Commit Graph

1934 Commits

Author SHA1 Message Date
Alexander Motin
ccf8a5688a Revert somewhat hackish geom_disk optimization, committed as part of r256880,
and the following r273143 commit, supposed to workaround introduced issue by
quite innocent-looking change.

While there is no clear understanding why, but r273143 is accused in data
corruption in some environments with high I/O load.  I personally don't see
any problem in that commit, and possibly it is just a trigger to some other
bug somewhere, but better safe then sorry for now.

Requested by:	scottl@
MFC after:	3 days
2014-10-25 15:16:19 +00:00
Colin Percival
66427784c1 Populate the GELI passphrase cache with the kern.geom.eli.passphrase
variable (if any) provided in the boot environment.  Unset it from
the kernel environment after doing this, so that the passphrase is
no longer present in kernel memory once we enter userland.

This will make it possible to provide a GELI passphrase via the boot
loader; FreeBSD's loader does not yet do this, but GRUB (and PCBSD)
will have support for this soon.

Tested by:	kmoore
2014-10-22 23:41:15 +00:00
Hans Petter Selasky
f0188618f2 Fix multiple incorrect SYSCTL arguments in the kernel:
- Wrong integer type was specified.

- Wrong or missing "access" specifier. The "access" specifier
sometimes included the SYSCTL type, which it should not, except for
procedural SYSCTL nodes.

- Logical OR where binary OR was expected.

- Properly assert the "access" argument passed to all SYSCTL macros,
using the CTASSERT macro. This applies to both static- and dynamically
created SYSCTLs.

- Properly assert the the data type for both static and dynamic
SYSCTLs. In the case of static SYSCTLs we only assert that the data
pointed to by the SYSCTL data pointer has the correct size, hence
there is no easy way to assert types in the C language outside a
C-function.

- Rewrote some code which doesn't pass a constant "access" specifier
when creating dynamic SYSCTL nodes, which is now a requirement.

- Updated "EXAMPLES" section in SYSCTL manual page.

MFC after:	3 days
Sponsored by:	Mellanox Technologies
2014-10-21 07:31:21 +00:00
Andrey V. Elsukov
52fa0beb0a Add provider's sectorsize and stripesize to confdot output.
Submitted by:	rpokala at panasas.com
2014-10-17 06:58:04 +00:00
Davide Italiano
2be111bf7d Follow up to r225617. In order to maximize the re-usability of kernel code
in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv().
This fixes a namespace collision with libc symbols.

Submitted by:   kmacy
Tested by:      make universe
2014-10-16 18:04:43 +00:00
Andrey V. Elsukov
0478dc0c16 Add an ability to set dumpdev via loader(8) tunable.
MFC after:	3 weeks
2014-10-08 12:18:16 +00:00
Hiroki Sato
d17183901f Fix a bug in r272297 which prevented dumpdev from setting.
!u is not equivalent to (u != 0).
2014-10-03 04:13:25 +00:00
Pawel Jakub Dawidek
227f68edbb Be prepared that set_dumper() might fail even when resetting it or prefix
the call with (void) to document that we intentionally ignore the return
value - no way to handle an error in case of device disappearing.
2014-09-30 12:00:50 +00:00
Pawel Jakub Dawidek
7f5b50719b Style fixes. 2014-09-30 11:51:32 +00:00
Colin Percival
835c4dd436 Cache GELI passphrases entered at the console during the boot process,
in order to improve user-friendliness when a system has multiple disks
encrypted using the same passphrase.

When examining a new GELI provider, the most recently used passphrase
will be attempted before prompting for a passphrase; and whenever a
passphrase is entered, it is cached for later reference.  When the root
disk is mounted, the cached passphrase is zeroed (triggered by the
"mountroot" event), in order to minimize the possibility of leakage
of passphrases.  (After root is mounted, the "taste and prompt for
passphrases on the console" code path is disabled, so there is no
potential for a passphrase to be stored after the zeroing takes place.)

This behaviour can be disabled by setting kern.geom.eli.boot_passcache=0.

