Aacraid (ARC) Linux driver release notes. Ascending (chronological) order
comments. Modifications performed by Mark_Salyzyn@adaptec.com and 
Achim_Leubner@adaptec.com if not otherwise noted. 
Document created 2003-05-15 at Version 0.9.10.

Streams support otc-main, avonpark and enzo

Version: 0.9.10

Version: 1.1.2

2003-05-15	Mark_Salyzyn@adaptec.com

Differences between 2.4.21-rc2-ac2 kernel and our 1.1.2 versioned driver,
changes as performed by Deanna Bonds, Bob Pasteur and Mark Salyzyn.

aachba.c:
	- If the state of a logical unit is hidden, then do not report. This
	  state is typically entered when a device is being cleared.
	- Added support for the Tallahassee project, where one channel is
	  dedicated to SCSI, and the other channel is dedicated to RAID.
	- Resolved some issues surrounding PAE support and IA64.
	- If the driver is a not a boot disk driver, then set the Removable
	  bit on the inquiry strings returned by the logical units to ensure
	  that any changes in the arrays will be acquired when the device is
	  re-attached.
	- mask the SRB status with 0x3F to deal with misbehaving devices.
	- Do not report DISKs to inquiry requests on the SCSI bus except if
	  the channel is designated as a SCSI only bus.
	- Propagate check conditions to the SCSI command result.
	- Add support for programmable timeouts to propagate down toe the
	  requests.
	- If we have pae mode enabled, right after we get the adapter
	  information and determine the pae mode capability, we enable the
	  system to issue 64 bit requests.
aacraid.h:
	- Had to drop from 512 commands to 100 commands because some versions
	  of the firmware would starve commands causing a timeout reaction
	  which lead to lost commands.
	- Added a global control variable for nondasd and paemode support.
	- Dealt with some 64 bit / 32 bit issues in list_head structures and
	  helper Macros, replacing them with our own more sensitive variants.
	- Differentiated virtual and physical references to the shared fib
	  allocations.
	- information structure not synchronized to firmware, needed to add
	  a clusterchannelmask.
	- Added definitions in support of the new configuration information
	  page bits in support of Tallahassee.
	- Changed to an allocated fib pool, rather than an array in the hba
	  structure as this affected the SCSI memory pool.
	- Added some AIF definitions to permit us to sniff for container
	  changes to permit a rescan to pick up new information or targets.
commctrl.c:
	- The fib reference was changed to a physical and a virtual address,
	  absorb the name changes.
	- The list_head structure handlers have been replaced with our own,
	  absorb the name changes.
	- The fib address reported in an AIF is a physical (32 bit) reference,
	  and not a virtual (possibly 64 bit) reference.
	- added the ioctl handling for sending a raw srb (FSACTL_SEND_RAW_SRB).
comminit.c:
	- Deal with IA64 issues.
	- Change to using the physical address (32 bit) for the AIF references.
	- The list_head structure handlers have been replaced with our own,
	  absorb the name changes.
	- Observed a memory leak, free up the queue resources should we fail
	  to initialize the adapter.
commsup.c:
	- The fib reference was changed to a physical and a virtual address,
	  absorb the name changes.
	- Instead of panicking the kernel when a fib allocation was available,
	  sleep until it is available.
	- Submitted fib pointers are physical (32 bit) rather than virtual
	  (possibly 64 bit) values.
	- producer and consumer indexes should be converted over to local
	  cpu endian before comparison.
	- aac_handle_aif now sniffs AIF events and takes plug and play action
	  for container changes.
	- The aif thread is set up to be a kernel thread, and not a user
	  thread. This permits us the ability to make plug and play calls
	  without prejudice.
	- Added instrumentation to the aif thread to confirm the plug and
	  play activity and as an aid to several other debug sessions.
	- Do not age an aif context based on the last received aif, but rather
	  the last poll.
dpcsup.c:
	- The fib reference was changed to a physical and a virtual address,
	  absorb the name changes.
	- Submitted fib pointers are physical (32 bit) rather than virtual
	  (possibly 64 bit) values.
linit.c:
	- Added paemode control.
	- Added various upcoming board products, and documented better the
	  existing board product ids. This includes SATA RAID products.
	- needed to take the io_request_lock during portions of initialization.
	- allocate the fib resource separately, rather than part of adapter
	  structure to aid in the precious SCSI resources.
	- cleanup of none dasd support options.
	- Added more details about the build date of the driver to the proc
	  information.
	- dropped a change that permitted 64 bit DMA resources to be generated
	  instead of through a 32 bit bounce buffer. (it was moved to aachba.c
	  where it can be turned on after we determine the adapter's
	  capabilities).
	- max_id, max_lun and max_channel parameters are set after the
	  adapter information has been picked up (the number of channels is
	  based on the product id table now).
sa.c:
	- Context of timeout handling was incorrect, only noticed in IA64
	  bit machines (due to lack of BIOS initialization).

