[Toybox] [landley/toybox] Help building toybox with the NDK/bionic (#43)

[Rob Landley]

On 08/17/2016 02:29 PM, enh wrote:
> On Wed, Aug 17, 2016 at 11:28 AM, Rob Landley <rob at landley.net> wrote:
>>> if you check out an AOSP tree, change the external/toybox/Android.mk
>>> to say "LOCAL_FORCE_STATIC_EXECUTABLE := true" and then mma -j in that
>>> directory, that should leave you with a tree with just enough built.
>>
>> Last time I tried this just the _checkout_ was something like 30 gigs.
> 
> make sure you sync with "-c" to not get the history.

Ok, trying again with that. I did an rm -rf on the previous data because
of the bloat problem from interrupted downloads never seeming to delete
old partial download data.

It does seem smaller, but it's hard to tell how much it's done...

Uh-huh. Did you know that if you do an apt-get update while this thing
is downloading, repo/git starts goin:

error: RPC failed; result=56, HTTP code = 200 | 68.00 KiB/s   B/s
fatal: The remote end hung up unexpectedly
fatal: early EOF
fatal: index-pack failed

On every package, and yet somehow this doesn't abort the sync?

Hmmm... that seemed worth a ctrl-c and a retry. Ok, doing so presumably
leaked some space, but it's 3.5 gigs into the _new_ download and I'm not
waiting around for that again...

>> I'm in San Diego at the moment with a 4 gig monthly phone tethering
>> limit (thanks t-mobile!) and ~120k/second coffee shop wifi.
>> (30*(1<<30))/(120000) is 268435 seconds, which is 4473 minutes, which is
>> 74 hours. (To download the NDK tarball, I sat at a McDonalds until it
>> closed and then sat outside at the concrete tables another 45 minutes.)
>>
>> I'll see what I can do. Maybe I can find a faster coffee shop. Or wait
>> until I get back to texas. (Looking like "february".)
> 
> if you can wait until february (and i'm sure you have enough other
> stuff to do in the meantime), this particular problem should have
> solved itself. we already know toybox can be built with the platform
> headers because it's checked in to the platform. afaics, the only bug
> here is "the NDK headers are three years out of date", and that's our
> bug, not yours :-)

If my current stab at it doesn't work out, I can try the NDK again in
februaryish, sure.

In general, I'd like to be able to test bionic stuff myself before it
gets to you. I've been poking at the test suite and need to do more,
that should be in a position where running "make tests" before a checkin
actually helps. (I'd also really like to document how OTHER people can
reproduce this build. Getting it to work for me is a prerequisite for me
to document how other people can do it.)

But yeah, I'm juggling this with a dozen other things, as usual...

Rob