Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Thu, 16 Dec 2010 07:30:14 -0500 (EST), Jim Meyering wrote: > Stephen Powell wrote: >> ... do you have an s390 environment in which to test? > > Yes, if an s390x will do. On one, uname -a reports this: > Linux xxx 2.6.32-19.el6.s390x #1 SMP Tue Mar 9 19:03:48 EST 2010 s390x s390x > s390x GNU/Linux I am treating s390 and s390x as synonymous for purposes of this discussion. In fact, in Debian, the implementation is a hybrid one. The kernel space is 64-bit (runs an s390x kernel) but the user space is 31-bit (uses s390 packages in 31-bit mode). >> Does it have both CKD and FBA DASD? >> Does it run as a guest under z/VM? Does your >> system have the DIAG driver? Do you have >> access to a CMS system from which you can FORMAT and/or >> RESERVE the disks in CMS format? The answer might not be >> "yes" to all those questions, and getting such an environment >> will take time. > > I don't know, for any of the above. > If you describe how to determine the answer > in a way that is easy to automate, the goal > would be to run a script that checks for required > pieces and then runs whichever tests it can. You can tell what the DASD type is by a combination of cat /proc/dasd/devices and poking around in the sysfs file system. For example: st...@debian3:~$ cat /proc/dasd/devices 0.0.0200(DIAG) at ( 94: 0) is dasda : active at blocksize: 4096, 54 blocks, 2109 MB 0.0.0201(ECKD) at ( 94: 4) is dasdb : active at blocksize: 4096, 13500 blocks, 52 MB 0.0.0202(DIAG) at ( 94: 8) is dasdc : active at blocksize: 4096, 9 blocks, 351 MB 0.0.0203(DIAG) at ( 94:12) is dasdd : active at blocksize: 4096, 9 blocks, 351 MB st...@debian3:~$ (The output of the above command shows devices that are online, but not necessarily mounted.) The output shows the correspondence between s390 device numbers (0.0., where is the hex device number), kernel-space major and minor device numbers, user-space block special file names (/dev/dasda, etc.) and drivers (dasd_eckd_mod, dasd_fba_mod, dasd_diag_mod). It also shows the effective block size, number of blocks, and size in MB. In the above example, three of the four DASD devices are using the DIAG driver and one is using the ECKD driver. If the ECKD driver is being used, it is CKD DASD. If the FBA driver is being used, it is FBA DASD. If the DIAG driver is being used, it could be either CKD or FBA DASD. You can't tell from here. For that, you need to probe the sysfs pseudo file system. For example: st...@debian3:~$ cat /sys/bus/ccw/devices/0.0.0200/devtype 3390/0a st...@debian3:~$ If the first four characters of output are 3370 or 9336, it's FBA DASD. Otherwise, it's CKD DASD. Make sure that your Linux kernel contains the FBA patch. Look at fs/partitions/ ibm.c, C function ibm_partition. About 40 lines down from the beginning of the function, you should see logic similar to the following: /* * Special case for FBA disks: label sector does not depend on * blocksize. */ if ((info->cu_type == 0x6310 && info->dev_type == 0x9336) || (info->cu_type == 0x3880 && info->dev_type == 0x3370)) labelsect = info->label_block; else labelsect = info->label_block * (blocksize >> 9); If that code isn't in there, you will need to update your kernel. If your kernel doesn't have this fix, then parted and the Linux kernel will report different starting sector numbers and number of sectors for FBA DASD in CMS RESERVED format under certain conditions, and the kernel will be wrong and parted will be right! See Debian bug report 582281 (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582281) for details. If I recall correctly, upstream kernel source 2.6.32-20 was the first to add this fix; so I don't know about your 2.6.32-19.el6.s390x distribution kernel. Unless it contains distribution specific patches to fix this problem, it probably won't work right. The basic testing idea is to compare the starting sector number and number of sectors for the partition as reported by parted and the starting sector number and number of sectors reported by the Linux kernel via the sysfs file system. You can also compare parted's reported file system type vs. what is reported by "mount" (if the partition is mounted) or "cat /proc/swaps" (if the partition is an active swap partition). For example, on my "debian3" test system, I have four DASD devices, all 3390 minidisks, formatted by CMS with a blocksize of 4096, and processed with the CMS RESERVE command. The number of cylinders varies from one disk to another: st...@debian3:~$ su Password: debian3:/home/steve# cd debian3:~# parted /dev/dasda unit s print free Model: IBM S390 DASD drive (dasd) Disk /dev/dasda: 432s Sector size (logical/physical): 512B/4096B Partition Table: dasd Number Start End Size File System Flags 1 4552s 4319991s 4315440s ex
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > On Tue, 14 Dec 2010 10:46:18 -0500 (EST), Jim Meyering wrote: >> Stephen Powell wrote: >>> How is your testing going? >> >> Finding time for parted has been a challenge recently. >>> >>> It's unclear to me from what you wrote. Are you asking me >>> to provide a test case? >> >> If you can, that'd be great and would accelerate the process. >> I would no longer have an excuse not to push your patches. >> I nearly pushed without a test case, but your changes are >> large enough that I cannot do that in good conscience. >> >> I didn't want to require you to write a test, but in the end >> my time constraints appear to be having the same effect. >> Sorry about that. > > Hmm. Well, as I see it, there are two main options. I can > write a test script, providing you with expected output, Thanks for persevering. > and you can run it too, checking to see that it produces the > same output for you. On the plus side, you testing it > yourself gives you the maximum assurances that it is correct. > On the minus side, you have to be able to reproduce my > testing environment. For example, do you have an s390 > environment in which to test? Yes, if an s390x will do. On one, uname -a reports this: Linux xxx 2.6.32-19.el6.s390x #1 SMP Tue Mar 9 19:03:48 EST 2010 s390x s390x s390x GNU/Linux > Does it have both CKD and FBA DASD? > Does it run as a guest under z/VM? Does your > system have the DIAG driver? Do you have > access to a CMS system from which you can FORMAT and/or > RESERVE the disks in CMS format? The answer might not be > "yes" to all those questions, and getting such an environment > will take time. I don't know, for any of the above. If you describe how to determine the answer in a way that is easy to automate, the goal would be to run a script that checks for required pieces and then runs whichever tests it can. > The other option is for me to test it for you. On the plus > side, that requires less time on your part. On the minus > side, I will need a copy of the exact source code that you > are using (which also carries a side benefit in that I will > get to see if my patches applied to your source as expected.) > Another minus is that you have to take my word for it, which > is not as reassuring to you. Which option do you choose? > Or do you have a third option in mind? The best is to put into version control some script that someone with the right equipment can run via "make check" (probably with a couple envvars specifying things that cannot or must not be guessed, like the name of a partition to clobber, or a device to format). But if it's really not feasible, I'm sure we'll find a reasonable middle ground. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Tue, 14 Dec 2010 10:46:18 -0500 (EST), Jim Meyering wrote: > Stephen Powell wrote: >> How is your testing going? > > Finding time for parted has been a challenge recently. >> >> It's unclear to me from what you wrote. Are you asking me >> to provide a test case? > > If you can, that'd be great and would accelerate the process. > I would no longer have an excuse not to push your patches. > I nearly pushed without a test case, but your changes are > large enough that I cannot do that in good conscience. > > I didn't want to require you to write a test, but in the end > my time constraints appear to be having the same effect. > Sorry about that. Hmm. Well, as I see it, there are two main options. I can write a test script, providing you with expected output, and you can run it too, checking to see that it produces the same output for you. On the plus side, you testing it yourself gives you the maximum assurances that it is correct. On the minus side, you have to be able to reproduce my testing environment. For example, do you have an s390 environment in which to test? Does it have both CKD and FBA DASD? Does it run as a guest under z/VM? Does your system have the DIAG driver? Do you have access to a CMS system from which you can FORMAT and/or RESERVE the disks in CMS format? The answer might not be "yes" to all those questions, and getting such an environment will take time. The other option is for me to test it for you. On the plus side, that requires less time on your part. On the minus side, I will need a copy of the exact source code that you are using (which also carries a side benefit in that I will get to see if my patches applied to your source as expected.) Another minus is that you have to take my word for it, which is not as reassuring to you. Which option do you choose? Or do you have a third option in mind? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > On Thu, 25 Nov 2010 10:49:06 -0500 (EST), Jim Meyering wrote: >> Thanks for the fix. >> Until today I had not compiled your changes on an s390x. >> When I did, there were 3 other minor problems, addressed >> by the patch below, just after the one you suggested. >> >> That got past "make check" as a non-privileged user. >> However, running "make check" as root exercises several >> other cases that cannot be tested as a regular user, and >> at least two of those are failing. Those failures >> are not due to your changes, so as soon as I have >> a test or two to exercise your changes I'll feel >> comfortable pushing the result. I expect to use >> the same mechanism as used in a few other tests >> whereby you specify a device and its size like this: >> >> sudo make check DEVICE_TO_ERASE=/dev/sdd DEVICE_TO_ERASE_SIZE=999MB >> >> But it won't be this weekend, so maybe next week... > > How is your testing going? It's unclear to me from what Finding time for parted has been a challenge recently. > you wrote. Are you asking me to provide a test case? If you can, that'd be great and would accelerate the process. I would no longer have an excuse not to push your patches. I nearly pushed without a test case, but your changes are large enough that I cannot do that in good conscience. I didn't want to require you to write a test, but in the end my time constraints appear to be having the same effect. Sorry about that. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Thu, 25 Nov 2010 10:49:06 -0500 (EST), Jim Meyering wrote: > > Thanks for the fix. > Until today I had not compiled your changes on an s390x. > When I did, there were 3 other minor problems, addressed > by the patch below, just after the one you suggested. > > That got past "make check" as a non-privileged user. > However, running "make check" as root exercises several > other cases that cannot be tested as a regular user, and > at least two of those are failing. Those failures > are not due to your changes, so as soon as I have > a test or two to exercise your changes I'll feel > comfortable pushing the result. I expect to use > the same mechanism as used in a few other tests > whereby you specify a device and its size like this: > > sudo make check DEVICE_TO_ERASE=/dev/sdd DEVICE_TO_ERASE_SIZE=999MB > > But it won't be this weekend, so maybe next week... How is your testing going? It's unclear to me from what you wrote. Are you asking me to provide a test case? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote:
...
> In the process of checking out the results, I discovered a bug in my code.
> In include/parted/vtoc.h, at or near line 105, as part of the structure
> declaration for ldl_volume_label, there is a line that says
>
> char ldl_version; /* Version number, valid for ldl format
> */
>
> It should say
>
> char ldl_version[1];/* Version number, valid for ldl format
> */
>
> I was clued into this by a bunch of compile warnings in
> libparted/labels/dasd.c
> complaining about conversion of integer to pointer without a cast,
> or something like that.
Hi Stephen,
Thanks for the fix.
Until today I had not compiled your changes on an s390x.
When I did, there were 3 other minor problems, addressed
by the patch below, just after the one you suggested.
That got past "make check" as a non-privileged user.
However, running "make check" as root exercises several
other cases that cannot be tested as a regular user, and
at least two of those are failing. Those failures
are not due to your changes, so as soon as I have
a test or two to exercise your changes I'll feel
comfortable pushing the result. I expect to use
the same mechanism as used in a few other tests
whereby you specify a device and its size like this:
sudo make check DEVICE_TO_ERASE=/dev/sdd DEVICE_TO_ERASE_SIZE=999MB
But it won't be this weekend, so maybe next week...
>From a0fd447e13ec76fa16a51bfd88b67af705d08c05 Mon Sep 17 00:00:00 2001
From: Stephen Powell
Date: Thu, 25 Nov 2010 15:52:21 +0100
Subject: [PATCH 1/2] tweak ldl_version member decl
* include/parted/vtoc.h [struct ldl_volume_label] (ldl_version):
Declare as array of length 1, not as a scalar.
---
include/parted/vtoc.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/include/parted/vtoc.h b/include/parted/vtoc.h
index 5ed40b5..d79ce39 100644
--- a/include/parted/vtoc.h
+++ b/include/parted/vtoc.h
@@ -102,7 +102,7 @@ struct __attribute__ ((packed)) ldl_volume_label {
char vollbl[4]; /* Label identifier ("LNX1" in EBCDIC) */
char volid[6]; /* Volume identifier */
char res3[69]; /* Reserved field*/
-char ldl_version; /* Version number, valid for ldl format */
+char ldl_version[1];/* Version number, valid for ldl format */
u_int64_t formatted_blocks; /* valid when ldl_version >= "2" (in
EBCDIC) */
};
--
1.7.3.2.765.g642a8
>From 668956027eebd9204e486fdd75e67a981ead5ac5 Mon Sep 17 00:00:00 2001
From: Jim Meyering
Date: Thu, 25 Nov 2010 15:57:00 +0100
Subject: [PATCH 2/2] s390: avoid warnings
* libparted/labels/vtoc.c (vtoc_read_volume_label): Remove decl
of unused var.
* libparted/labels/fdasd.c (fdasd_valid_vtoc_pointer): Return
a value also when successful.
* libparted/labels/dasd.c (dasd_alloc_metadata): Initialize "part"
to NULL solely to avoid a spurious used-uninitialized warning.
---
libparted/labels/dasd.c |2 +-
libparted/labels/fdasd.c |1 +
libparted/labels/vtoc.c |1 -
3 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/libparted/labels/dasd.c b/libparted/labels/dasd.c
index 8114347..bca1662 100644
--- a/libparted/labels/dasd.c
+++ b/libparted/labels/dasd.c
@@ -906,7 +906,7 @@ dasd_alloc_metadata (PedDisk* disk)
PedSector vtoc_end;
LinuxSpecific* arch_specific;
DasdDiskSpecific* disk_specific;
- PedPartition* part;
+ PedPartition* part = NULL; /* initialize solely to placate gcc */
PedPartition* new_part2;
PedSector trailing_meta_start, trailing_meta_end;
struct fdasd_anchor anchor;
diff --git a/libparted/labels/fdasd.c b/libparted/labels/fdasd.c
index 8da7b1f..6d708f6 100644
--- a/libparted/labels/fdasd.c
+++ b/libparted/labels/fdasd.c
@@ -707,6 +707,7 @@ fdasd_valid_vtoc_pointer(fdasd_anchor_t *anc, unsigned long
b, int fd)
return 0;
fdasd_error(anc, wrong_disk_format, _("Invalid VTOC."));
+ return 1;
}
/*
diff --git a/libparted/labels/vtoc.c b/libparted/labels/vtoc.c
index dab5181..cf2990e 100644
--- a/libparted/labels/vtoc.c
+++ b/libparted/labels/vtoc.c
@@ -290,7 +290,6 @@ vtoc_read_volume_label (int f, unsigned long vlabel_start,
bogus_label_t *bogus_ptr = &mybogus;
vollabel_t *union_ptr = &bogus_ptr->actual_label;
volume_label_t *cdl_ptr = &union_ptr->cdl;
- cms_volume_label_t *cms_ptr = &union_ptr->cms;
PDEBUG
int rc;
--
1.7.3.2.765.g642a8
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Thu, 28 Oct 2010 12:26:00 -0400 (EDT), Jim Meyering wrote: > Stephen Powell wrote: >> >> OK, here is the new patch: >> >>http://www.wowway.com/~zlinuxman/parted/parted_fba_diag_patch.diff > > Thank you. > I confirmed that it applies and doesn't cause any trouble on x86_64/linux. > I'll actually look at the code and make sure tests pass on an > s390x next week, assuming I can get access. Debian came out with an update to parted version 2.3 recently. Their package version is called 2.3-4. I fetched their updated source package and reapplied the original four context diff patches to their respective source files: http://www.wowway.com/~zlinuxman/parted/dasd.c.diff -> libparted/labels/dasd.c http://www.wowway.com/~zlinuxman/parted/fdasd.c.diff -> libparted/labels/fdasd.c http://www.wowway.com/~zlinuxman/parted/vtoc.c.diff -> libparted/labels/vtoc.c http://www.wowway.com/~zlinuxman/parted/vtoc.h.diff -> include/parted/vtoc.h then applied the unified diff file http://www.wowway.com/~zlinuxman/parted/parted_fba_diag_patch.diff and I ran into some apply problems. The first four patches applied OK, but the first two hunks of the unified patch did not apply for some reason. Not knowing enough about the internals of the structure of a diff file to fix the problem, I manually re-made the changes. In the process of checking out the results, I discovered a bug in my code. In include/parted/vtoc.h, at or near line 105, as part of the structure declaration for ldl_volume_label, there is a line that says char ldl_version; /* Version number, valid for ldl format */ It should say char ldl_version[1];/* Version number, valid for ldl format */ I was clued into this by a bunch of compile warnings in libparted/labels/dasd.c complaining about conversion of integer to pointer without a cast, or something like that. So, how is your testing coming? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Thu, 28 Oct 2010 12:26:00 -0400 (EDT), Jim Meyering wrote: > Thank you. > I confirmed that it applies and doesn't cause any trouble on x86_64/linux. > I'll actually look at the code and make sure tests pass on an > s390x next week, assuming I can get access. > > Speaking of tests, can you outline a couple > that will exercise some (more the merrier) of this new code? I primarily use the print command for testing. For example, suppose that device number 0200 is /dev/dasda. I would issue the command parted /dev/dasda unit s print free and compare the starting sector and number of sectors to the output of cat /sys/bus/ccw/devices/0.0.0200/block/dasda/dasda1/start and cat /sys/bus/ccw/devices/0.0.0200/block/dasda/dasda1/size to check to see if parted is calculating the size of the partition the same way as the Linux kernel does. (Believe it or not I found a bug in the Linux kernel this way about five months ago. See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582281 for details.) The key, of course, is to try as many different combinations of DASD type (CKD and FBA), DASD format (CDL, LDL, CMS non-reserved, and CMS reserved), and DASD driver (ECKD, FBA, and DIAG) as possible. Only the following combinations are supported by the Linux kernel: DASD type DASD format DASD driver - --- --- CKD CDL ECKD CKD LDL ECKD CKD LDL DIAG CKD CMS-NR ECKD CKD CMS-NR DIAG CKD CMS-R ECKD CKD CMS-R DIAG FBA CMS-NR FBA FBA CMS-NR DIAG FBA CMS-R FBA FBA CMS-R DIAG The DIAG driver can only be used when Linux for s390 is running in a virtual machine under z/VM. And the CMS DASD format can only be created by the CMS operating system, which only runs in a virtual machine under z/VM. So nine of the eleven combinations will require z/VM in some way to test them. parted only supports the first combination for read-write operations. The others are supported read-only. By read-only I mean that a new partition cannot be created, the pre-existing partition cannot be deleted, and the pre-existing partition cannot be moved or changed in size. I do not mean that write operations cannot be performed on the data in the partition. They can. For example, a file system or swap space can be created on the partition. The other variant is the block size of the disk. The block size can be 512, 1024, 2048, or 4096. With 11 combinations of DASD type, DASD format, and driver times 4 possible block sizes that yields 44 possible configurations for exhaustive testing. And that does not take into account variations in the size of the disk. I have tested all combinations that do not involve the CDL format, since I didn't touch that code. One thing I did discover during testing: it is essential to take the disk offline echo 0 >/sys/bus/ccw/devices/0.0./online where is the device number, and bring it online again echo 1 >/sys/bus/ccw/devices/0.0./online after changing the block size of the disk. Simply refreshing the operating system's view of the partition table is not sufficient. (Of course, all partitions on the disk that are currently mounted must be unmounted before the disk can be taken offline; but you wouldn't normally re-format the disk while partitions are mounted anyway. Similarly, any active swap spaces on the disk must be deactivated before taking the disk offline.) -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote:
> As promised earlier today, I now have the following patch files on my
> web site ready for download:
>
> http://www.wowway.com/~zlinuxman/parted/vtoc.h.diff (apply to
> include/parted/vtoc.h)
> http://www.wowway.com/~zlinuxman/parted/dasd.c.diff (apply to
> libparted/labels/dasd.c)
> http://www.wowway.com/~zlinuxman/parted/fdasd.c.diff (apply to
> libparted/labels/fdasd.c)
> http://www.wowway.com/~zlinuxman/parted/vtoc.c.diff (apply to
> libparted/labels/vtoc.c)
>
> I have not made any updates to NEWS. You can edit that as you see
> fit. These are ordinary "context diffs", with three lines of context.
