All of lore.kernel.org
 help / color / mirror / Atom feed
From: Trevor Woerner <trevor.woerner@linaro.org>
To: Gunnar Roth <gunnar.roth@gmx.de>,
	 "meta-freescale@yoctoproject.org"
	<meta-freescale@yoctoproject.org>
Subject: Re: imxq6 sabre sd locks up when starting Qt5_Cinematic experience
Date: Sat, 04 Oct 2014 22:21:26 -0400	[thread overview]
Message-ID: <5430AB26.100@linaro.org> (raw)
In-Reply-To: <E3C823DB-E609-4DF4-A7DB-1B6340F6C0E0@gmx.de>

(old thread... sorry, just catching up)

On 08/02/14 04:00, Gunnar Roth wrote:
> Btw how can i define the size of the rootfs , so to have more space available? Right now i always did a resize of the partition afterwards,
> but somehow i must be able to give the size of the roots including free space. 

There are a couple variables you can play with in your conf/local.conf
to affect the size of the final image.

First off you can set "IMAGE_OVERHEAD_FACTOR" to a number, say 20
(IMAGE_OVERHEAD_FACTOR = "20"). This will ask that an extra 20% size be
added to your final image.

You can also specify a size (in KB) for your image using IMAGE_ROOTFS_SIZE.

In my experience, however, sometimes these variables get ignored.
Sometimes they're ignored because the code that is used to calculate
these sizes (for the specific build you're doing) doesn't take these
variables into account. If the size you specify is too small, the build
will ignore your request and make the output the size it would have
anyway without any input from you. Sometimes different types of output
will ignore the suggestions you make in your configuration file (e.g.
vmdk versus hddimage).

Nominally you could look in poky/meta/classes/image_types.bbclass and
see how things are calculated. You could also make use of "bitbake -e |
grep <some variable>" to see what your build thinks of the various
things calculated in this bbclass file.

But at the end of the day, unless you have a specific need for the extra
space, all you're doing is increasing the time it takes your "dd"
command to write your cfcard.  If you have a 2GB card and are only using
200MB the "dd" will only take one tenth the time it would take to write
out the entire 2GB card... and all you're doing with all that extra time
is writing empty file space!

There are some intelligent tools [1] that can help you write a 2GB card
with a 200MB image and only take the amount of time that it would take
to write a 200MB card, but those tools aren't integrated into the build
system (yet?). But they certainly work/help if you use them manually
after your build completes but before you write your cfcard (if your
image has lots of empty space added to it, otherwise there's no savings).

Best regards,
    Trevor


[1] https://source.tizen.org/documentation/reference/bmaptool


      parent reply	other threads:[~2014-10-05  2:21 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-01 11:18 imxq6 sabre sd locks up when starting Qt5_Cinematic experience Gunnar Roth
2014-08-01 11:52 ` Konstantin Kovalev
2014-08-01 14:36   ` Gunnar Roth
2014-08-01 15:06     ` Otavio Salvador
2014-08-01 15:33       ` Gunnar Roth
2014-08-01 17:19         ` Otavio Salvador
2014-08-02  8:00           ` Gunnar Roth
2014-08-08 18:12             ` Gunnar Roth
2014-10-05  2:21             ` Trevor Woerner [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5430AB26.100@linaro.org \
    --to=trevor.woerner@linaro.org \
    --cc=gunnar.roth@gmx.de \
    --cc=meta-freescale@yoctoproject.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.