<HTML dir=ltr><HEAD><TITLE>Re: [openfabrics-ewg] Gen2 basic and resource tracking test fails for Pathscale HCA</TITLE>
<META http-equiv=Content-Type content="text/html; charset=unicode">
<META content="MSHTML 6.00.2900.2963" name=GENERATOR></HEAD>
<BODY>
<DIV id=idOWAReplyText332 dir=ltr>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2>Hi Robert:</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2></FONT> </DIV>
<DIV dir=ltr><FONT face=Arial size=2>Thanks for acknowledging the issue.</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2>I didn't reported other failures with gen2_basic tests, because Mellanox HCA card also fails those tests. Also, these are known failures mentioned in readme file.</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2></FONT> </DIV>
<DIV dir=ltr><FONT face=Arial size=2>Thanks</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2>Karun</FONT></DIV></DIV>
<DIV dir=ltr><BR>
<HR tabIndex=-1>
<FONT face=Tahoma size=2><B>From:</B> Robert Walsh [mailto:rjwalsh@pathscale.com]<BR><B>Sent:</B> Tue 9/19/2006 7:34 PM<BR><B>To:</B> Sharma, Karun<BR><B>Cc:</B> dotanb@mellanox.co.il; openfabrics-ewg@openib.org<BR><B>Subject:</B> Re: [openfabrics-ewg] Gen2 basic and resource tracking test fails for Pathscale HCA<BR></FONT><BR></DIV>
<DIV>
<P><FONT size=2>-----BEGIN PGP SIGNED MESSAGE-----<BR>Hash: SHA1<BR><BR>Sharma, Karun wrote:<BR>> Hi Dotan:<BR>> <BR>> I am testing Pathscale HCA with OFED1.1 RC5. I performed gen2_basic,<BR>> qp_test, multicast_test, mem_lock_check and resource_tracking tests.<BR>> Following are the failures encountered.<BR>> <BR>> 1. gen2_basic test fails for POLL_POST component. With Mellanox HCA,<BR>> this is passing.<BR>> 2. When I specify -n option with resource_tracking_test, the test fails<BR>> for Pathscale HCA. Again it passes for Mellanox HCA.<BR>> <BR>> I have attached the logs below.<BR><BR>Hi Karun,<BR><BR>We've run the gen2_basic tests here and have fixed a number of problems<BR>in the test suite itself. I will be sending patches out shortly, after<BR>I've had time to review them. We now have most tests passing. The<BR>tests that still fails are:<BR><BR> AV (try to create max ah + 1)<BR> CQ (resize CQ to more than initial size)<BR> CQ (modify with outstanding WR to more than outstanding)<BR> POLL_POST (post sr to RC QP)<BR> QP (test 3, test 5 and test 6)<BR><BR>The first test fails because the ib_mad module has an AH already<BR>created, so we cannot get as far as creating max AH. I'm not sure how<BR>to fix this one, but that seems like a test shortcoming.<BR><BR>The second and third tests fail because, for reasons I haven't figured<BR>out yet, resize_cq support got dropped from our libipathverbs. I'm<BR>still tracking this down and hope to have a fix for RC6.<BR><BR>I haven't figured out yet why the fourth test fails. It's looking like<BR>a memory corruption problem (stack smashing), and I think it's in the<BR>test itself, but I haven't figured it out for certain yet.<BR><BR>The QP tests are failing because of core dumps in the tests. Again, I<BR>think these are failures in the tests themselves, but I haven't figured<BR>them out yet.<BR><BR>We have not yet run a full resource test, as we had some early problems<BR>with that running out of memory. We are still working on this.<BR><BR>Regards,<BR> Robert.<BR>-----BEGIN PGP SIGNATURE-----<BR>Version: GnuPG v1.4.5 (GNU/Linux)<BR>Comment: Using GnuPG with Fedora - <A href="http://enigmail.mozdev.org/">http://enigmail.mozdev.org</A><BR><BR>iQEVAwUBRRB+b/zvnpzTd9fxAQKcjwgApa3l92R6FLv9D37zlHf6hPwqJNoOtGgG<BR>uAGYfMWpfOai0cHx0fgTNTEFBxxeFPdTcG6CEEmyO53piNDUxEpB2OqBJSd5dBys<BR>sKjHNscmmGxktHaUPnuK1MF9defvCoGv0Wd+yGBhjuzhazNaOlLBRVkXPXpe9fX7<BR>UMeCdiRgGOoTbcAUt+Rwq8eDYNIkCO7U+EapsRxeIsgpFDkiVQ/RsQsh+QSfLcQb<BR>SX6sY0tm+WZx7WHesOcHGP9i8JtIb7rUNf1n05xhmdaaaEgp/kR/cYIROvwoI+Oq<BR>63+z95vDGFKG0BM3IAvelSCQL/wqLm9QlzsPTLGOgPyE5ymnyF4uhw==<BR>=2Lk6<BR>-----END PGP SIGNATURE-----<BR></FONT></P></DIV></BODY></HTML>