2019-05-19 15:08:20 +03:00
// SPDX-License-Identifier: GPL-2.0-only
2005-04-17 02:20:36 +04:00
/*
* scsi . c Copyright ( C ) 1992 Drew Eckhardt
* Copyright ( C ) 1993 , 1994 , 1995 , 1999 Eric Youngdale
* Copyright ( C ) 2002 , 2003 Christoph Hellwig
*
* generic mid - level SCSI driver
* Initial versions : Drew Eckhardt
* Subsequent revisions : Eric Youngdale
*
* < drew @ colorado . edu >
*
* Bug correction thanks go to :
* Rik Faith < faith @ cs . unc . edu >
* Tommy Thorn < tthorn >
* Thomas Wuensche < tw @ fgb1 . fgb . mw . tu - muenchen . de >
*
* Modified by Eric Youngdale eric @ andante . org or ericy @ gnu . ai . mit . edu to
* add scatter - gather , multiple outstanding request , and other
* enhancements .
*
* Native multichannel , wide scsi , / proc / scsi and hot plugging
* support added by Michael Neuffer < mike @ i - connect . net >
*
* Added request_module ( " scsi_hostadapter " ) for kerneld :
* ( Put an " alias scsi_hostadapter your_hostadapter " in / etc / modprobe . conf )
* Bjorn Ekwall < bj0rn @ blox . se >
* ( changed to kmod )
*
* Major improvements to the timeout , abort , and reset processing ,
* as well as performance modifications for large queue depths by
* Leonard N . Zubkoff < lnz @ dandelion . com >
*
* Converted cli ( ) code to spinlocks , Ingo Molnar
*
* Jiffies wrap fixes ( host - > resetting ) , 3 Dec 1998 Andrea Arcangeli
*
* out_of_space hacks , D . Gilbert ( dpg ) 990608
*/
# include <linux/module.h>
# include <linux/moduleparam.h>
# include <linux/kernel.h>
# include <linux/timer.h>
# include <linux/string.h>
# include <linux/slab.h>
# include <linux/blkdev.h>
# include <linux/delay.h>
# include <linux/init.h>
# include <linux/completion.h>
# include <linux/unistd.h>
# include <linux/spinlock.h>
# include <linux/kmod.h>
# include <linux/interrupt.h>
# include <linux/notifier.h>
# include <linux/cpu.h>
2006-01-11 15:16:10 +03:00
# include <linux/mutex.h>
2012-07-10 06:33:25 +04:00
# include <linux/async.h>
2012-09-18 20:19:30 +04:00
# include <asm/unaligned.h>
2005-04-17 02:20:36 +04:00
# include <scsi/scsi.h>
# include <scsi/scsi_cmnd.h>
# include <scsi/scsi_dbg.h>
# include <scsi/scsi_device.h>
Revert "scsi: revert "[SCSI] Get rid of scsi_cmnd->done""
This reverts commit ac40532ef0b8649e6f7f83859ea0de1c4ed08a19, which gets
us back the original cleanup of 6f5391c283d7fdcf24bf40786ea79061919d1e1d.
It turns out that the bug that was triggered by that commit was
apparently not actually triggered by that commit at all, and just the
testing conditions had changed enough to make it appear to be due to it.
The real problem seems to have been found by Peter Osterlund:
"pktcdvd sets it [block device size] when opening the /dev/pktcdvd
device, but when the drive is later opened as /dev/scd0, there is
nothing that sets it back. (Btw, 40944 is possible if the disk is a
CDRW that was formatted with "cdrwtool -m 10236".)
The problem is that pktcdvd opens the cd device in non-blocking mode
when pktsetup is run, and doesn't close it again until pktsetup -d is
run. The effect is that if you meanwhile open the cd device,
blkdev.c:do_open() doesn't call bd_set_size() because
bdev->bd_openers is non-zero."
In particular, to repeat the bug (regardless of whether commit
6f5391c283d7fdcf24bf40786ea79061919d1e1d is applied or not):
" 1. Start with an empty drive.
2. pktsetup 0 /dev/scd0
3. Insert a CD containing an isofs filesystem.
4. mount /dev/pktcdvd/0 /mnt/tmp
5. umount /mnt/tmp
6. Press the eject button.
7. Insert a DVD containing a non-writable filesystem.
8. mount /dev/scd0 /mnt/tmp
9. find /mnt/tmp -type f -print0 | xargs -0 sha1sum >/dev/null
10. If the DVD contains data beyond the physical size of a CD, you
get I/O errors in the terminal, and dmesg reports lots of
"attempt to access beyond end of device" errors."
which in turn is because the nested open after the media change won't
cause the size to be set properly (because the original open still holds
the block device, and we only do the bd_set_size() when we don't have
other people holding the device open).
The proper fix for that is probably to just do something like
bdev->bd_inode->i_size = (loff_t)get_capacity(disk)<<9;
in fs/block_dev.c:do_open() even for the cases where we're not the
original opener (but *not* call bd_set_size(), since that will also
change the block size of the device).
Cc: Peter Osterlund <petero2@telia.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-01-06 21:17:12 +03:00
# include <scsi/scsi_driver.h>
2005-04-17 02:20:36 +04:00
# include <scsi/scsi_eh.h>
# include <scsi/scsi_host.h>
# include <scsi/scsi_tcq.h>
# include "scsi_priv.h"
# include "scsi_logging.h"
2010-04-01 15:41:40 +04:00
# define CREATE_TRACE_POINTS
# include <trace/events/scsi.h>
2005-04-17 02:20:36 +04:00
/*
* Definitions and constants .
*/
/*
* Note - the initial logging level can be set here to log events at boot time .
* After the system is up , you may enable logging via the / proc interface .
*/
unsigned int scsi_logging_level ;
# if defined(CONFIG_SCSI_LOGGING)
EXPORT_SYMBOL ( scsi_logging_level ) ;
# endif
2014-04-11 02:30:35 +04:00
/*
2019-05-01 00:39:18 +03:00
* Domain for asynchronous system resume operations . It is marked ' exclusive '
* to avoid being included in the async_synchronize_full ( ) that is invoked by
* dpm_resume ( ) .