Reviewed by:	pjd, dteske, allanjude
MFC after:	7 days
2014-09-16 08:40:52 +00:00
Sean Bruno
5f23eb4d9c Add device name used in geom_map verbose output. This helps when using
geom_map with multiple flash/spi devices.

Phabric:  https://reviews.freebsd.org/D766
Reviewed by:	adrian
MFC after:	2 weeks
2014-09-11 22:39:27 +00:00
John-Mark Gurney
89fac384c8 use a straight buffer instead of an iov w/ 1 segment... The aesni
driver when it hits a mbuf/iov buffer, it mallocs and copies the data
for processing..  This improves perf by ~8-10% on my machine...

I have thoughts of fixing AES-NI so that it can better handle segmented
buffers, which should help improve IPSEC performance, but that is for
the future...
2014-09-04 23:53:51 +00:00
Scott Long
274919e965 Deal explicitly with possible failures of make_dev_alias_p() in GEOM.
Submitted by:   Mariusz Zaborski <oshogbo@FreeBSD.org>
MFC after:      3 days
2014-08-18 19:27:47 +00:00
Andrey V. Elsukov
36b16d1f7d Turn off kern.geom.part.mbr.enforce_chs by default. 2014-08-12 10:31:31 +00:00
Andrey V. Elsukov
fb86534cb1 Add sysctl and loader tunable kern.geom.part.mbr.enforce_chs that is set
by default. It can be used to disable automatic alignment to CHS geometry,
that GEOM_PART_MBR does.

Reviewed by:	wblock
MFC after:	1 week
2014-08-12 09:10:13 +00:00
Warner Losh
cba7d97b61 cswitch is unsigned, so don't compare it < 0. Any negative numbers
will look huge and be caught by > 100.
2014-08-07 21:56:42 +00:00
Warner Losh
86e26cb154 Unsigned values can never be less than 0. 2014-08-07 21:56:37 +00:00
Marcel Moolenaar
6c25615f39 In r264504, we prevented doing I/O for more than MAXPHYS by making
the assumption that consumers would respect bio_completed and/or
bio_resid to detect short reads. This assumption proved false and
file corruption was the result.
Create as many bios as we need to satisfy the original request.
Check the cached chunk every time we need to do I/O to increase the
hit rate.

Obtained from:	junipre Networks, Inc.
MFC after:	1 week
2014-07-22 17:30:05 +00:00
Nathan Whitehorn
1ee0f08975 After EFI support was added to the installer, it needed to allow boot
partitions of types other than "freebsd-boot" (in particular, "efi").
This allows the removal of some nasty hacks for supporting PowerPC systems,
in particular aliasing freebsd-boot to apple-boot on APM and an IBM-specific
code on MBR.

This changes the installer to use the correct names, which also breaks a
degeneracy in the meaning of "freebsd-boot" that allows the addition
of support for some newer IBM systems that can boot from GPT in addition to
MBR. Since I have no idea how to detect which those systems are, leave
the default on IBM PPC systems as MBR for now.
2014-07-04 15:55:32 +00:00
Hans Petter Selasky
af3b2549c4 Pull in r267961 and r267973 again. Fix for issues reported will follow. 2014-06-28 03:56:17 +00:00
Glen Barber
37a107a407 Revert r267961, r267973:
These changes prevent sysctl(8) from returning proper output,
such as:

 1) no output from sysctl(8)
 2) erroneously returning ENOMEM with tools like truss(1)
    or uname(1)
 truss: can not get etype: Cannot allocate memory
2014-06-27 22:05:21 +00:00
Hans Petter Selasky
3da1cf1e88 Extend the meaning of the CTLFLAG_TUN flag to automatically check if
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.

Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.

