From: Petr Vorel <pvorel@suse.cz>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH 0/3] Build android fixes
Date: Tue, 5 Feb 2019 02:22:52 +0100 [thread overview]
Message-ID: <20190205012252.GA11738@dell5510> (raw)
In-Reply-To: <20190204192004.GB222456@google.com>
Hi Sandeep,
> > > How did you come across these issues? We have yet to see them. Did you merge
> > > tip of LTP into an Android AOSP tree and build it that way?
> > Any sane person who wants to tests Android kernel uses LTP build inside of AOSP
> > to get all your custom (not yet upstreamed) fixes, skip non relevant test etc.
> > But I don't want to store whole AOSP on my disc, so I tried to cross compile
> > upstream LTP with android standalone toolchain (with make ANDROID=1).
> That is correct and I really want to be able to do the same. Admittedly the
> current AOSP LTP build is "meant for the harness it normally runs under". It
> really should not be the case and we'd love for AOSP LTP to build standalone.
Great :). I see your bionic-compat.h can handle some errors (e.g. missing
getdtablesize(void)). Since we have already some support, this could be probably
merged (assuming Cyril is not against).
That layer allows you to build more tests than it's currently allowed in our
Makefiles (we filter quite a lot syscalls).
> I think I asked about the "make ANDROID" option a while ago and my
> understanding was that its building LTP against the Android NDK. Steve, may
> be this is the path we go? I do remember however that with that build, the
> number of tests that we can build against Android was far less than what we
> do right now (haven't tried in a while though).
I do build with configure --host triplet add --with-sysroot and SYSROOT
+ add -I and -L flags to sysroot dirs to CFLAGS and LDFLAGS.
But this way autoconf does detect wrongly some options (at least HAVE_PREADV and
HAVE_PWRITEV).
BTW which compiler option allows you to ignore missing <string.h> and
<strings.h> ?
> Also, FWIW, there are almost zero non-upstream tests in AOSP LTP.
Great!
> I can only
> think of one test suite but that is just very *android specific* and still
> being worked on.
> - ssp
Kind regards,
Petr
next prev parent reply other threads:[~2019-02-05 1:22 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-30 11:28 [LTP] [PATCH 0/3] Build android fixes Petr Vorel
2019-01-30 11:28 ` [LTP] [PATCH 1/3] lib: Add missing <string.h> and <strings.h> includes Petr Vorel
2019-01-30 11:28 ` [LTP] [PATCH 2/3] lib/tst_kernel.c: Use <sys/personality.h> Petr Vorel
2019-01-30 11:28 ` [LTP] [PATCH 3/3] lib: Build tst_path_has_mnt_flags.c also on Android Petr Vorel
2019-01-30 19:01 ` [LTP] [PATCH 0/3] Build android fixes Jan Stancek
2019-02-02 17:22 ` Petr Vorel
2019-01-30 23:02 ` Steve Muckle
2019-02-02 17:22 ` Petr Vorel
2019-02-04 19:20 ` Sandeep Patil
2019-02-05 0:06 ` Steve Muckle
2019-02-05 1:22 ` Petr Vorel [this message]
2019-02-05 23:40 ` Steve Muckle
2019-02-05 0:00 ` Steve Muckle
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=20190205012252.GA11738@dell5510 \
--to=pvorel@suse.cz \
--cc=ltp@lists.linux.it \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox