* Re: Atari label false positives [not found] ` <2d6f5455-963d-b044-a64d-67634df46071-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> @ 2018-05-11 15:29 ` John Paul Adrian Glaubitz [not found] ` <35ad9889-bc13-56bd-7831-f7bb3eea2d1e-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org> 0 siblings, 1 reply; 10+ messages in thread From: John Paul Adrian Glaubitz @ 2018-05-11 15:29 UTC (permalink / raw) To: Phil Susi; +Cc: Debian m68k, Linux/m68k, Parted development Hi Phil! On 05/11/2018 04:18 PM, Phil Susi wrote: > It seems that the atari label likes to falsely recognize dos labels as > atari. Looking at atari_probe, it doesn't seem like this is very hard > to do. All that it needs is one word for the size of the disk to be > less than the physical size of the disk, two bsl words to be zero or > less than the disk size, the partition start and length words to fit > within the disk, and 3 bytes to appear to be ascii characters. Is there > any way that this can be beefed up at all? Thanks for digging this up. I wasn't aware of this particular issue before and I agree, this needs to be addressed. I hope that I didn't cause too many DOS partitions to be misdetected. Sorry for the inconvenience if I did. The Atari partition table code has been working great for Debian/m68k so far though. > For instance, the name field.. is it really free form ascii or is there > a specific set of strings that could be limited to? Also this XGM and > ICD parts; are they completely optional, so we can't require one of > those be detected too? I don't know by heart, but chapter 9 in this Atari filesystem reference might help us: > https://people.debian.org/~glaubitz/Atari_HD_File_Sytem_Reference_Guide.pdf I have also CC'ed the Linux/m68k and Debian/m68k mailing lists in case some of the guys there knows the answer faster. Again, sorry for the inconvenience this may have caused. I will look into how we can improve atari_probe(). Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaubitz-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org `. `' Freie Universitaet Berlin - glaubitz-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 ^ permalink raw reply [flat|nested] 10+ messages in thread
[parent not found: <35ad9889-bc13-56bd-7831-f7bb3eea2d1e-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org>]
* Re: Atari label false positives [not found] ` <35ad9889-bc13-56bd-7831-f7bb3eea2d1e-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org> @ 2018-05-11 15:56 ` Phillip Susi [not found] ` <874bf955-d5bc-5d9c-0f87-445da67f3bdd-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> 2018-05-11 23:17 ` Michael Schmitz 0 siblings, 2 replies; 10+ messages in thread From: Phillip Susi @ 2018-05-11 15:56 UTC (permalink / raw) To: John Paul Adrian Glaubitz; +Cc: Parted development, Linux/m68k, Debian m68k [-- Attachment #1.1: Type: text/plain, Size: 819 bytes --] On 5/11/2018 11:29 AM, John Paul Adrian Glaubitz wrote: > Thanks for digging this up. I wasn't aware of this particular issue before > and I agree, this needs to be addressed. I hope that I didn't cause too > many DOS partitions to be misdetected. Sorry for the inconvenience if I > did. The Atari partition table code has been working great for > Debian/m68k so > far though. I have seen at least one or two reports of Ubuntu users getting false positives, and then I saw the bug report on the upstream parted bug tracker the other day where someone claimed to have found a logical error that caused atari_probe to return true when it shouldn't. I finally realized it was not an error ( at least I think not ). I think I will at least move atari to be probed after all other labels have failed. [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 473 bytes --] ^ permalink raw reply [flat|nested] 10+ messages in thread
[parent not found: <874bf955-d5bc-5d9c-0f87-445da67f3bdd-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org>]
* Re: Atari label false positives [not found] ` <874bf955-d5bc-5d9c-0f87-445da67f3bdd-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> @ 2018-05-11 16:10 ` John Paul Adrian Glaubitz 2018-05-11 18:30 ` [parted-devel] " Richard Z 0 siblings, 1 reply; 10+ messages in thread From: John Paul Adrian Glaubitz @ 2018-05-11 16:10 UTC (permalink / raw) To: Phillip Susi; +Cc: Parted development, Linux/m68k, Debian m68k On 05/11/2018 05:56 PM, Phillip Susi wrote: > On 5/11/2018 11:29 AM, John Paul Adrian Glaubitz wrote: >> Thanks for digging this up. I wasn't aware of this particular issue before >> and I agree, this needs to be addressed. I hope that I didn't cause too >> many DOS partitions to be misdetected. Sorry for the inconvenience if I >> did. The Atari partition table code has been working great for >> Debian/m68k so >> far though. > > I have seen at least one or two reports of Ubuntu users getting false > positives, and then I saw the bug report on the upstream parted bug > tracker the other day where someone claimed to have found a logical > error that caused atari_probe to return true when it shouldn't. I > finally realized it was not an error ( at least I think not ). According to the documentation, p.43 [1], I think offset $01 should always contain either "GEM", "BGM" or "XGM". They don't seem to be optional. I guess adding those to atari_probe() should help improve the detection quite a bit. But I would have to test that first. > I think I will at least move atari to be probed after all other labels > have failed. This sounds like a reasonable thing to do until I have worked out a patch to improve atari_probe(). Adrian > [1] https://people.debian.org/~glaubitz/Atari_HD_File_Sytem_Reference_Guide.pdf -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaubitz-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org `. `' Freie Universitaet Berlin - glaubitz-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [parted-devel] Atari label false positives 2018-05-11 16:10 ` John Paul Adrian Glaubitz @ 2018-05-11 18:30 ` Richard Z 0 siblings, 0 replies; 10+ messages in thread From: Richard Z @ 2018-05-11 18:30 UTC (permalink / raw) To: John Paul Adrian Glaubitz Cc: Phillip Susi, Debian m68k, Linux/m68k, Parted development [-- Attachment #1: Type: text/plain, Size: 1283 bytes --] On Fri, May 11, 2018 at 06:10:37PM +0200, John Paul Adrian Glaubitz wrote: > According to the documentation, p.43 [1], I think offset $01 should always > contain either "GEM", "BGM" or "XGM". They don't seem to be optional. The Atari partitioning tables are also used on the Q40/Q60 which breaks many some assumptions.. no GEM here but instead one of two Q40 specific types such as "QWA" or "QBD" in the first slot. BTW I have never seen a low-case or camel-case string in this place, this should be also used. > I guess adding those to atari_probe() should help improve the detection > quite a bit. But I would have to test that first. It would, however I am not sure what happens if a hard disk is repurposed from Atari to PC or the other way around partitioning. There may be some strings or other data left which could be misinterpreted. > >I think I will at least move atari to be probed after all other labels > >have failed. in principle both should be probed, prompting user action if both find something. They could also return a "score" - based on number of partitions recognised, their sizes and other info to give a hint which one is likely a non-random hit. Richard -- Name and OpenPGP keys available from pgp key servers [-- Attachment #2: Type: application/pgp-signature, Size: 811 bytes --] ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [parted-devel] Atari label false positives 2018-05-11 15:56 ` Phillip Susi [not found] ` <874bf955-d5bc-5d9c-0f87-445da67f3bdd-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> @ 2018-05-11 23:17 ` Michael Schmitz [not found] ` <80a28dd6-ce19-e6ed-fc7a-c5561fecf989-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 1 sibling, 1 reply; 10+ messages in thread From: Michael Schmitz @ 2018-05-11 23:17 UTC (permalink / raw) To: Phillip Susi, John Paul Adrian Glaubitz Cc: Debian m68k, Linux/m68k, Parted development Hi Phillip, Am 12.05.2018 um 03:56 schrieb Phillip Susi: > On 5/11/2018 11:29 AM, John Paul Adrian Glaubitz wrote: >> Thanks for digging this up. I wasn't aware of this particular issue before >> and I agree, this needs to be addressed. I hope that I didn't cause too >> many DOS partitions to be misdetected. Sorry for the inconvenience if I >> did. The Atari partition table code has been working great for >> Debian/m68k so >> far though. > > I have seen at least one or two reports of Ubuntu users getting false > positives, and then I saw the bug report on the upstream parted bug > tracker the other day where someone claimed to have found a logical > error that caused atari_probe to return true when it shouldn't. I > finally realized it was not an error ( at least I think not ). > > I think I will at least move atari to be probed after all other labels > have failed. The link order in the kernel (atari before msdos) has not been chosen at random. The msdos partition format probe will always succeed on any valid Atari partition table, while Atari format probe should not succeed on plain MSDOS partitions. Do the reporters if these bugs have Atari partition format enabled in their kernels? I would expect the same false positives to happen there, if the atari_probe() code is functionally equivalent to the kernel partition probe code. Do you have a dump of the root sector (plus additional eventual extended partition root sector) from these bugs? Regarding some of your other questions: The presence of extended partitions is entirely optional (though highly