Differences that need further investigation and could be viewed as regressions
and added after submission:

rx.c:
	- Dropped detection of failure to generate kernel command thread.
sa.c:
	- Dropped detection of failure to generate kernel command thread.

Version: 1.1.3

2003-07-01	Mark_Salyzyn@adaptec.com

aachba.c:
	- Added aac_get_container_name to permit override of array inquiry
	  string with the set name.

2003-07-08	Mark_Salyzyn@adaptec.com

aachba.c:
	- Return 0 (success) for unsupported commands, the check condition
	  should perform the necessary action of error handling.

2003-07-10	Mark_Salyzyn@adaptec.com

aachba.c:
	- The pass-through SCSI SCB command in PAE mode was getting the fib
	  size count wrong, by using the 32 bit command, then doing an (n-1)
	  times the size of the 64 bit scatter gather. Resolution was to
	  subtract the 32 bit scatter gather, then do an n times the 64 scatter
	  gather size.
	- Only go into PAE mode if more than 4MB of memory in the system.

2003-07-10	Mark_Salyzyn@adaptec.com

linit.c:
	- Added `Family' product codes and reordered the product discovery code
	  to produce devices in PCI order rather than in product order.
	  Dell, Legend and Adaptec Families were produced with the assumption
	  of 2 available busses.

2003-07-24	Mark_Salyzyn@adaptec.com

linit.c:
	- Added Bearcat (6 ch SATA) and a commented entry for Lancer where
	  future workarounds may be necessary due to hardware constraints.
	- Set highmem_io (for kernels of 2.4.18 and above).

aachba.c:
	- Set highmem_io (for kernels of 2.4.18 and above; and when the
	  adapter is guaranteed to handle the possible address ranges it
	  will be provided).

Version: 1.1.4:

2003-07-28	Mark_Salyzyn@adaptec.com

aacraid.h+common/include/fsaioctl.h+aachba.c
	- Added the FSACTL_REGISTER_FIB_SEND function to the ioctl. This ioctl
	  is *not* a user accessible ioctl, meant only for driver use to permit
	  stacking a filter driver just ahead of the hardware layer. The call
	  to register is:

		typedef void (*fib_callback)(void *ctxt, struct fib *fibctx);
		typedef struct {
			int (*fib_send)(u16 command,
					struct fib * context,
					unsigned long fib_size,
					int priority,
					int wait,
					int reply
					fib_callback callback,
					void * ctxt);
		} fib_send_t;
		. . .
		fib_send_t original;
		int dummy_fib_send (u16 command,
				    struct fib * context,
				    unsigned long fib_size,
				    int priority,
				    int wait,
				    int reply
				    fib_callback callback,
				    void * ctxt)
		{
			return (*original->fib_send)(command, context, fib_size, priority, wait, reply, callback, ctxt);
		}
		. . .
		Scsi_Host_Template * host;
		Scsi_Device * adapter;
		original->fib_send = dummy_fib_send;
		host->ioctl(adapter, FSACTL_REGISTER_FIB_SEND, &original);

	  Return value from the ioctl include ENOTTY (not supported), EINVAL
	  (invalid argument pointer) and EBUSY (another function already
	  registered) and the original fib_send function is returned in the
	  ioctl argument structure. A NULL value for the fib_send member of the
	  structure deregisters the filter driver. The fib_callback function is
	  issued at interrupt priority and should follow all the constraints of
	  interrupt operation. It is the responsibility of the registered
	  fib_send function to ensure that the original fib_callback function
	  is called with the ctxt value when completing the command (this
	  subtlety is lost in the above dummy function).

