On Fri, Jun 28, 2013 at 2:47 PM, Jan Engelhardt <jengelh@inai.de> wrote:
On Friday 2013-06-28 12:05, Andrey Borzenkov wrote:
FWIW, I've expanded my 32MB initrd and looked for the top 10 largest files So indeed the branding takes up a lot of space, but I'm suprised to see libicu49 take take the lion's share.
bor@opensuse:~> rpm -qf /usr/lib64/libicudata.so.49.1 libicu49-49.1-6.1.1.x86_64 bor@opensuse:~> LC_ALL=C rpm -q --whatrequires libicu49-49.1-6.1.1.x86_64 no package requires libicu49-49.1-6.1.1.x86_64
But, just WHAT program / lib pulls libicudata.so.49 into initrd (cant't lay the blame atm.)?
It is pulled in through libpancocairo.so which in turn is pulled in by
libpangocairo.so of course
/usr/lib/plymouth/label.so
I can offer a solution whereby libicudata.so is replaced by a stub and the actual data is provided in a separate (noarch) file. Not sure what happens though if the data file is not actually present when a program runs... any testers?
How does it affect i18n? We do want to be able to display in native language, assuming we want plymouth in initrd at all. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org