> I don't understand this "git" stuff. I've never used it. Sorry.
> My starting point was the Debian source package for parted, version 2.2-5.
...
Thanks for your patience, Stephen.
I've applied your patches locally, made some tiny changes
and included the result below.
(I moved 2 decls "down", and made indentation more consistent)
Two things remain to be done:
- add a sentence or two for NEWS; the idea is to be as brief as possible
while still giving enough of the details so people don't have to
go look at the commit logs or -- gasp! -- the code.
- a recipe or two (the more the merrier) for tests that fail
without your patch, but pass with it.
For a change of this size and complexity, I would be remiss
not to include at least one or two patches. It's ok (expected
even) that the tests will run only on an s390 and will be skipped
on other systems.
I've lightly modified words from your email to make the log
message below. If you'd like something different there, let me know.
I haven't tested these changes at all, but should be able
to do so next week.
Jim
>From 7fd0768feda02e0d51956cddac0eaf6d80fb4ebe Mon Sep 17 00:00:00 2001
From: Stephen Powell
Date: Sat, 8 May 2010 09:39:24 +0200
Subject: [PATCH] s390: improve/correct DASD support
The long title would be "corrections to partition size and location
calculations for type 1 partitions for s390 dasd".
This could be treated as two separate fixes, one to make corrections
for LDL formatted disks and one to add support for CMS formatted
disks. I see CMS formatted disks as a variant of LDL formatted
disks, with the additional twist that CMS formatted disks can
be reserved or recomped, which LDL formatted disks cannot be.
This affects the size and location of the partition.
With these patches, parted matches the behavior of the Linux kernel
in recognizing partitions on CMS- and LDL-formatted disks, as documented
in the Linux kernel source code in routine fs/partitions/ibm.c.
Calculation of the metadata has also been changed so that parted will
show no free space on such a disk. In some cases there are now two
non-contiguous metadata extents: one at the beginning of the disk and
one at the end.
As before, parted only supports CKD DASD using the ECKD driver.
FBA DASD and CKD DASD using the DIAG driver are still not supported.
In my regression testing I have discovered some problems in the area
of recognizing file system options. However, since I can duplicate
these errors on a version of parted which does not contain my changes,
I have concluded that my changes did not cause this and therefore
this is an unrelated bug.
---
include/parted/vtoc.h| 56 -
libparted/labels/dasd.c | 106 -
libparted/labels/fdasd.c | 15 ---
libparted/labels/vtoc.c | 14 --
4 files changed, 166 insertions(+), 25 deletions(-)
diff --git a/include/parted/vtoc.h b/include/parted/vtoc.h
index 6b41584..5ed40b5 100644
--- a/include/parted/vtoc.h
+++ b/include/parted/vtoc.h
@@ -49,6 +49,8 @@ typedef struct cchhbcchhb_t;
typedef struct cchh cchh_t;
typedef struct labeldatelabeldate_t;
typedef struct volume_label volume_label_t;
+typedef struct cms_volume_label cms_volume_label_t;
+typedef struct ldl_volume_label ldl_volume_label_t;
typedef struct extent extent_t;
typedef struct dev_constdev_const_t;
typedef struct format1_labelformat1_label_t;
@@ -81,7 +83,7 @@ struct __attribute__ ((packed)) labeldate {
struct __attribute__ ((packed)) volume_label {
char volkey[4]; /* volume key = volume label */
- char vollbl[4]; /* volume label */
+ char vollbl[4]; /* volume label ("VOL1" in EBCDIC) */
char volid[6]; /* volume identifier */
u_int8_t security; /* security byte */
cchhb_t vtoc; /* VTOC address */
@@ -93,6 +95,58 @@ struct __attribute__ ((packed)) volume_label {
char res2[4]; /* reserved */
char lvtoc[14]; /* owner code for LVTOC */
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > I haven't heard anything for a while. I just thought I would touch base > after a couple of months. Is anything happening? Hi Stephen. I'll be able to review your patches in 2-3 weeks. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
I haven't heard anything for a while. I just thought I would touch base after a couple of months. Is anything happening? -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Sat, 05 Jun 2010 05:51:52 -0400 (EDT), Jim Meyering wrote: > Stephen Powell wrote: >> Well, my paperwork just came through today for assignment of copyright. >> And it's just in time. Here is a link to my second round of enhancements. >> >>http://www.wowway.com/~zlinuxman/parted/dasd_complete.diff >> >> These changes require my previous changes as a prerequisite. Let me >> know if you have any trouble with them. The enhancement is >> >> * Support all valid combinations of DASD format (CDL, LDL, and CMS), >> DASD type (CKD and FBA), and DASD driver (ECKD, FBA, and DIAG) supported >> by the Linux kernel. Only CDL format on CKD DASD using the ECKD >> driver is supported for read/write operations on partitions >> (create, delete, move, and resize). For all other combinations >> of DASD format, DASD type, and DASD driver that are supported by the >> Linux kernel, parted recognizes the pre-existing implicit partition; >> but it does not allow that partition to be deleted, moved, or resized; >> nor does it allow the creation of another partition on that DASD device. > > Hi Stephen, > > Now that your paperwork has gone through and I've released parted-2.3, > we can work on getting your changes committed. Great! > > Would you please outline a few sets of shell/parted commands > that I can run on a dasd disk to exercise as much of this new > code as possible? > During development, I tested things primarily with the command parted /dev/dasde print free This assumes that /dev/dasde is the device being looked at. Substitute the appropriate Linux device name as appropriate. There are variables in the /sys file system in the Linux kernel that can be used for comparison purposes to see if parted is returning the same partition start and size as the Linux kernel. For example, if the device number is 0400 and the device name is /dev/dasde, then the disk size can be examined by means of cat /sys/bus/ccw/devices/0.0.0400/block/dasde/size the starting sector of the first partition can be examined by means of cat /sys/bus/ccw/devices/0.0.0400/block/dasde/dasde1/start and the size of the first partition can be examined by means of cat /sys/bus/ccw/devices/0.0.0400/block/dasde/dasde1/size (This testing was done using a 2.6.32 kernel and version 1.8.3 of s390-tools.) The units used for measurement in all three of the above variables are 512-byte pseudo-sectors, with the sectors numbered starting with zero. Remember that parted reports output in decimal K (10**3), decimal M (10**6) and decimal G (10**9); not binary K (2**10), binary M (2**20), or binary G (2**30). The only other command that I issued was parted /dev/dasde print 1 but it was that command that brought my attention to the problems that parted has with file systems. I didn't touch the file system code, only code related to recognizing partitions. I got errors related to file systems even on disk formats, such as CDL, that I didn't touch. Someone needs to look at that. To summarize, here are all the supported combinations of dasd type, driver, and disk format that parted now recognizes. These are the same combinations recognized by the Linux kernel. Dasd type Driver Disk formatType of support = == ====== CKD ECKD CDLread-write* CKD ECKD LDLread-only* CKD ECKD CMS non-reserved read-only CKD ECKD CMS reserved read-only CKD DIAG LDLread-only CKD DIAG CMS non-reserved read-only CKD DIAG CMS reserved read-only FBA FBA CMS non-reserved read-only FBA FBA CMS reserved read-only FBA DIAG CMS non-reserved read-only FBA DIAG CMS reserved read-only * supported by parted 2.2 read-write means that partitions can be created, deleted, moved, or resized, and that up to three partitions may exist on the device. read-only means that the single pre-existing partition is recognized, but it cannot be deleted, moved, or resized, nor can another partition be added. read-only does not mean that no write operations can be performed. For example, parted should be able to make a file system on the partition. It just can't alter the definition of the partition itself or create another partition. Also, keep in mind that there is a bug in the kernel that keeps certain types of devices from working properly. To be specific, CMS reserved disks on FBA DASD using the DIAG driver with a logical block size greater than 512 are not properly recognized by the Linux kernel. See Debian bug report http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582281 for details. Parted assumes that this kernel bug has been fixed. There is a link to a patch file for this kernel bug in the Debian bug report log. > Thanks for all your work. > > Jim You're welcome. > > P.S. Please