likely). A partition ID of XGM for one of the primary partitions, and for at least one of the extended subpartitions (if extended partitions are present) is distinctive for AHDI extended partition format. Seeing this condition fulfilled by a MSDOS partition would seem quite unlikely. I run a mix of AHDI and MSDOS partition format disks on my Atari and have never seen one misidentified by the kernel. Presence of the ICD extended partitions block at offset 0x156 of the device root sector (with only IDs of GEM, BGM, RAW, LNX or SWP expected) is distinctive for ICD partitions, and even more unlikely to encounter at random. Leaves disks without extended partitions prone to misidentifcation. If you want to make doubly sure there, look for at least one of the partition IDs commonly used in Atari format partitions (GEM, BGM, RAW, LNX, SWP), and additionally those identified by Richard for Q40. If that's still not enough, score both probe results according to the number of partitions identified, and allow the user to override. Repurposing disks that had formerly been used on Atari elsewhere should begin by overwriting the root sector IMO. Cheers, Michael ^ permalink raw reply [flat|nested] 10+ messages in thread
[parent not found: <80a28dd6-ce19-e6ed-fc7a-c5561fecf989-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]
* Re: Atari label false positives [not found] ` <80a28dd6-ce19-e6ed-fc7a-c5561fecf989-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> @ 2018-05-12 7:59 ` John Paul Adrian Glaubitz 2018-05-12 8:41 ` [parted-devel] " Michael Schmitz 0 siblings, 1 reply; 10+ messages in thread From: John Paul Adrian Glaubitz @ 2018-05-12 7:59 UTC (permalink / raw) To: Michael Schmitz, Phillip Susi; +Cc: Parted development, Linux/m68k, Debian m68k On 05/12/2018 01:17 AM, Michael Schmitz wrote: > The link order in the kernel (atari before msdos) has not been chosen at > random. The msdos partition format probe will always succeed on any > valid Atari partition table, while Atari format probe should not succeed > on plain MSDOS partitions. > > Do the reporters if these bugs have Atari partition format enabled in > their kernels? I would expect the same false positives to happen there, > if the atari_probe() code is functionally equivalent to the kernel > partition probe code. > Do you have a dump of the root sector (plus additional eventual extended > partition root sector) from these bugs? This is a very good point. We should have a look at the probing code in the kernel because apparently the problem doesn't exist there. And, yes, it would be interesting to get a dump of these boot sectors as I haven't run into the problem myself and would be very curios why the misdetection happens. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaubitz-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org `. `' Freie Universitaet Berlin - glaubitz-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [parted-devel] Atari label false positives 2018-05-12 7:59 ` John Paul Adrian Glaubitz @ 2018-05-12 8:41 ` Michael Schmitz [not found] ` <c1bf0748-f7e0-b109-8c9e-26f8a2c70bf3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 0 siblings, 1 reply; 10+ messages in thread From: Michael Schmitz @ 2018-05-12 8:41 UTC (permalink / raw) To: John Paul Adrian Glaubitz, Phillip Susi Cc: Debian m68k, Linux/m68k, Parted development Hi Adrian, Am 12.05.2018 um 19:59 schrieb John Paul Adrian Glaubitz: > On 05/12/2018 01:17 AM, Michael Schmitz wrote: >> The link order in the kernel (atari before msdos) has not been chosen at >> random. The msdos partition format probe will always succeed on any >> valid Atari partition table, while Atari format probe should not succeed >> on plain MSDOS partitions. I had that the wrong way - link order is not what matters here, the checks order has msdos before atari in fact. >> >> Do the reporters if these bugs have Atari partition format enabled in >> their kernels? I would expect the same false positives to happen there, >> if the atari_probe() code is functionally equivalent to the kernel >> partition probe code. >> Do you have a dump of the root sector (plus additional eventual extended >> partition root sector) from these bugs? > > This is a very good point. We should have a look at the probing code in > the kernel because apparently the problem doesn't exist there. It may not exist because the MSDOS partition detection is specific enough to reject Atari partition tables. But still, compare the kernel partition parsers to see what may be missing. Cheers, Michael > And, yes, it would be interesting to get a dump of these boot sectors as > I haven't run into the problem myself and would be very curios why > the misdetection happens. > > Adrian > ^ permalink raw reply [flat|nested] 10+ messages in thread
[parent not found: <c1bf0748-f7e0-b109-8c9e-26f8a2c70bf3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]