2003-07-28	Mark_Salyzyn@adaptec.com

linit.c:
	- Added Kernel, Firmware and BIOS revision and build number to proc
	  information.
	- Added board serial number to proc information.

aachba.c:
	- Do not set removable bit in the inquiry command, the aif delivery
	  of array status change will handle the reasons for the removable
	  bit (capacity change and differences in the partition table). Some
	  customers take issue with the fact our arrays appear as removable.

commctrl.c:
	- Reported driver version and build number instead of Firmware version
	  and build number for the Miniport Version Check ioctl. ADPmp57715

2003-08-06	Mark_Salyzyn@adaptec.com and a cast of thousands

all files:
	- Added appropriate ifdefs, or merged in additions, in support of the
	  2.6.0-test2 kernels as follows:

Makefile:
	- Added ifdefs for 2.4 and 2.6 kernels so we can use a common Makefile
	  for both kernel build environments.

aachba.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- define aac_spin_* macros to differentiate between lock requirements
	  in 2.5+ and 2.4 kernels.
	- Use the SCSI layers definitions of the SCSI commands, rather than
	  our own internal SS_* manifests.
	- Define SCSICMD_TO_* macros to acquire the SCSI target host, channel,
	  id and lun.
	- Use the 2.6 SAM_* status codes for return, the 2.4 system will
	  redefine the SAM_* codes to 2.4 variants.
	- Change to devname instead of devno when referencing devices to
	  simplify conversions.
	- MAXIMUM_NUM_CONTAINERS references were +/- 1 in comparisons, made
	  this value a `number' rather than a mix of `number' and `limit'.
	- Resolved `Cast of pointer from integer of different size' by
	  (void *)(ulong)dma_addr_t.
	- Change to `id' rather than `target' to match SCSI subsystem
	  references name for consistency.

aacraid.h:
	- MAXIMUM_NUM_CONTAINERS references were +/- 1 in comparisons, made
	  this value a `number' rather than a mix of `number' and `limit'.
	- Removed AAC_MAX_TARGET, as it is no longer used.
	- Added CONTAINER_TO_* macros to simplify references.
	- Change to `id' rather than `target' to match SCSI subsystem
	  references name for consistency.
	- Change to devname instead of devno when referencing devices.
	- Use cap_to_cyls inline to handle 64 bit calculation correctly.

commctrl.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- Change to `id' rather than `target' to match SCSI subsystem
	  references name for consistency.

comminit.c:
	- use linux/blkdev.h in 2.5+ kernels.

commsup.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- Moved CONTAINER_TO_* macros to aacraid.h to simplify references.
	- Device Discovery loops are different for 2.4 and 2.5+ kernels,
	  use list_for_each_entry siblings instead of host_queue loop.
	- daemonize adds the process name as a parameter, and requires
	  SIGKILL to be enabled to permit kernel shutdown in 2.5+ kernels.

dpcsup.c:
	- use linux/blkdev.h in 2.5+ kernels.

linit.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- added aacids to provide a table hint for installers.
	- changed over to utilize ANSI structure initialization.
	- aac_biosparm and aac_procinfo change parameters in 2.5+ kernels.
	- aac_slave_configure replaces aac_queuedepth in 2.5+ kernels.
	- detect no longer needs to unlock io_request_lock to do it's duty
	  in 2.5+ kernels.
	- use SCSI_set_device in 2.5+ kernels rather than scsi_set_pci_device.
	- Change to devname instead of devno when referencing devices to
	  simplify conversions.
	- Use MAXIMUM_NUM_CONTAINERS rather than AAC_MAX_TARGET
	- Use cap_to_cyls inline to handle 64 bit calculation correctly in
	  aac_biosparm.
	- Use minor in 2.5+ kernels instead of MINOR macro.

rx.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- interrupts now return irqreturn_t.

sa.c:
	- use linux/blkdev.h in 2.5+ kernels.
	- interrupts now return irqreturn_t.

