All of lore.kernel.org
 help / color / mirror / Atom feed
From: Robert Yang <liezhi.yang@windriver.com>
To: Richard Purdie <richard.purdie@linuxfoundation.org>,
	<bitbake-devel@lists.openembedded.org>
Subject: Re: [PATCH 4/8] bitbake: cooker: don't preserve BB_CONSOLELOG
Date: Thu, 8 Feb 2018 12:18:43 +0800	[thread overview]
Message-ID: <9da1f8e4-f56b-f88e-00cb-b97327b195c8@windriver.com> (raw)
In-Reply-To: <1517823574.3090.79.camel@linuxfoundation.org>

Hi RP,

On 02/05/2018 05:39 PM, Richard Purdie wrote:
> On Thu, 2018-02-01 at 23:15 +0800, Robert Yang wrote:
>> Fixed:
>> $ bitbake --server-only -T -1
>> Set MACHINE = "qemux86" in conf/local.conf
>> $ bitbake quilt
>> Set MACHINE = "qemuppc" in conf/local.conf
>> $ bitbake quilt
>>
>> The log still goes into tmp/log/cooker/qemux86 in the second run,
>> this is
>> incorrect (should be tmp/log/cooker/qemuppc). I checked the code, the
>> ui
>> initializes it every time when it starts, so let it use the up-to-
>> date
>> one rather than old one.
>>
>> Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
>> ---
>>   bitbake/lib/bb/cooker.py | 8 --------
>>   1 file changed, 8 deletions(-)
> 
> This effectively reverts:
> 
> http://git.yoctoproject.org/cgit.cgi/poky/commit/bitbake/lib/bb/cooker.py?id=3ebf7617d6c869f798807792918e1030b3ab66de
> 
> Sadly I didn't put a bug number in that commit message but I do
> remember there being a good reason to do this.
> 
> I think the problem is a "UI" can run multiple commands with reset
> events within that command stream and it would only have one console
> log. I also believe there may have been issues where toaster would have
> been unable to find the logs if they kept moving around.

I've found the bug:
https://bugzilla.yoctoproject.org/show_bug.cgi?id=8411

I think that it has been fixed by:

https://bugzilla.yoctoproject.org/show_bug.cgi?id=8373

http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=a0791c19db7d5b33ce7a4e3614a574ad4a319a84

I've tried the following commands with my patch:
$ . /workspace1/lyang1/poky/bitbake/bin/toaster start noweb
$ MACHINE=qemuppc bitbake quilt-native

The log is tmp/log/cooker/qemuppc/build_20180208_120722.942.log

$ MACHINE=qemux86 bitbake quilt-native

The log is tmp/log/cooker/qemux86/build_20180208_121605.382.log

And the log contents are correct.

So I think that we don't need preserve BB_CONSOLELOG anymore.

// Robert

> 
> The console log handling probably needs more thought as there are
> clearly multiple issues here. We may want to move it out of a directory
> which is machine specific?
> 
> Cheers,
> 
> Richard
> 
>> diff --git a/bitbake/lib/bb/cooker.py b/bitbake/lib/bb/cooker.py
>> index af482f9..3f113ae 100644
>> --- a/bitbake/lib/bb/cooker.py
>> +++ b/bitbake/lib/bb/cooker.py
>> @@ -326,11 +326,6 @@ class BBCooker:
>>           self.state = state.initial
>>           self.caches_array = []
>>   
>> -        # Need to preserve BB_CONSOLELOG over resets
>> -        consolelog = None
>> -        if hasattr(self, "data"):
>> -            consolelog = self.data.getVar("BB_CONSOLELOG")
>> -
>>           if CookerFeatures.BASEDATASTORE_TRACKING in self.featureset:
>>               self.enableDataTracking()
>>   
>> @@ -358,9 +353,6 @@ class BBCooker:
>>           self.data_hash = self.databuilder.data_hash
>>           self.extraconfigdata = {}
>>   
>> -        if consolelog:
>> -            self.data.setVar("BB_CONSOLELOG", consolelog)
>> -
>>           self.data.setVar('BB_CMDLINE', self.ui_cmdline)
>>   
>>           #
>> -- 
>> 2.7.4
>>
> 


  reply	other threads:[~2018-02-08  4:17 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-01 15:15 [PATCH 0/8] cooker: fixes and parsing performance improve Robert Yang
2018-02-01 15:15 ` [PATCH 1/8] bitbake: cooker: skip when BBFILE_PATTERN is empty Robert Yang
2018-02-01 15:15 ` [PATCH 2/8] bitbake: cooker: fix for BBFILE_PATTERN matches bbappend Robert Yang
2018-02-01 15:15 ` [PATCH 3/8] bitbake: cooker: don't stop file notifier when cooker is shutdown Robert Yang
2018-02-01 15:15 ` [PATCH 4/8] bitbake: cooker: don't preserve BB_CONSOLELOG Robert Yang
2018-02-05  9:39   ` Richard Purdie
2018-02-08  4:18     ` Robert Yang [this message]
2018-02-01 15:15 ` [PATCH 5/8] bitbake: cache: improve debug message Robert Yang
2018-02-01 15:15 ` [PATCH 6/8] bitbake: parse: fixes for resolve_file() Robert Yang
2018-02-01 15:15 ` [PATCH 7/8] bitbake: cooker: fix event notifications Robert Yang
2018-02-02 11:19   ` Richard Purdie
2018-02-06  8:46     ` Robert Yang
2018-02-01 15:15 ` [PATCH 8/8] bitbake: cooker: improve inotify handling Robert Yang

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=9da1f8e4-f56b-f88e-00cb-b97327b195c8@windriver.com \
    --to=liezhi.yang@windriver.com \
    --cc=bitbake-devel@lists.openembedded.org \
    --cc=richard.purdie@linuxfoundation.org \
    /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.