2014-04-11 02:30:35 +04:00
*/
ASYNC_DOMAIN_EXCLUSIVE ( scsi_sd_pm_domain ) ;
EXPORT_SYMBOL ( scsi_sd_pm_domain ) ;
2005-04-17 02:20:36 +04:00
# ifdef CONFIG_SCSI_LOGGING
void scsi_log_send ( struct scsi_cmnd * cmd )
{
unsigned int level ;
/*
* If ML QUEUE log level is greater than or equal to :
*
* 1 : nothing ( match completion )
*
2014-10-24 16:27:01 +04:00
* 2 : log opcode + command of all commands + cmd address
2005-04-17 02:20:36 +04:00
*
2014-10-24 16:27:01 +04:00
* 3 : same as 2
2005-04-17 02:20:36 +04:00
*
2015-01-08 09:43:49 +03:00
* 4 : same as 3
2005-04-17 02:20:36 +04:00
*/
if ( unlikely ( scsi_logging_level ) ) {
level = SCSI_LOG_LEVEL ( SCSI_LOG_MLQUEUE_SHIFT ,
SCSI_LOG_MLQUEUE_BITS ) ;
if ( level > 1 ) {
2014-10-24 16:27:01 +04:00
scmd_printk ( KERN_INFO , cmd ,
" Send: scmd 0x%p \n " , cmd ) ;
2005-04-17 02:20:36 +04:00
scsi_print_command ( cmd ) ;
}
}
}
void scsi_log_completion ( struct scsi_cmnd * cmd , int disposition )
{
unsigned int level ;
/*
* If ML COMPLETE log level is greater than or equal to :
*
* 1 : log disposition , result , opcode + command , and conditionally
* sense data for failures or non SUCCESS dispositions .
*
* 2 : same as 1 but for all command completions .
*
2014-10-24 16:27:01 +04:00
* 3 : same as 2
2005-04-17 02:20:36 +04:00
*
* 4 : same as 3 plus dump extra junk
*/
if ( unlikely ( scsi_logging_level ) ) {
level = SCSI_LOG_LEVEL ( SCSI_LOG_MLCOMPLETE_SHIFT ,
SCSI_LOG_MLCOMPLETE_BITS ) ;
if ( ( ( level > 0 ) & & ( cmd - > result | | disposition ! = SUCCESS ) ) | |
( level > 1 ) ) {
2015-01-08 09:43:47 +03:00
scsi_print_result ( cmd , " Done " , disposition ) ;
2005-04-17 02:20:36 +04:00
scsi_print_command ( cmd ) ;
2021-04-27 11:30:14 +03:00
if ( scsi_status_is_check_condition ( cmd - > result ) )
2014-10-24 16:26:45 +04:00
scsi_print_sense ( cmd ) ;
2007-02-28 06:40:27 +03:00
if ( level > 3 )
scmd_printk ( KERN_INFO , cmd ,
" scsi host busy %d failed %d \n " ,
2018-06-24 17:03:26 +03:00
scsi_host_busy ( cmd - > device - > host ) ,
2007-02-28 06:40:27 +03:00
cmd - > device - > host - > host_failed ) ;
2005-04-17 02:20:36 +04:00
}
}
}
# endif
2007-11-03 21:30:39 +03:00
/**
* scsi_finish_command - cleanup and pass command back to upper layer
* @ cmd : the command
2005-04-17 02:20:36 +04:00
*
2007-11-03 21:30:39 +03:00
* Description : Pass command off to upper layer for finishing of I / O
2005-04-17 02:20:36 +04:00
* request , waking processes that are waiting on results ,
* etc .
*/
void scsi_finish_command ( struct scsi_cmnd * cmd )
{
struct scsi_device * sdev = cmd - > device ;
[SCSI] Add helper code so transport classes/driver can control queueing (v3)
SCSI-ml manages the queueing limits for the device and host, but
does not do so at the target level. However something something similar
can come in userful when a driver is transitioning a transport object to
the the blocked state, becuase at that time we do not want to queue
io and we do not want the queuecommand to be called again.
The patch adds code similar to the exisiting SCSI_ML_*BUSY handlers.
You can now return SCSI_MLQUEUE_TARGET_BUSY when we hit
a transport level queueing issue like the hw cannot allocate some
resource at the iscsi session/connection level, or the target has temporarily
closed or shrunk the queueing window, or if we are transitioning
to the blocked state.
bnx2i, when they rework their firmware according to netdev
developers requests, will also need to be able to limit queueing at this
level. bnx2i will hook into libiscsi, but will allocate a scsi host per
netdevice/hba, so unlike pure software iscsi/iser which is allocating
a host per session, it cannot set the scsi_host->can_queue and return
SCSI_MLQUEUE_HOST_BUSY to reflect queueing limits on the transport.
The iscsi class/driver can also set a scsi_target->can_queue value which
reflects the max commands the driver/class can support. For iscsi this
reflects the number of commands we can support for each session due to
session/connection hw limits, driver limits, and to also reflect the
session/targets's queueing window.
Changes:
v1 - initial patch.
v2 - Fix scsi_run_queue handling of multiple blocked targets.
Previously we would break from the main loop if a device was added back on
the starved list. We now run over the list and check if any target is
blocked.
v3 - Rediff for scsi-misc.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-08-18 00:24:38 +04:00
struct scsi_target * starget = scsi_target ( sdev ) ;
2005-04-17 02:20:36 +04:00
struct Scsi_Host * shost = sdev - > host ;
Revert "scsi: revert "[SCSI] Get rid of scsi_cmnd->done""
This reverts commit ac40532ef0b8649e6f7f83859ea0de1c4ed08a19, which gets
us back the original cleanup of 6f5391c283d7fdcf24bf40786ea79061919d1e1d.
It turns out that the bug that was triggered by that commit was
apparently not actually triggered by that commit at all, and just the
testing conditions had changed enough to make it appear to be due to it.
The real problem seems to have been found by Peter Osterlund:
"pktcdvd sets it [block device size] when opening the /dev/pktcdvd
device, but when the drive is later opened as /dev/scd0, there is
nothing that sets it back. (Btw, 40944 is possible if the disk is a
CDRW that was formatted with "cdrwtool -m 10236".)
The problem is that pktcdvd opens the cd device in non-blocking mode
when pktsetup is run, and doesn't close it again until pktsetup -d is
run. The effect is that if you meanwhile open the cd device,
blkdev.c:do_open() doesn't call bd_set_size() because
bdev->bd_openers is non-zero."
In particular, to repeat the bug (regardless of whether commit
6f5391c283d7fdcf24bf40786ea79061919d1e1d is applied or not):
" 1. Start with an empty drive.
2. pktsetup 0 /dev/scd0
3. Insert a CD containing an isofs filesystem.
4. mount /dev/pktcdvd/0 /mnt/tmp
5. umount /mnt/tmp
6. Press the eject button.
7. Insert a DVD containing a non-writable filesystem.
8. mount /dev/scd0 /mnt/tmp
9. find /mnt/tmp -type f -print0 | xargs -0 sha1sum >/dev/null
10. If the DVD contains data beyond the physical size of a CD, you
get I/O errors in the terminal, and dmesg reports lots of
"attempt to access beyond end of device" errors."
which in turn is because the nested open after the media change won't
cause the size to be set properly (because the original open still holds
the block device, and we only do the bd_set_size() when we don't have
other people holding the device open).