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > Well, my paperwork just came through today for assignment of copyright. > And it's just in time. Here is a link to my second round of enhancements. > >http://www.wowway.com/~zlinuxman/parted/dasd_complete.diff > > These changes require my previous changes as a prerequisite. Let me > know if you have any trouble with them. The enhancement is > > * Support all valid combinations of DASD format (CDL, LDL, and CMS), > DASD type (CKD and FBA), and DASD driver (ECKD, FBA, and DIAG) supported > by the Linux kernel. Only CDL format on CKD DASD using the ECKD > driver is supported for read/write operations on partitions > (create, delete, move, and resize). For all other combinations > of DASD format, DASD type, and DASD driver that are supported by the > Linux kernel, parted recognizes the pre-existing implicit partition; > but it does not allow that partition to be deleted, moved, or resized; > nor does it allow the creation of another partition on that DASD device. Hi Stephen, Now that your paperwork has gone through and I've released parted-2.3, we can work on getting your changes committed. Would you please outline a few sets of shell/parted commands that I can run on a dasd disk to exercise as much of this new code as possible? Thanks for all your work. Jim P.S. Please excuse the high latency. I'll be available only sporadically for the next couple months. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Well, my paperwork just came through today for assignment of copyright. And it's just in time. Here is a link to my second round of enhancements. http://www.wowway.com/~zlinuxman/parted/dasd_complete.diff These changes require my previous changes as a prerequisite. Let me know if you have any trouble with them. The enhancement is * Support all valid combinations of DASD format (CDL, LDL, and CMS), DASD type (CKD and FBA), and DASD driver (ECKD, FBA, and DIAG) supported by the Linux kernel. Only CDL format on CKD DASD using the ECKD driver is supported for read/write operations on partitions (create, delete, move, and resize). For all other combinations of DASD format, DASD type, and DASD driver that are supported by the Linux kernel, parted recognizes the pre-existing implicit partition; but it does not allow that partition to be deleted, moved, or resized; nor does it allow the creation of another partition on that DASD device. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Mon, 24 May 2010 03:26:43 -0400 (EDT), Jim Meyering wrote: > Stephen Powell wrote: >> Praise the Lord! I do believe I've got it now. Parted now supports all >> DASD types (CKD and FBA), drivers (ECKD, FBA, and DIAG) and disk formats >> (CDL, LDL, and CMS) that are supported by the Linux kernel on the s390 >> architecture. > > Nice. > Sounds promising. >> >> ... >> I intend to put the changed code through some fairly extensive >> final tests next week, and after that I will upload a unified diff >> with all my changes. >> ... > > Considering the number of different formats involved, > I'd guess that your changes include several bug fixes. Not really. The only bug fix is one that is included in my first set of updates. (LDL support only works for a blocksize of 4096). The rest was enhancements. Before my changes, parted supported only CDL and LDL format, supported only CKD DASD, and supported only the ECKD driver. (All of this discussion is assumed to be in the context of the s390 architecture.) And the LDL support only worked for a blocksize of 4096. My first set of changes fixed that bug and added support for CMS disk format. But it still supported only CKD DASD and only the ECKD driver. This second set of changes supports FBA DASD and the FBA and DIAG drivers. My goal was to support everything that the Linux kernel supports in the way of mainframe DASD. And I believe that I have accomplished my goal. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > Praise the Lord! I do believe I've got it now. Parted now supports all > DASD types (CKD and FBA), drivers (ECKD, FBA, and DIAG) and disk formats Nice. Sounds promising. > (CDL, LDL, and CMS) that are supported by the Linux kernel on the s390 ... > I intend to put the changed code through some fairly extensive > final tests next week, and after that I will upload a unified diff > with all my changes. It will, however, have my original changes Considering the number of different formats involved, I'd guess that your changes include several bug fixes. If you can find a way to make this into a series of patches, I would be most grateful. Teasing one big unified diff into a series of largely-independent patches can be tedious, but when done properly, it makes reviewing and maintenance (e.g., future bisection) a lot easier. > published via the context diffs as a prerequisite. Sorry. I'm > a novice at this. In fact, my only qualifications for making these > changes seems to be that (a) I need the new functionality, and > (b) I have the necessary hardware with which to test. Programming > skills, Linux knowledge, C knowledge, knowledge of "how things are > done procedurally", etc. are probably all sub-par. I ask for your > patience and forbearance as I slowly and painfully learn all that > stuff. Also, please include as much as you can in the way of test recipes. I.e., for each bug, what commands did you run to trigger it, and what were the symptoms before it was fixed. Thanks! -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Praise the Lord! I do believe I've got it now. Parted now supports all DASD types (CKD and FBA), drivers (ECKD, FBA, and DIAG) and disk formats (CDL, LDL, and CMS) that are supported by the Linux kernel on the s390 architecture. I even have that last remaining case covered: old LDL disk format (pre version 2) on CKD DASD using the DIAG driver. It was the only case left that required knowledge of device geometry when device geometry was bogus. disk->dev->length proved to be the key to solving that problem. This is the size of the disk in 512-byte sectors and is correct even when device geometry (cylinders, heads, and sectors per track) is bogus. Only CDL format on CKD DASD using the ECKD driver is supported for actual partition editing (create, delete, move, re-size, etc.) The other combinations are read-only. That is, parted recognizes the preexisting partition, but it cannot be deleted, moved, or re-sized. New partitions cannot be added either. You should be able to make a file system on the pre-existing partition with parted though. Speaking of file systems, I think there may be a bug in the file system code. I haven't touched that portion of the code, but when I issue a command like parted /dev/dasde print 1 for a disk on which a file system has been created on the partition directly by Linux (such as with mke2fs) I get funny errors. (For some reason I have to take the device offline and bring it back online again after I run mke2fs in order for parted to see the file system.) For example, the output of the above command is Error: File system has an incompatible feature enabled. Compatible features are has_journal, dir_index, filetype, sparse_super, and large_file. Use tune2fs or debugfs to remove features. I just made this file system seconds ago. And it was made with a native file system tool. How can it be invalid already? Someone ought to take a look at that. I have the same problem with the original code; so I don't think I broke it. I intend to put the changed code through some fairly extensive final tests next week, and after that I will upload a unified diff with all my changes. It will, however, have my original changes published via the context diffs as a prerequisite. Sorry. I'm a novice at this. In fact, my only qualifications for making these changes seems to be that (a) I need the new functionality, and (b) I have the necessary hardware with which to test. Programming skills, Linux knowledge, C knowledge, knowledge of "how things are done procedurally", etc. are probably all sub-par. I ask for your patience and forbearance as I slowly and painfully learn all that stuff. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
