X-Sylpheed-Account-Id:1
S:andy.sharp@onstor.com
SCF:#mh/Mailbox/sent
X-Sylpheed-Sign:0
X-Sylpheed-Encrypt:0
X-Sylpheed-Privacy-System:
RMID:#imap/andys@onstor.net@onstor-exch02.onstor.net/INBOX	0	BB375AF679D4A34E9CA8DFA650E2B04E05DA1D1A@onstor-exch02.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Mon, 7 Apr 2008 14:17:15 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: "Henry Lau" <henry.lau@onstor.com>
Cc: "Jobi Ariyamannil" <jobi.ariyamannil@onstor.com>
Subject: Re: PERFORCE change 28718 for review
Message-ID: <20080407141715.0a3cd3df@ripper.onstor.net>
References: <20080407140236.1a06e52d@ripper.onstor.net>
	<BB375AF679D4A34E9CA8DFA650E2B04E05DA1D1A@onstor-exch02.onstor.net>
Organization: Onstor
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Mon, 7 Apr 2008 14:06:57 -0700 "Henry Lau" <henry.lau@onstor.com>
wrote:

> Hi Andy,
> 
> Please see below.
> 
> Thanks,
> Henry
> 
> -----Original Message-----
> From: Andy Sharp 
> Sent: Monday, April 07, 2008 2:03 PM
> To: Henry Lau
> Cc: Jobi Ariyamannil
> Subject: Re: PERFORCE change 28718 for review
> 
> In your change, quite a few lines are modified, but with no behavioral
> changes to routine at all. 
> >> Alignment changes.

Right.  No change to the actual behavior whatsoever.  So you couldn't
have fixed anything.

>  On top of that you changed 'done' to be a
> byte wide, instead of an integer, for no apparent reason, but
> with the result that it is slowing the whole routine down.
> >> Not sure how it can slow the whole routine.

Byte level operations are slower on mips than word level.  What would
be the reason for changing it?
 
> If you're getting segfaults, you're the only one.  Most likely you are
> not running the code you think you are.
> >> Running from the top of the tree with latest dev built.

That's what you think you're running.  But I suspect it isn't what you
are actually running.  I just tried a system copy all yesterday, with
no segfaults.


> On Mon, 7 Apr 2008 13:54:00 -0700 "Henry Lau" <henry.lau@onstor.com>
> wrote:
> 
> > HI Andy,
> > 
> > I hit this all the times in Bobcat, check the crash below. I need to
> > understand what wrong with the changes.
> > 
> > 1 #0  0x50ac9b78 in mkpath (path=0x50ac2d88 "/mnt1", mode=493,
> > dir_mode=493)
> >   2     at fs.c:637
> >   3     fs.c:637: No such file or directory.
> >   4     (gdb) dir /mnt/ssc-genlib
> >   5     Source directories searched: /mnt/ssc-genlib:$cdir:$cwd
> >   6 632         while (!done) {
> >   7     633             slash += strspn(slash, "/");
> >   8     634             slash += strcspn(slash, "/");
> >   9     635
> >  10     636             done = (*slash == '\0');
> >  11     637             *slash = '\0';
> >  12     638
> >  13     639             if (stat(path, &sb)) {
> >  14     640                 if (errno != ENOENT ||
> >  15     641                     (mkdir(path, done ? mode : dir_mode)
> > && 16 (gdb) bt
> >  17 #0  0x50ac9b78 in mkpath (path=0x50ac2d88 "/mnt1", mode=493,
> > dir_mode=493)
> >  18             at fs.c:637
> >  19 #1  0x50ac9d98 in mkpath1 (path=0x50ac2d88 "/mnt1", mode=493) at
> > fs.c:667
> >  20 #2  0x50ac93d8 in modify_readonly_file (path=0x50ac2d88 "/mnt1",
> >  21                 vfun=0x50ac9d30, arg=0x1ed) at fs.c:465
> >  22 #3  0x50ac9e6c in ensure_dir_exists (path=0x50ac2d88 "/mnt1",
> > mode=493)
> >  23                     at fs.c:689
> >  24 #4  0x50ae4fec in make_secondary_fstabs (verbose=1) at
> > fs-vector.c:91
> >  25 #5  0x5023cd3c in copy_to_secondary (
> >  26                         files=0x500dd000 ".cshrc .profile
> > 4.4BSD.dist
> >  27                         altroot bin bsd dev etc home mnt root
> > sbin txrx_bc
> >  28                         .bin fp_bc.bin fc_bc.bin stand tmp usr
> > onstor var
> >  29                         version", tar_only=1 '\001') at
> > cmd_flash.c:536
> >  30 #6  0x5023d268 in do_copy_all (argc=0, argv=0x7ffff468) at
> > cmd_flash.c:632
> >  31 #7  0x501d7f84 in system_copy_all (self=0x50268998,
> > vty=0x10047200, argc=0,
> >  32                             argv=0x7ffff468) at
> > cmd_system.c:2411 33 #8  0x5021ea70 in cmd_execute_command
> > (vline=0x100e5080, vty=0x10047200,
> >  34                                 cmd=0x7ffff530) at
> > command.c:1912 35
> > 
> > -----Original Message-----
> > From: Andy Sharp 
> > Sent: Monday, April 07, 2008 1:49 PM
> > To: Henry Lau
> > Cc: Jobi Ariyamannil
> > Subject: Re: PERFORCE change 28718 for review
> > 
> > On 7 Apr 2008 13:25:01 -0700 henryl <henry.lau@onstor.com> wrote:
> > 
> > > Change 28718 by henryl@henryl-linux on 2008/04/07 13:20:37
> > > 
> > > 	   Fixed segmentation fault like system copy all.
> > > 	   In mkpath(), do not add any char to the string after
> > > strcspn() once it reaches the end, ie done is set to TRUE.
> > > Reviewed by Jobia.
> > > 
> > > Affected files ...
> > > 
> > > ... //depot/dev/nfx-tree/code/ssc-genlib/fs.c#9 edit
> > 
> > This change is quite broken, please revert it.  And please send the
> > review to the right person in the future?
> > 
> > Thanks,
> > 
> > a
> > 
