|
From: Paul A. <pal...@ea...> - 2003-08-28 02:02:56
|
Actually, I just was trying to get an error on ctrl-C. 1. First "tree -a /mnt/1wire no error. Perhaps problem is not multithreaded request? 2/ Second 'cat /mnt/1wire/*/tempC no error. 3. cd /mnt/1wire ahha! fusermount: failed to unmount /mnt/1wire: Device or resource busy. 4. From the author of FUSE: > From: > Miklos Szeredi > <Mik...@et...> > To: > pal...@ea... > Subject: > Re: fuse > Date: > Wed, 27 Aug 2003 > 10:02:14 +0200 > (MEST) > > > > I'm getting many questions, now, on FUSE: > > > > 1. mounting/unmounting seems a little awkward. mount points do not > seem > > to be cleanly released. > > Yes, that is something that will be cleanly solved by the forced > unmount thing in linux 2.6. So this may not be easy to fix. Paul On Wed, 2003-08-27 at 21:12, Gregg C Levine wrote: > Hello from Gregg C Levine > Is that what it is? Nice to know we've got an idea what it is. Next > time I run things here, I'll make an effort to trap the debugging > output via a script file, from the one-wire program. Now the joker, > what part of the decoded oops gave you the clew? It looked like > typical Pentium assembler language to me. > ------------------- > Gregg C Levine han...@wo... > ------------------------------------------------------------ > "The Force will be with you...Always." Obi-Wan Kenobi > "Use the Force, Luke." Obi-Wan Kenobi > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > (This company dedicates this E-Mail to Master Yoda ) > > > > > -----Original Message----- > > From: Paul Alfille [mailto:pal...@ea...] > > Sent: Wednesday, August 27, 2003 8:51 PM > > To: Gregg C Levine > > Subject: RE: OWFS > > > > Ahha, > > Yes, when you quit OWFS, all sort of nasty messages appear. Seems > > harmless, but ugly. Something about device not open, etc. I suspect > I'm > > trying to close the com port in each of the threads. > > > > I'd certainly like to chase down each of these awkward bugs. It > would > > certainly give me more confidence of the program. > > > > Paul > > > > On Tue, 2003-08-26 at 23:53, Gregg C Levine wrote: > > > Hello from Gregg C Levine > > > I haven't as of right now had the chance to run the oops decoder > that > > > I've got squirreled away someplace, on that oops message. As far > as I > > > can tell, this happened, when I interrupted the owfs program with > a > > > control-C, so that I could close the windows, and leave the KDE > > > desktop. I saw it, when I ran dmesg to see if anything was printed > > > there, via that action. > > > Tomorrow on the other hand, I'll have more information for you. On > the > > > third hand (?), I don't think my manipulating a DS2406 at the file > in > > > named directory level caused it. At the shell prompt I would type > > > either a "#echo "1" > PIO_A", to turn the output on, or a "#echo > "0" > > > > PIO_A" to turn it off. So, as I stated earlier I'll run the oops > > > through the decoder, and then post the output on a second reply to > > > your message. This is with the latest version of the ow program. > > > ------------------- > > > Gregg C Levine han...@wo... > > > ------------------------------------------------------------ > > > "The Force will be with you...Always." Obi-Wan Kenobi > > > "Use the Force, Luke." Obi-Wan Kenobi > > > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > > > (This company dedicates this E-Mail to Master Yoda ) > > > > > > > > > > > > > -----Original Message----- > > > > From: Paul Alfille [mailto:pal...@ea...] > > > > Sent: Tuesday, August 26, 2003 11:07 PM > > > > To: Gregg C Levine > > > > Cc: owfs-developers > > > > Subject: RE: OWFS > > > > > > > > Please explain what you mean. > > > > > > > > Is it a particular device? (Not all are really tested, I don't > have > > > > all the chips and in any case didn't exhaustively test > everything.) > > > > > > > > Is it memory access? I do split large memory requests into many > > > > smaller ones in a not very efficient manner. Perhaps we have a > stack > > > > overflow. > > > > > > > > That would be easy to fix, > > > > > > > > Instead of > > > > if (size>16) return GetMem( 16) || GetMem( size-16) > > > > > > > > I could do a binary split, or tail recusion, or iterative > > > techniques. > > > > > > > > Paul > > > > > > > > On Tue, 2003-08-26 at 21:51, Gregg C Levine wrote: > > > > > Hello again from Gregg C Levine > > > > > Oh, and yes, and no. Yes, because when I commented it out it > > > compiled > > > > > okay, but through an oops instead. I tried an earlier version, > and > > > it > > > > > through one as well. Its a fuse issue, it seems the loaded > module > > > > > can't support the paging aspects of the program. > > > > > ------------------- > > > > > Gregg C Levine han...@wo... > > > > > ------------------------------------------------------------ > > > > > "The Force will be with you...Always." Obi-Wan Kenobi > > > > > "Use the Force, Luke." Obi-Wan Kenobi > > > > > (This company dedicates this E-Mail to General Obi-Wan Kenobi > ) > > > > > (This company dedicates this E-Mail to Master Yoda ) > > > > > > > > > > > > > > > > > > > > > -----Original Message----- > > > > > > From: Paul Alfille [mailto:pal...@ea...] > > > > > > Sent: Tuesday, August 26, 2003 8:35 PM > > > > > > To: Gregg C Levine > > > > > > Subject: OWFS > > > > > > > > > > > > Hi Gregg, > > > > > > > > > > > > Usually, I comment out the line in main.c that calls dfork > when > > > I > > > > > debug > > > > > > so I can see console messages. Is that the problem? > > > > > > > > > > > > Paul > > > > > > > > > > > > > > > > > > > > |