<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">Dear Wei, <br>
      <br>
      thank you a lot for the update! I'll try to package the new
      version next week.<br>
      <br>
      Best regards, <br>
      Alex<br>
      <br>
      On 09/04/2015 03:27 AM, Wei Shi wrote:<br>
    </div>
    <blockquote
      cite="mid:A773617D-B94A-4B06-9A48-92494B2EA5E7@wehi.edu.au"
      type="cite">Dear Alex,
      <div><br>
      </div>
      <div>Hope you are well. Just let you know that we have included
        complete documentation in our latest release 1.4.6-p5.</div>
      <div><br>
      </div>
      <div>Best wishes,</div>
      <div><br>
      </div>
      <div>Wei<br>
        <div><br>
        </div>
        <div><br>
        </div>
        <div><br>
          <div>
            <div>On May 22, 2015, at 7:08 PM, Alex Mestiashvili wrote:</div>
            <br class="Apple-interchange-newline">
            <blockquote type="cite">
              <meta content="text/html; charset=windows-1252"
                http-equiv="Content-Type">
              <div text="#000000" bgcolor="#FFFFFF">
                <div class="moz-cite-prefix">Hi Wei, <br>
                  <br>
                  On 05/20/2015 01:09 AM, Wei Shi wrote:<br>
                </div>
                <blockquote
                  cite="mid:CED5503E-3372-40D1-ADEC-100C1424B17F@wehi.edu.au"
                  type="cite">Hi Alex,
                  <div><br>
                  </div>
                  <div>Thanks for testing our package and providing us
                    the feedback.</div>
                  <div><br>
                  </div>
                  <div>We will look into the getopt issue. If that is
                    the only issue that caused problem for running
                    Subread on ARM, I think we can fix it.</div>
                </blockquote>
                <br>
                The good news - ARM issues are solved by adding
                -fsigned-char to CCFLAGS.<br>
                In this case subread works on architectures where char
                is by default unsigned such as armel, armhf and so on.<br>
                And in this case there is no need to patch getopt.<br>
                <br>
                Though there are still problems on powerpc, mips and
                s390x:<br>
                - subjunc test return Segmentation fault, <br>
                - exactSNP test complains that it can not read the file
                <br>
                <br>
                here is the output:<br>
                exactSNP FEOF 0!<br>
                SAM file does not exist or is not accessible:
                'data/test-in.BAM'<br>
                <br>
                <blockquote
                  cite="mid:CED5503E-3372-40D1-ADEC-100C1424B17F@wehi.edu.au"
                  type="cite"><br>
                  <div>Could you also please provide us your cppcheck
                    results? We would like to know where memory leaks
                    occur in the code and we can then fix them.</div>
                </blockquote>
                <br>
                Please see the attachment for the cppcheck output.<br>
                <br>
                Thank you, <br>
                Alex<br>
                <br>
                <blockquote
                  cite="mid:CED5503E-3372-40D1-ADEC-100C1424B17F@wehi.edu.au"
                  type="cite">
                  <div>Best regards,</div>
                  <div><br>
                  </div>
                  <div>Wei</div>
                </blockquote>
                <br>
                <blockquote
                  cite="mid:CED5503E-3372-40D1-ADEC-100C1424B17F@wehi.edu.au"
                  type="cite">
                  <div><br>
                    <div>
                      <div>On May 15, 2015, at 10:59 PM, Alex
                        Mestiashvili wrote:</div>
                      <br class="Apple-interchange-newline">
                      <blockquote type="cite">
                        <meta content="text/html; charset=windows-1252"
                          http-equiv="Content-Type">
                        <div text="#000000" bgcolor="#FFFFFF">
                          <div class="moz-cite-prefix">Dear Wei, <br>
                            <br>
                            thank you for the quick answer.<br>
                            <br>
                            On 05/12/2015 01:59 AM, Wei Shi wrote:<br>
                          </div>
                          <blockquote
                            cite="mid:F0200EB6-835B-4F01-AFAA-1D2BF4E9048E@wehi.edu.au"
                            type="cite">Hi Alex,
                            <div><br>
                            </div>
                            <div>We have tested subread on both 32-bit
                              and 64-bit Intel and AMD platforms and it
                              works fine (the gcc flags you mentioned
                              were all included). So I do not know what
                              caused the installation failure. Maybe you
                              can show us the errors you have
                              encountered?</div>
                          </blockquote>
                          <br>
                          I was not correct in the last email, the
                          failure on i386 was related to the debian
                          specific part of the build process.<br>
                          So both AMD64 and i386 are fine.<br>
                          The full build logs are located here: <a
                            moz-do-not-send="true"
                            class="moz-txt-link-freetext"
                            href="https://buildd.debian.org/status/logs.php?pkg=subread">https://buildd.debian.org/status/logs.php?pkg=subread</a><br>
                          <br>
                          Regarding the -mtune option, My understanding
                          is that tuning for a specific CPU shouldn't be
                          used for a <br>
                          binary package which can be used on a wide
                          range of CPUs.<br>
                          Currently I removed -mtune completely, but
                          maybe -mtune=generic should be used instead
                          of  -mtune=core2.<br>
                          <br>
                           "generic" <br>
                           Produce code optimized for the most common
                          IA32/AMD64/EM64T processors.<br>
                           If you know the CPU on which your code will
                          run, then you should use the <br>
                           corresponding <samp><span class="option">-mtune</span></samp>
                          or <samp><span class="option">-march</span></samp>
                          option instead of <samp><span class="option">-mtune=generic</span></samp>.
                          <br>
                           But, if you do not know exactly what CPU
                          users of your application will have,  <br>
                           then you should use this option. <br>
                          <br>
                          <a moz-do-not-send="true"
                            class="moz-txt-link-freetext"
                            href="https://gcc.gnu.org/onlinedocs/gcc/x86-Options.html">https://gcc.gnu.org/onlinedocs/gcc/x86-Options.html</a><br>
                          <br>
                          The -msse and -msse2 flags are enabled by
                          default for x86-64 compiler, so it should be
                          no problem to strip them, <br>
                          and AFAIK SSE also implies support for MMX, so
                          -mmmx flag is not needed too.<br>
                          <br>
                          <blockquote
