CVS log for src/sys/bus/cam/scsi/scsi_da.c

Up to [DragonFly] / src / sys / bus / cam / scsi

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

MFC numerous features from HEAD.

* Disk flush support in scsi and nata subsystems.

Add BUF_CMD_FLUSH support - issue flush command to mass storage device.

MFC CAM fixes for the 2.0 release.

Fix multiple bugs in CAM related devices which go away unexpectedly.  This
fixes numerous panics when pulling a USB mass media device in the midst of
heavy I/O.

* The SIM lock was being unlocked via the periph->sim path after periph
  was unheld.  periph can become free and blow up the unlock, so get
  the sim into a local variable first, then release periph.

* The code which waits for CCB completion needs to be a while loop, not
  an if.  It worked anyway, but wasn't very robust.

* Add CAM_SIM_DEREGISTERED to flag when a sim is undergoing deregistration.

* Beef up cam_dead_sim so it works more like a real sim.

* Properly install &cam_dead_sim in the device and periph structures
  related to a SCSI bus, when deregistering the bus.

* Disallow the addition of new devices when deregistering a bus.

* NULL out periph->softc when freeing it.

Add quirk for the Samsung YP-U3.

Obtained-from: FreeBSD

Sync CAM with FreeBSD using lockmgr locks instead of mutexes.
Note:  This is mostly a code sync with FreeBSD which improves stability
in addition to the items listed below.  This provides a framework for
releasing the mplock, but for now it's still there.

Add an xpt_print function to reduce most of the xpt_print_path/printf
pairs. Convert the core code to use it.

Initial cut at Basic Domain Validation.

Make cam_xpt's pronouncements match camcontrol (Tagged -> Command) Queueing.

Pay attention to return value from xpt_bus_register in xpt_init.

Add an xpt_rescan function and a thread that will field rescan requests.
The purpose of this is to allow a SIM (or other entities) to request a bus
rescan and have it then fielded in a different (process) context from the
caller.

Check the return value from cam_periph_acquire.

Drop the periph/sim lock when calling disk_destroy().

Drop the topology lock before calling the periph oninvalidate and dtor
vectors.

For the XPT_SASYNC_CB operation, only decouple the broadcast to the bus
and device lists instead of decoupling the whole operation.  This avoids
problems with SIMs going away.

Split the camisr into per-SIM done queues.  This optimizes the locking a
little bit and allows for direct dispatch of the doneq from certain
contexts that would otherwise face recursive locking problems.

Zero the CCBs when mallocing them.

Only schedule the xpt_finishconfig_task once.

Eliminate the use of M_TEMP.

Add a helper function for registering async callbacks.

Release the bus reference that is acquired when doing a CAMIOCOMMAND ioctl.

Zero scsi_readcapacity allocations so we can really tell if there has
been data returned.

Remove duplicate includes and fix typos.

Add a bunch of definitions and structures to support newer drivers.

When probing a newly found device, don't automatically assume that the
device supports retrieving a serial number.  Instead, first query the
list of VPD pages it does support, and only query the serial number if
it's supported, else silently move on.  This eliminates a lot of noise
during verbose booting, and will likely eliminate the need for most
NOSERIAL quirks.

Reduce diffs from FreeBSD.

Obtained-from: FreeBSD

Nuke unneded include.

Submitted-by: Robert Gauthier <321.robert@gmail.com>

Add quirk for an USB Flash Disk Media Player.

Submitted-by: Steve O'Hara-Smith <steve@sohara.org>

o Correct quirks entry for a MP3 player
o Add new quirk for Netak OnlyDisk drive

Obtained-from: FreeBSD

Sync quirk tables with FreeBSD.

Obtained-from: FreeBSD

Add a tunable that allows one to turn off the automatic sending of
the ORDERED tag. This recoups significant performance gains for many
arrays.

The default is still to send out the ORDERED tag periodically.

Obtained-from: FreeBSD

Actually check to see if XPT_CALC_GEOMETRY had an error
and pick a bogus geometry if it had so we don't later
get a divide by zero trap.

Obtained-from: FreeBSD

Remove unused ioctl routine.

Obtained-from: FreeBSD

Don't use the devstat->busy_count for state decisions in the device
drivers.  Doing so imposes atomicity and locking constraints.