MFC after:	2 weeks
Sponsored by:	Mellanox Technologies
2014-06-27 16:33:43 +00:00
Andrey V. Elsukov
91ca76a590 Add disklabel64 support to GEOM_PART class.
This partitioning scheme is used in DragonFlyBSD. It is similar to
BSD disklabel, but has the following improvements:
* metadata has own dedicated place and isn't accessible through partitions;
* all offsets are 64-bit;
* supports 16 partitions by default (has reserved place for more);
* has reserved place for backup label (but not yet implemented);
* has UUIDs for partitions and partition types;

No objections from:	geom
MFC after:	2 weeks
Relnotes:	yes
2014-06-11 10:42:34 +00:00
Andrey V. Elsukov
4042ab48c7 Allow swapping to DragonFlyBSD's swap partition.
MFC after:	2 weeks
2014-06-11 10:23:49 +00:00
Andrey V. Elsukov
0640b71dfe Add aliases for DragonFlyBSD's partition types.
MFC after:	2 weeks
2014-06-11 10:19:11 +00:00
Brad Davis
ebd05adab8 - Fix the keyfile being cleared prematurely after r259428
PR:		185084
Submitted by:	fk@fabiankeil.de
Reviewed by:	pjd@
2014-06-06 03:17:37 +00:00
Andrey V. Elsukov
39dcac849e Use g_conf_printf_escaped() to escape symbols, which can break
an XML tree.

MFC after:	1 week
2014-05-30 10:35:51 +00:00
Andrey V. Elsukov
17e0c43319 Add a topology trace to the g_spoil_event.
MFC after:	1 week
2014-05-19 16:08:15 +00:00
Andrey V. Elsukov
362073c089 We have two functions from where a geom orphan method could be called:
g_orphan_register and g_resize_provider_event. Both are called from the
event queue. Also we have GEOM_DEV class, which does deferred destroy
for its consumers via g_dev_destroy (also called from the event queue).
So it is possible, that for some consumers an orphan method will be
called twice. This triggers panic in g_dev_orphan.
Check that consumer isn't already orphaned before call orphan method.

MFC after:	2 weeks
2014-05-19 16:05:42 +00:00
Alexander Motin
413037c8e7 Make GEOM DISK to account also BIO_FLUSH operations. 2014-05-17 15:07:00 +00:00
Andrey V. Elsukov
579259ea0d It is safe to allow shrinking, when aligned size is bigger than current.
Tested by:	jmg
MFC after:	1 week
2014-05-07 11:18:27 +00:00
Edward Tomasz Napierala
c7c7d7d0f0 Make r242379 - the fix for UFS labels disappearing after resizing
the provider - also apply to UFS1 filesystems.  This should help with
resizing filesystems created by makefs(8), which still uses UFS1.

Tested by:	jmg@
Sponsored by:	The FreeBSD Foundation
2014-05-05 09:20:30 +00:00
Andrey V. Elsukov
4f31a94bd2 Add an advice what to do when partition was automatically resized.
X-MFC after:	r256690
2014-05-04 20:00:08 +00:00
Andrey V. Elsukov
c778397f26 Add better error description for case when we are doing resize and
scheme-specific method returns EBUSY.

MFC after:	1 week
2014-05-04 16:55:51 +00:00
Andrey V. Elsukov
0dd7f00cee Prevent an unexpected shrinking on resizing due to alignment for MBR,
PC98 and VTOC8 schemes.

Reported by:	jmg
MFC after:	1 week
2014-05-04 16:43:57 +00:00
Andrey V. Elsukov
bc1e8f56ff For schemes that do an automatic partition aligning move this code to
separate function.

MFC after:	1 week
2014-05-04 10:14:25 +00:00
Luiz Otavio O Souza
81694cde44 Fix a leak in g_uzip_taste(). After retrieve all the block offsets from
the uzip image, free the last data read.
2014-05-01 15:23:20 +00:00
Luiz Otavio O Souza
ccb7284af1 Actually the FEATURE() macro is defined on sys/sysctl.h.
Pointyhat to:	loos
2014-05-01 14:59:04 +00:00
Luiz Otavio O Souza
6d8beede60 Some style and whitespace fixes. Reduce the difference between geom_uzip(4)
and geom_uncompress(4).  Now, they produce an almost clean diff(1) output.

Remove a duplicated variable from g_uncompress.c and an unnecessary header
from g_uzip.c.

