<div dir="ltr"><div><div><div><div>Hi Barry and Jed,<br></div>just a curiosity and maybe a good point to learn something:<br><br></div>1- why are those<b> inconvenient </b>"convenience routines" ?<br><br></div>2 - why those routines <b>should not be used</b> when writing a fresh code or a large re-factoring?<br>
<br></div><div><br></div>Thanks in advance.<br><div><div><br></div></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sun, Sep 29, 2013 at 3:52 PM, Jed Brown <span dir="ltr"><<a href="mailto:jedbrown@mcs.anl.gov" target="_blank">jedbrown@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">Barry Smith <<a href="mailto:bsmith@mcs.anl.gov">bsmith@mcs.anl.gov</a>> writes:<br>
<br>
> We can do all this, but how much effort do we really want to make<br>
> for supporting these inconvenient "convenience routines"?<br>
<br>
</div>Not much, but this is a tiny amount of code.<br>
<div class="im"><br>
> BTW: We should really label these function with<br>
><br>
> Level: Recommended only for legacy applications being quickly ported to PETSc<br>
><br>
> I really only want people who already have a working non-PETSc<br>
> code that uses CSR directly already to use these routines so they<br>
> can quickly try out PETSc, I never want someone writing fresh code<br>
> or doing a large refactorization to EVER use these routines.<br>
<br>
</div>Agreed.<br>
<div class="im"><br>
>> On second thought, why don't we just document that<br>
>> MatMPIBAIJSetPreallocationCSR respects MAT_ROW_ORIENTED within each<br>
>> block and let the user choose which they want to use?<br>
><br>
> Also there is no way to get this flag into MatCreateMPIBAIJWithArrays()<br>
<br>
</div>The user calls MatSetOption before making that call and changes it<br>
afterward if they want.<br>
</blockquote></div><br><br clear="all"><br>-- <br><div dir="ltr">Matteo<br></div>
</div>