Hi, I have three comps with SUSE 10.0 installed. All of them desktops and all of them can copy a 500kb file in about _20 minutes_ to a floppy. No joke, but seemingly "common feature". Reading floppies is no problem at all, only writing to a floppy is like a nightmare. Anyone else seeing the same strange behaviour?! All of the systems fully updated and the automounter is active for floppy/CD. The corresponding fstab things are defaults from the installations themselves... Thanks, Pelibali
On Tuesday 01 November 2005 04:21 pm, pelibali wrote:
Hi,
I have three comps with SUSE 10.0 installed. All of them desktops and all of them can copy a 500kb file in about _20 minutes_ to a floppy. No joke, but seemingly "common feature". Reading floppies is no problem at all, only writing to a floppy is like a nightmare. Anyone else seeing the same strange behaviour?! All of the systems fully updated and the automounter is active for floppy/CD. The corresponding fstab things are defaults from the installations themselves...
Thanks, Pelibali
Pelibali, Mine doesn't take quite 20 minutes, but it is a very long time to copy any files to the floppy with SuSE 10! It seems to read as quickly as ever, but writing to a floppy is something SuSe has managed to mangle quite badly in their efforts! What's going on SuSE guys? Granted not many people use floppies these days, heck, not many computers even come with one anymore! I still have need of one, as I'm sure others do, so can someone help in fixing this problem? I've tried changing from the subfs to standard without making any difference. Lee
On 11/1/05, BandiPat
What's going on SuSE guys?
Granted not many people use floppies these days, heck, not many computers even come with one anymore! I still have need of one, as I'm sure others do, so can someone help in fixing this problem? I've tried changing from the subfs to standard without making any difference.
Lee
As far as SuSE already took care to provide a way for user's feedback, the right way to go is to file a bug report at opensuse.org. Please, spend some time to log it. -- -- Svetoslav Milenov (Sunny)
BandiPat wrote:
On Tuesday 01 November 2005 04:21 pm, pelibali wrote:
Hi,
I have three comps with SUSE 10.0 installed. All of them desktops and all of them can copy a 500kb file in about _20 minutes_ to a floppy. No joke, but seemingly "common feature". Reading floppies is no problem at all, only writing to a floppy is like a nightmare. Anyone else seeing the same strange behaviour?! All of the systems fully updated and the automounter is active for floppy/CD. The corresponding fstab things are defaults from the installations themselves...
Thanks, Pelibali
Pelibali, Mine doesn't take quite 20 minutes, but it is a very long time to copy any files to the floppy with SuSE 10! It seems to read as quickly as ever, but writing to a floppy is something SuSe has managed to mangle quite badly in their efforts!
What's going on SuSE guys?
Granted not many people use floppies these days, heck, not many computers even come with one anymore! I still have need of one, as I'm sure others do, so can someone help in fixing this problem? I've tried changing from the subfs to standard without making any difference.
Are those with USB attached floppies? There's a severe bug, that causes extremely slow writes, to USB drives.
On Tuesday 01 November 2005 05:11 pm, BandiPat wrote:
On Tuesday 01 November 2005 04:21 pm, pelibali wrote:
Hi,
I have three comps with SUSE 10.0 installed. All of them desktops and all of them can copy a 500kb file in about _20 minutes_ to a floppy. No joke, but seemingly "common feature". Reading floppies is no problem at all, only writing to a floppy is like a nightmare. Anyone else seeing the same strange behaviour?! All of the systems fully updated and the automounter is active for floppy/CD. The corresponding fstab things are defaults from the installations themselves...
Thanks, Pelibali
***********
Pelibali, Mine doesn't take quite 20 minutes, but it is a very long time to copy any files to the floppy with SuSE 10! It seems to read as quickly as ever, but writing to a floppy is something SuSe has managed to mangle quite badly in their efforts!
Ok Pelibali, there seems to be a bug report already submitted for this problem, so it's well known by SuSE already. The bug #105871 covers mostly external floppies or other usb devices, but the fix for now works with internal floppies as well. Seems there is a bug in using "sync" with any of these devices. Simple fix is to edit your /etc/fstab and remove the word sync from your floppy line. I kept mine manual mount and unmount as well, but it doesn't matter, because subfs works without the sync. You'll be amazed at the speed it writes to your floppy after that! It was unbelievable! Anything under about 300-400Kb doesn't even open the transfer window and 500+ only shows for a few seconds! You can check to see if sync is active on the device after you insert a disk or plug a usb device up with "cat /proc/mounts | grep vfat". If you see "sync" in the result, it's going to write slow! hope that helps Lee
Pat, On Tuesday 01 November 2005 18:43, BandiPat wrote:
...