The proper fix for that is probably to just do something like
bdev->bd_inode->i_size = (loff_t)get_capacity(disk)<<9;
in fs/block_dev.c:do_open() even for the cases where we're not the
original opener (but *not* call bd_set_size(), since that will also
change the block size of the device).
Cc: Peter Osterlund <petero2@telia.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-01-06 21:17:12 +03:00
struct scsi_driver * drv ;
unsigned int good_bytes ;
2005-04-17 02:20:36 +04:00
2019-10-25 09:58:55 +03:00
scsi_device_unbusy ( sdev , cmd ) ;
2005-04-17 02:20:36 +04:00
2014-01-23 15:07:41 +04:00
/*
* Clear the flags that say that the device / target / host is no longer
* capable of accepting new commands .
*/
if ( atomic_read ( & shost - > host_blocked ) )
atomic_set ( & shost - > host_blocked , 0 ) ;
if ( atomic_read ( & starget - > target_blocked ) )
atomic_set ( & starget - > target_blocked , 0 ) ;
if ( atomic_read ( & sdev - > device_blocked ) )
atomic_set ( & sdev - > device_blocked , 0 ) ;
2005-04-17 02:20:36 +04:00
2005-10-25 02:04:06 +04:00
SCSI_LOG_MLCOMPLETE ( 4 , sdev_printk ( KERN_INFO , sdev ,
" Notifying upper driver of completion "
" (result %x) \n " , cmd - > result ) ) ;
2005-04-17 02:20:36 +04:00
2008-04-11 14:56:52 +04:00
good_bytes = scsi_bufflen ( cmd ) ;
2021-08-10 02:03:05 +03:00
if ( ! blk_rq_is_passthrough ( scsi_cmd_to_rq ( cmd ) ) ) {
2008-03-09 03:24:17 +03:00
int old_good_bytes = good_bytes ;
Revert "scsi: revert "[SCSI] Get rid of scsi_cmnd->done""
This reverts commit ac40532ef0b8649e6f7f83859ea0de1c4ed08a19, which gets
us back the original cleanup of 6f5391c283d7fdcf24bf40786ea79061919d1e1d.
It turns out that the bug that was triggered by that commit was
apparently not actually triggered by that commit at all, and just the
testing conditions had changed enough to make it appear to be due to it.
The real problem seems to have been found by Peter Osterlund:
"pktcdvd sets it [block device size] when opening the /dev/pktcdvd
device, but when the drive is later opened as /dev/scd0, there is
nothing that sets it back. (Btw, 40944 is possible if the disk is a
CDRW that was formatted with "cdrwtool -m 10236".)
The problem is that pktcdvd opens the cd device in non-blocking mode
when pktsetup is run, and doesn't close it again until pktsetup -d is
run. The effect is that if you meanwhile open the cd device,
blkdev.c:do_open() doesn't call bd_set_size() because
bdev->bd_openers is non-zero."
In particular, to repeat the bug (regardless of whether commit
6f5391c283d7fdcf24bf40786ea79061919d1e1d is applied or not):
" 1. Start with an empty drive.
2. pktsetup 0 /dev/scd0
3. Insert a CD containing an isofs filesystem.
4. mount /dev/pktcdvd/0 /mnt/tmp
5. umount /mnt/tmp
6. Press the eject button.
7. Insert a DVD containing a non-writable filesystem.
8. mount /dev/scd0 /mnt/tmp
9. find /mnt/tmp -type f -print0 | xargs -0 sha1sum >/dev/null
10. If the DVD contains data beyond the physical size of a CD, you
get I/O errors in the terminal, and dmesg reports lots of
"attempt to access beyond end of device" errors."
which in turn is because the nested open after the media change won't
cause the size to be set properly (because the original open still holds
the block device, and we only do the bd_set_size() when we don't have
other people holding the device open).
The proper fix for that is probably to just do something like
bdev->bd_inode->i_size = (loff_t)get_capacity(disk)<<9;
in fs/block_dev.c:do_open() even for the cases where we're not the
original opener (but *not* call bd_set_size(), since that will also
change the block size of the device).
Cc: Peter Osterlund <petero2@telia.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-01-06 21:17:12 +03:00
drv = scsi_cmd_to_driver ( cmd ) ;
if ( drv - > done )
good_bytes = drv - > done ( cmd ) ;
2008-03-09 03:24:17 +03:00
/*
* USB may not give sense identifying bad sector and
* simply return a residue instead , so subtract off the
* residue if drv - > done ( ) error processing indicates no
* change to the completion length .
*/
if ( good_bytes = = old_good_bytes )
good_bytes - = scsi_get_resid ( cmd ) ;
Revert "scsi: revert "[SCSI] Get rid of scsi_cmnd->done""
This reverts commit ac40532ef0b8649e6f7f83859ea0de1c4ed08a19, which gets
us back the original cleanup of 6f5391c283d7fdcf24bf40786ea79061919d1e1d.
It turns out that the bug that was triggered by that commit was
apparently not actually triggered by that commit at all, and just the
testing conditions had changed enough to make it appear to be due to it.
The real problem seems to have been found by Peter Osterlund:
"pktcdvd sets it [block device size] when opening the /dev/pktcdvd
device, but when the drive is later opened as /dev/scd0, there is
nothing that sets it back. (Btw, 40944 is possible if the disk is a
CDRW that was formatted with "cdrwtool -m 10236".)
The problem is that pktcdvd opens the cd device in non-blocking mode
when pktsetup is run, and doesn't close it again until pktsetup -d is
run. The effect is that if you meanwhile open the cd device,
blkdev.c:do_open() doesn't call bd_set_size() because
bdev->bd_openers is non-zero."
In particular, to repeat the bug (regardless of whether commit
6f5391c283d7fdcf24bf40786ea79061919d1e1d is applied or not):
" 1. Start with an empty drive.
2. pktsetup 0 /dev/scd0
3. Insert a CD containing an isofs filesystem.
4. mount /dev/pktcdvd/0 /mnt/tmp
5. umount /mnt/tmp
6. Press the eject button.
7. Insert a DVD containing a non-writable filesystem.
8. mount /dev/scd0 /mnt/tmp
9. find /mnt/tmp -type f -print0 | xargs -0 sha1sum >/dev/null
10. If the DVD contains data beyond the physical size of a CD, you
get I/O errors in the terminal, and dmesg reports lots of
"attempt to access beyond end of device" errors."
which in turn is because the nested open after the media change won't
cause the size to be set properly (because the original open still holds
the block device, and we only do the bd_set_size() when we don't have
other people holding the device open).
The proper fix for that is probably to just do something like
bdev->bd_inode->i_size = (loff_t)get_capacity(disk)<<9;
in fs/block_dev.c:do_open() even for the cases where we're not the
original opener (but *not* call bd_set_size(), since that will also
change the block size of the device).