In the process of testing and verifying my changes to parted, I have discovered a bona-fide bug in the Linux kernel, one that prevents it from reading the disk label in some cases. See Debian bug report 582281 for details. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582281 I will write the parted code the correct way. In other words, it will assume that this kernel bug has been fixed. If it hasn't, the kernel and parted will disagree about the starting location and size of the partition in some circumstances. But parted will be right and the kernel will be wrong. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
OK, I have FBA DASD working with the DIAG driver now. The only thing that doesn't yet work is LDL disks on CKD DASD using the DIAG driver that have been formatted with an old version of dasdfmt (which does not include formatted_blocks in the label). That's the only remaining case where the partition size calculation depends on device geometry, which is bogus when using the DIAG driver. Nevertheless, parted does know the correct size of the disk in sectors when using the DIAG driver. It uses that information to calculate free space, which it annoyingly displays if I don't create the metadata partitions properly. So it may be possible to eliminate that last remaining dependency on device geometry. I'll give it a shot. (Of course, I am speaking in the context of LDL and CMS formatted disks. CDL formatted disks still have a dependency on device geometry, and always will. But CDL formatted disks can only be created on CKD DASD and they cannot use the DIAG driver; so obtaining the device geometry is no problem.) -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
OK, I have LDL disks working with the DIAG driver in parted now. The reason I was having trouble yesterday was that the LDL disk I was using for testing was formatted by a very old version of dasdfmt. Debian Squeeze currently uses version 1.6.2 of s390-tools. I upgraded to version 1.8.3 from Sid and reformatted the disk. This version uses the "version 2" format of an LDL disk label, which contains the number of formatted blocks in the label. Without this information in the label, which the old version of dasdfmt did not supply, parted relies on device geometry to calculate the size of the partition. And device geometry is all bogus when using the DIAG driver. Upgrading to a newer version of s390-tools and reformatting the disk solved the problem. Now on to the final frontier: FBA DASD. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Well, I have parted working with the DIAG driver now, at least for CMS disks on CKD devices. LDL disks don't work with the DIAG driver yet. I haven't done anything with FBA disks yet. My efforts continue. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Now that CMS minidisk support on CKD DASD using the ECKD driver is working, there are three remaining frontiers: CMS minidisk support on FBA DASD using the FBA driver, CMS minidisk support on either CKD or FBA DASD using the DIAG driver, and LDL disk support on CKD DASD using the DIAG driver. If I can get those three things working, then parted on the s390 architecture will support all DASD devices and all drivers supported by the Linux kernel on the s390 architecture, which of course is the ultimate goal. (CMS format is the only format supported on FBA DASD by the Linux kernel. CDL and LDL both require CKD DASD.) I don't have any FBA DASD with which to test. However, z/VM can emulate FBA DASD by using something called "Virtual Disks in Storage". That should be good enough. I have no idea whether or not I will be successful. I may have bitten off more than I can chew. But I'm going to give it the good old college try, and we'll see what happens. Maybe I'll get lucky. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: ... > Now, for the specific "bullet points" for NEWS: > > o Fixed a bug in which the starting block for the implicit partition > of LDL disks was miscalculated if the disk was formatted at a block > size other than 4096. The program now calculates the starting block > of the partition correctly, regardless of whether the disk has a > block size of 512, 1024, 2048, or 4096. (These are all the valid > block sizes.) > > o Enhanced LDL support to use the "formatted blocks" field of an LDL- > format disk label (for version 2 and above LDL formats) to determine > the size of the partition, rather than relying on disk geometry. > This will allow parted to determine the correct partition size if, > for example, an image copy backup of a 3390-3 volume is taken at > the home site and restored to a 3390-9 volume at a disaster recovery > hot site. The extra space at the end of the 3390-9 volume is not > part of the partition, and parted now knows that. > > o Added support for CMS-formatted minidisks, both non-reserved and > reserved, provided that they are on CKD DASD. CMS minidisks on > FBA DASD are not supported at this time. Also, CMS minidisks on > CKD DASD which use the DIAG driver are not supported either. The > ECKD driver must be used. > > The changes to function dasd_alloc_metadata in dasd.c are necessary to > support both the bug fix and the enhancements. It is important that > no metadata/partition overlap occurs, and it is important that all > sectors on an LDL or CMS disk be covered either by the partition or by > metadata. Thanks for the detailed explanation. I'll add it to NEWS once the paperwork has gone through. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
On Sat, 08 May 2010 04:06:16 -0400 (EDT), Jim Meyering wrote: > Stephen Powell wrote: >> As promised earlier today, I now have the following patch files on my >> web site ready for download: >> ... > > I don't often see context diffs, as opposed to unified diffs. > Most people prefer to read the latter. You can generate them with diff -u > if you're using GNU diff, or with git diff if you ever get around > to using git. Jonathan Nieder explained to me how to make unified diffs in more detail in a private e-mail. Thanks to both you and Jonathan. I apologize for not following the normal conventions. This stuff is all new to me. I'm used to applications programming in an IBM mainframe shop using PL/I. The whole GNU/Linux world is a total paradigm shift for me. The operating system, the programming language, the tools, etc. are all completely new to me. What I do have to offer is basic programming aptitudes and a knowledge of IBM mainframe hardware architecture, especially when running in a virtual machine under z/VM. But everything else is totally different. Please be patient with me as I learn. > > I've confirmed that those patches do apply locally and removed a couple > of trialing spaces so "make syntax-check" still passes. That's another tool I know nothing about. :-( I use front-ends like dpkg-buildpackage, which no doubt use make "under the