Re: growisofs minus_rw_quickgrow question
> I was trying to write arbitrary files to DVD-RW using growisofs. When I > use > dumpsomething | growisofs -use-the-force-luke=seek:N -Z /dev/dvd=/dev/fd/0, > the DVD can not be play back. I think it was because the media wasn't > closed, Define "can not be played back." Point is that under most circumstances recorder unit is perfectly capable of accessing open media and it's essential to understand how does "can not be played back" manifest itself in order to answer the question. Always(!) complement problem reports with dvd+rw-mediainfo output (as well as exact growisofs command line and exact output). > but I don't know how to set "quickgrown=1" to active "Close session" > cmd in minus_rw_finalize. "quickgrown" is DVD-RW Restricted Overwrite thing, so I assume we're talking about DVD-RW Restricted Overwrite. Note that in order for the referred procedure to work with DVD-RW media, the latter has to be *fully* formatted prior recording(*), e.g. with 'dvd+rw-format -force=full /dev/dvd'(**). Why? Here is how DVD-RW Restricted Overwrite works. Imagine you've just formatted your DVD-RW disc with 'dvd+rw-format -force /dev/dvd'. The only thing you can do with it at this point is *write* from LBA 0 in sequential(!) manner. After you've written some data (in sequential manner) you write a lead-out, let's say at LBA X. After that [and only after that] you can intermix reads *and* [aligned] writes with arbitrary LBA less than X. Whenever you want to add more data you have to remove the lead-out at LBA X, after which the only thing you can do is *write* again and from LBA X in strictly sequential(!) manner. Before you can intermix reads and writes again, you have to write lead-out at the point sequential recording has proceeded to beyond X. What does it mean for the referred command? Given that growisofs does not support "on-the-fly(***) DVD-RW Quick Grow," it means that if N is less than X, but the stream is larger than X-N, then recording is bound to fail. As you most likely don't know the exact values of stream size and X in advance, the only way to assure reliable results is to fully format the media. The fact that Quick Grow procedure is not properly applied with -use-the-force-luke=seek:N might as well be a bug, and I'll look into it(), but meanwhile do make sure the media is fully formatted. (*) Keep in mind that we're talking about DVD-RW, DVD+RW does not have to be fully formatted and the referred command should always work. (**) Another way to make sure media is fully formatted is to 'growisofs -Z /dev/dvd=/dev/zero'; (***) Not with -use-the-force-luke=seek:N that is. Quick Grow procedure is properly applied for growisofs -M recordings automatically/transparently. () But I need your dvd+rw-mediainfo output for this. > So, I checked the minus_rw_quickgrow function, and > get confused. Can Andy or anyone please explain how can type ==2 be true > when format was not initialized? See above and imagine yourself at the point right after you've written lead-out at LBA X. > If I just add quickgrown=1, what will happen? Isn't it easier to just try? We're talking about rewritable media, so you can always just start from scratch with 'dvd+rw-format -force /dev/dvd':-) A. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: occasional problem formatting/rewriting DVD+RW disk with Sony DRU530A burner.
> I have 2 DVD burners on the same machine, the SONY DRU530A (primary) and the > SONY 510A (slave). I have no problem burning DVD+R and DVD+RW using DRU510A. I > have occasional problem formatting and rewriting data (use -Z option) on DVD+RW > disks using the DRU530A. The problem only happens with the DVD+RW disks. I have > no problem formatting a virgin DVD+RW disk the first time using DRU530A. The > problem occurs when rewriting data (or ISO image) the second time or after with > different size of data from the previous one. ^ You mean "larger," right? If yes, then transparent resuming of background formatting is likely to be broken. Note that "transparent resuming of background formatting" is a firmware procedure. > The message on the terminal shows > that it keeps the 0% (data written) for a while and frozen at the step of > de-icing Exact output from growisofs is always preferred. I mean for example here I'm uncertain if "step of de-icing" refers to "stopping de-icing" at the end. Must be, but point is that I don't want spill time wondering and assuming:-) > (I have to turn off PC machine and reboot after). Does it mean that OS stops responding/locks up and computer is rendered useless or does it mean that it works, but the only way to get rid of hanging growisofs is to cycle power? If latter for how long did you left it running? How large are recordings? > It also fails to > reformat the disk when this problem occurs. However, I can reformat the same > disk using the SONY DRU510A. In either case it smells very much like 530 firmware deficiency... Check for firmware update... Then see if -dvd-compat option makes it work reliably (it might *mitigate* the impact from bug in "transparent resuming of background formatting," if any that is). > I notice that the 2 machine firmware format the DVD+RW disks differently. The > DRU530A formats very fast. I don't even see the % of disk formatted. When I > check (run dv+rw-format without the -force option), the DRU530A only format a > disk to 0.1% of capacity while the DRU510A formats to 11% of disk capacity. > Any idea about this problem? It's not necessarily a problem. 0.1% is most likely lowest value ever observed, but it doesn't necessarily mean it's wrong. Formally initial format procedure doesn't have to touch the data area. It does it to provide some limited DVD-ROM compatibility, but it's not necessary for recording per se. A. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: occasional problem formatting/rewriting DVD+RW disk with Sony DRU530A burner.
