Re: [Linux-usb-users] Writing to SD-Card very slow if mounted with 'sync'
On Wed, 2005-09-28 at 20:24 +0200, Marc Schiffbauer wrote: > * Alan Stern schrieb am 28.09.05 um 18:58 Uhr: > > On Wed, 28 Sep 2005, Marc Schiffbauer wrote: > > > > > Hi all, > > > > > > I have an usb-storage device in my Computer hat can handle several > > > memory-cards like CF, SD, MMC, MS and SM Cards. > > > > > > I often read/write to my 1 gig SD card (USB 2.0). Reading is always > > > fast (~ 5MB/s) using Kernel 2.6.12 or 2.6.13. > > > > > > But: Writing to the card is very very slow: < 40 kb/s. > > > > > > My kubuntu system mounts that kind of devices like that: > > > /dev/sda1 on /media/sda1 type vfat > > > > > > (rw,noexec,nosuid,nodev,sync,quiet,uid=1000,gid=1000,umask=077,iocharset=utf8) > > > > > > Now I discovered that if I do a "mount -o remount,async /meda/sda1", > > > the writing speed increases to a normal value of about 5 megs/s. > > > > > > I remember, that the problem did not exist with earlier kernels > > > (and this version of kubuntu) > > > > > > I think this might be a kernel bug introduced by some patch applied before > > > 2.6.12. > > > > It's not a bug; it's deliberate. :-) > > > > See this bug report: > > > > http://bugzilla.kernel.org/show_bug.cgi?id=4882 > > > > Thanks Alan for pointing to it. Seems like using sync on FAT devices > is a very bad idea... now the question is how to I make hald - or > whatever mounts the drive in the end - not to use the sync option... > but that is Offtopic I guess. If you are using a recent version of Fedora, just update hald. They realized what a serious mistake that was and fixed it in an update. I had that option destroy a couple of USB flash memory cards just by copying a 700 Meg file to them. It ended up beating on the FAT so bad that even the wear leveling failed to protect it against flash wear-out. Thanks to a suggestion one person made on the kernel mailing list I was able to recover the burned out device (it only had a dozen or so bad sectors - mostly the fat tables) by blind writing across the entire drive using dd against the device and flawing out the burned sectors (which then results in a good drive with a smaller internal spares table, I guess). > Thanks again! > > Cheers > -Marc Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] /\/\|=mhw=|\/\/ | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! signature.asc Description: This is a digitally signed message part
Re: [Linux-usb-users] Writing to SD-Card very slow if mounted with 'sync'
On Wed, 28 Sep 2005, Marc Schiffbauer wrote: > Thanks Alan for pointing to it. Seems like using sync on FAT devices > is a very bad idea... now the question is how to I make hald - or > whatever mounts the drive in the end - not to use the sync option... > but that is Offtopic I guess. How to adjust hald is discussed in one of the links from that bug report. Alan Stern --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
Re: [Linux-usb-users] Writing to SD-Card very slow if mounted with 'sync'
* Alan Stern schrieb am 28.09.05 um 18:58 Uhr: > On Wed, 28 Sep 2005, Marc Schiffbauer wrote: > > > Hi all, > > > > I have an usb-storage device in my Computer hat can handle several > > memory-cards like CF, SD, MMC, MS and SM Cards. > > > > I often read/write to my 1 gig SD card (USB 2.0). Reading is always > > fast (~ 5MB/s) using Kernel 2.6.12 or 2.6.13. > > > > But: Writing to the card is very very slow: < 40 kb/s. > > > > My kubuntu system mounts that kind of devices like that: > > /dev/sda1 on /media/sda1 type vfat > > > > (rw,noexec,nosuid,nodev,sync,quiet,uid=1000,gid=1000,umask=077,iocharset=utf8) > > > > Now I discovered that if I do a "mount -o remount,async /meda/sda1", > > the writing speed increases to a normal value of about 5 megs/s. > > > > I remember, that the problem did not exist with earlier kernels > > (and this version of kubuntu) > > > > I think this might be a kernel bug introduced by some patch applied before > > 2.6.12. > > It's not a bug; it's deliberate. :-) > > See this bug report: > > http://bugzilla.kernel.org/show_bug.cgi?id=4882 > Thanks Alan for pointing to it. Seems like using sync on FAT devices is a very bad idea... now the question is how to I make hald - or whatever mounts the drive in the end - not to use the sync option... but that is Offtopic I guess. Thanks again! Cheers -Marc -- +-O . . . o . . . O . . . o . . . O . . . ___ . . . O . . . o .-+ | Ein Service von Links2Linux.de: / o\ RPMs for SuSE | | --> PackMan! <-- naeheres unter| __| and others| | http://packman.links2linux.de/ . . . O \__\ . . . O . . . O . | --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
Re: [Linux-usb-users] Writing to SD-Card very slow if mounted with 'sync'
On Wed, 28 Sep 2005, Marc Schiffbauer wrote: > Hi all, > > I have an usb-storage device in my Computer hat can handle several > memory-cards like CF, SD, MMC, MS and SM Cards. > > I often read/write to my 1 gig SD card (USB 2.0). Reading is always > fast (~ 5MB/s) using Kernel 2.6.12 or 2.6.13. > > But: Writing to the card is very very slow: < 40 kb/s. > > My kubuntu system mounts that kind of devices like that: > /dev/sda1 on /media/sda1 type vfat > > (rw,noexec,nosuid,nodev,sync,quiet,uid=1000,gid=1000,umask=077,iocharset=utf8) > > Now I discovered that if I do a "mount -o remount,async /meda/sda1", > the writing speed increases to a normal value of about 5 megs/s. > > I remember, that the problem did not exist with earlier kernels > (and this version of kubuntu) > > I think this might be a kernel bug introduced by some patch applied before > 2.6.12. It's not a bug; it's deliberate. :-) See this bug report: http://bugzilla.kernel.org/show_bug.cgi?id=4882 Alan Stern --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
