Ack. I was just telling jcm yesterday (at the SF RT summit), that this needed to be done. We have definite evidence of major page faults on cyclictest, and consequent spikes. (its opaque for far, why a small, and high prio process should become swap fodder to begin with) Sven Sent via BlackBerry by AT&T -----Original Message----- From: Clark Williams <clark.williams@xxxxxxxxx> Date: Fri, 09 May 2008 13:55:40 To:Thomas Gleixner <tglx@xxxxxxxxxxxxx> Cc:Arnaldo Carvalho de Melo <acme@xxxxxxxxxxxxxxxxxx>,Luis Claudio Goncalves <lclaudio@xxxxxxxx>,RT <linux-rt-users@xxxxxxxxxxxxxxx> Subject: add mlockall option to cyclictest, et al -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thomas, Got a couple of requests for mlockall() in cyclictest, so I added an --mlockall option to cyclictest, signaltest and pi_stress. Patches attached. Clark -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkgknisACgkQqA4JVb61b9e+5wCgkfnFdvymSXLsxQ9AwhVU9te+ J9YAniZv12Xj1r6epjft/L/CLv3P1Yl4 =ismw -----END PGP SIGNATURE----- ��.n��������+%������w��{.n�����{�����ǫ���ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