> Then see if -dvd-compat option makes it work > reliably (it might *mitigate* the impact from bug in "transparent > resuming of background formatting," if any that is). Just one little detail. You need 5.17 for the above, because versions prior 5.15 don't implement explicit background formatting resume. A. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Updated DVD patch for cdrecord
Warly wrote: Available at http://people.mandrakesoft.com/~warly/files/cdrtools/ Updated DVD patch (at present against a25) with a better DVD+RW formatting and burning support. scanbus will also display dev=ATA scanning if no dev option is passed on the command line, mostly because we are defaulting on ide-cd with 2.6 linux kernel. One question: is this against the Mandrake modified cdrecord in the same directory, or the unmodified source from Joerg? Or are the 'mdk' versions just stock source labeled to indicate the source? -- E. Robert Bogusta It seemed like a good idea at the time -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: extension of mkisofs for incremental backups
On Thu, 2004-03-04 at 22:03, Joerg Schilling wrote:
> >Old-Return-Path: <[EMAIL PROTECTED]>
>
> >fixed the places that you mentioned manually. Please
> >be patient if I didn't get all of them right and perhaps
> >consider making cstyle available on your web site if you
> >expect patches to be cstyle-clean (especially the correct
> >line continuation isn't clear to me).
>
> Looks better now (only one deviation):
>
> multi.c: 1615: continuation line not indented by 4 spaces
> strcmp(new_orig_contents[i]->name,
> reloc_old_root) != 0) {
>
> But it seems that you have a strange editor:
>
> if (merge_image != NULL) {
> if (merge_previouS_session(ROOT, MROOTP,
>
>reloc_root, reloc_old_root) < 0) {
I was experimenting with different tab sizes in emacs, for
me it looked okay. But this is one of those continuation
lines that I don't know how they should be handled. Anyway,
thanks for fixing it and including the patch (and let's
not get into a discussion about editors, please ;-)
> Didn't you see this:
>
> #define OPTION_CHRP_BOOT 2042
>
> + #define OPTION_RELOC_ROOT 2042
> + #define OPTION_RELOC_OLD_ROOT 2043
> +
Sorry, I forgot that these magic constants need to be unique.
Perhaps now is a good time to change this code over to enums...
just an idea, though.