* Re: Atari label false positives [not found] ` <c1bf0748-f7e0-b109-8c9e-26f8a2c70bf3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> @ 2018-05-12 11:24 ` John Paul Adrian Glaubitz 2018-05-12 19:22 ` [parted-devel] " Michael Schmitz 2018-05-14 13:02 ` Phillip Susi 1 sibling, 1 reply; 10+ messages in thread From: John Paul Adrian Glaubitz @ 2018-05-12 11:24 UTC (permalink / raw) To: Michael Schmitz, Phillip Susi; +Cc: Parted development, Linux/m68k, Debian m68k On 05/12/2018 10:41 AM, Michael Schmitz wrote: >>> Do you have a dump of the root sector (plus additional eventual extended >>> partition root sector) from these bugs? >> >> This is a very good point. We should have a look at the probing code in >> the kernel because apparently the problem doesn't exist there. > > It may not exist because the MSDOS partition detection is specific > enough to reject Atari partition tables. But still, compare the kernel > partition parsers to see what may be missing. Which would mean that the kernel probes for DOS partition tables before Atari partition tables, wouldn't it? Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaubitz-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org `. `' Freie Universitaet Berlin - glaubitz-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [parted-devel] Atari label false positives 2018-05-12 11:24 ` John Paul Adrian Glaubitz @ 2018-05-12 19:22 ` Michael Schmitz 0 siblings, 0 replies; 10+ messages in thread From: Michael Schmitz @ 2018-05-12 19:22 UTC (permalink / raw) To: John Paul Adrian Glaubitz, Phillip Susi Cc: Debian m68k, Linux/m68k, Parted development Hi Adrian, Am 12.05.2018 um 23:24 schrieb John Paul Adrian Glaubitz: > On 05/12/2018 10:41 AM, Michael Schmitz wrote: >>>> Do you have a dump of the root sector (plus additional eventual >>>> extended >>>> partition root sector) from these bugs? >>> >>> This is a very good point. We should have a look at the probing code in >>> the kernel because apparently the problem doesn't exist there. >> >> It may not exist because the MSDOS partition detection is specific >> enough to reject Atari partition tables. But still, compare the kernel >> partition parsers to see what may be missing. > > Which would mean that the kernel probes for DOS partition tables before > Atari partition tables, wouldn't it? Yes, it does (as I had stated earlier - block/partitions/checks.c has only a few rather more specific formats listed before MSDOS. Sorry if that wasn't clear.). > > Adrian > ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: Atari label false positives [not found] ` <c1bf0748-f7e0-b109-8c9e-26f8a2c70bf3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 2018-05-12 11:24 ` John Paul Adrian Glaubitz @ 2018-05-14 13:02 ` Phillip Susi 1 sibling, 0 replies; 10+ messages in thread From: Phillip Susi @ 2018-05-14 13:02 UTC (permalink / raw) To: Michael Schmitz, John Paul Adrian Glaubitz Cc: Parted development, Linux/m68k, Debian m68k [-- Attachment #1.1: Type: text/plain, Size: 491 bytes --] On 5/12/2018 4:41 AM, Michael Schmitz wrote: > I had that the wrong way - link order is not what matters here, the > checks order has msdos before atari in fact. Ok, I was going to say the dos table requires the 55AA signature that is not expected to be found in an atari table. One other thing I thought of is the atari table lists the size of the disk. Right now it just has to be less than or equal to the actual disk size. What if it was required to be strictly equal? [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 473 bytes --] ^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2018-05-14 13:02 UTC | newest] Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- [not found] <2d6f5455-963d-b044-a64d-67634df46071@ubuntu.com> [not found] ` <2d6f5455-963d-b044-a64d-67634df46071-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> 2018-05-11 15:29 ` Atari label false positives John Paul Adrian Glaubitz [not found] ` <35ad9889-bc13-56bd-7831-f7bb3eea2d1e-1Olz3AKvcsuAKZTfuerNgRvVK+yQ3ZXh@public.gmane.org> 2018-05-11 15:56 ` Phillip Susi [not found] ` <874bf955-d5bc-5d9c-0f87-445da67f3bdd-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org> 2018-05-11 16:10 ` John Paul Adrian Glaubitz 2018-05-11 18:30 ` [parted-devel] " Richard Z 2018-05-11 23:17 ` Michael Schmitz [not found] ` <80a28dd6-ce19-e6ed-fc7a-c5561fecf989-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 2018-05-12 7:59 ` John Paul Adrian Glaubitz 2018-05-12 8:41 ` [parted-devel] " Michael Schmitz [not found] ` <c1bf0748-f7e0-b109-8c9e-26f8a2c70bf3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 2018-05-12 11:24 ` John Paul Adrian Glaubitz 2018-05-12 19:22 ` [parted-devel] " Michael Schmitz 2018-05-14 13:02 ` Phillip Susi
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).