Obtained-from: FreeBSD

Don't call dkpart() just to print the result, it is constant.

Obtained-from: FreeBSD

Be very generous with timeouts for synchronize cache. We may wait a very long
time in the cases where it really sends the drive out to lunch, but it also
allows us to catch very weird edge cases of strange drives that might take
a very long time.

Obtained-from: FreeBSD

Check the void * argument in the AC_FOUND_DEV case against NULL.

Obtained-from: FreeBSD

Fix up unit attention and selection timeout handling.

- change daprevent() to set CAM_RETRY_SELTO and SF_RETRY_UA when it calls
  cam_periph_runccb().
- change the pt(4) driver to ignore unit attentions

Obtained-from: FreeBSD

Rewrite of the CAM error recovery code.

Some of the major changes include:

	- The SCSI error handling portion of cam_periph_error() has
	  been broken out into a number of subfunctions to better
	  modularize the code that handles the hierarchy of SCSI errors.
	  As a result, the code is now much easier to read.

	- String handling and error printing has been significantly
	  revamped.  We now use sbufs to do string formatting instead
	  of using printfs (for the kernel) and snprintf/strncat (for
	  userland) as before.

	  There is a new catchall error printing routine,
	  cam_error_print() and its string-based counterpart,
	  cam_error_string() that allow the kernel and userland
	  applications to pass in a CCB and have errors printed out
	  properly, whether or not they're SCSI errors.  Among other
	  things, this helped eliminate a fair amount of duplicate code
	  in camcontrol.

	  We now print out more information than before, including
	  the CAM status and SCSI status and the error recovery action
	  taken to remedy the problem.

Obtained-from: FreeBSD

Change the peripheral driver list from a linker set to module driven
driver registration.  This should allow things like da, sa, cd etc to be
in seperate KLD's to the cam core and make them preloadable.

Obtained-from: FreeBSD

Switch to sys/queue.h macros.

When CAM creates the disk layer, set the underlying raw device's
si_iosize_max to MAXPHYS.  For now assume that underlying real scsi
devices are not further limited.

Note that NATA's CAM layer will chain I/Os coming from the CAM subsytem
which are too large.

Use TAILQ_FOREACH to enhance readibility.

Disklabel separation work - more.

dev_dopen() can be called multiple times with only one dev_dclose() when
D_TRACKCLOSE is not specified.  Make sure multiple opens only reference
the CAM peripheral once and complain if dev_dclose() is called more then
once.

This also prevents bugs in higher layers from propogating to lower ones.

Port 16 byte SCSI command support from FreeBSD.  This adds support for
detecting and performing I/O on disks greater then 2TB (with more
then 2^32 blocks).

Remove KKASSERT from dual user/kernel compiled file - reported by Sascha

Obtained-from: FreeBSD / Kenneth D. Merry did the original FreeBSD work.

Continue untangling the disklabel.  Have most disk device drivers fill out
and install a generic disk_info structure instead of filling out random
fields in the disklabel.

The generic disk_info structure uses a 64 bit integer to represent
the media size in bytes or total sector count.

Rename printf -> kprintf in sys/ and add some defines where necessary
(files which are used in userland, too).

Rename sprintf  -> ksprintf
Rename snprintf -> knsprintf

Make allowances for source files that are compiled for both userland and
the kernel.

Change the kernel dev_t, representing a pointer to a specinfo structure,
to cdev_t.  Change struct specinfo to struct cdev.  The name 'cdev' was taken
from FreeBSD.  Remove the dev_t shim for the kernel.

This commit generally removes the overloading of 'dev_t' between userland and
the kernel.

Also fix a bug in libkvm where a kernel dev_t (now cdev_t) was not being
properly converted to a userland dev_t.

Rename malloc->kmalloc, free->kfree, and realloc->krealloc.  Pass 1

MASSIVE reorganization of the device operations vector.  Change cdevsw
to dev_ops.  dev_ops is a syslink-compatible operations vector structure
similar to the vop_ops structure used by vnodes.

Remove a huge number of instances where a thread pointer is still being
passed as an argument to various device ops and other related routines.
The device OPEN and IOCTL calls now take a ucred instead of a thread pointer,
and the CLOSE call no longer takes a thread pointer.