Simple fix is to edit your /etc/fstab and remove the word sync from your floppy line. I kept mine manual mount and unmount as well, but it doesn't matter, because subfs works without the sync. You'll be amazed at the speed it writes to your floppy after that! It was unbelievable! Anything under about 300-400Kb doesn't even open the transfer window and 500+ only shows for a few seconds!
You can check to see if sync is active on the device after you insert a disk or plug a usb device up with "cat /proc/mounts | grep vfat". If you see "sync" in the result, it's going to write slow!
That's mostly illusory. If the floppy is mounted w/o the sync option, then the kernel buffer pool can easily hold the entire floppie's contents. In that case, any user-level software writing to the device will appear to return immediately, but at that point almost all of the data is in memory, not on the device. On the other hand, with the synchrounous writing mode in effect, the command or program that writes to the floppy will not return until all the data has made it all the way to the diskette. But in either case the amount of time it takes to write a given amount of data to the device is the same.
hope that helps Lee
Randall Schulz
At 08:01 PM 11/1/2005 -0800, Randall R Schulz wrote:
Content-Disposition: inline
Pat,
On Tuesday 01 November 2005 18:43, BandiPat wrote:
...
Simple fix is to edit your /etc/fstab and remove the word sync from your floppy line. I kept mine manual mount and unmount as well, but it doesn't matter, because subfs works without the sync. You'll be amazed at the speed it writes to your floppy after that! It was unbelievable! Anything under about 300-400Kb doesn't even open the transfer window and 500+ only shows for a few seconds!
You can check to see if sync is active on the device after you insert a disk or plug a usb device up with "cat /proc/mounts | grep vfat". If you see "sync" in the result, it's going to write slow!
That's mostly illusory.
If the floppy is mounted w/o the sync option, then the kernel buffer pool can easily hold the entire floppie's contents. In that case, any user-level software writing to the device will appear to return immediately, but at that point almost all of the data is in memory, not on the device.
On the other hand, with the synchrounous writing mode in effect, the command or program that writes to the floppy will not return until all the data has made it all the way to the diskette.
But in either case the amount of time it takes to write a given amount of data to the device is the same.
hope that helps Lee
Randall Schulz
It's been a while since I copied a whole floppy (in Windows or DOS) but it certainly didn't take 10 minutes. I've never done it in Linux, but why should it take longer? You can format the floppy (in MS) in about 2 minutes or less. It shouldn't take much longer to write data, instead of eeee, etc. --doug -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.1.362 / Virus Database: 267.12.6/152 - Release Date: 10/31/2005
Pat,
On Tuesday 01 November 2005 18:43, BandiPat wrote:
...
Simple fix is to edit your /etc/fstab and remove the word sync from your floppy line. I kept mine manual mount and unmount as well, but it doesn't matter, because subfs works without the sync. You'll be amazed at the speed it writes to your floppy after that! It was unbelievable! Anything under about 300-400Kb doesn't even open the transfer window and 500+ only shows for a few seconds!
You can check to see if sync is active on the device after you insert a disk or plug a usb device up with "cat /proc/mounts | grep vfat". If you see "sync" in the result, it's going to write slow!
That's mostly illusory.
If the floppy is mounted w/o the sync option, then the kernel buffer pool can easily hold the entire floppie's contents. In that case, any user-level software writing to the device will appear to return immediately, but at that point almost all of the data is in memory, not on the device.
On the other hand, with the synchrounous writing mode in effect, the command or program that writes to the floppy will not return until all the data has made it all the way to the diskette.
But in either case the amount of time it takes to write a given amount of data to the device is the same.
hope that helps Lee
Randall Schulz ======== Randy, Maybe in your mind many things are illusions, but the points made about
On Tuesday 01 November 2005 11:01 pm, Randall R Schulz wrote: the floppy writing is tested and factual. If in fact the synchronous writing were being done correctly, certainly the times should be about equal, but considering it's taking from 5-20 minutes with it turned on to finish writing a 500+Kb file to the floppy, I doubt one can say minutes to write versus seconds to write are anywhere close. hopefully helpful, Lee
Lee, On Tuesday 01 November 2005 20:40, BandiPat wrote:
... Randy,
My name is Randall. It's not your place to shorten it.
Maybe in your mind many things are illusions, but the points made about the floppy writing is tested and factual.
