From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Stancek Date: Mon, 15 May 2017 09:00:19 -0400 (EDT) Subject: [LTP] [PATCH v2] [RFC] pselect01: Tune thresholds In-Reply-To: <20170515123556.GA8280@rei.lan> References: <20170512141658.26810-1-chrubis@suse.cz> <495691788.11461692.1494842630946.JavaMail.zimbra@redhat.com> <20170515123556.GA8280@rei.lan> Message-ID: <1832031935.11522951.1494853219630.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! > > while looking at fs/select.c I noticed that it also takes > > current->timer_slack_ns into account and uses that if > > slack is smaller. Would it make sense to add ...? > > slack_per_scall = MAX(slack_per_scall, prctl(PR_GET_TIMERSLACK) / 1000); > > The default is 50us, so maybe the 250us in formula covers this already? > > I guess that the 50us is included in the 250us already. So what about > adding the line you sugests and decreasing the 250us to 200us? We can do that, but looking again at PR_GET_TIMERSLACK, it is supported since 2.6.28, so we'd need to skip older kernels or fall back to a default of 50us. Regards, Jan