2003-08-15	Mark_Salyzyn@adaptec.com

install.sh:
	- increased range of kernel version reports in the magic file to 30.

2003-08-19	Mark_Salyzyn@adaptec.com & ijohns@elipsan.com

aachba.c:
	- status_byte in the result is shifted down by one.
	- set_sense spoof was not immediately followed by a copy of the check
	  condition results into the SCSI command.

2003-08-20	Mark_Salyzyn@adaptec.com, Scott_Long@adaptec.com & Alan Cox

commctrl.c:
	- The raw SCSI SCB ioctl command in PAE mode was getting the fib
	  size count wrong, by using the 32 bit command, then doing an (n-1)
	  times the size of the 64 bit scatter gather. Resolution was to
	  subtract the 32 bit scatter gather, then do an n times the 64 scatter
	  gather size.

aacraid.h:
	- Added definition of CT_FLUSH_CACHE command and structures.
	- Added AAC_QUIRK_31BIT for ROMB based adapters.

linit.c:
	- Added AAC_QUIRK_31BIT for ROMB based adapters.
	- Check return from scsi_register.

aachba.c:
	- Added support for issuing CT_FLUSH_CACHE command when the SCSI
	  SYNCHRONIZE command is issued to a container.
	- Restored mask after adding AAC_QUIRK_31BIT for ROMB based adapters.

2003-08-21	Mark_Salyzyn@adaptec.com

aachba.c:
	- Changed aac_get_container_name to be a none-blocking function,
	  completing the incoming scsicmd with the adapter response.

2003-08-26	Mark_Salyzyn@adaptec.com

commsup.c + aacraid.h:
	- Altered handling of AIF messages from Firmware to differentiate
	  events in a finer grained manner.

2003-08-29	Mark_Salyzyn@adaptec.com

aachba.c + aacraid.h
	- Driver too noisy, undefined AAC_DETAILD_STATUS_INFO and incorporated
	  check condition report into the AAC_DETAILED_STATUS_INFO ifdef.

2003-09-03	Mark_Salyzyn@adaptec.com

aachba.c:
	- Check if the device is in use and report that as a locked device
	  to both the FSACTL_QUERY_DISK and FSACTL_DELETE_ARRAY ioctls.
	- unlock/lock around probe_container as this is a blocking function.
	  This change addresses a deadlock issue that surfaced in SMP only
	  environments.

Version: 1.1.4-2172

2003-09-04	Mark_Salyzyn@adaptec.com

commsup.c:
	- References to the Status Job update structure were at incorrect
	  offsets causing incorrect operation during an Array Clear with
	  regards to plug and play actions.

Version: 1.1.4-2177

2003-09-05	Mark_Salyzyn@adaptec.com

aachba.c:
	- Cleanup request from the SCSI list maintainers.
	- Dropped use of SCSICMD_TO_CHANNEL & friends since
	  scsicmd->device->channel is available in all versions of the
	  operating system.
	- Removed deprecated code and/or comments related to deprecation.
	- include <linux/blkdev.h> works in all versions of the operating
	  system.

2003-09-09	Mark_Salyzyn@adaptec.com

aacraid.h:
	- NUM_FIBs should be 64 larger (AIFS) larger than the NUM_IO_FIBS.

commsup.c:
	- efficiency improved if we hold on to the aac_queue variable, aims
	  towards better code compliance and consistency.

2003-09-15	Mark_Salyzyn@adaptec.com

rkt.c:
	- Copy if rx.c with rx = rkt

aacraid.h:
	- Added definition for rkt interface structures, copy of rx, but a
	  larger reserved region.

linit.c:
	- Added product code for ROC (Lancer/Rocket) U320 two channel, use rkt
	  interface.

2003-09-16	Mark_Salyzyn@adaptec.com

linit.c:
	- Show Adapter vendor and model in proc information.

