Commit graph

20 commits

Author SHA1 Message Date
David Howells
7d12e780e0 IRQ: Maintain regs pointer globally rather than passing to IRQ handlers
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.

The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around.  On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).

Where appropriate, an arch may override the generic storage facility and do
something different with the variable.  On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.

Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions.  Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller.  A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.

I've build this code with allyesconfig for x86_64 and i386.  I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.

This will affect all archs.  Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:

	struct pt_regs *old_regs = set_irq_regs(regs);

And put the old one back at the end:

	set_irq_regs(old_regs);

Don't pass regs through to generic_handle_irq() or __do_IRQ().

In timer_interrupt(), this sort of change will be necessary:

	-	update_process_times(user_mode(regs));
	-	profile_tick(CPU_PROFILING, regs);
	+	update_process_times(user_mode(get_irq_regs()));
	+	profile_tick(CPU_PROFILING);

I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().

Some notes on the interrupt handling in the drivers:

 (*) input_dev() is now gone entirely.  The regs pointer is no longer stored in
     the input_dev struct.

 (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking.  It does
     something different depending on whether it's been supplied with a regs
     pointer or not.

 (*) Various IRQ handler function pointers have been moved to type
     irq_handler_t.

Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 15:10:12 +01:00
James Smart
c01f320879 [SCSI] lpfc 8.1.10 : Add support for dev_loss_tmo_callbk and fast_io_fail_tmo_callbk
Add support for new dev_loss_tmo callback
  Goodness is that it removes code for a parallel nodev timer that
  existed in the driver
Add support for the new fast_io_fail callback

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-04 21:25:21 -05:00
James Smart
a309a6b6e6 [SCSI] lpfc 8.1.7 : Fix failing firmware download due to mailbox delays needing to be longer
Fix failing firmware download due to mailbox delays needing to be longer.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-06 11:28:03 -05:00
James Smart
b4c026520f [SCSI] lpfc 8.1.7: Add lpfc_sli_flush_mbox_queue() function
Add lpfc_sli_flush_mbox_queue() function and use it in lpfc_offline() call
to avoid deadlock on thread block.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-09 10:48:38 -05:00
James Smart
4b0b91d461 [SCSI] lpfc 8.1.5 : Use asynchronous ABTS completion to speed up abort completions
Use asynchronous ABTS completion to speed up abort completions

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-19 20:48:41 -04:00
James Smart
9290831f00 [SCSI] lpfc 8.1.4 : Introduce lpfc_reset_barrier() function for resets on dual channel adapters
Introduce lpfc_reset_barrier() function for resets on dual channel adapters

Workaround for a hardware errata on dual channel asics. There is a
potential for the chip to lock up on a reset if a shared dma engine is in
use. The (ugly) work around requires a reset process which uses a mailbox
command to synchronize the independent channels prior to the reset to
avoid the issue. Unfortunately, the timing windows required to ensure this
workaround succeeds are very specific, meaning we can't release the cpu
during the barrier.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-12 08:56:58 -06:00
James Smart
fdcebe282f [SCSI] lpfc 8.1.4 : Fixed RSCN handling when a PLOGI is in retry
Fixed RSCN handling when a PLOGI is in retry.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-12 08:56:13 -06:00
James Smart
488d1469b3 [SCSI] lpfc 8.1.4 : Fix Discovery processing for NPorts that change their NPortId on the fly
Fix Discovery processing for NPorts that change their NPortId on the fly
due to a cable swap.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-12 08:54:29 -06:00
Jamie Wellnitz
41415862a2 [SCSI] lpfc 8.1.2: Add ERROR and WARM_START modes for diagnostic purposes.
Add ERROR and WARM_START modes for diagnostic purposes.

Signed-off-by: Jamie Wellnitz <Jamie.Wellnitz@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-28 19:00:36 -06:00
Jamie Wellnitz
7bb3b137ab [SCSI] lpfc 8.1.2: Handling of ELS commands RRQ, RPS, RPL and LIRR correctly
Handling of ELS commands RRQ, RPS, RPL and LIRR correctly

Signed-off-by: Jamie Wellnitz <Jamie.Wellnitz@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-28 18:52:50 -06:00
Jamie Wellnitz
3bbae37ab4 [SCSI] lpfc 8.1.2: Remove unused prototypes from lpfc_crtn.h
Remove unused prototypes from lpfc_crtn.h

Signed-off-by: Jamie Wellnitz <Jamie.Wellnitz@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-28 18:50:44 -06:00
James.Smart@Emulex.Com
875fbdfe9b [SCSI] lpfc 8.1.1 : Add polled-mode support
- Add functionality to run in polled mode only. Includes run time
  attribute to enable mode.
- Enable runtime writable hba settings for coallescing and delay parameters

Customers have requested a mode in the driver to run strictly polled.
This is generally to support an environment where the server is extremely
loaded and is looking to reclaim some cpu cycles from adapter interrupt
handling.

