Well that's pretty bizarre! So for some reason, attempting to read an incomplete SSL packet takes 3x as long as reading and decoding a packet? Your results point heavily toward a bad design inside libssl itself.

Just to be sure though, what if you make a quick change to your SSLeay.xs to verify that the problem is in SSL_read... 

    PREINIT:
        char *buf;
        int got;
        int succeeded = 1;
        struct timespec ts1, ts2;
        long dT;
    PPCODE:
        New(0, buf, max, char);
 
        clock_gettime(CLOCK_MONOTONIC, &ts1);
        got = SSL_read(s, buf, max);
        clock_gettime(CLOCK_MONOTONIC, &ts2);
        dT= (ts2.tv_nsec - ts1.tv_nsec)/1000
          + (ts2.tv_sec - ts1.tv_sec)*1000000;
        warn("SSL_read(s, buf, %ld) => %ld  (%ld us)\n",
             (long)max, (long)got, dT);
        if (got <= 0 && SSL_ERROR_ZERO_RETURN != SSL_get_error(s, got)
+)
               succeeded = 0;

[download]

This probably only works on Linux (haven't checked if MSWin32 added clock_gettime yet) and if you aren't able to test, let me know and I will.

It will also completely throw off the CPU usage report because now it prints on every read, but it should still be able to measure the duration of the SSL_read calls. (the other option is to capture the timings in a persistent buffer and print them at the end, but that's a lot of hassle.)

If the timing of SSL_read matches your results above, then it would be time to head to the openssl project with a bug report. But, if it doesn't, then maybe the slowness is inside SSL_get_error or something else in the surrounding code that only happens on an unproductive read.

And the other option is to build against a debug-symbols version of libssl, and go all the way down into it with gdb.

In reply to Re^5: IO::Socket::SSL / Net::SSLeay inefficient in non-blocking mode ? by NERDVANA
in thread IO::Socket::SSL / Net::SSLeay inefficient in non-blocking mode ? by Yaribz

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.