From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53609) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XKSAa-0004fF-Ol for qemu-devel@nongnu.org; Thu, 21 Aug 2014 09:12:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XKSAT-00071Y-JW for qemu-devel@nongnu.org; Thu, 21 Aug 2014 09:12:48 -0400 Received: from mx1.redhat.com ([209.132.183.28]:58139) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XKSAT-00071P-AS for qemu-devel@nongnu.org; Thu, 21 Aug 2014 09:12:41 -0400 Received: from int-mx13.intmail.prod.int.phx2.redhat.com (int-mx13.intmail.prod.int.phx2.redhat.com [10.5.11.26]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id s7LDCdhH006852 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for ; Thu, 21 Aug 2014 09:12:40 -0400 Date: Thu, 21 Aug 2014 14:12:37 +0100 From: Stefan Hajnoczi Message-ID: <20140821131237.GA22844@stefanha-thinkpad.redhat.com> References: <87oave58q3.fsf@blackfin.pond.sub.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="SLDf9lqlvOQaIe6s" Content-Disposition: inline In-Reply-To: <87oave58q3.fsf@blackfin.pond.sub.org> Subject: Re: [Qemu-devel] What tests should "make check-block" run? List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Markus Armbruster Cc: Kevin Wolf , qemu-devel@nongnu.org --SLDf9lqlvOQaIe6s Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Aug 21, 2014 at 02:27:00PM +0200, Markus Armbruster wrote: > Should "make check-block" run more tests? Which ones? Let's add image format specific test cases to "make check-block" (e.g. VHDX and VMDK). For these formats we would not run the standard tests, only the format-specific test cases. That way "make check-block" stays fast but gets a lot more coverage. > Should we have a variant of "make check-block" for testing other > (format, protocol) combinations? I don't think variants are useful. If you need control, use ./check. > Should we have a variant of "make check-block" that tests everything > that's testable without special setup? >=20 > How can we make advanced iotest features more discoverable for > developers not already intimately familiar with it? The limiting factor is test execution speed. If tests are very fast we can run more of them without annoying the user. If anyone wants to optimize test run times, that would be very welcome. Here is a list of the longest test cases: 026 116s 041 56s 023 43s 015 29s 014 29s There are some more test cases that take more than 10s which I have not listed. Ideally all test cases would take less than 5s. Stefan --SLDf9lqlvOQaIe6s Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJT9fBFAAoJEJykq7OBq3PIlosH/RK5eDKsqKMhxOOxjeA71cHd 5/bPU9YxiczgB+3JWPcrES3X5/slLpO9O3KgXvSvBvT8rcZ8m/sRqcabelQ+WKSl CTcft9SwXEWIeIrQbLjBRxvQmwnd0SOFc+yLLCHLdvkRvr6SSHTPSKOZbr2/cDjV 6Kt0lSXbtf/DRo7FHgSIqC1RY9JCs7gCep/XS5d4HsBcU3qk476heAN74U85d9gY gBq2regBnPnYVfkxfSBWgYS+yPJOSP8G14aSTKELqx+MDc3sq3YiB34yU0x4+9VZ OzEDMClD/zS6guRHldDwrXCM65Sbfjz57ZVv3zW5LLmySBkxTtUzlKGpU2PZ0sM= =g0pe -----END PGP SIGNATURE----- --SLDf9lqlvOQaIe6s--