AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@lsi.com
RQ:
SSV:mhbs.lsil.com
NSV:
SSH:
R:<Larry.Scheer@lsi.com>
MAID:2
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/LSI/INBOX	0	DEC609CD0E54B2448DAF023C89AE9755EB50C4F2@cosmail02.lsi.com
X-Sylpheed-End-Special-Headers: 1
Date: Mon, 22 Mar 2010 13:31:10 -0700
From: Andrew Sharp <andy.sharp@lsi.com>
To: "Scheer, Larry" <Larry.Scheer@lsi.com>
Subject: Re: Please review change 34893 -- nighlty build script
 updates/fixes
Message-ID: <20100322133110.47795edf@ripper.onstor.net>
In-Reply-To: <DEC609CD0E54B2448DAF023C89AE9755EB50C4F2@cosmail02.lsi.com>
References: <DEC609CD0E54B2448DAF023C89AE9755EB50C4CA@cosmail02.lsi.com>
	<20100317132037.056faedf@ripper.onstor.net>
	<DEC609CD0E54B2448DAF023C89AE9755EB50C4F2@cosmail02.lsi.com>
Organization: LSI
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

ok

On Mon, 22 Mar 2010 13:15:11 -0600 "Scheer, Larry"
<Larry.Scheer@lsi.com> wrote:

> My comments to your comments are below...
> ________________________________________
> From: Andrew Sharp [andy.sharp@lsi.com]
> Sent: Wednesday, March 17, 2010 1:20 PM
> To: Scheer, Larry
> Subject: Re: Please review change 34893 -- nighlty build script
> updates/fixes
> 
> On Mon, 15 Mar 2010 14:31:13 -0600 "Scheer, Larry"
> <Larry.Scheer@lsi.com> wrote:
> 
> > Change 34893 by larrys@larrys on 2010/03/15 13:29:52 *pending*
> >
> >       1. Time each make so I can track if build performance.
> >       2. Skip checking for warnings for a nonextant build
> > (test-dbg.) 3. Fix a bug where the nfx-tree for multiple build
> > branches (dev and r403rel) were being copied to the bobcat version
> > directory versus the cougar version.  Links to nfx-tree/code were
> > also reversed. This was due to the build location being selected
> > from the last build to be run. This assumed the builds would follow
> > a strict order (i. e. cougar last.) Changed the selection of build
> > location to be the highest version number in the list of builds.
> > This allows build order to be arbitrary and allows our latest
> > development version to be the primary location for code and Build.
> >       4. Developers wanted a link to the Build directory to make
> > running the debugger easier.
> >       Reviewed by:
> >
> > Affected files ...
> >
> > ... //depot/cm/build_scripts/buildver/nfxbuild.sh#57 edit
> 
> 
> 
> 
> = Change 34893 by larrys@larrys on 2010/03/15 13:29:52 *pending*
> =
> =       1. Time each make so I can track if build performance.
> =       2. Skip checking for warnings for a nonextant build
> (test-dbg.) =       3. Fix a bug where the nfx-tree for multiple
> build branches =          (dev and r403rel) were being copied to the
> bobcat version directory =          versus the cougar version.  Links
> to nfx-tree/code were also reversed. =          This was due to the
> build location being selected from the last build =          to be
> run. This assumed the builds would follow a strict order =
> (i. e. cougar last.) Changed the selection of build location to be
> the =          highest version number in the list of builds. This
> allows build order =          to be arbitrary and allows our latest
> development version to be =          the primary location for code
> and Build. =       4. Developers wanted a link to the Build directory
> to make running the =          debugger easier. =       Reviewed by:
> 
> 
> I didn't say this in the comments below, but don't we actually want
> (3) for all the branches?  I thought that's what the code was doing,
> just for each branch, regardless of bobcat or cougar.
> 
> [LS] Yes, it is being done for all branches.. I changed the
> description so it is a bit more clear.
> 
> build_scripts/buildver/nfxbuild.sh
> 
>      line 739-747 this is racy.  needs to be unique to *this* instance
>      of build.sh, which is why it was in buildlogs before.  we might
>      want to run simultaneous builds.
> 
> [LS] Good catch. I added pid to the temporary file. This will  allow
> simultaneous builds. optional:
> 
>      line 505 would you want to make this more like ($TIME) make test,
>      and allow TIME to be defined or not defined depending on whether
>      or not you're trying to debug something?
> 
> [LS} Made it a variable, but didn't place it in a conditional
> statement that can be done later, if ever. 
>      line 644 normally i like to put parens in compound if statements
>      like this.  just to be sure.  i don't wear belt and suspenders,
>      but i do wear tight pants and a belt.
> 
>  [LS] Done.
> 
>      line 838 you sure you don't want a uniq here?  if not here,
>      then probably back around line 739ish
> 
> [LS] The uniquification was already done on line 748 with a sort -u.
> 