Version: 1.1.4-2185

2003-09-16	Mark_Salyzyn@adaptec.com

aacraid.h:
	- Added definition of nblank() to assist us in determining if
	  dprintk(x) is defined as a blank definition to enable us to ifdef
	  debug code that ends up calling only dprintk functions.

commsup.c:
	- Ignore events that refer to containers > MAXIMUM_NUM_CONTAINERS
	- include <linux/blkdev.h> works in all versions of the operating
	  system.

linit.c:
	- print more details about outstanding commands when a SCSI hang
	  occurs (first use of nblank() macro just defined in aacraid.h)

2003-09-19	Mark_Salyzyn@adaptec.com & Mark Haverkamp <markh@osdi.org>

commsup.c & aachba.c:
	- valid flag has added support for a value of 2, which means target
	  is still valid, but needs a probe_container.

commsup.c:
	- fib_alloc should not go to sleep, but return NULL if there are no
	  available entries in the pool.

dpcsup.c:
	- print a message if the fib kmalloc fails when forwarding AIFs

comminit.c:
	- check fib_alloc return, and report -ENOMEM should the pool be
	  empty.

aachba.c:
	- Check value of scsicmd->scsi_done in aac_io_done as we can get
	  errant firmware which returns commands twice (no released firmware
	  does this, this is a driver hardening issue only).
	- When a fib_alloc fails, return -1 to SCSI layer. Formerly, we would
	  send the command with DID_ERROR.

Version: 1.1.4-2192

2003-09-25	Mark_Salyzyn@adaptec.com

linit.c:
	- Moved debug variables into block to reduce impact on none-debug
	  environments.

dpcsup.c + commsup.c:
	- Use the fib pool instead of a kmalloc to allocate a fib for the
	  processing of an AIF.

install.sh:
	- Install driver into any forgotten /lib/modules directories.

2003-09-26	Mark_Salyzyn@adaptec.com

commctrl.c + aacraid.h:
	- AMD-64 and IA-64 management applications will fail, need to change
	  fibctx to a 32 bit unique value.

Version: 1.1.4-2194

2003-09-29	Mark_Salyzyn@adaptec.com & Mark Haverkamp <markh@osdi.org>

aachba.c:
	- use linux/blkdev.h for all variants on Linux.
	- hold on to the host pointer in aac_io_done, because it's reference
	  in the device and scsicmd can go away after scsi_done is called.
	- check return value of pci_set_dma_mask.

commctrl.c:
	- use linux/blkdev.h for all variants on Linux.

comminit.c:
	- use linux/blkdev.h for all variants on Linux.

commsup.c:
	- use linux/blkdev.h for all variants on Linux.
	- drop linux/smp_lock.h include as it was added in a debug test from
	  some time ago.
	- Added current 2.6 kernel support routines for rescanning.

dpcsup.c:
	- use linux/blkdev.h for all variants on Linux.

linit.c:
	- use linux/blkdev.h for all variants on Linux.
	- check return value of pci_set_dma_mask.
	- template->present is no longer relevant in 2.6 based kernels.

rx.c:
	- use linux/blkdev.h for all variants on Linux.

sa.c:
	- use linux/blkdev.h for all variants on Linux.

rkt.c:
	- use linux/blkdev.h for all variants on Linux.

2003-10-01	Mark_Salyzyn@adaptec.com

commsup.c:
	- needed a fib_dealloc call ahead of the fib_free call added when
	  we moved over to the fib pool to handle the AIFs.

dpcsup.c:
	- need to use the `local' fibctx so that the AIF command can be
	  acknowledged.

commctrl.c:
	- return error status from the send_fib function in ioctl_send_fib.

2003-10-07	Mark_Salyzyn@adaptec.com

aachba.c + linit.c:
	- serial number contains the cookie (fafa0001) that is at index 1
	  of the serial number element. Only show the serial number which
	  is at index 0.

linit.c:
	- Added registration to receive 32 bit ioctls.

commsup.c + dpcsup.c + aacraid.h:
	- Dropped code to acquire AIF's from the general FIB pool, it was a
	  fool's errand. However, we kept the code that limits the AIF's
	  received and allocated to the AdapterFibsSize / sizeof(hw_fib).
	  The `last' AIF hw_fib is used to quickly acknowledge the entries,
	  and drop the results on the floor.

rx.c + rkt.c:
	- Cache the OIMR data in dev->OIMR, it looks remarkably like irq_mask,
	  which is really unused, but we can clean that up later.