This patch adds a new "poll" attribute, and the following behavior:

if value is 0 (default):
  The driver uses the normal method for i/o completion. It uses the
  firmware feature of interrupt coalesing. The firmware allows a
  minimum number of i/o completions before an interrupt, or a maximum
  time delay between interrupts.  By default, the driver sets these
  to no delay (disabled) or 1 i/o - meaning coalescing is disabled.

  Attributes were provided to change the coalescing values, but it was
  a module-load time only and global across all adapters.
  This patch allows them to be writable on a per-adapter basis.

if value is 1 :
  Interrupts are left enabled, expecting that the user has tuned the
  interrupt coalescing values. When this setting is enabled, the driver
  will attempt to service completed i/o whenever new i/o is submitted
  to the adapter. If the coalescing values are large, and the i/o
  generation rate steady, an interrupt will be avoided by servicing
  completed i/o prior to the coalescing thresholds kicking in. However,
  if the i/o completion load is high enough or i/o generation slow, the
  coalescion values will ensure that completed i/o is serviced in a timely
  fashion.

if value is 3 :
  Turns off FCP i/o interrupts altogether. The coalescing values now have
  no effect. A new attribute "poll_tmo" (default 10ms) exists to set
  the polling interval for i/o completion. When this setting is enabled,
  the driver will attempt to service completed i/o and restart the
  interval timer whenever new i/o is submitted. This behavior allows for
  servicing of completed i/o sooner than the interval timer, but ensures
  that if no i/o is being issued, then the interval timer will kick in
  to service the outstanding i/o.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-12-13 18:34:14 -07:00
James.Smart@Emulex.Com
0bd4ca25ad [SCSI] lpfc: Fix eh_ return codes for commands
Return FAILED from eh_ routines if command(s) is(are) not completed

There were scenarios where we may have returned from the error
handlers prior to all affected commands being flushed to the midlayer.
Add changes to ensure this doesn't happen.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-29 10:31:48 -05:00
James.Smart@Emulex.Com
68876920f4 [SCSI] lpfc: Replace lpfc_sli_issue_iocb_wait_high_priority
Replace lpfc_sli_issue_iocb_wait_high_priority with lpfc_sli_issue_iocb_wait.

Simplify code paths, as there really wasn't a "priority"

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-29 10:29:09 -05:00
James Bottomley
604a3e3042 [SCSI] lpfc: Fix for "command completion for iotax x?? not found"
From: James Smart <James.Smart@emulex.com>

There were scenarios where the error handlers could reuse an iotag
value of an active io.  Remove all possibility of this by
pre-assigning iotag resources to command resources.

Signed-off-by: James Smart <James.Smart@emulex.com>

Rejections fixed up and
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-29 10:28:33 -05:00
James.Smart@Emulex.Com
21568f5387 [SCSI] lpfc: Remove RPI hash from the driver
Table was not providing a lot of value and injected a couple of
errors. Removed it and made functionality inline.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-29 10:22:50 -05:00
James.Smart@Emulex.Com
c44ce17374 [SCSI] lpfc: Update copyright notices
Update copyright notice text and include year 2005.

Add Copyright notice for Christoph Hellwig to several files: lpfc.h
lpfc_attr.c lpfc_els.c lpfc_hbadisc.c lpfc_init.c lpfc_mbox.c
lpfc_mem.c lpfc_nportdisc.c lpfc_scsi.c lpfc_sli.c

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-07-02 19:08:29 -05:00
James.Smart@Emulex.Com
5542134f61 [SCSI] lpfc: Remove $Id$ keyword strings.
Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-07-02 19:06:16 -05:00
James.Smart@Emulex.Com
5eb95af086 [SCSI] lpfc: Add completion handler to the abort iocbs
Add completion handler to the abort iocbs to close a hole where we
could reuse an iotag.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-07-02 18:57:22 -05:00
dea3101e0a lpfc: add Emulex FC driver version 8.0.28
From: 	James.Smart@Emulex.Com

Modified for kernel import and
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-04-18 13:50:53 -05:00