Re: [PATCHv3 5/6] feature-arm-vfp.inc: respect vfpv4 when setting -mfloat-abi and ARMPKGSFX_EABI

[Martin Jansa <martin.jansa@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1405 bytes --]

On Tue, Dec 08, 2015 at 10:34:47AM +0000, Richard Purdie wrote:
> On Mon, 2015-12-07 at 23:35 +0100, Martin Jansa wrote:
> > It's highly recommended to use the script together with memres bitbake,
> > the execution time is at least 10 times shorted, but the changes aren't
> > picked between runs, so make sure to restart bitbake after changing the
> > tunes.
> 
> That sounds like something it should detect. Could you file a bug
> please, preferably with a simpler test case that shows the problem?

Once I finish the tune testing I'll try to get simpler test case.

FWIW: yesterday I've changed the test to set the DEFAULTTUNE directly in
MACHINE.conf instead of using DEFAULTTUNE_MACHINE = "foo" overrides in
.inc file included in local.conf and since then even with memres bitbake
it's terribly slow, because each combination causes all metadata to be
reparsed.

So after 12 hours, I'm not even in half:
Testing DEFAULTTUNE armv5te (18/82) for fake MACHINE fake-cortexa9 (14/44)

I haven't confirmed it, but I think it weren't re-parsing anything
yesterday, that's probably why it was so fast, but IIRC it was
generating different output (with bitbake -e) so it was noticing
something (but maybe it was only for different MACHINEs not different
DEFAULTTUNEs.. I need to reproduce that and confirm).

Regards,

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 188 bytes --]