AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20070418174407.5d15c06e@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:onstor-exch02.onstor.net
NSV:
SSH:
R:<maxim.kozlovsky@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/andys@onstor.net@onstor-exch02.onstor.net/INBOX	0	BB375AF679D4A34E9CA8DFA650E2B04E0351BA33@onstor-exch02.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Wed, 18 Apr 2007 17:45:03 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: "Maxim Kozlovsky" <maxim.kozlovsky@onstor.com>
Subject: Re: Please review
Message-ID: <20070418174503.23b0e555@ripper.onstor.net>
In-Reply-To: <BB375AF679D4A34E9CA8DFA650E2B04E0351BA33@onstor-exch02.onstor.net>
References: <BB375AF679D4A34E9CA8DFA650E2B04E032A6A27@onstor-exch02.onstor.net>
	<20070405141235.508a5be7@ripper.onstor.net>
	<BB375AF679D4A34E9CA8DFA650E2B04E032A6A40@onstor-exch02.onstor.net>
	<20070418163503.24a3dab5@ripper.onstor.net>
	<BB375AF679D4A34E9CA8DFA650E2B04E0351BA33@onstor-exch02.onstor.net>
Organization: Onstor
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

On Wed, 18 Apr 2007 17:16:38 -0700 "Maxim Kozlovsky"
<maxim.kozlovsky@onstor.com> wrote:

> inux/include/net/neteee/eee.h
> 
>      line 37, you changed the values here, but the comment still
> mentions
>      arbitrary values.  should the comment be updated (and r. beck
> removed).
> 
> MK> OK
> 
> nfx-tree/Includes/rpc/rpc.h
> 
>      line 57 ifdef.  i thought you fixed all ifdefs.
> 
> nfx-tree/Includes/rpc/xdr.h
> 
>      line 54 ifdef
> 
> MK> Let's make an exception on those two. This is so messed up that I
> don't know how to fix it, who in there right mind would define the
> includes that match the system include files

This is not an exception.

If they don't belong, get rid of them all together, and fix any code
as necessary. this isn't just an exercise in porting code but a time to
clean things up.  The best thing is to axe this include file, and see
what code stops compiling.  Probably very little?

Alternatively, you could just have these files be in the include list
only on BSD makes.  That's probably the least effort method.

I thought you wrote all this code?  ~:^)  But seriously, this is pretty
messed up.

> nfx-tree/Tools/defs.mk
> 
>      line 324, comment should be "if bobcat-linux"
> 
> MK> OK
> 
> 
> nfx-tree/code/sm-event/linux.h
> 
>      geez, at least do the work of making it a C file, as I suspect
> this function is called in many places.  we don't want to make it an
> inline.
> 
> MK> One thing at a time, OK? This is harder than you think - why would
> the person who wrote this code make it inline instead of doing the
> right thing?

Well then, are you keeping a list of things you're definitely going to
go back and fix?  BTW, it's not harder than I think.  If it was easy,
we would have given the job to Jon.

Are you saying it was originally inline?  Again, this is our big
opportunity to make things better.  It's only a few key strokes, and we
all know how fast you type.

> nfx-tree/code/sm-event/openbsd.h
> 
>      ditto with the inline.  isn't this function called a lot?
> MK> See above. 
> 
> 
> nfx-tree/code/ssc-bsdcompat/openbsd-fgetln.c
> 
>      uh, this file is empty.  no empty files.
> 
>      just have the makefile include local OS variants of an include
> file,
>      in which the list of files to compile is defined.  so if linux
> has one or two more or one or two less source files, the respective
>      list of source files would be defined in the included makefile,
>      like linux.mk, openbsd.mk, etc.
> MK> I don't think building a library with an empty list of files
> MK> works.

hmm.  i don't get it.  why would it be empty?

> nfx-tree/code/ssc-bsdcompat/openbsd.h
> 
>      ditto for this essentially empty file as well
> MK> Seems to be necessary, otherwise you'll get an error on missing
> include file when doing '#include OS_INCL'

hmm

> nfx-tree/code/ssc-cluster/Makefile
> 
>      looks good
> 
> 
> nfx-tree/code/ssc-cluster/openbsd.h
> 
>      no empty files.  hmm, actually, maybe it's ok since inevitably
> something
>      will go in here.
> 
>      but no DOS format files, please.  that goes for all your files.
> 
> nfx-tree/code/ssc-nfxnis/Makefile
> 
>      i think i covered this one in other comments
> MK> Covered what?

No idea.  looks fine.

> nfx-tree/code/ssc-nfxnis/nfxdns-openbsd.c
> 
>      line 729 is whitespace, and looks like a bug to me.  and why one
> should
>      always use curly braces, even for one line.
> 
> MK> Right. Go tell this to BSD guys. Doesn't look like a bug, just
> messed up indentation.

but you're the BSD guy.  isn't this our code?

> nfx-tree/code/ssc-nfxnis/nfxnis.c
> 
>      line 676, what is wrong with isblank?
> MK> does not compile. 

problem with our cross environment then.  i'll inform the media.  i
mean, i'll inform larry.

> nfx-tree/code/ssc-nfxnis/openbsd/resolv-internal.h
> 
>      err, empty file?  no-no.  unless it will eventually get stuff in
> it.
> MK> What's the problem with empty files? I can put a copyright notice
> there, this will make it non-empty.

the answer to this question is not immediately coming to mind.  i just
know we shouldn't have any.  any ideas?

> nfx-tree/code/ssc-openafs-lwp/process.S
> 
>      line 18, forbidden ifdef
> MK> Let's leave this one for a while. I am not sure that this is 

Well it looks like you already fixed it ~:^)

> nfx-tree/code/ssc-pm/Makefile
> 
>      line 66, illegal ifdef
> OK
> 
> nfx-tree/code/ssc-pm/linux.c
> 
>      line 57, 81 whitespace
> OK
> 
> nfx-tree/code/ssc-pm/openbsd.c
> 
>      line 102, 111 113, whitespace
> 
> OK
> 
> nfx-tree/code/ssc-vsd/vs-ipm-openbsd.c
> 
>      gar, bombarded by whitespace!  think of the disk space you'll
> save.
> 
>      but seriously, well, i was already, but seriously, none of this
> code
>      is portable?
> MK> Some is, but it will require additional effort to separate it.
> Starting from the assumption that none is portable is safe.

throw it on the list.  what is vserver-ipm anyway?