Replace the the buffer cache's B_READ, B_WRITE, B_FORMAT, and B_FREEBUF
b_flags with a separate b_cmd field.  Use b_cmd to test for I/O completion
as well (getting rid of B_DONE in the process).  This further simplifies
the setup required to issue a buffer cache I/O.

Remove a redundant header file, bus/isa/i386/isa_dma.h and merge any
discrepancies into bus/isa/isavar.h.

Give ISADMA_READ/WRITE/RAW their own independant flag definitions instead of
trying to overload them on top of B_READ, B_WRITE, and B_RAW.  Add a
routine isa_dmabp() which takes a struct buf pointer and returns the ISA
dma flags associated with the operation.

Remove the 'clear_modify' argument to vfs_busy_pages().  Instead,
vfs_busy_pages() asserts that the buffer's b_cmd is valid and then uses
it to determine the action it must take.

Major BUF/BIO work commit.  Make I/O BIO-centric and specify the disk or
file location with a 64 bit offset instead of a 32 bit block number.

* All I/O is now BIO-centric instead of BUF-centric.

* File/Disk addresses universally use a 64 bit bio_offset now.  bio_blkno
  no longer exists.

* Stackable BIO's hold disk offset translations.  Translations are no longer
  overloaded onto a single structure (BUF or BIO).

* bio_offset == NOOFFSET is now universally used to indicate that a
  translation has not been made.  The old (blkno == lblkno) junk has all
  been removed.

* There is no longer a distinction between logical I/O and physical I/O.

* All driver BUFQs have been converted to BIOQs.

* BMAP, FREEBLKS, getblk, bread, breadn, bwrite, inmem, cluster_*,
  and findblk all now take and/or return 64 bit byte offsets instead
  of block numbers.  Note that BMAP now returns a byte range for the before
  and after variables.

Make the entire BUF/BIO system BIO-centric instead of BUF-centric.  Vnode
and device strategy routines now take a BIO and must pass that BIO to
biodone().  All code which previously managed a BUF undergoing I/O now
manages a BIO.

The new BIO-centric algorithms allow BIOs to be stacked, where each layer
represents a block translation, completion callback, or caller or device
private data.  This information is no longer overloaded within the BUF.
Translation layer linkages remain intact as a 'cache' after I/O has completed.

The VOP and DEV strategy routines no longer make assumptions as to which
translated block number applies to them.  The use the block number in the
BIO specifically passed to them.

Change the 'untranslated' constant to NOOFFSET (for bio_offset), and
(daddr_t)-1 (for bio_blkno).  Rip out all code that previously set the
translated block number to the untranslated block number to indicate
that the translation had not been made.

Rip out all the cluster linkage fields for clustered VFS and clustered
paging operations.  Clustering now occurs in a private BIO layer using
private fields within the BIO.

Reformulate the vn_strategy() and dev_dstrategy() abstraction(s).  These
routines no longer assume that bp->b_vp == the vp of the VOP operation, and
the dev_t is no longer stored in the struct buf.  Instead, only the vp passed
to vn_strategy() (and related *_strategy() routines for VFS ops), and
the dev_t passed to dev_dstrateg() (and related *_strategy() routines for
device ops) is used by the VFS or DEV code.  This will allow an arbitrary
number of translation layers in the future.

Create an independant per-BIO tracking entity, struct bio_track, which
is used to determine when I/O is in-progress on the associated device
or vnode.

NOTE: Unlike FreeBSD's BIO work, our struct BUF is still used to hold
the fields describing the data buffer, resid, and error state.

Major-testing-by: Stefan Krueger

* Move function types to a separate line.

* Ansify function definitions.

* Remove (void) casts for discarded return values.

In collaboration with: Alexey Slynko <slynko@tronet.ru>

Give each CPU its own taskqueue thread so per-cpu data can be
safely manipulated in taskqueue-defered routines.

Conditionalize an #include so libcam can build the file.

Remove spl*() calls from the bus/ infrastructure, replacing them with
critical sections.  Remove splusb() from everywhere, replacing it with
critical sections.

CISS quirk.

