All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tom Rini <trini@konsulko.com>
To: u-boot@lists.denx.de
Subject: [PATCH 19/20] test/py: Use buildman to build U-Boot
Date: Thu, 12 Mar 2020 10:03:16 -0400	[thread overview]
Message-ID: <20200312140316.GQ12423@bill-the-cat> (raw)
In-Reply-To: <CAPnjgZ24EocgdmsHSMr0M44X6tksEhROkMV8uG5_fARAZdKRHg@mail.gmail.com>

On Tue, Mar 10, 2020 at 08:27:47PM -0600, Simon Glass wrote:
> Hi Tom,
> 
> On Mon, 9 Mar 2020 at 11:42, Tom Rini <trini@konsulko.com> wrote:
> >
> > On Mon, Mar 09, 2020 at 11:10:55AM -0600, Stephen Warren wrote:
> > > On 3/6/20 8:07 PM, Simon Glass wrote:
> > > > It is a pain to have to set the ARCH and CROSS_COMPILE environment
> > > > variables when using test.py's --build option. It is possible to get these
> > > > using the -A and -a options from buildman. But it seems better to just use
> > > > buildman to do the build.
> > > >
> > > > Remove the manual 'make' logic in test/py and use buildman instead.
> > >
> > > I far prefer using make here; this requires zero setup of buildman (e.g. the
> > > config file and specific toolchains), and so it much *less* of a pain.
> >
> > I have to agree here.  Keeping our test suite as dependency-free as
> > possible is important.  But... that's also not what's going on in the
> > code.  We don't set ARCH from what I can see, and of course don't use
> > it.  We don't set the CROSS_COMPILER from the snippet in question, only
> > the output directory.  Today, looking at the Travis/GitLab CI scripts we
> > don't even build via test.py but rather buildman prior to calling
> > test.py.  And I don't think I saw that changing in this series either.
> 
> I mean that to run pytest I have to do:
> 
> PATH=$PATH:tools/buildman ARCH=`buildman -a zynq_zybo`
> CROSS_COMPILE=`buildman -A zynq_zybo` \
> test/py/test.py -B zynq_zybo --id sjg-zynq_zybo --build-dir
> ../current/zynq_zybo --build
> 
> which is a bit of a pain.
> 
> With this change I can do:
> 
> test/py/test.py -B zynq_zybo --id sjg-zynq_zybo --build-dir
> ../current/zynq_zybo --build

Right.  The commit message isn't clear as the CI loops build the board
with buildman first.  Second, we don't use ARCH= when building U-Boot,
so we could just drop that from buildman I suspect.  Third, no, I think
it's important to NOT require buildman to be builder here and setting
CROSS_COMPILE in the environment is fine and makes integration with
other systems easier.  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200312/650cf31a/attachment.sig>

  reply	other threads:[~2020-03-12 14:03 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-07  3:07 [PATCH 00/20] gitlab: Simplify the test script Simon Glass
2020-03-07  3:07 ` [PATCH 01/20] sandbox: Add documentation about required/useful packages Simon Glass
2020-03-07  3:07 ` [PATCH 02/20] main: Drop show_boot_progress() prototype Simon Glass
2020-03-07  3:07 ` [PATCH 03/20] buildman: Document the members of BuilderJob Simon Glass
2020-03-07  3:07 ` [PATCH 04/20] bulidman: Add support for a simple build Simon Glass
2020-03-07  3:07 ` [PATCH 05/20] gitlab: Use the -w option for sandbox_spl Simon Glass
2020-03-09 17:50   ` Tom Rini
2020-03-07  3:07 ` [PATCH 06/20] azure: " Simon Glass
2020-03-07  3:07 ` [PATCH 07/20] gitlab: Use the --board buildman flag Simon Glass
2020-03-09 17:46   ` Tom Rini
2020-03-10 23:22     ` Simon Glass
2020-03-07  3:07 ` [PATCH 08/20] gitlab: Drop the BUILDMAN variable Simon Glass
2020-03-07  3:07 ` [PATCH 09/20] buildman: Update help for -d Simon Glass
2020-03-07  3:07 ` [PATCH 10/20] gitlab: Drop the buildman -d flag Simon Glass
2020-03-07  3:07 ` [PATCH 11/20] gitlab: Drop unnecessary if..fi Simon Glass
2020-03-07  3:07 ` [PATCH 12/20] gitlab: Use -w flag for all builds Simon Glass
2020-03-09 17:58   ` Tom Rini
2020-03-15  3:10     ` Simon Glass
2020-03-15 13:03       ` Tom Rini
2020-03-15 15:07         ` Simon Glass
2020-03-15 15:23           ` Tom Rini
2020-03-15 15:50             ` Simon Glass
2020-03-15 16:18               ` Tom Rini
2020-03-15 16:43                 ` Simon Glass
2020-03-07  3:07 ` [PATCH 13/20] gitlab: Use bash to avoid needing a_test_which_does_not_exist Simon Glass
2020-03-09 17:56   ` Tom Rini
2020-03-07  3:07 ` [PATCH 14/20] buildman: Allow ignoring warnings in the return code Simon Glass
2020-03-07  3:07 ` [PATCH 15/20] gitlab: Use the buildman -W flag Simon Glass
2020-03-09 18:01   ` Tom Rini
2020-03-07  3:07 ` [PATCH 16/20] gitlab: Enable test_handoff Simon Glass
2020-03-07  3:07 ` [PATCH 17/20] buildman: Be more selective about which directories to remove Simon Glass
2020-03-07  3:07 ` [PATCH 18/20] buildman: Allow building within a subdir of the current dir Simon Glass
2020-03-07  3:07 ` [PATCH 19/20] test/py: Use buildman to build U-Boot Simon Glass
2020-03-09 17:10   ` Stephen Warren
2020-03-09 17:41     ` Tom Rini
2020-03-11  2:27       ` Simon Glass
2020-03-12 14:03         ` Tom Rini [this message]
2020-03-10 23:22     ` Simon Glass
2020-03-07  3:07 ` [PATCH 20/20] gitlab: Simplify the exit code for test.py Simon Glass
2020-03-09 17:55 ` [PATCH 00/20] gitlab: Simplify the test script Tom Rini
2020-03-15  3:10   ` Simon Glass
2020-03-15 13:02     ` Tom Rini
2020-03-15 15:07       ` Simon Glass
2020-03-15 15:17         ` Tom Rini

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=20200312140316.GQ12423@bill-the-cat \
    --to=trini@konsulko.com \
    --cc=u-boot@lists.denx.de \
    /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.