Excuse me? All I'm saying is that synchronous mode on or off does not change the hardware data transfer rate. In normal operation, the amount of time it takes the data to get from the program writing it to the device does not depend on whether the file system is mounted in synchronous mode. Only the time one must wait for programs writing to the device to complete is changed, and that is the illusory aspect to which I refer. Bugs notwithstanding.
...
hopefully helpful,
Well, I'm not the one seeking help, only trying to prevent misunderstanding. There's a lot of confusion, misinformation and poorly stated facts in this forum and when I see it and recognize it, I feel compelled to correct or clarify it.
Lee
Randall Schulz
On Tuesday 01 November 2005 11:49 pm, Randall R Schulz wrote: [...]
Well, I'm not the one seeking help, only trying to prevent misunderstanding. There's a lot of confusion, misinformation and poorly stated facts in this forum and when I see it and recognize it, I feel compelled to correct or clarify it.
Randall Schulz
************* Randill, Well, yes you do. Although not always necessary, not always factual, but you do feel compelled to voice an opinion. Clarity is in the eye of the receiver. Sometimes the clarity one seeks is not the clarity you see, but you do try, sometimes too hard. Lee
On Tuesday 01 November 2005 21:07, BandiPat wrote:
On Tuesday 01 November 2005 11:49 pm, Randall R Schulz wrote: [...]
Well, I'm not the one seeking help, only trying to prevent misunderstanding. There's a lot of confusion, misinformation and poorly stated facts in this forum and when I see it and recognize it, I feel compelled to correct or clarify it.
Randall Schulz
************* Randill, Well, yes you do. Although not always necessary, not always factual, but you do feel compelled to voice an opinion.
I distinguish opinion from fact. You seem to want to declare something an opinion if it does not correspond to your understanding of things.
Clarity is in the eye of the receiver. Sometimes the clarity one seeks is not the clarity you see, but you do try, sometimes too hard.
And this is gibberish.
Randill, Well, yes you do. Although not always necessary, not always factual, but you do feel compelled to voice an opinion.
I distinguish opinion from fact. You seem to want to declare something an opinion if it does not correspond to your understanding of things. -------------- You took the words right out of my mouth! I knew you understood some
Clarity is in the eye of the receiver. Sometimes the clarity one seeks is not the clarity you see, but you do try, sometimes too hard.
And this is gibberish. =========== Yes, Ran, with one, who in his mind, has supreme knowledge of all
On Wednesday 02 November 2005 12:25 am, Randall R Schulz wrote: [...] things. ;-) things, I can understand how you might think that. Have a great day! Of should that be an enlightened day? Or should that be a quiet day? Whatever! Lee
On Wed, Nov 02, 2005 at 12:07:56AM -0500, BandiPat wrote:
On Tuesday 01 November 2005 11:49 pm, Randall R Schulz wrote: [...]
Well, I'm not the one seeking help, only trying to prevent misunderstanding. There's a lot of confusion, misinformation and poorly stated facts in this forum and when I see it and recognize it, I feel compelled to correct or clarify it.
Randall Schulz
************* Randill,
I believe he said RANDALL. And ummm, I have a USB ZIP Drive 100 hooked up tot his box, and an 80 GB USB HD attached to another and have yet to ever have these problems you guys are talking about. -Allen.
Well, yes you do. Although not always necessary, not always factual, but you do feel compelled to voice an opinion. Clarity is in the eye of the receiver. Sometimes the clarity one seeks is not the clarity you see, but you do try, sometimes too hard.
Lee
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
[Randall R Schulz]
[BandiPat]
Simple fix is to edit your /etc/fstab and remove the word sync from your floppy line. [...] You'll be amazed at the speed it writes to your floppy after that!
That's mostly illusory. [...] But in either case the amount of time it takes to write a given amount of data to the device is the same.
It's true that the memory cache may give the illusion of speed, and that one undoubtedly has to be careful, waiting that all of the floppy has been written for real before removing it. On the other hand, when a flurry of files are copied to the floppy, those floppy areas holding directories, and information about sector allocation and linking, are being read and rewritten many times, at least once per file written. So, letting the system use the cache brings a real benefit, because all those updates are all done in memory first, and the combined result physically written only once to the floppy. -- François Pinard http://pinard.progiciels-bpi.ca
François, On Wednesday 02 November 2005 06:31, François Pinard wrote:
[Randall R Schulz]
...
It's true that the memory cache may give the illusion of speed, and that one undoubtedly has to be careful, waiting that all of the floppy has been written for real before removing it.