covers", but I never directly invoke make. > > Once I have confirmation that the FSF has received your paperwork, > I'll review and test. I mailed the paperwork today. I can send you an e-mail with a PDF file attached that has copies of the paperwork, if that will speed things up. > > One way that you can help in the mean time is by writing a slightly > more detailed summary of what changed. For example > * what is it that would fail without your patch, but works with it? > * how does your improvement make things better? > That is the sort of information I'd like to put in NEWS. > Does this count as a bug fix? It seems so, but there's also > the free-space change that looks more like an improvement than a bug fix. > I haven't looked at details yet... Necessity is the mother of invention. What prompted me to look into this in the first place was a desire to install Debian GNU/Linux for s390 onto reserved CMS minidisks in a virtual machine under z/VM. Reserved CMS minidisks are advantageous to use when Linux is run in a virtual machine under z/VM primarily for two reasons: (1) reserved CMS minidisks can be accessed and manipulated read/write by both CMS and the Linux kernel, and (2) reserved CMS minidisks are eligible for use by the DIAG driver, a high-performance disk driver usable only when GNU/Linux is running in a virtual machine under IBM's z/VM operating system. But the Debian installer would not work with CMS minidisks. I opened a wishlist bug report against the Debian installer years ago, but no action ensued until recently. Someone looked into it and said that the reason that the Debian installer didn't work with CMS minidisks is that parted didn't recognize them. So my goal was to enhance parted so that I could install Debian directly to CMS minidisks. (See http://www.wowway.com/~zlinuxman/diag250.htm for the convoluted way I have to go about installing Debian to CMS minidisks due to the lack of CMS minidisk support in parted.) I really didn't feel adequate for the job of changing parted, a program written in C, a language that I don't know. But Debian is a volunteer organization. They don't have the money to go out and buy IBM mainframes, license z/VM, and do their own development and testing. I already had all that, but did not possess C knowledge. It was easier for me to acquire enough C knowledge to make the changes than it was for Debian to acquire a real mainframe and a z/VM license. In the process of making these changes I discovered some bugs in the existing code for LDL disks, to which CMS disks are closely related. Now, for the specific "bullet points" for NEWS: o Fixed a bug in which the starting block for the implicit partition of LDL disks was miscalculated if the disk was formatted at a block size other than 4096. The program now calculates the starting block of the partition correctly, regardless of whether the disk has a block size of 512, 1024, 2048, or 4096. (These are all the valid block sizes.) o Enhanced LDL support to use the "formatted blocks" field of an LDL- format disk label (for version 2 and above LDL formats) to determine the size of the partition, rather than relying on disk geometry. This will allow parted to determine the correct partition size if, for example, an image copy backup of a 3390-3 volume is taken at the home site and restored to a 3390-9 volume at a disaster recovery hot site. The extra space at the end of the 3390-9 volume is not part of the partition, and parted now knows that. o Added support for CMS-formatted minidisks, both non-reserved and
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
Stephen Powell wrote: > As promised earlier today, I now have the following patch files on my > web site ready for download: > > http://www.wowway.com/~zlinuxman/parted/vtoc.h.diff (apply to > include/parted/vtoc.h) > http://www.wowway.com/~zlinuxman/parted/dasd.c.diff (apply to > libparted/labels/dasd.c) > http://www.wowway.com/~zlinuxman/parted/fdasd.c.diff (apply to > libparted/labels/fdasd.c) > http://www.wowway.com/~zlinuxman/parted/vtoc.c.diff (apply to > libparted/labels/vtoc.c) > > I have not made any updates to NEWS. You can edit that as you see > fit. These are ordinary "context diffs", with three lines of context. > I don't understand this "git" stuff. I've never used it. Sorry. > My starting point was the Debian source package for parted, version 2.2-5. > > To my way of thinking, these patches are logically one fix, which I > would describe as "corrections to partition size and location > calculations for type 1 partitions for s390 dasd". If you want to I don't often see context diffs, as opposed to unified diffs. Most people prefer to read the latter. You can generate them with diff -u if you're using GNU diff, or with git diff if you ever get around to using git. I've confirmed that those patches do apply locally and removed a couple of trialing spaces so "make syntax-check" still passes. Once I have confirmation that the FSF has received your paperwork, I'll review and test. One way that you can help in the mean time is by writing a slightly more detailed summary of what changed. For example * what is it that would fail without your patch, but works with it? * how does your improvement make things better? That is the sort of information I'd like to put in NEWS. Does this count as a bug fix? It seems so, but there's also the free-space change that looks more like an improvement than a bug fix. I haven't looked at details yet... -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
Stephen Powell wrote: > On Wed, 21 Apr 2010 13:44:59 -0400 (EDT), Jim Meyering wrote: >> >> Your patch is large enough that we have to require >> a signed copyright assignment. >> >> Here's the relevant section from HACKING: >> >> http://git.debian.org/?p=parted/parted.git;a=blob;f=HACKING#l428 >> >> Sorry for the hassle (it'll mean you sending an email, >> waiting for snail/paper mail, and signing and sending that >> back to the FSF), but it's for a good cause. > > Well it took a while, and some lobbying, and two days ago I didn't > think it was going to fly. But a miracle has occurred. The CIO > signed the "Employer Disclaimer of Rights" form today! I have the > original signed form in my possession! I will send you my final > patches this evening. (Regression testing has revealed additional > problems, which I have fixed.) Thank you for your patience and your > encouragement. Good news! Thanks for pursuing this. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks
As promised earlier today, I now have the following patch files on my web site ready for download: http://www.wowway.com/~zlinuxman/parted/vtoc.h.diff (apply to include/parted/vtoc.h) http://www.wowway.com/~zlinuxman/parted/dasd.c.diff (apply to libparted/labels/dasd.c) http://www.wowway.com/~zlinuxman/parted/fdasd.c.diff (apply to libparted/labels/fdasd.c) http://www.wowway.com/~zlinuxman/parted/vtoc.c.diff (apply to libparted/labels/vtoc.c) I have not made any updates to NEWS. You can edit that as you see fit. These are ordinary "context diffs", with three lines of context. I don't understand this "git" stuff. I've never used it. Sorry. My starting point was the Debian source package for parted, version 2.2-5. To my way of thinking, these patches are logically one fix, which I would describe as "corrections to partition size and location calculations for type 1 partitions for s390 dasd". If you want to treat it as two separate fixes, one to make corrections for LDL formatted disks and one to add support for CMS formatted disks, that's up to you. I see CMS formatted disks as a variant of LDL formatted disks, with the additional twist that CMS formatted disks can be reserved or recomped, which LDL formatted disks cannot be. This affects the size and location of the partition. With these patches, parted matches the behavior of the Linux kernel in recognizing partitions on CMS- and LDL-formatted disks, as documented in the Linux kernel source code in routine fs/partitions/ibm.c. Calculation of the metadata has also been changed so that parted will show no free space on such a disk. In some cases there are now two non-contiguous metadata extents: one at the beginning of the disk and one at the end. As before, parted only supports CKD DASD using the ECKD driver. FBA DASD and CKD DASD using the DIAG driver are still not supported. Someday I may work on that. Maybe. But not today. In my regression testing I have discovered some problems in the area of recognizing file system options. However, since I can duplicate these errors on a version of parted which does not contain my changes, I have concluded that my changes did not cause this and therefore this is an unrelated bug. I make no claim to be a C programmer. I'm just starting to learn C. You C experts might want to take a look at my code. It does seem to work, but that doesn't necessarily mean that it's well written. There may be a better way to do what I did. Feel free to change it, as long as the end result is the same. If you have any questions, please feel free to ask. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