Obtained-from: FreeBSD

Improve seperation between kernel code and userland code by requiring
that source files that #include kernel headers also #define _KERNEL or
_KERNEL_STRUCTURES as appropriate.

With-suggestions-from: joerg
Still-todo: systimer.h, and clean up scsi_da.c
Tested-by: cd /usr/src/nrelease && make installer_release
Approved-by: dillon
If-anything-breaks-yell-at: cpressey

timeout/untimeout ==> callout_*

add in support Frontier Labs NEX IA+ Digital Audio Player with USB CF
card reader/writer

Merged from fbsd.

Add in support for the iRiver iFP MP3 player.

Merged from fbsd.

Device layer rollup commit.

* cdevsw_add() is now required.  cdevsw_add() and cdevsw_remove() may specify
  a mask/match indicating the range of supported minor numbers.  Multiple
  cdevsw_add()'s using the same major number, but distinctly different
  ranges, may be issued.  All devices that failed to call cdevsw_add() before
  now do.

* cdevsw_remove() now automatically marks all devices within its supported
  range as being destroyed.

* vnode->v_rdev is no longer resolved when the vnode is created.  Instead,
  only v_udev (a newly added field) is resolved.  v_rdev is resolved when
  the vnode is opened and cleared on the last close.

* A great deal of code was making rather dubious assumptions with regards
  to the validity of devices associated with vnodes, primarily due to
  the persistence of a device structure due to being indexed by (major, minor)
  instead of by (cdevsw, major, minor).  In particular, if you run a program
  which connects to a USB device and then you pull the USB device and plug
  it back in, the vnode subsystem will continue to believe that the device
  is open when, in fact, it isn't (because it was destroyed and recreated).

  In particular, note that all the VFS mount procedures now check devices
  via v_udev instead of v_rdev prior to calling VOP_OPEN(), since v_rdev
  is NULL prior to the first open.

* The disk layer's device interaction has been rewritten.  The disk layer
  (i.e. the slice and disklabel management layer) no longer overloads
  its data onto the device structure representing the underlying physical
  disk.  Instead, the disk layer uses the new cdevsw_add() functionality
  to register its own cdevsw using the underlying device's major number,
  and simply does NOT register the underlying device's cdevsw.  No
  confusion is created because the device hash is now based on
  (cdevsw,major,minor) rather then (major,minor).

  NOTE: This also means that underlying raw disk devices may use the entire
  device minor number instead of having to reserve the bits used by the disk
  layer, and also means that can we (theoretically) stack a fully
  disklabel-supported 'disk' on top of any block device.

* The new reference counting scheme prevents this by associating a device
  with a cdevsw and disconnecting the device from its cdevsw when the cdevsw
  is removed.  Additionally, all udev2dev() lookups run through the cdevsw
  mask/match and only successfully find devices still associated with an
  active cdevsw.

* Major work on MFS:  MFS no longer shortcuts vnode and device creation.  It
  now creates a real vnode and a real device and implements real open and
  close VOPs.  Additionally, due to the disk layer changes, MFS is no longer
  limited to 255 mounts.  The new limit is 16 million.  Since MFS creates a
  real device node, mount_mfs will now create a real /dev/mfs<PID> device
  that can be read from userland (e.g. so you can dump an MFS filesystem).

* BUF AND DEVICE STRATEGY changes.  The struct buf contains a b_dev field.
  In order to properly handle stacked devices we now require that the b_dev
  field be initialized before the device strategy routine is called.  This
  required some additional work in various VFS implementations.  To enforce
  this requirement, biodone() now sets b_dev to NODEV.  The new disk layer
  will adjust b_dev before forwarding a request to the actual physical
  device.

* A bug in the ISO CD boot sequence which resulted in a panic has been fixed.

Testing by: lots of people, but David Rhodus found the most aggregious bugs.

device switch 1/many: Remove d_autoq, add d_clone (where d_autoq was).

d_autoq was used to allow the device port dispatch to mix old-style synchronous
calls with new style messaging calls within a particular device.  It was never
used for that purpose.

d_clone will be more fully implemented as work continues.  We are going to
install d_port in the dev_t (struct specinfo) structure itself and d_clone
will be needed to allow devices to 'revector' the port on a minor-number
by minor-number basis, in particular allowing minor numbers to be directly
dispatched to distinct threads.  This is something we will be needing later
on.