                            cite="mid:F0200EB6-835B-4F01-AFAA-1D2BF4E9048E@wehi.edu.au"
                            type="cite">
                            <div>Currently, Subread only supports AMD
                              and Intel architectures and we do not have
                              a plan to support other platforms such as
                              ARM.</div>
                          </blockquote>
                          <br>
                          I see, In such case I'll limit the
                          architectures for the package to the ones
                          where it builds successfully and passes the
                          tests.<br>
                          <br>
                          Just for your information, I tried to fix the
                          getopt issue by defining "c" as int instead of
                          char.<br>
                          The binaries were build successfully on ARM
                          and other architectures, but the tests fail,
                          you can see the results here:<br>
                           <br>
                           <a moz-do-not-send="true"
                            class="moz-txt-link-freetext"
href="https://buildd.debian.org/status/logs.php?pkg=subread&ver=1.4.6-p2%2Bdfsg-3&suite=experimental">https://buildd.debian.org/status/logs.php?pkg=subread&ver=1.4.6-p2%2Bdfsg-3&suite=experimental</a><br>
                          <br>
                          Also the cppcheck reveals a lot of memory
                          leaks in the code.<br>
                          <br>
                          Best regards, <br>
                          Alex<br>
                          <br>
                          <blockquote
                            cite="mid:F0200EB6-835B-4F01-AFAA-1D2BF4E9048E@wehi.edu.au"
                            type="cite">
                            <div><br>
                              <div>
                                <div>On May 11, 2015, at 5:34 PM, Alex
                                  Mestiashvili wrote:</div>
                                <br class="Apple-interchange-newline">
                                <blockquote type="cite">
                                  <meta http-equiv="content-type"
                                    content="text/html;
                                    charset=windows-1252">
                                  <div text="#000000" bgcolor="#FFFFFF">
                                    Dear Dr. Wei Shi, <br>
                                    <br>
                                    Debian supports many CPU
                                    architectures and building of
                                    subread failed on some of these
                                    architectures.<br>
                                    <br>
                                    1. Due to quite specific -mtune and
                                    -mmmx -msse* gcc flags, subread
                                    failed to build almost on any
                                    architecture except amd64.<br>
                                    This issue was fixed by removing
                                    -mtune and -mmmx -msse* flags from
                                    the src/Makefile.Linux. <br>
                                    <a moz-do-not-send="true"
                                      class="moz-txt-link-freetext"
href="http://anonscm.debian.org/cgit/debian-med/subread.git/tree/debian/patches/gcc_flags_cleanup.patch">http://anonscm.debian.org/cgit/debian-med/subread.git/tree/debian/patches/gcc_flags_cleanup.patch</a><br>
                                    <br>
                                    2. We still have an issue, now with
                                    exactSNP, please see the bug report
                                    below.<br>
                                    <br>
                                    Thank you, <br>
                                    Alex<br>
                                    <div class="moz-forward-container"><br>
                                      <br>
                                      -------- Forwarded Message
                                      --------
                                      <table
                                        class="moz-email-headers-table"
                                        border="0" cellpadding="0"
                                        cellspacing="0">
                                        <tbody>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Subject: </th>
                                            <td>[Debian-med-packaging]
                                              Bug#784929: subread: FTBFS
                                              on arm64, armel, armhf,
                                              ...</td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Resent-Date:
                                            </th>
                                            <td>Sun, 10 May 2015
                                              19:54:01 +0000</td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Resent-From:
                                            </th>
                                            <td>Edmund Grimley Evans <a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-rfc2396E"
href="mailto:edmund.grimley.evans@gmail.com"><edmund.grimley.evans@gmail.com></a></td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Resent-To: </th>
                                            <td><a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-abbreviated"
href="mailto:debian-bugs-dist@lists.debian.org">debian-bugs-dist@lists.debian.org</a></td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Resent-CC: </th>
                                            <td>Debian Med Packaging
                                              Team <a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-rfc2396E"
href="mailto:debian-med-packaging@lists.alioth.debian.org"><debian-med-packaging@lists.alioth.debian.org></a></td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Date: </th>
                                            <td>Sun, 10 May 2015
                                              20:51:41 +0100</td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">From: </th>
                                            <td>Edmund Grimley Evans <a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-rfc2396E"
href="mailto:edmund.grimley.evans@gmail.com"><edmund.grimley.evans@gmail.com></a></td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">Reply-To: </th>
                                            <td>Edmund Grimley Evans <a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-rfc2396E"
href="mailto:edmund.grimley.evans@gmail.com"><edmund.grimley.evans@gmail.com></a>,
                                              <a moz-do-not-send="true"
class="moz-txt-link-abbreviated" href="mailto:784929@bugs.debian.org">784929@bugs.debian.org</a></td>
                                          </tr>
                                          <tr>
                                            <th nowrap="nowrap"
                                              valign="BASELINE"
                                              align="RIGHT">To: </th>
                                            <td><a
                                                moz-do-not-send="true"
                                                class="moz-txt-link-abbreviated"
href="mailto:submit@bugs.debian.org">submit@bugs.debian.org</a></td>
                                          </tr>
                                        </tbody>
                                      </table>
                                      <br>
                                      <br>
                                      <pre>Source: subread
Version: 1.4.6-p2+dfsg-2