2003-10-08	Matthew Wilcox <willy@debian.org>

aachba.c:
	- Use SCp.dma_handle instead of SCp.ptr for holding on to the physical
	  address of the allocated pci_map_single as part of the request.

compat.h:
	- define dma_handle to be ptr (in support of SCp.dma_handle change
	  above) for kernels that do not define this member.

2003-10-08	Christoph Hellwig <hch@infradead.org>

aachba.c:
	- drop use of scsi_to_pci_dma_dir() as it is a pass-through in all
	  versions of the kernel.

2003-10-09	Mark_Salyzyn@adaptec.com

linit.c:
	- When an Adapter Reset is requested, wait up to 60 seconds for all
	  outstanding commands to complete and report SUCCESS.

Version: 1.1.4-2221

2003-10-09	Mark_Salyzyn@adaptec.com

linit.c:
	- Waited for *all* commands to complete for *all* devices on the
	  controller when an Adapter Reset is requested.

Version: 1.1.4-2222

2003-10-10	Mark_Salyzyn@adaptec.com

aacraid.h + rx.c + rkt.c + sa.c + linit.c:
	- Added a aac_adapter_check_health, make sure the adapter is healthy
	  when performing and Adapter Reset request, report error codes.

aachba.c:
	- revert to use of scsi_to_pci_dma_dir() as it is not a pass-through in
	  all versions of the kernel.

linit.c:
	- SCSI_HAS_HOST_LOCK means that we should be working with releasing
	  host->lock or host->host_lock instead of io_request_lock surrounding
	  scsi_sleep.

aacraid.h:
	- Added definition for AAC_MAX_HOSTPHYSMEMPAGES

comminit.c:
	- Utilized AAC_MAX_HOSTPHYSMEMPAGES to limit the number of open DMA
	  4096 byte PAGES of memory requested by the operating system.

2003-10-16	Mark_Salyzyn@adaptec.com

install.sh:
	- Added support for x86_64 installs

aachba.c:
	- used SENSE KEYS from scsi.h rather than our own definitions.

2003-10-20	Xose Vazquez Perez <xose@wanadoo.es>

linit.c:
	- Added pci_ids for 0x10110046/0x90050365

Version: 1.1.4-2265

2003-10-23	Mark_Salyzyn@adaptec.com

linit.c:
	- no need to set template->present as this is done by the SCSI layer.

2003-10-24	Mark_Salyzyn@adaptec.com

install.sh
	- Added support for SuSE kernel determination for finer selection
	  of modules
	- If the kernel is compiled for athlon, use that instead of
	  /proc/cpuinfo
	- if /proc/cpuinfo is not present, don't show any errors during
	  install

2003-10-28	Mark_Salyzyn@adaptec.com

install.sh
	- The entire class of SuSE OS releases (sles7, sles8, suse7, suse8,
	  suse8.1, suse8.2, ul1, ul1-sp2a) place the driver module results into
	  /lib/modules/[kernel]/kernel/drivers/scsi/aacraid/aacraid.o. The
	  package places updates in ...//scsi/aacraid.o (note, one directory
	  up). The module selected for use in the mkinitrd is fed via a `find'
	  command which reports files in raw directory order which in the
	  reiser file system would be in the .../scsi directory, but for EXT2
	  since the file was added later, would prefer the previously placed
	  product in ../scsi/aacraid/aacraid.o. The fix is to have the driver
	  disk post-install remove the older .../scsi/aacraid directory.

2003-10-30	Mark_Salyzyn@adaptec.com

install.sh
	- For the installations to `extra' /lib/modules directories beyond
	  the boot set, take the processor clue from the postscript (-athlon,
	  -x86_64 or -ia64) rather than from /proc/cpuinfo.

Version: 1.1.4-2282
Version: 1.1.4-2292 (Debug)

2003-10-31	Mark_Salyzyn@adaptec.com

aacraid.h + aachba.c:
	- Added a nested count to the fsa_scsi_dev structure since some kernels
	  before 2.4.19 have troubles overflowing their stack when a device
	  goes offline. The issue is that the SCSI done call nests into sending
	  another queued command, which in turn spoofs a response back
	  indicating failure which in turn calls SCSI done. We limit the
	  nesting to 64 commands before we respond with a busy instead.