Do some M_WAITOK<->M_INTWAIT cleanups.  Code entered from userland, such as
device open and device ioctl, generally use M_WAITOK, while low level
structures such as the capacity structure are allocated using M_INTWAIT.

Change M_NOWAIT to M_INTWAIT or M_WAITOK.  CAM does a mediocre job checking
for NULL returns from malloc() and even when it does it generally causes
the device operation to fail instead of retrying, resulting in unacceptable
behavior.  M_NOWAIT semantics allow NULL to be returned during normal system
operation.  This is especially true in DragonFly.

Also remove much of the code that previously checked for NULL.  By using
M_INTWAIT or M_WAITOK, malloc() will panic rather then return NULL.  Only
the addition of M_NULLOK allows a blocking malloc() to return NULL, and we
do not use that flag in CAM.

Add M_ZERO to a number of malloc()'s and remove subsequent bzero()'s, and
add M_ZERO to a few mallocs (primarily for the read capacity data structure)
that did not bother zeroing out the structure before.  While the data is
supposed to be overwritten read-capacity is often quite fragile due to the
SCSI simulation layer, so we do not take any chances.

Remove duplicate declarations in preparation for adding -fno-common to
the kernel compile.

Synchronize the USB, CAM, and TASKQUEUE subsystems with FreeBSD RELENG_4.
Also update the $FreeBSD$ ids in the files to the synchronized rev numbers.

This has the side effect of bringing in some additional SCSI robustness
checks, bug fixes, quirk inheritance between subsystems (e.g. USB now sets
PIM_NO_6_BYTE by default and CAM now understands it).

This also brings in a huge amount of SCSI CD code that had been MFCd to
FreeBSD-4 from FreeBSD-5.

*	Add quirk for LEXAR 256MB JUMPDRIVE

Do some fairly major include file cleanups to further separate kernelland
from userland.

    * Do not allow userland to include sys/proc.h directly, it must use
      sys/user.h instead.   This is because sys/proc.h has a huge number
      of kernel header file dependancies.

    * Do cleanups and work in lwkt_thread.c and lwkt_msgport.c to allow
      these files to be directly compiled in an upcoming userland thread
      support library.

    * sys/lock.h is inappropriately included by a number of third party
      programs so we can't disallow its inclusion, but do not include
      any kernel structures unless _KERNEL or _KERNEL_STRUCTURES are
      defined.

    * <ufs/ufs/inode.h> is often included by userland to get at the
      on-disk inode structure.  Only include the on-disk components and do
      not include kernel structural components unless _KERNEL or
      _KERNEL_STRUCTURES is defined

    * Various usr.bin programs include sys/proc.h unnecessarily.

    * The slab allocator has no concept of malloc buckets.  Remove malloc
      buckets structures and VMSTAT support from the system.

    * Make adjustments to sys/thread.h and sys/msgport.h such that the
      upcoming userland thread support library can include these files
      directly rather then copy them.

    * Use low level __int types in sys/globaldata.h, sys/msgport.h,
      sys/slaballoc.h, sys/thread.h, and sys/malloc.h, instead of
      high level sys/types.h types, reducing include dependancies.

Make sure a DiskOnKey never sends a sync_cache.  This fixes the error
message when using my DiskOnKey.

Incidentally it also fixes FreeBSD/PR: 47793

kernel tree reorganization stage 1: Major cvs repository work (not logged as
commits) plus a major reworking of the #include's to accomodate the
relocations.

    * CVS repository files manually moved.  Old directories left intact
      and empty (temporary).

    * Reorganize all filesystems into vfs/, most devices into dev/,
      sub-divide devices by function.

    * Begin to move device-specific architecture files to the device
      subdirs rather then throwing them all into, e.g. i386/include

    * Reorganize files related to system busses, placing the related code
      in a new bus/ directory.  Also move cam to bus/cam though this may
      not have been the best idea in retrospect.

    * Reorganize emulation code and place it in a new emulation/ directory.

    * Remove the -I- compiler option in order to allow #include file
      localization, rename all config generated X.h files to use_X.h to
      clean up the conflicts.