Cc: Peter Osterlund <petero2@telia.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-01-06 21:17:12 +03:00
}
scsi_io_completion ( cmd , good_bytes ) ;
2005-04-17 02:20:36 +04:00
}
2021-01-22 05:33:16 +03:00
/*
* 1024 is big enough for saturating the fast scsi LUN now
*/
2021-01-22 05:33:17 +03:00
int scsi_device_max_queue_depth ( struct scsi_device * sdev )
2021-01-22 05:33:16 +03:00
{
return max_t ( int , sdev - > host - > can_queue , 1024 ) ;
}
2007-11-03 21:30:39 +03:00
/**
2014-11-13 17:08:42 +03:00
* scsi_change_queue_depth - change a device ' s queue depth
2007-11-03 21:30:39 +03:00
* @ sdev : SCSI Device in question
2014-11-13 17:08:42 +03:00
* @ depth : number of commands allowed to be queued to the driver
2005-04-17 02:20:36 +04:00
*
2014-11-13 17:08:42 +03:00
* Sets the device queue depth and returns the new value .
2005-04-17 02:20:36 +04:00
*/
2014-11-13 17:08:42 +03:00
int scsi_change_queue_depth ( struct scsi_device * sdev , int depth )
2005-04-17 02:20:36 +04:00
{
2021-01-22 05:33:16 +03:00
depth = min_t ( int , depth , scsi_device_max_queue_depth ( sdev ) ) ;
2015-10-08 11:28:04 +03:00
if ( depth > 0 ) {
sdev - > queue_depth = depth ;
wmb ( ) ;
2008-07-12 04:50:32 +04:00
}
2005-04-17 02:20:36 +04:00
2016-03-30 19:21:08 +03:00
if ( sdev - > request_queue )
blk_set_queue_depth ( sdev - > request_queue , depth ) ;
2021-01-22 05:33:17 +03:00
sbitmap_resize ( & sdev - > budget_map , sdev - > queue_depth ) ;
2014-11-13 17:08:42 +03:00
return sdev - > queue_depth ;
2005-04-17 02:20:36 +04:00
}
2014-11-13 17:08:42 +03:00
EXPORT_SYMBOL ( scsi_change_queue_depth ) ;
2005-04-17 02:20:36 +04:00
2007-11-03 21:30:39 +03:00
/**
* scsi_track_queue_full - track QUEUE_FULL events to adjust queue depth
* @ sdev : SCSI Device in question
* @ depth : Current number of outstanding SCSI commands on this device ,
* not counting the one returned as QUEUE_FULL .
2005-04-17 02:20:36 +04:00
*
2007-11-03 21:30:39 +03:00
* Description : This function will track successive QUEUE_FULL events on a
2005-04-17 02:20:36 +04:00
* specific SCSI device to determine if and when there is a
* need to adjust the queue depth on the device .
*
2007-11-03 21:30:39 +03:00
* Returns : 0 - No change needed , > 0 - Adjust queue depth to this new depth ,
2005-04-17 02:20:36 +04:00
* - 1 - Drop back to untagged operation using host - > cmd_per_lun
* as the untagged command depth
*
* Lock Status : None held on entry
*
* Notes : Low level drivers may call this at any time and we will do
* " The Right Thing. " We are interrupt context safe .
*/
int scsi_track_queue_full ( struct scsi_device * sdev , int depth )
{
2009-10-23 02:46:33 +04:00
/*
* Don ' t let QUEUE_FULLs on the same
* jiffies count , they could all be from
* same event .
*/
if ( ( jiffies > > 4 ) = = ( sdev - > last_queue_full_time > > 4 ) )
2005-04-17 02:20:36 +04:00
return 0 ;
2009-10-23 02:46:33 +04:00
sdev - > last_queue_full_time = jiffies ;
2005-04-17 02:20:36 +04:00
if ( sdev - > last_queue_full_depth ! = depth ) {
sdev - > last_queue_full_count = 1 ;
sdev - > last_queue_full_depth = depth ;
} else {
sdev - > last_queue_full_count + + ;
}
if ( sdev - > last_queue_full_count < = 10 )
return 0 ;
2014-10-30 13:54:58 +03:00
2014-11-13 17:08:42 +03:00
return scsi_change_queue_depth ( sdev , depth ) ;
2005-04-17 02:20:36 +04:00
}
EXPORT_SYMBOL ( scsi_track_queue_full ) ;
2008-12-31 21:12:46 +03:00
/**
* scsi_vpd_inquiry - Request a device provide us with a VPD page
* @ sdev : The device to ask
* @ buffer : Where to put the result
* @ page : Which Vital Product Data to return
* @ len : The length of the buffer
*
* This is an internal helper function . You probably want to use
* scsi_get_vpd_page instead .
*
2014-03-15 12:51:48 +04:00
* Returns size of the vpd page on success or a negative error number .
2008-12-31 21:12:46 +03:00
*/
static int scsi_vpd_inquiry ( struct scsi_device * sdev , unsigned char * buffer ,
u8 page , unsigned len )
{
int result ;
unsigned char cmd [ 16 ] ;
2014-03-15 12:51:48 +04:00
if ( len < 4 )
return - EINVAL ;
2008-12-31 21:12:46 +03:00
cmd [ 0 ] = INQUIRY ;
cmd [ 1 ] = 1 ; /* EVPD */
cmd [ 2 ] = page ;
cmd [ 3 ] = len > > 8 ;
cmd [ 4 ] = len & 0xff ;
cmd [ 5 ] = 0 ; /* Control byte */
/*
* I ' m not convinced we need to try quite this hard to get VPD , but
* all the existing users tried this hard .
*/
result = scsi_execute_req ( sdev , cmd , DMA_FROM_DEVICE , buffer ,
2009-08-11 19:59:09 +04:00
len , NULL , 30 * HZ , 3 , NULL ) ;
2008-12-31 21:12:46 +03:00
if ( result )
2014-03-15 12:51:48 +04:00
return - EIO ;
2008-12-31 21:12:46 +03:00
/* Sanity check that we got the page back that we asked for */
if ( buffer [ 1 ] ! = page )
return - EIO ;
2014-03-15 12:51:48 +04:00
return get_unaligned_be16 ( & buffer [ 2 ] ) + 4 ;
2008-12-31 21:12:46 +03:00
}
/**
* scsi_get_vpd_page - Get Vital Product Data from a SCSI device
* @ sdev : The device to ask
* @ page : Which Vital Product Data to return
2010-03-01 04:32:45 +03:00
* @ buf : where to store the VPD
* @ buf_len : number of bytes in the VPD buffer area
2008-12-31 21:12:46 +03:00
*
* SCSI devices may optionally supply Vital Product Data . Each ' page '
* of VPD is defined in the appropriate SCSI document ( eg SPC , SBC ) .