On Wed, 21 Apr 2010 13:44:59 -0400 (EDT), Jim Meyering wrote: > > Your patch is large enough that we have to require > a signed copyright assignment. > > Here's the relevant section from HACKING: > > http://git.debian.org/?p=parted/parted.git;a=blob;f=HACKING#l428 > > Sorry for the hassle (it'll mean you sending an email, > waiting for snail/paper mail, and signing and sending that > back to the FSF), but it's for a good cause. Well it took a while, and some lobbying, and two days ago I didn't think it was going to fly. But a miracle has occurred. The CIO signed the "Employer Disclaimer of Rights" form today! I have the original signed form in my possession! I will send you my final patches this evening. (Regression testing has revealed additional problems, which I have fixed.) Thank you for your patience and your encouragement. (Jim, please note the CC list is back on.) -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
On Wed, 21 Apr 2010 13:44:59 -0400 (EDT), Jim Meyering wrote: > Your patch is large enough that we have to require > a signed copyright assignment. > > Here's the relevant section from HACKING: > > http://git.debian.org/?p=parted/parted.git;a=blob;f=HACKING#l428 > > Sorry for the hassle (it'll mean you sending an email, > waiting for snail/paper mail, and signing and sending that > back to the FSF), but it's for a good cause. Well, I started the paperwork process, but I'm a little concerned about one of the questions on the form: Do you have an employer who might claim to own your changes? (Or words to that effect.) Yes, I do have an employer. And yes, I did test the changes on their mainframe; so if they want to be obnoxious about it, they could, in theory, cause trouble. I guarantee you they don't care. They don't know, and they don't want to know. But if FSF insists on my employer signing a copyright assignment or a quit claim, then we may have trouble. I guarantee my employer doesn't care. But managers tends to think like this: "If I sign this form, I might possibly get in trouble later with my boss. But if I don't sign this form, there's no way I'll ever get in trouble. There's no benefit to me if I sign it, and there's no hurt to me if I don't sign it. Hmm. What should I do?" What would you do? I'll continue working on this, but you may have to derive the enhancements yourself from scratch based on the published specs in "Device Drivers, Features, and Commands" and the Linux kernel source if FSF or my employer gives me trouble. It's not that I don't want to help. It's just that ... well, I think you understand the situation I'm in. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
Stephen Powell wrote: ... > OK, I'll do the paperwork. But did you start with my raw changes, or Thanks. So far I have applied no changes, but rather looked through the patches Jonathan posted. From what he said, his modifications were insubstantial, but I did not try to verify that. > did you start with Jonathan Nieder's modifications to my changes? > If the latter, perhaps he needs to sign off too? I submitted > my changes to Debian. Jonathan saw the bug report, massaged those > changes a little bit, then submitted them to upstream. If you > got the changes directly from my web site via the links in the > Debian bug report, then Jonathan is not involved, except as an > interested party. But if you started with Jonathan's modifications > to my changes, then he may need to sign off too. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
On Wed, 21 Apr 2010 13:44:59 -0400 (EDT), Jim Meyering wrote: > Stephen Powell wrote: >> I'm not a lawyer, but I hereby grant to FSF all rights that they want or need >> in order to include any of my code under the same license that parted >> currently >> uses. If you're worried about IBM, the closest thing to IBM code that I >> used was the structure mapping of the CMS volume label, which was taken >> from the linux kernel source tree (arch/s390/include/asm/vtoc.h). I changed >> the data types to correspond to the user-defined data types that parted uses, >> but the structure mapping and field names were taken from there. So there >> should be no problem with IBM. If there's anything else you need, let me >> know. > > Your patch is large enough that we have to require > a signed copyright assignment. > > Here's the relevant section from HACKING: > > http://git.debian.org/?p=parted/parted.git;a=blob;f=HACKING#l428 > > Sorry for the hassle (it'll mean you sending an email, > waiting for snail/paper mail, and signing and sending that > back to the FSF), but it's for a good cause. OK, I'll do the paperwork. But did you start with my raw changes, or did you start with Jonathan Nieder's modifications to my changes? If the latter, perhaps he needs to sign off too? I submitted my changes to Debian. Jonathan saw the bug report, massaged those changes a little bit, then submitted them to upstream. If you got the changes directly from my web site via the links in the Debian bug report, then Jonathan is not involved, except as an interested party. But if you started with Jonathan's modifications to my changes, then he may need to sign off too. -- .''`. Stephen Powell : :' : `. `'` `- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#578097: [parted-devel] [Parted-maintainers] Debian Bug #578097: No support for CMS-formatted disks in parted
Stephen Powell wrote: > On Sat, 17 Apr 2010 11:45:58 -0400 (EDT), Otavio Salvador wrote: >> Hello Jonathan and Stephen, >> >> First, thank for both of you. >> >> Stephen, from what I saw in your patch it will require copyright assignment >> to FSF so while Jim and others with more experience in s390 take a look on >> that, it would be good to start the paper work to get the copyright >> assignment done and allow it to be included in upstream repository. > > I'm not a lawyer, but I hereby grant to FSF all rights that they want or need > in order to include any of my code under the same license that parted > currently > uses. If you're worried about IBM, the closest thing to IBM code that I > used was the structure mapping of the CMS volume label, which was taken > from the linux kernel source tree (arch/s390/include/asm/vtoc.h). I changed > the data types to correspond to the user-defined data types that parted uses, > but the structure mapping and field names were taken from there. So there > should be no problem with IBM. If there's anything else you need, let me > know. Hi Stephen, Your patch is large enough that we have to require a signed copyright assignment. Here's the relevant section from HACKING: http://git.debian.org/?p=parted/parted.git;a=blob;f=HACKING#l428 Sorry for the hassle (it'll mean you sending an email, waiting for snail/paper mail, and signing and sending that back to the FSF), but it's for a good cause. Jim -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
