From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Stancek Date: Mon, 18 Feb 2019 13:33:46 -0500 (EST) Subject: [LTP] [PATCH] adding s390x tests to ltp In-Reply-To: <20190218182137.GA27553@dell5510> References: <20190204161328.4616-1-elas@linux.vnet.ibm.com> <1740320205.101841684.1549543929404.JavaMail.zimbra@redhat.com> <20190218182137.GA27553@dell5510> Message-ID: <1043798029.1537735.1550514826801.JavaMail.zimbra@redhat.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it ----- Original Message ----- > Hi Jan, Cyril, > > > Maybe also: TST_NEEDS_CMDS="vmcp" ? > > Not sure if all s390x distros have this command installed. > > > But that should also be sufficient check to run this on any arch, > > if command doesn't exist it will TCONF. > > The reason, why I suggested to move these tests to upstream [1] > is that testing "vmcp --version" or "vmcp --help" does not really depend on > anyhow on SUT. Adding it to upstream into make check (there are none check > yet) > would bring 1) earlier catch of potencial bugs as upstream developers can run > it > for each commit (CI) and distros during package build. > > We had some discussion before about removing some user space packages [2]. > Cyril suggested to to keep tests that are working with files (cp/df/ln) > and remove at least unzip and wc. > Doing some real testing (really using vmcp.ko) would make this tests useful > to > run on LTP, but current code IMHO does nothing with kernel. Or do I miss > something? > > Anyway, it looks like you have no objection about adding it to LTP I don't. My only question was if runtest/commands would be more fitting for the test, but people seemed to prefer s390x specific runtest-file instead. Regards, Jan >, so I > changed > state of this test New. In that case, I suggest to > > remove debugging: > > +#set -x > > +# > > use tst_res TINFO instead of echo > + echo "verifyBasicVMCPCommands" > + echo "verifyErrorCondition" > > Kind regards, > Petr > > [1] https://github.com/ibm-s390-tools/s390-tools > [2] https://patchwork.ozlabs.org/patch/905064/#1902070 >