* If the device supports this VPD page , this routine returns a pointer
* to a buffer containing the data from that page . The caller is
* responsible for calling kfree ( ) on this pointer when it is no longer
* needed . If we cannot retrieve the VPD page this routine returns % NULL .
*/
2009-11-03 21:33:07 +03:00
int scsi_get_vpd_page ( struct scsi_device * sdev , u8 page , unsigned char * buf ,
int buf_len )
2008-12-31 21:12:46 +03:00
{
int i , result ;
2013-07-31 06:58:34 +04:00
if ( sdev - > skip_vpd_pages )
goto fail ;
2008-12-31 21:12:46 +03:00
/* Ask for all the pages supported by this device */
2009-11-03 21:33:07 +03:00
result = scsi_vpd_inquiry ( sdev , buf , 0 , buf_len ) ;
2014-03-15 12:51:48 +04:00
if ( result < 4 )
2008-12-31 21:12:46 +03:00
goto fail ;
/* If the user actually wanted this page, we can skip the rest */
if ( page = = 0 )
2010-09-10 09:20:27 +04:00
return 0 ;
2008-12-31 21:12:46 +03:00
2014-03-15 12:51:48 +04:00
for ( i = 4 ; i < min ( result , buf_len ) ; i + + )
if ( buf [ i ] = = page )
2008-12-31 21:12:46 +03:00
goto found ;
2009-11-03 21:33:07 +03:00
2014-03-15 12:51:48 +04:00
if ( i < result & & i > = buf_len )
2009-11-03 21:33:07 +03:00
/* ran off the end of the buffer, give us benefit of doubt */
goto found ;
2008-12-31 21:12:46 +03:00
/* The device claims it doesn't support the requested page */
goto fail ;
found :
2009-11-03 21:33:07 +03:00
result = scsi_vpd_inquiry ( sdev , buf , page , buf_len ) ;
2014-03-15 12:51:48 +04:00
if ( result < 0 )
2008-12-31 21:12:46 +03:00
goto fail ;
2009-11-03 21:33:07 +03:00
return 0 ;
2008-12-31 21:12:46 +03:00
fail :
2009-11-03 21:33:07 +03:00
return - EINVAL ;
2008-12-31 21:12:46 +03:00
}
EXPORT_SYMBOL_GPL ( scsi_get_vpd_page ) ;
2017-08-29 18:50:12 +03:00
/**
* scsi_get_vpd_buf - Get Vital Product Data from a SCSI device
* @ sdev : The device to ask
* @ page : Which Vital Product Data to return
*
* Returns % NULL upon failure .
*/
2017-08-29 18:50:13 +03:00
static struct scsi_vpd * scsi_get_vpd_buf ( struct scsi_device * sdev , u8 page )
2017-08-29 18:50:12 +03:00
{
2017-08-29 18:50:13 +03:00
struct scsi_vpd * vpd_buf ;
2017-08-29 18:50:12 +03:00
int vpd_len = SCSI_VPD_PG_LEN , result ;
retry_pg :
2017-08-29 18:50:13 +03:00
vpd_buf = kmalloc ( sizeof ( * vpd_buf ) + vpd_len , GFP_KERNEL ) ;
2017-08-29 18:50:12 +03:00
if ( ! vpd_buf )
return NULL ;
2017-08-29 18:50:13 +03:00
result = scsi_vpd_inquiry ( sdev , vpd_buf - > data , page , vpd_len ) ;
2017-08-29 18:50:12 +03:00
if ( result < 0 ) {
kfree ( vpd_buf ) ;
return NULL ;
}
if ( result > vpd_len ) {
vpd_len = result ;
kfree ( vpd_buf ) ;
goto retry_pg ;
}
2017-08-29 18:50:13 +03:00
vpd_buf - > len = result ;
2017-08-29 18:50:12 +03:00
return vpd_buf ;
}
static void scsi_update_vpd_page ( struct scsi_device * sdev , u8 page ,
2017-08-29 18:50:13 +03:00
struct scsi_vpd __rcu * * sdev_vpd_buf )
2017-08-29 18:50:12 +03:00
{
2017-08-29 18:50:13 +03:00
struct scsi_vpd * vpd_buf ;
2017-08-29 18:50:12 +03:00
2017-08-29 18:50:13 +03:00
vpd_buf = scsi_get_vpd_buf ( sdev , page ) ;
2017-08-29 18:50:12 +03:00
if ( ! vpd_buf )
return ;
mutex_lock ( & sdev - > inquiry_mutex ) ;
2019-09-24 01:26:28 +03:00
vpd_buf = rcu_replace_pointer ( * sdev_vpd_buf , vpd_buf ,
lockdep_is_held ( & sdev - > inquiry_mutex ) ) ;
2017-08-29 18:50:12 +03:00
mutex_unlock ( & sdev - > inquiry_mutex ) ;
2017-08-29 18:50:13 +03:00
if ( vpd_buf )
kfree_rcu ( vpd_buf , rcu ) ;
2017-08-29 18:50:12 +03:00
}
2014-03-15 12:51:49 +04:00
/**
* scsi_attach_vpd - Attach Vital Product Data to a SCSI device structure
* @ sdev : The device to ask
*
* Attach the ' Device Identification ' VPD page ( 0x83 ) and the
* ' Unit Serial Number ' VPD page ( 0x80 ) to a SCSI device
* structure . This information can be used to identify the device
* uniquely .
*/
void scsi_attach_vpd ( struct scsi_device * sdev )
{
2017-08-29 18:50:13 +03:00
int i ;
struct scsi_vpd * vpd_buf ;
2014-03-15 12:51:49 +04:00
2016-04-01 09:57:36 +03:00
if ( ! scsi_device_supports_vpd ( sdev ) )
2014-03-15 12:51:49 +04:00
return ;
2016-04-01 09:57:36 +03:00
2014-03-15 12:51:49 +04:00
/* Ask for all the pages supported by this device */
2017-08-29 18:50:13 +03:00
vpd_buf = scsi_get_vpd_buf ( sdev , 0 ) ;
2017-08-29 18:50:12 +03:00
if ( ! vpd_buf )
2014-03-15 12:51:49 +04:00
return ;
2017-08-29 18:50:13 +03:00
for ( i = 4 ; i < vpd_buf - > len ; i + + ) {
2019-09-26 19:22:17 +03:00
if ( vpd_buf - > data [ i ] = = 0x0 )
scsi_update_vpd_page ( sdev , 0x0 , & sdev - > vpd_pg0 ) ;
2017-08-29 18:50:13 +03:00
if ( vpd_buf - > data [ i ] = = 0x80 )
scsi_update_vpd_page ( sdev , 0x80 , & sdev - > vpd_pg80 ) ;
if ( vpd_buf - > data [ i ] = = 0x83 )
scsi_update_vpd_page ( sdev , 0x83 , & sdev - > vpd_pg83 ) ;
2019-09-26 19:22:17 +03:00
if ( vpd_buf - > data [ i ] = = 0x89 )
scsi_update_vpd_page ( sdev , 0x89 , & sdev - > vpd_pg89 ) ;
2014-03-15 12:51:49 +04:00
}
kfree ( vpd_buf ) ;
}
2012-09-18 20:19:30 +04:00
/**
* scsi_report_opcode - Find out if a given command opcode is supported
* @ sdev : scsi device to query
* @ buffer : scratch buffer ( must be at least 20 bytes long )
* @ len : length of buffer
* @ opcode : opcode for command to look up
*
* Uses the REPORT SUPPORTED OPERATION CODES to look up the given
2013-06-07 06:15:55 +04:00
* opcode . Returns - EINVAL if RSOC fails , 0 if the command opcode is
* unsupported and 1 if the device claims to support the command .