--
Bye, Patrick Ohly
--
[EMAIL PROTECTED]
[EMAIL PROTECTED] (MakeCD related mails)
http://home.pages.de/~ohly/
http://makecd.core.de/ (MakeCD home page)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Updated DVD patch for cdrecord
Rob Bogus <[EMAIL PROTECTED]> writes: > Warly wrote: > >>Available at http://people.mandrakesoft.com/~warly/files/cdrtools/ >> >>Updated DVD patch (at present against a25) with a better DVD+RW >>formatting and burning support. >> >>scanbus will also display dev=ATA scanning if no dev option is passed on >>the command line, mostly because we are defaulting on ide-cd with 2.6 >>linux kernel. >> >> >> > One question: is this against the Mandrake modified cdrecord in the > same directory, or the unmodified source from Joerg? Or are the 'mdk' > versions just stock source labeled to indicate the source? Unmodified source from Joerg. This patch is the one used into the mandrake modified cdrecord. -- Warly -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: growisofs minus_rw_quickgrow question
> I was trying to write arbitrary files to DVD-RW using growisofs. When I > use > dumpsomething | growisofs -use-the-force-luke=seek:N -Z > /dev/dvd=/dev/fd/0, > the DVD can not be play back. I think it was because the media wasn't > closed, Define "can not be played back." Point is that under most circumstances recorder unit is perfectly capable of accessing open media and it's essential to understand how does "can not be played back" manifest itself in order to answer the question. Always(!) complement problem reports with dvd+rw-mediainfo output (as well as exact growisofs command line and exact output). > but I don't know how to set "quickgrown=1" to active "Close session" > cmd in minus_rw_finalize. "quickgrown" is DVD-RW Restricted Overwrite thing, so I assume we're talking about DVD-RW Restricted Overwrite. Note that in order for the referred procedure to work with DVD-RW media, the latter has to be *fully* formatted prior recording(*), e.g. with 'dvd+rw-format -force=full /dev/dvd'(**). Why? Here is how DVD-RW Restricted Overwrite works. Imagine you've just formatted your DVD-RW disc with 'dvd+rw-format -force /dev/dvd'. The only thing you can do with it at this point is *write* from LBA 0 in sequential(!) manner. After you've written some data (in sequential manner) you write a lead-out, let's say at LBA X. After that [and only after that] you can intermix reads *and* [aligned] writes with arbitrary LBA less than X. Whenever you want to add more data you have to remove the lead-out at LBA X, after which the only thing you can do is *write* again and from LBA X in strictly sequential(!) manner. Before you can intermix reads and writes again, you have to write lead-out at the point sequential recording has proceeded to beyond X. What does it mean for the referred command? Given that growisofs does not support "on-the-fly(***) DVD-RW Quick Grow," it means that if N is less than X, but the stream is larger than X-N, then recording is bound to fail. As you most likely don't know the exact values of stream size and X in advance, the only way to assure reliable results is to fully format the media. The fact that Quick Grow procedure is not properly applied with -use-the-force-luke=seek:N might as well be a bug, and I'll look into it(), but meanwhile do make sure the media is fully formatted. (*) Keep in mind that we're talking about DVD-RW, DVD+RW does not have to be fully formatted and the referred command should always work. (**) Another way to make sure media is fully formatted is to 'growisofs -Z /dev/dvd=/dev/zero'; (***) Not with -use-the-force-luke=seek:N that is. Quick Grow procedure is properly applied for growisofs -M recordings automatically/transparently. () But I need your dvd+rw-mediainfo output for this. > So, I checked the minus_rw_quickgrow function, and > get confused. Can Andy or anyone please explain how can type ==2 be true > when format was not initialized? See above and imagine yourself at the point right after you've written lead-out at LBA X. > If I just add quickgrown=1, what will happen? Isn't it easier to just try? We're talking about rewritable media, so you can always just start from scratch with 'dvd+rw-format -force /dev/dvd':-) A.
Re: occasional problem formatting/rewriting DVD+RW disk with Sony DRU530A burner.
> I have 2 DVD burners on the same machine, the SONY DRU530A (primary) and the > SONY 510A (slave). I have no problem burning DVD+R and DVD+RW using DRU510A. I > have occasional problem formatting and rewriting data (use -Z option) on > DVD+RW > disks using the DRU530A. The problem only happens with the DVD+RW disks. I > have > no problem formatting a virgin DVD+RW disk the first time using DRU530A. The > problem occurs when rewriting data (or ISO image) the second time or after > with > different size of data from the previous one. ^ You mean "larger," right? If yes, then transparent resuming of background formatting is likely to be broken. Note that "transparent resuming of background formatting" is a firmware procedure. > The message on the terminal shows > that it keeps the 0% (data written) for a while and frozen at the step of > de-icing Exact output from growisofs is always preferred. I mean for example here I'm uncertain if "step of de-icing" refers to "stopping de-icing" at the end. Must be, but point is that I don't want spill time wondering and assuming:-) > (I have to turn off PC machine and reboot after). Does it mean that OS stops responding/locks up and computer is rendered useless or does it mean that it works, but the only way to get rid of hanging growisofs is to cycle power? If latter for how long did you left it running? How large are recordings? > It also fails to > reformat the disk when this problem occurs. However, I can reformat the same > disk using the SONY DRU510A. In either case it smells very much like 530 firmware deficiency... Check for firmware update... Then see if -dvd-compat option makes it work reliably (it might *mitigate* the impact from bug in "transparent resuming of background formatting," if any that is). > I notice that the 2 machine firmware format the DVD+RW disks differently. The > DRU530A formats very fast. I don't even see the % of disk formatted. When I > check (run dv+rw-format without the -force option), the DRU530A only format a > disk to 0.1% of capacity while the DRU510A formats to 11% of disk capacity. > Any idea about this problem? It's not necessarily a problem. 0.1% is most likely lowest value ever observed, but it doesn't necessarily mean it's wrong. Formally initial format procedure doesn't have to touch the data area. It does it to provide some limited DVD-ROM compatibility, but it's not necessary for recording per se. A.