    * Remove /usr/src/include (or /usr/include) dependancies during the
      kernel build, beyond what is normally needed to compile helper
      programs.

    * Make config create 'machine' softlinks for architecture specific
      directories outside of the standard <arch>/include.

    * Bump the config rev.

    WARNING! after this commit /usr/include and /usr/src/sys/compile/*
    should be regenerated from scratch.

DEV messaging stage 2/4: In this stage all DEV commands are now being
funneled through the message port for action by the port's beginmsg function.
CONSOLE and DISK device shims replace the port with their own and then
forward to the original.  FB (Frame Buffer) shims supposedly do the same
thing but I haven't been able to test it.   I don't expect instability
in mainline code but there might be easy-to-fix, and some drivers still need
to be converted.  See primarily: kern/kern_device.c (new dev_*() functions and
inherits cdevsw code from kern/kern_conf.c), sys/device.h, and kern/subr_disk.c
for the high points.

In this stage all DEV messages are still acted upon synchronously in the
context of the caller.  We cannot create a separate handler thread until
the copyin's (primarily in ioctl functions) are made thread-aware.

Note that the messaging shims are going to look rather messy in these early
days but as more subsystems are converted over we will begin to use
pre-initialized messages and message forwarding to avoid having to constantly
rebuild messages prior to use.

Note that DEV itself is a mess oweing to its 4.x roots and will be cleaned
up in subsequent passes.  e.g. the way sub-devices inherit the main device's
cdevsw was always a bad hack and it still is, and several functions
(mmap, kqfilter, psize, poll) return results rather then error codes, which
will be fixed since now we have a message to store the result in :-)

DEV messaging stage 1/4: Rearrange struct cdevsw and add a message port
and auto-queueing mask.  The mask will tell us which message functions
can be safely queued to another thread and which still need to run in the
context of the caller.   Primary configuration fields (name, cmaj, flags,
port, autoq mask) are now at the head of the structure.  Function vectors,
which may eventually go away, are at the end.  The port and autoq fields
are non-functional in this stage.

The old BDEV device major number support has also been removed from cdevsw,
and code has been added to translate the bootdev passed from the boot code
(the boot code has always passed the now defunct block device major numbers
and we obviously need to keep that compatibility intact).

Remove the priority part of the priority|flags argument to tsleep().  Only
flags are passed now.  The priority was a user scheduler thingy that is not
used by the LWKT subsystem.  For process statistics assume sleeps without
P_SINTR set to be disk-waits, and sleeps with it set to be normal sleeps.

This commit should not contain any operational changes.

proc->thread stage 2: MAJOR revamping of system calls, ucred, jail API,
and some work on the low level device interface (proc arg -> thread arg).
As -current did, I have removed p_cred and incorporated its functions
into p_ucred.  p_prison has also been moved into p_ucred and adjusted
accordingly.  The jail interface tests now uses ucreds rather then processes.

The syscall(p,uap) interface has been changed to just (uap).  This is inclusive
of the emulation code.  It makes little sense to pass a proc pointer around
which confuses the MP readability of the code, because most system call code
will only work with the current process anyway.  Note that eventually
*ALL* syscall emulation code will be moved to a kernel-protected userland
layer because it really makes no sense whatsoever to implement these
emulations in the kernel.

suser() now takes no arguments and only operates with the current process.
The process argument has been removed from suser_xxx() so it now just takes
a ucred and flags.

The sysctl interface was adjusted somewhat.

thread stage 5: Separate the inline functions out of sys/buf.h, creating
sys/buf2.h (A methodology that will continue as time passes).  This solves
inline vs struct ordering problems.

Do a major cleanup of the globaldata access methodology.  Create a
gcc-cacheable 'mycpu' macro & inline to access per-cpu data.  Atomicy is not
required because we will never change cpus out from under a thread, even if
it gets preempted by an interrupt thread, because we want to be able to
implement per-cpu caches that do not require locked bus cycles or special
instructions.

Add the DragonFly cvs id and perform general cleanups on cvs/rcs/sccs ids.  Most
ids have been removed from !lint sections and moved into comment sections.

import from FreeBSD RELENG_4 1.42.2.36