2012-09-18 20:19:30 +04:00
*/
int scsi_report_opcode ( struct scsi_device * sdev , unsigned char * buffer ,
unsigned int len , unsigned char opcode )
{
unsigned char cmd [ 16 ] ;
struct scsi_sense_hdr sshdr ;
int result ;
if ( sdev - > no_report_opcodes | | sdev - > scsi_level < SCSI_SPC_3 )
2013-06-07 06:15:55 +04:00
return - EINVAL ;
2012-09-18 20:19:30 +04:00
memset ( cmd , 0 , 16 ) ;
cmd [ 0 ] = MAINTENANCE_IN ;
cmd [ 1 ] = MI_REPORT_SUPPORTED_OPERATION_CODES ;
cmd [ 2 ] = 1 ; /* One command format */
cmd [ 3 ] = opcode ;
put_unaligned_be32 ( len , & cmd [ 6 ] ) ;
memset ( buffer , 0 , len ) ;
result = scsi_execute_req ( sdev , cmd , DMA_FROM_DEVICE , buffer , len ,
& sshdr , 30 * HZ , 3 , NULL ) ;
2021-04-27 11:30:12 +03:00
if ( result < 0 )
return result ;
2012-09-18 20:19:30 +04:00
if ( result & & scsi_sense_valid ( & sshdr ) & &
sshdr . sense_key = = ILLEGAL_REQUEST & &
( sshdr . asc = = 0x20 | | sshdr . asc = = 0x24 ) & & sshdr . ascq = = 0x00 )
2013-06-07 06:15:55 +04:00
return - EINVAL ;
2012-09-18 20:19:30 +04:00
if ( ( buffer [ 1 ] & 3 ) = = 3 ) /* Command supported */
return 1 ;
return 0 ;
}
EXPORT_SYMBOL ( scsi_report_opcode ) ;
2005-04-17 02:20:36 +04:00
/**
2007-11-03 21:30:39 +03:00
* scsi_device_get - get an additional reference to a scsi_device
2005-04-17 02:20:36 +04:00
* @ sdev : device to get a reference to
*
2007-11-03 21:30:39 +03:00
* Description : Gets a reference to the scsi_device and increments the use count
2005-04-17 02:20:36 +04:00
* of the underlying LLDD module . You must hold host_lock of the
* parent Scsi_Host or already have a reference when calling this .
2015-02-02 16:01:26 +03:00
*
* This will fail if a device is deleted or cancelled , or when the LLD module
* is in the process of being unloaded .
2005-04-17 02:20:36 +04:00
*/
int scsi_device_get ( struct scsi_device * sdev )
{
2015-02-02 16:01:26 +03:00
if ( sdev - > sdev_state = = SDEV_DEL | | sdev - > sdev_state = = SDEV_CANCEL )
goto fail ;
2005-04-17 02:20:36 +04:00
if ( ! get_device ( & sdev - > sdev_gendev ) )
2015-02-02 16:01:26 +03:00
goto fail ;
if ( ! try_module_get ( sdev - > host - > hostt - > module ) )
goto fail_put_device ;
2005-04-17 02:20:36 +04:00
return 0 ;
2015-02-02 16:01:26 +03:00
fail_put_device :
put_device ( & sdev - > sdev_gendev ) ;
fail :
return - ENXIO ;
2005-04-17 02:20:36 +04:00
}
EXPORT_SYMBOL ( scsi_device_get ) ;
/**
* scsi_device_put - release a reference to a scsi_device
* @ sdev : device to release a reference on .
*
2007-11-03 21:30:39 +03:00
* Description : Release a reference to the scsi_device and decrements the use
* count of the underlying LLDD module . The device is freed once the last
2005-04-17 02:20:36 +04:00
* user vanishes .
*/
void scsi_device_put ( struct scsi_device * sdev )
{
2021-10-08 08:01:18 +03:00
struct module * mod = sdev - > host - > hostt - > module ;
2005-04-17 02:20:36 +04:00
put_device ( & sdev - > sdev_gendev ) ;
2021-10-08 08:01:18 +03:00
module_put ( mod ) ;
2005-04-17 02:20:36 +04:00
}
EXPORT_SYMBOL ( scsi_device_put ) ;
2007-11-03 21:30:39 +03:00
/* helper for shost_for_each_device, see that for documentation */
2005-04-17 02:20:36 +04:00
struct scsi_device * __scsi_iterate_devices ( struct Scsi_Host * shost ,
struct scsi_device * prev )
{
struct list_head * list = ( prev ? & prev - > siblings : & shost - > __devices ) ;
struct scsi_device * next = NULL ;
unsigned long flags ;
spin_lock_irqsave ( shost - > host_lock , flags ) ;
while ( list - > next ! = & shost - > __devices ) {
next = list_entry ( list - > next , struct scsi_device , siblings ) ;
/* skip devices that we can't get a reference to */
if ( ! scsi_device_get ( next ) )
break ;
next = NULL ;
list = list - > next ;
}
spin_unlock_irqrestore ( shost - > host_lock , flags ) ;
if ( prev )
scsi_device_put ( prev ) ;
return next ;
}
EXPORT_SYMBOL ( __scsi_iterate_devices ) ;
/**
* starget_for_each_device - helper to walk all devices of a target
* @ starget : target whose devices we want to iterate over .
2007-11-03 21:30:39 +03:00
* @ data : Opaque passed to each function call .
* @ fn : Function to call on each device
2005-04-17 02:20:36 +04:00
*
2007-12-11 02:49:31 +03:00
* This traverses over each device of @ starget . The devices have
2005-04-17 02:20:36 +04:00
* a reference that must be released by scsi_host_put when breaking
* out of the loop .