No functional changes.
2014-05-01 14:47:27 +00:00
Bryan Drewery
74679c6a99 Remove redundant include
MFC after:	3 days
2014-04-29 01:17:43 +00:00
Alexander Motin
dea1e22600 Reduce number of opens by REOM RAID during provider taste.
Instead opening/closing provider by each of metadata classes, do it only
once in core code.  Since for SCSI disks open/close means sending some
SCSI commands to the device, this change reduces taste time.

MFC after:	2 weeks
Sponsored by:	iXsystems, Inc.
2014-04-28 15:03:52 +00:00
Luiz Otavio O Souza
6f05733a1f Keep geom_uncompress(4) in line with geom_uzip(4), bring in the r264504 fix.
Make sure not to start I/O bigger than MAXPHYS bytes.

Quoting r264504:

When we detect the condition, we'll reduce the block count and perform
a "short" read.  In g_uncompress_done() we need to consider the original
I/O length and stop early if we're about to deflate a block that we didn't
read.  By using bio_completed in the cloned BIO and not bio_length to
check for this, we automatically and gracefully handle short reads that
our providers may be doing on top of the short reads we may initiate
ourselves.

Reviewed by:	marcel
2014-04-22 18:08:34 +00:00
Marcel Moolenaar
855be5b2c1 Make sure not to do I/O for more than MAXPHYS bytes. Doing so can cause
problems in our providers, such as a KASSERT in md(4). We can initiate
I/O for more than MAXPHYS bytes if we've been given a BIO for MAXPHYS
bytes, the blocks from which we're reading couldn't be compressed and
we had compression in preceeding blocks resulting in misalignment of
the blocks we're trying to read relative to the sector. We're forced to
round up the I/O length to make it an multiple of the sector size.

When we detect the condition, we'll reduce the block count and perform
a "short" read. In g_uzip_done() we need to consider the original I/O
length and stop early if we're about to deflate a block that we didn't
read. By using bio_completed in the cloned BIO and not bio_length to
check for this, we automatically and gracefully handle short reads that
our providers may be doing on top of the short reads we may initiate
ourselves.

Obtained from:	Juniper Networks, Inc.
2014-04-15 15:41:57 +00:00
Bryan Drewery
87bc328d63 Make g_access() KASSERT() more useful.
Sponsored by:	EMC / Isilon Storage Division
Obtained from:	Isilon OneFS
MFC after:	2 weeks
2014-04-15 14:41:41 +00:00
Marcel Moolenaar
4787115d04 Align and round the partitionable disk space to 4K by default.
Since this would also apply when recovering, make sure not to
align or round when that would have a partition fall outside
the partitionable area.
2014-04-12 20:28:39 +00:00
Bryan Drewery
1e4b22b44b Fix spelling error in g_trace() call.
Sponsored by:	EMC / Isilon Storage Division
MFC after:	1 week
2014-04-10 17:00:44 +00:00
Alexander Motin
1229e83d2b Fix wrong sizes used to access PD_Type and PD_State DDF metadata fields.
This caused incorrect behavior of arrays with big-endian DDF metadata.
Little-endian (like used by Adaptec controllers) should not be harmed.
Add workaround should be enough to manage compatibility.

MFC after:	2 weeks
2014-04-10 16:00:33 +00:00
Alexander Motin
66b92c07fe Do not increment bio_data in case of BIO_DELETE.
This fixes KASSERT() panic in g_io_request().
2014-04-10 10:12:56 +00:00
Marcel Moolenaar
e8c166e85a An all-or-nothing approach to labels isn't flexible enough. Embedded
systems need fine-grained control over what's in and what's out.
That's ideal. For now, separate GPT labels from the rest and allow
g_label to be built with just GPT labels.

Obtained from:	Juniper Networks, Inc.
2014-04-06 02:44:37 +00:00
Marcel Moolenaar
12b2d77da9 Make sure we don't free memory that's already been freed by setting
the geom->softc pounter to NULL before freeing the g_slicer softc.
In g_slicer_free() the pointer is checked first.

Obtained from:	Juniper Networks, Inc.
2014-04-06 02:20:42 +00:00