On the other hand, when a flurry of files are copied to the floppy, those floppy areas holding directories, and information about sector allocation and linking, are being read and rewritten many times, at least once per file written. So, letting the system use the cache brings a real benefit, because all those updates are all done in memory first, and the combined result physically written only once to the floppy.
The system will still use the cache, it's just that writes to the device will not be held indefinitely _only_ in the cache, but rather will be flushed to device as soon as the data is written from the application level. The only real problem (again, bugs notwithstanding) occurs when the application is not writing in sector-aligned quantities. The worst case, of course, is unbuffered, single-byte writing. Line-by-line I/O, while not as bad, will still be far worse than proper sector-aligned writes. To the OP (Pelibali): What program is doing the writing that's so slow? If it's "cp" or "mv" or the default bs= / obs= values of "dd", then things ought to be OK. On the other hand, many other applications will not write in sector-aligned quantities and could trigger the symptom of very slow writing without there being any bug afoot. Randall Schulz
Hi, On Wed, 2 Nov 2005 06:57:35 -0800 Randall R Schulz <.> wrote:
To the OP (Pelibali): What program is doing the writing that's so slow? If it's "cp" or "mv" or the default bs= / obs= values of "dd", then things ought to be OK. On the other hand, many other applications will not write in sector-aligned quantities and could trigger the symptom of very slow writing without there being any bug afoot.
I tried to copy stuff under icewm and then remotely through ssh with cp, but at the end I confirmed the 20 minute e.g. under the KDE session of my girly. She has tons of floppies, so probably will push me to give her back 9.x, where she never had this problem... Thanks anyway, Pelibali PS. If you would have thoughts, why cp could be theoretically slow, please consider to tell us in few words. I would be interested to listen to your explanation, but I'm afraid, ~20 minutes can't be easily explained. Hmmm, just realizing, that all of the three comps are AMDs... Anyway, the sync/nosync options I will try asap, just to be sure that it is (or it is not) so simple.
Pelibali, On Wednesday 02 November 2005 11:54, pelibali wrote:
Hi,
...
PS. If you would have thoughts, why cp could be theoretically slow, please consider to tell us in few words. I would be interested to listen to your explanation, but I'm afraid, ~20 minutes can't be easily explained. Hmmm, just realizing, that all of the three comps are AMDs... Anyway, the sync/nosync options I will try asap, just to be sure that it is (or it is not) so simple.
1.44 MB in 20 minutes is not normal. It's about 1.2 KB per second--glacial by computer standards. You should file a bug report. Randall Schulz
On Tuesday 01 November 2005 01:21 pm, pelibali wrote:
Hi,
I have three comps with SUSE 10.0 installed. All of them desktops and all of them can copy a 500kb file in about _20 minutes_ to a floppy. No joke, but seemingly "common feature". Reading floppies is no problem at all, only writing to a floppy is like a nightmare.
What's a floppy? I think I used to have those on my Apple IIe. -- kai www.perfectreign.com linux - genuine windows replacement part
On Wednesday 02 November 2005 12:36 am, Kai Ponte wrote:
What's a floppy?
I think I used to have those on my Apple IIe. Does your IIe run Linux :-) BTW: I had an original Apple II (no e).
--
Jerry Feldman
Hi, As previously predicted, removing the 'sync' tag from fstab didn't solve my issue, when several comps copied stuff to floppies extremely slowly. I could confirm, that with AND without sync it takes about 20mins to put a single 500kb file to a floppy, independently which method/window manager/etc I attempted to use for copying the file(s). In the meantime I have 3 AMD comps _and_ one PIII with this feature, meaning, that I have _only_ older SUSE-s without this issue. I tried also various floppy-disks (, but all with FAT), several floppy-drives. Bug report will follow soon. Best, Pelibali
pelibali wrote:
Hi, As previously predicted, removing the 'sync' tag from fstab didn't solve my issue, when several comps copied stuff to floppies extremely slowly. I could confirm, that with AND without sync it takes about 20mins to put a single 500kb file to a floppy, independently which method/window manager/etc I attempted to use for copying the file(s). In the meantime I have 3 AMD comps _and_ one PIII with this feature, meaning, that I have _only_ older SUSE-s without this issue. I tried also various floppy-disks (, but all with FAT), several floppy-drives. Bug report will follow soon.
how about (from konsole): mcopy file a: This should be fast.
participants (11)
-
Allen -
BandiPat -
Doug McGarrett -
François Pinard -
James Knott -
Jerry Feldman -
Kai Ponte -
pelibali -
Randall R Schulz -
Silviu Marin-Caea -
Sunny