*/
2007-12-11 02:49:31 +03:00
void starget_for_each_device ( struct scsi_target * starget , void * data ,
2005-04-17 02:20:36 +04:00
void ( * fn ) ( struct scsi_device * , void * ) )
{
struct Scsi_Host * shost = dev_to_shost ( starget - > dev . parent ) ;
struct scsi_device * sdev ;
shost_for_each_device ( sdev , shost ) {
if ( ( sdev - > channel = = starget - > channel ) & &
( sdev - > id = = starget - > id ) )
fn ( sdev , data ) ;
}
}
EXPORT_SYMBOL ( starget_for_each_device ) ;
2007-12-11 02:49:31 +03:00
/**
2008-02-04 02:06:36 +03:00
* __starget_for_each_device - helper to walk all devices of a target ( UNLOCKED )
2007-12-11 02:49:31 +03:00
* @ starget : target whose devices we want to iterate over .
2008-02-04 02:06:36 +03:00
* @ data : parameter for callback @ fn ( )
* @ fn : callback function that is invoked for each device
2007-12-11 02:49:31 +03:00
*
* This traverses over each device of @ starget . It does _not_
* take a reference on the scsi_device , so the whole loop must be
* protected by shost - > host_lock .
*
* Note : The only reason why drivers would want to use this is because
* they need to access the device list in irq context . Otherwise you
* really want to use starget_for_each_device instead .
* */
void __starget_for_each_device ( struct scsi_target * starget , void * data ,
void ( * fn ) ( struct scsi_device * , void * ) )
{
struct Scsi_Host * shost = dev_to_shost ( starget - > dev . parent ) ;
struct scsi_device * sdev ;
__shost_for_each_device ( sdev , shost ) {
if ( ( sdev - > channel = = starget - > channel ) & &
( sdev - > id = = starget - > id ) )
fn ( sdev , data ) ;
}
}
EXPORT_SYMBOL ( __starget_for_each_device ) ;
2005-04-17 02:20:36 +04:00
/**
* __scsi_device_lookup_by_target - find a device given the target ( UNLOCKED )
* @ starget : SCSI target pointer
* @ lun : SCSI Logical Unit Number
*
2007-11-03 21:30:39 +03:00
* Description : Looks up the scsi_device with the specified @ lun for a given
* @ starget . The returned scsi_device does not have an additional
2005-04-17 02:20:36 +04:00
* reference . You must hold the host ' s host_lock over this call and
2009-01-13 18:50:37 +03:00
* any access to the returned scsi_device . A scsi_device in state
* SDEV_DEL is skipped .
2005-04-17 02:20:36 +04:00
*
2007-11-16 02:42:30 +03:00
* Note : The only reason why drivers should use this is because
2007-11-03 21:30:39 +03:00
* they need to access the device list in irq context . Otherwise you
2005-04-17 02:20:36 +04:00
* really want to use scsi_device_lookup_by_target instead .
* */
struct scsi_device * __scsi_device_lookup_by_target ( struct scsi_target * starget ,
2014-06-25 17:27:36 +04:00
u64 lun )
2005-04-17 02:20:36 +04:00
{
struct scsi_device * sdev ;
list_for_each_entry ( sdev , & starget - > devices , same_target_siblings ) {
2009-01-13 18:50:37 +03:00
if ( sdev - > sdev_state = = SDEV_DEL )
continue ;
2005-04-17 02:20:36 +04:00
if ( sdev - > lun = = lun )
return sdev ;
}
return NULL ;
}
EXPORT_SYMBOL ( __scsi_device_lookup_by_target ) ;
/**
* scsi_device_lookup_by_target - find a device given the target
* @ starget : SCSI target pointer
* @ lun : SCSI Logical Unit Number
*
2009-04-27 22:54:22 +04:00
* Description : Looks up the scsi_device with the specified @ lun for a given
* @ starget . The returned scsi_device has an additional reference that
2007-11-03 21:30:39 +03:00
* needs to be released with scsi_device_put once you ' re done with it .
2005-04-17 02:20:36 +04:00
* */
struct scsi_device * scsi_device_lookup_by_target ( struct scsi_target * starget ,
2014-06-25 17:27:36 +04:00
u64 lun )
2005-04-17 02:20:36 +04:00
{
struct scsi_device * sdev ;
struct Scsi_Host * shost = dev_to_shost ( starget - > dev . parent ) ;
unsigned long flags ;
spin_lock_irqsave ( shost - > host_lock , flags ) ;
sdev = __scsi_device_lookup_by_target ( starget , lun ) ;
if ( sdev & & scsi_device_get ( sdev ) )
sdev = NULL ;
spin_unlock_irqrestore ( shost - > host_lock , flags ) ;
return sdev ;
}
EXPORT_SYMBOL ( scsi_device_lookup_by_target ) ;
/**
2007-11-03 21:30:39 +03:00
* __scsi_device_lookup - find a device given the host ( UNLOCKED )
2005-04-17 02:20:36 +04:00
* @ shost : SCSI host pointer
* @ channel : SCSI channel ( zero if only one channel )
2007-11-03 21:30:39 +03:00
* @ id : SCSI target number ( physical unit number )
2005-04-17 02:20:36 +04:00
* @ lun : SCSI Logical Unit Number
*
2007-11-03 21:30:39 +03:00
* Description : Looks up the scsi_device with the specified @ channel , @ id , @ lun
* for a given host . The returned scsi_device does not have an additional
* reference . You must hold the host ' s host_lock over this call and any access
* to the returned scsi_device .
2005-04-17 02:20:36 +04:00
*
* Note : The only reason why drivers would want to use this is because
2007-11-03 21:30:39 +03:00
* they need to access the device list in irq context . Otherwise you
2005-04-17 02:20:36 +04:00
* really want to use scsi_device_lookup instead .
* */
struct scsi_device * __scsi_device_lookup ( struct Scsi_Host * shost ,
2014-06-25 17:27:36 +04:00
uint channel , uint id , u64 lun )
2005-04-17 02:20:36 +04:00
{
struct scsi_device * sdev ;
list_for_each_entry ( sdev , & shost - > __devices , siblings ) {
2017-04-28 12:43:04 +03:00
if ( sdev - > sdev_state = = SDEV_DEL )
continue ;
2005-04-17 02:20:36 +04:00
if ( sdev - > channel = = channel & & sdev - > id = = id & &
sdev - > lun = = lun )
return sdev ;
}
return NULL ;
}
EXPORT_SYMBOL ( __scsi_device_lookup ) ;
/**
* scsi_device_lookup - find a device given the host
* @ shost : SCSI host pointer
* @ channel : SCSI channel ( zero if only one channel )
* @ id : SCSI target number ( physical unit number )
* @ lun : SCSI Logical Unit Number
*
2007-11-03 21:30:39 +03:00
* Description : Looks up the scsi_device with the specified @ channel , @ id , @ lun
* for a given host . The returned scsi_device has an additional reference that
* needs to be released with scsi_device_put once you ' re done with it .
2005-04-17 02:20:36 +04:00
* */
struct scsi_device * scsi_device_lookup ( struct Scsi_Host * shost ,
2014-06-25 17:27:36 +04:00
uint channel , uint id , u64 lun )
2005-04-17 02:20:36 +04:00
{
struct scsi_device * sdev ;
unsigned long flags ;
spin_lock_irqsave ( shost - > host_lock , flags ) ;
sdev = __scsi_device_lookup ( shost , channel , id , lun ) ;
if ( sdev & & scsi_device_get ( sdev ) )
sdev = NULL ;
spin_unlock_irqrestore ( shost - > host_lock , flags ) ;
return sdev ;
}
EXPORT_SYMBOL ( scsi_device_lookup ) ;
MODULE_DESCRIPTION ( " SCSI core " ) ;
MODULE_LICENSE ( " GPL " ) ;
module_param ( scsi_logging_level , int , S_IRUGO | S_IWUSR ) ;
MODULE_PARM_DESC ( scsi_logging_level , " a bit mask of logging levels " ) ;
static int __init init_scsi ( void )
{
2006-09-30 22:31:11 +04:00
int error ;
2005-04-17 02:20:36 +04:00
error = scsi_init_procfs ( ) ;
if ( error )
goto cleanup_queue ;
error = scsi_init_devinfo ( ) ;
if ( error )
goto cleanup_procfs ;
error = scsi_init_hosts ( ) ;
if ( error )
goto cleanup_devlist ;
error = scsi_init_sysctl ( ) ;
if ( error )
goto cleanup_hosts ;
error = scsi_sysfs_register ( ) ;
if ( error )
goto cleanup_sysctl ;
[SCSI] SCSI and FC Transport: add netlink support for posting of transport events
This patch formally adds support for the posting of FC events via netlink.
It is a followup to the original RFC at:
http://marc.theaimsgroup.com/?l=linux-scsi&m=114530667923464&w=2
and the initial posting at:
http://marc.theaimsgroup.com/?l=linux-scsi&m=115507374832500&w=2
The patch has been updated to optimize the send path, per the discussions
in the initial posting.
Per discussions at the Storage Summit and at OLS, we are to use netlink for
async events from transports. Also per discussions, to avoid a netlink
protocol per transport, I've create a single NETLINK_SCSITRANSPORT protocol,
which can then be used by all transports.
This patch:
- Creates new files scsi_netlink.c and scsi_netlink.h, which contains the
single and shared definitions for the SCSI Transport. It is tied into the
base SCSI subsystem intialization.
Contains a single interface routine, scsi_send_transport_event(), for a
transport to send an event (via multicast to a protocol specific group).
- Creates a new scsi_netlink_fc.h file, which contains the FC netlink event
messages
- Adds 3 new routines to the fc transport:
fc_get_event_number() - to get a FC event #
fc_host_post_event() - to send a simple FC event (32 bits of data)
fc_host_post_vendor_event() - to send a Vendor unique event, with
arbitrary amounts of data.
Note: the separation of event number allows for a LLD to send a standard
event, followed by vendor-specific data for the event.
Note: This patch assumes 2 prior fc transport patches have been installed:
http://marc.theaimsgroup.com/?l=linux-scsi&m=115555807316329&w=2
http://marc.theaimsgroup.com/?l=linux-scsi&m=115581614930261&w=2
Sorry - next time I'll do something like making these individual
patches of the same posting when I know they'll be posted closely
together.
Signed-off-by: James Smart <James.Smart@emulex.com>
Tidy up configuration not to make SCSI always select NET
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-19 01:30:09 +04:00
scsi_netlink_init ( ) ;
2005-04-17 02:20:36 +04:00
printk ( KERN_NOTICE " SCSI subsystem initialized \n " ) ;
return 0 ;
cleanup_sysctl :
scsi_exit_sysctl ( ) ;
cleanup_hosts :
scsi_exit_hosts ( ) ;
cleanup_devlist :
scsi_exit_devinfo ( ) ;
cleanup_procfs :
scsi_exit_procfs ( ) ;
cleanup_queue :
scsi_exit_queue ( ) ;
printk ( KERN_ERR " SCSI subsystem failed to initialize, error = %d \n " ,
- error ) ;
return error ;
}
static void __exit exit_scsi ( void )
{
[SCSI] SCSI and FC Transport: add netlink support for posting of transport events
This patch formally adds support for the posting of FC events via netlink.
It is a followup to the original RFC at:
http://marc.theaimsgroup.com/?l=linux-scsi&m=114530667923464&w=2
and the initial posting at:
http://marc.theaimsgroup.com/?l=linux-scsi&m=115507374832500&w=2
The patch has been updated to optimize the send path, per the discussions
in the initial posting.
Per discussions at the Storage Summit and at OLS, we are to use netlink for
async events from transports. Also per discussions, to avoid a netlink
protocol per transport, I've create a single NETLINK_SCSITRANSPORT protocol,
which can then be used by all transports.
This patch:
- Creates new files scsi_netlink.c and scsi_netlink.h, which contains the
single and shared definitions for the SCSI Transport. It is tied into the
base SCSI subsystem intialization.
Contains a single interface routine, scsi_send_transport_event(), for a
transport to send an event (via multicast to a protocol specific group).
- Creates a new scsi_netlink_fc.h file, which contains the FC netlink event
messages
- Adds 3 new routines to the fc transport:
fc_get_event_number() - to get a FC event #
fc_host_post_event() - to send a simple FC event (32 bits of data)
fc_host_post_vendor_event() - to send a Vendor unique event, with
arbitrary amounts of data.
Note: the separation of event number allows for a LLD to send a standard
event, followed by vendor-specific data for the event.
Note: This patch assumes 2 prior fc transport patches have been installed:
http://marc.theaimsgroup.com/?l=linux-scsi&m=115555807316329&w=2
http://marc.theaimsgroup.com/?l=linux-scsi&m=115581614930261&w=2
Sorry - next time I'll do something like making these individual
patches of the same posting when I know they'll be posted closely
together.
Signed-off-by: James Smart <James.Smart@emulex.com>
Tidy up configuration not to make SCSI always select NET
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-19 01:30:09 +04:00
scsi_netlink_exit ( ) ;
2005-04-17 02:20:36 +04:00
scsi_sysfs_unregister ( ) ;
scsi_exit_sysctl ( ) ;
scsi_exit_hosts ( ) ;
scsi_exit_devinfo ( ) ;
scsi_exit_procfs ( ) ;
scsi_exit_queue ( ) ;
}
subsys_initcall ( init_scsi ) ;
module_exit ( exit_scsi ) ;