<a moz-do-not-send="true"
            href="mailto:manishrma@gmail.com">manishrma@gmail.com</a>&gt;</span>
        wrote:<br>
        <blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt
          0.8ex; border-left: 1px solid rgb(204, 204, 204);
          padding-left: 1ex;">Dear Eduardo,<br>
          <br>
          &nbsp;&nbsp; I am using squashfs filesystem. So i am more concern with
          the read speed.<br>
          &nbsp; &nbsp; whereas below are the details <br>
          &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; read
          speed&nbsp;&nbsp; write speed <br>
          &nbsp;&nbsp;&nbsp; ELEVATOR_INSERT_SORT&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 8 MBps&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 5MBps<br>
          &nbsp;&nbsp;&nbsp; ELEVATOR_INSERT_BACK&nbsp; &nbsp; &nbsp; 10 MBps&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 7.2MBps<br>
          <br>
          &nbsp;&nbsp; used the following command for measurement for both the
          cases. <br>
          &nbsp;&nbsp; reading <br>
          &nbsp;&nbsp; hdparm -t /dev/mtdblock3<br>
          &nbsp;&nbsp; <br>
          &nbsp;&nbsp; writing <br>
          &nbsp;&nbsp; dd if=/dev/zero of=/dev/mtd3 bs=4096 count=100k <br>
          <br>
          &nbsp;&nbsp; As of now everything is working fine with those changes <br>
          &nbsp;&nbsp; but i am worried if these changes would have any adverse
          effect anywhere ? <br>
          &nbsp;<br>
          &nbsp;&nbsp; Thanks. <br>
          <div>
            <div class="h5">&nbsp;<br>
              <br>
              <div class="gmail_quote">On Fri, May 27, 2011 at 8:37 PM,
                Eduardo Silva <span dir="ltr">&lt;<a
                    moz-do-not-send="true"
                    href="mailto:edsiper@gmail.com" target="_blank">edsiper@gmail.com</a>&gt;</span>
                wrote:<br>
                <blockquote class="gmail_quote" style="margin: 0pt 0pt
                  0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204);
                  padding-left: 1ex;">
                  <div>On Fri, May 27, 2011 at 8:29 AM, mani &lt;<a
                      moz-do-not-send="true"
                      href="mailto:manishrma@gmail.com" target="_blank">manishrma@gmail.com</a>&gt;
                    wrote:<br>
                    &gt;<br>
                    &gt; Dear All,<br>
                    &gt;<br>
                    &gt; I am working on linux kernel 2.6.32.9 tegra
                    NVIDIA board.<br>
                    &gt;<br>
                    &gt; I am getting ~8MBps speed of the Nand disk if i
                    use hdparm<br>
                    &gt; hdparm -t /dev/mtdblock3<br>
                    &gt;<br>
                    &gt; i made changes in block layer of kernel as
                    below:-<br>
                    &gt;<br>
                    &gt; block/blk-core.c<br>
                    &gt; static inline void add_request(struct
                    request_queue *q, struct request *req)<br>
                    &gt; {<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; drive_stat_acct(req, 1);<br>
                    &gt;<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; /*<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; * elevator indicated where it wants
                    this request to be<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; * inserted at elevator_merge time<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; */<br>
                    &gt;<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; __elv_add_request(q, req,
                    ELEVATOR_INSERT_BACK, 0);<br>
                    &gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; //__elv_add_request(q, req,
                    ELEVATOR_INSERT_SORT, 0);<br>
                    &gt; }<br>
                    &gt;<br>
                    &gt;<br>
                    <br>
                  </div>
                  What are the results for read and write for both cases
                  ?<br>
                  <div><br>
                  </div>
                </blockquote>
                <blockquote class="gmail_quote" style="margin: 0pt 0pt
                  0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204);
                  padding-left: 1ex;">
                  <div>
                    <br>
                    &gt; changed ELEVATOR_INSERT_SORT to
                    ELEVATOR_INSERT_BACK<br>
                    &gt; it improves my NAND speed to 10MBps.<br>
                    &gt;<br>
                    &gt; I am using "noop" I/O scheduler.<br>
                    &gt;<br>
                    &gt; Will this change have any adverse effect in
                    kernel ? or any other side<br>
                    &gt; effect as far as i am using only Nand no Hard
                    disk.<br>
                    &gt;<br>
                    &gt;<br>
                    &gt; Thanks<br>
                    &gt; Mani<br>
                    &gt;<br>
                    &gt;<br>
                  </div>
                  &gt; _______________________________________________<br>
                  &gt; Kernelnewbies mailing list<br>
                  &gt; <a moz-do-not-send="true"
                    href="mailto:Kernelnewbies@kernelnewbies.org"
                    target="_blank">Kernelnewbies@kernelnewbies.org</a><br>
                  &gt; <a moz-do-not-send="true"
                    href="http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies"
                    target="_blank">http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies</a><br>
                  &gt;<br>
                  &gt;<br>
                  <font color="#888888"><br>
                    <br>
                    <br>
                    --<br>
                    Eduardo Silva<br>
                    <a moz-do-not-send="true"
                      href="http://edsiper.linuxchile.cl"
                      target="_blank">http://edsiper.linuxchile.cl</a><br>
                    <a moz-do-not-send="true"
                      href="http://www.monkey-project.com"
                      target="_blank">http://www.monkey-project.com</a><br>
                  </font></blockquote>
              </div>
              <br>
            </div>
          </div>
        </blockquote>
      </div>
      <br>
      <pre wrap="">
<fieldset class="mimeAttachmentHeader"></fieldset>
_______________________________________________
Kernelnewbies mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Kernelnewbies@kernelnewbies.org">Kernelnewbies@kernelnewbies.org</a>
<a class="moz-txt-link-freetext" href="http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies">http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>