pgtesttiming.sgml, FreeBSD command output
I noticed this little snippet claiming to be output on FreeBSD in
pgtesttiming.sgml:
<screen>
dmesg | grep "Timecounter"
sysctl kern.timecounter.hardware=TSC
</screen>
First, grep is case-sensitive by default, so that line of output
shouldn't have matched at all, right? Second, if I actually try the
above command on a FreeBSD 9 VM I have, I get (copying by hand, since
I can't seem to copy-and-paste out of this VM without X, please excuse
any typos):
Timecounter "ACPI-fast" frequency 3579545 Hz quality 900
Timecounter "i8254" frequency 1193182 Hz quality 0
Timecounters tick every 10.000 msec
Timecounter "TSC" frequency 2531787134 Hz quality 800
Anyway, if the original "sysctl ..." output is valid for a FreeBSD
machine, maybe only the grep command needs to be fixed (i.e. to "grep
-i ...") ?
Josh
On fre, 2012-07-06 at 13:21 -0700, Josh Kupershmidt wrote:
I noticed this little snippet claiming to be output on FreeBSD in
pgtesttiming.sgml:<screen>
dmesg | grep "Timecounter"
sysctl kern.timecounter.hardware=TSC
</screen>First, grep is case-sensitive by default, so that line of output
shouldn't have matched at all, right?
I don't think this was meant to say that the second line is the output
of the first. I read this as two separate commands. But I agree this
part is a bit unclear.
On Fri, Jul 6, 2012 at 1:33 PM, Peter Eisentraut <peter_e@gmx.net> wrote:
On fre, 2012-07-06 at 13:21 -0700, Josh Kupershmidt wrote:
<screen>
dmesg | grep "Timecounter"
sysctl kern.timecounter.hardware=TSC
</screen>First, grep is case-sensitive by default, so that line of output
shouldn't have matched at all, right?I don't think this was meant to say that the second line is the output
of the first. I read this as two separate commands. But I agree this
part is a bit unclear.
Oh, I see now. The <screen> sections above and below that one were
including output, and I was just expecting the same here.
One way to fix would be to include shell prompts and outputs for both
commands (patch Option 1, attached). Or, if we don't want to bother
including output, as the section does now, maybe we could at least
include the shell prompts and mention that output is excluded (patch
Option 2).
I've also attached an unrelated typofix I noticed for the same page.
Josh
Attachments:
pgtesttiming_typofix.diffapplication/octet-stream; name=pgtesttiming_typofix.diffDownload+2-2
pgtesttiming_output_Option2.diffapplication/octet-stream; name=pgtesttiming_output_Option2.diffDownload+7-7
pgtesttiming_output_Option1.diffapplication/octet-stream; name=pgtesttiming_output_Option1.diffDownload+9-9
Patch attached and applied for PG 9.3.
---------------------------------------------------------------------------
On Fri, Jul 6, 2012 at 02:09:00PM -0700, Josh Kupershmidt wrote:
On Fri, Jul 6, 2012 at 1:33 PM, Peter Eisentraut <peter_e@gmx.net> wrote:
On fre, 2012-07-06 at 13:21 -0700, Josh Kupershmidt wrote:
<screen>
dmesg | grep "Timecounter"
sysctl kern.timecounter.hardware=TSC
</screen>First, grep is case-sensitive by default, so that line of output
shouldn't have matched at all, right?I don't think this was meant to say that the second line is the output
of the first. I read this as two separate commands. But I agree this
part is a bit unclear.Oh, I see now. The <screen> sections above and below that one were
including output, and I was just expecting the same here.One way to fix would be to include shell prompts and outputs for both
commands (patch Option 1, attached). Or, if we don't want to bother
including output, as the section does now, maybe we could at least
include the shell prompts and mention that output is excluded (patch
Option 2).I've also attached an unrelated typofix I noticed for the same page.
--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://enterprisedb.com
+ It's impossible for everything to be true. +