Re: fix: lock order reversal proc/filedesc.

2003-02-15 Thread David O'Brien
On Sat, Feb 15, 2003 at 11:30:09AM +1030, Greg 'groggy' Lehey wrote: > > It is becoming increasingly clear to me that the majority of FreeBSD > > developers don't really care if their code works, as long as they get > > the credit (and / or paycheck) for committing it. > > I think that's unnecessa

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Greg 'groggy' Lehey
On Friday, 14 February 2003 at 12:23:48 +0100, Dag-Erling Smorgrav wrote: > Alfred Perlstein <[EMAIL PROTECTED]> writes: >> What exactly is broken about dumps for you on ata? > > Well, after you told me that "call dumpsys" is no longer kosher (when > did that happen, and where was it documented?),

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread phk
In message <[EMAIL PROTECTED]>, Dag-Erling Smorgrav writes: >Makes me want to get my Norwegian Sword [tm] and make a short trip to >Denmark. I my be reading my email out of order here, but I guess you changed your mind and tried to bury use with boatloads of useless asterix ('*') instead ? :-)

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Garance A Drosihn
At 3:04 AM -0800 2/14/03, Alfred Perlstein wrote: * Dag-Erling Smorgrav <[EMAIL PROTECTED]> [030214 02:52] wrote: Alfred Perlstein <[EMAIL PROTECTED]> writes: > > Thanks to Paul Saab's work on fixing twe(4) I was able to > > get a crash dump from my box > > How? I can't get a crash dump in

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Daniel O'Connor
On Fri, 2003-02-14 at 21:53, Dag-Erling Smorgrav wrote: > Alfred Perlstein <[EMAIL PROTECTED]> writes: > > What exactly is broken about dumps for you on ata? > > Well, after you told me that "call dumpsys" is no longer kosher (when > did that happen, and where was it documented?), I tried 'call do

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Alfred Perlstein
* Dag-Erling Smorgrav <[EMAIL PROTECTED]> [030214 03:23] wrote: > Alfred Perlstein <[EMAIL PROTECTED]> writes: > > What exactly is broken about dumps for you on ata? > > Well, after you told me that "call dumpsys" is no longer kosher (when > did that happen, and where was it documented?), I tried

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Kris Kennaway
On Fri, Feb 14, 2003 at 12:23:48PM +0100, Dag-Erling Smorgrav wrote: > > What exactly is broken about dumps for you on ata? > > Well, after you told me that "call dumpsys" is no longer kosher (when > did that happen, and where was it documented?), I tried 'call doadump': > > # Debugger("manual e

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Dag-Erling Smorgrav
Alfred Perlstein <[EMAIL PROTECTED]> writes: > What exactly is broken about dumps for you on ata? Well, after you told me that "call dumpsys" is no longer kosher (when did that happen, and where was it documented?), I tried 'call doadump': # Debugger("manual escape to debugger") Stopped at D

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Alfred Perlstein
* Dag-Erling Smorgrav <[EMAIL PROTECTED]> [030214 02:52] wrote: > Alfred Perlstein <[EMAIL PROTECTED]> writes: > > Thanks to Paul Saab's work on fixing twe(4) I was able to get a > > crash dump from my box > > How? I can't get a crash dump in -CURRENT, even on a plain jane ata > disk, and it's be

Re: fix: lock order reversal proc/filedesc.

2003-02-14 Thread Dag-Erling Smorgrav
Alfred Perlstein <[EMAIL PROTECTED]> writes: > Thanks to Paul Saab's work on fixing twe(4) I was able to get a > crash dump from my box How? I can't get a crash dump in -CURRENT, even on a plain jane ata disk, and it's been months since I last managed to get one. DES -- Dag-Erling Smorgrav - [E

fix: lock order reversal proc/filedesc.

2003-02-14 Thread Alfred Perlstein
Thanks to Paul Saab's work on fixing twe(4) I was able to get a crash dump from my box and figure out why my filedesc locking patch was panic'ing. kevent_close was dereferencing the proc's filedesc pointer during close, that doesn't work so well when you need to do what I had to do. :) The gist o