[systemd-devel] [RFC][PATCH v2] analyze: draw kernel and initrd boot time as well

[Kay Sievers]

On Mon, Sep 26, 2011 at 20:29, Paul Menzel
<paulepanter at users.sourceforge.net> wrote:
> Am Montag, den 26.09.2011, 14:34 +0200 schrieb Kay Sievers:
>> On Sun, Sep 25, 2011 at 20:41, Koen Kooi <koen at dominion.thruhere.net> wrote:
>> > Sample output without initrd: http://dominion.thruhere.net/koen/angstrom/systemd/omap4430-panda-201109221422.svg
>> >
>> > Signed-off-by: Koen Kooi <koen at dominion.thruhere.net>
>> > ---
>> >
>> > Second attempt at this, please check the logic in this patch, a quick test on F15 looks OK, but I'd like more datapoints from initrd users.
>>
>> Looks a bit wrong. The initramfs bar should end when systemd starts,
>> and the overall length should not exceed systemd's finish time.
>>
>> Mangled your patch a bit. Added a 3rd userspace bar, made them all
>> grey. Renamed initrd to initramfs, because initramdisks do not exist
>> since many years.
>>
>> Output is here:
>>   http://people.freedesktop.org/~kay/f16.svg
>>
>> Please check.
>
> Does the grey userspace bar have to end as the last of the processes(?)
> below. I am not sure if it is an illusion, but it does not look like
> that.

It's the end of the 'initial transaction'. We should not print
anything after that point, unless we explicitly support such numbers.

> The 4 s mark would be nice to have.

Like this?
  http://people.freedesktop.org/~kay/f16-own-kernel.svg

Kay
-------------- next part --------------
A non-text attachment was scrubbed...
Name: analyze.patch
Type: text/x-patch
Size: 5264 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/systemd-devel/attachments/20110926/6c25f5ea/attachment.bin>