It failed to build on several architectures:

<a moz-do-not-send="true" class="moz-txt-link-freetext" href="https://buildd.debian.org/status/package.php?p=subread&suite=sid">https://buildd.debian.org/status/package.php?p=subread&suite=sid</a>

There seems to be a problem with (unsigned) char. See this file, for
example:

<a moz-do-not-send="true" class="moz-txt-link-freetext" href="https://sources.debian.net/src/subread/1.4.6-p2%2Bdfsg-2/src/SNPCalling.c/">https://sources.debian.net/src/subread/1.4.6-p2%2Bdfsg-2/src/SNPCalling.c/</a>

This line:

while ((c = getopt_long (argc, argv, "7:N:a:i:g:o:bQ:p:f:n:r:x:w:s:t:T:v4",snp_long_options, &optindex))!=-1)

The function getopt_long returns int, but the value is assigned to a
char c, which is unsigned on ARM architectures, so it will never be
equal to -1. That is probably the immediate cause of the failure.
However, there may be other char-related problems in there. I noticed,
for example, that mask_snp_bitmap may be incorrect according to the C
standard (signed overflow).

_______________________________________________
Debian-med-packaging mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:Debian-med-packaging@lists.alioth.debian.org">Debian-med-packaging@lists.alioth.debian.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging">http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging</a>
</pre>
                                      <br>
                                    </div>
                                    <br>
                                  </div>
                                </blockquote>
                              </div>
                              <br>
                            </div>
                            <br clear="all">
______________________________________________________________________<br>
                            The information in this email is
                            confidential and intended solely for the
                            addressee.<br>
                            You must not disclose, forward, print or use
                            it without the permission of the sender.<br>
______________________________________________________________________<br>
                          </blockquote>
                          <br>
                        </div>
                      </blockquote>
                    </div>
                    <br>
                  </div>
                  <br clear="all">
______________________________________________________________________<br>
                  The information in this email is confidential and
                  intended solely for the addressee.<br>
                  You must not disclose, forward, print or use it
                  without the permission of the sender.<br>
______________________________________________________________________<br>
                </blockquote>
                <br>
              </div>
              <span><cppcheck4subread.txt></span></blockquote>
          </div>
          <br>
        </div>
      </div>
      <br clear="all">
______________________________________________________________________<br>
      The information in this email is confidential and intended solely
      for the addressee.<br>
      You must not disclose, forward, print or use it without the
      permission of the sender.<br>
______________________________________________________________________<br>
    </blockquote>
    <br>
  </body>
</html>