Re: occasional problem formatting/rewriting DVD+RW disk with Sony DRU530A burner.
> Then see if -dvd-compat option makes it work > reliably (it might *mitigate* the impact from bug in "transparent > resuming of background formatting," if any that is). Just one little detail. You need 5.17 for the above, because versions prior 5.15 don't implement explicit background formatting resume. A.
Re: Updated DVD patch for cdrecord
Warly wrote: Available at http://people.mandrakesoft.com/~warly/files/cdrtools/ Updated DVD patch (at present against a25) with a better DVD+RW formatting and burning support. scanbus will also display dev=ATA scanning if no dev option is passed on the command line, mostly because we are defaulting on ide-cd with 2.6 linux kernel. One question: is this against the Mandrake modified cdrecord in the same directory, or the unmodified source from Joerg? Or are the 'mdk' versions just stock source labeled to indicate the source? -- E. Robert Bogusta It seemed like a good idea at the time
Re: extension of mkisofs for incremental backups
On Thu, 2004-03-04 at 22:03, Joerg Schilling wrote:
> >Old-Return-Path: <[EMAIL PROTECTED]>
>
> >fixed the places that you mentioned manually. Please
> >be patient if I didn't get all of them right and perhaps
> >consider making cstyle available on your web site if you
> >expect patches to be cstyle-clean (especially the correct
> >line continuation isn't clear to me).
>
> Looks better now (only one deviation):
>
> multi.c: 1615: continuation line not indented by 4 spaces
> strcmp(new_orig_contents[i]->name,
> reloc_old_root) != 0) {
>
> But it seems that you have a strange editor:
>
> if (merge_image != NULL) {
> if (merge_previouS_session(ROOT, MROOTP,
>
> reloc_root, reloc_old_root) < 0) {
I was experimenting with different tab sizes in emacs, for
me it looked okay. But this is one of those continuation
lines that I don't know how they should be handled. Anyway,
thanks for fixing it and including the patch (and let's
not get into a discussion about editors, please ;-)
> Didn't you see this:
>
> #define OPTION_CHRP_BOOT 2042
>
> + #define OPTION_RELOC_ROOT 2042
> + #define OPTION_RELOC_OLD_ROOT 2043
> +
Sorry, I forgot that these magic constants need to be unique.
Perhaps now is a good time to change this code over to enums...
just an idea, though.
--
Bye, Patrick Ohly
--
[EMAIL PROTECTED]
[EMAIL PROTECTED] (MakeCD related mails)
http://home.pages.de/~ohly/
http://makecd.core.de/ (MakeCD home page)
Re: Updated DVD patch for cdrecord
Rob Bogus <[EMAIL PROTECTED]> writes: > Warly wrote: > >>Available at http://people.mandrakesoft.com/~warly/files/cdrtools/ >> >>Updated DVD patch (at present against a25) with a better DVD+RW >>formatting and burning support. >> >>scanbus will also display dev=ATA scanning if no dev option is passed on >>the command line, mostly because we are defaulting on ide-cd with 2.6 >>linux kernel. >> >> >> > One question: is this against the Mandrake modified cdrecord in the > same directory, or the unmodified source from Joerg? Or are the 'mdk' > versions just stock source labeled to indicate the source? Unmodified source from Joerg. This patch is the one used into the mandrake modified cdrecord. -- Warly