Version: 1.1.4-2296 (Debug)

linit.c & .version:
	- Versioning is defined by the structure:
	    struct {
		unsigned char dash; // Dash version number
		unsigned char type; // Type, 1=Devo, 2=Alpha, 3=Beta, 4=Release
		unsigned char minor;// Minor version minor
		unsigned char major;// Major version number
	    }
	  Adjusted version data to match this definition for generation and
	  support.

Version: 1.1.4-2299
Version: 1.1.4-2301
Version: 1.1.4-2302
Version: 1.1.4-2303

linit.c & aacraid.h:
	- Allow 64 bit apps to call GET_NEXT_ADAPTER_FIB ioctl directly,
	  promoting 32 bit apps when they call.

aachba.c & aacraid.h:
	- Set MAX_NESTED to 1, and improve code to reflect this simplicity.

install.sh:
	- Handle name change of products from *-athlon-athlon to *-athlon.
	- Warn the user if the initrd shrinks too much

Version: 1.1.4-2308

install.sh:
	- Add support for identifying 2.4.19-340 kernels.

2003-12-12	Mark Haverkamp <markh@osdl.org>

linit.c:
	- updated aac_eh_reset to use __shost_for_each_device now that the
	  device element is now private and we're supposed to use the helper
	  function for access.

Version: 1.1.4-2309
Version: 1.1.4-2310 (debug)

2003-12-18	Mark Salyzyn <Mark_Salyzyn@adaptec.com>

linit.c:
	- suppress unused variable warning in debug code.
	- cast sys_ioctl when registering as it does not match prototype
	  argument for ioctl32 registration.

Version: 1.1.4-2311

2003-12-22	Mark Haverkamp <markh@osdl.org>

aachba.c:
	- change from pae to dac as this is the more public understanding of
	  the 64 bit support concepts.
aacraid.h:
	- Remove padding and SavedIrql
commsup.c + aachba.c:
	- use atomic_read when accessing access_count member of device
	  structure.
linit.c & aacraid.h
	- iminor takes the inode, not the inode->i_rdev member.

Version: 1.1.4-2313

aachba.c + commsup.c:
	- use device_busy, shost_status, in_recovery instead of just
	  access_count. Adjust for each OS release variant.

Version: 1.1.4-2314

2003-12-22: Ken Beaty <ken@nova.org>

aachba.c + commsup.c:
	- Adjusted ifdefs for kernel version to make more sense.

2004-01-24: Mark Salyzyn <Mark_Salyzyn@adaptec.com>

install.sh:
	- Altered script to discover prebuilt binaries from the classic
	  Adaptec rpm package, the Red Hat install disk format, or the
	  SuSE install disk format.

2004-02-09: Christoph Hellwig <hch@lst.de>

aachba.c:
	- Remove fsa_dev structure since fsa_dev is always available.

Version: 1.1.4-2324

2004-02-10: Submit to scsi list for inclusion

2004-02-17: Herve MORILLON <hmorillon@doremilabs.fr> + Mark_Salyzyn@adaptec.com

rx.c + rkt.c:
	- hit doorbell before processing host_command_normal

aachba.c:
	- Permit requests larger than 64KB

aacraid.h:
	- Permit 512 outstanding requests

Version: 1.1.5-2326

linit.c + build:
	- Added support for vary_io, unfortunately the build system also needed
	  to be adjusted to generate the SCSI_HAS_VARY_IO if the member is
	  seen in the drivers/scsi/hosts.h file.

build + install.sh:
	- Added support for 2.4.19-189, 2.4.19-191 and 2.4.19-201 SuSE Kernels

Version: 1.1.5-2327

rkt.c + rx.c:
	- Added support to issue the Temperature sync command. Since the
	  cost of the sync command should not increase, the decision was
	  made to support a `varargs' approach to dealing with the additional
	  temperature elements *only* for this command.

linit.c:
	- Added a proc write that accepts the string "Temperature=[0-9.],..."
	  to send the off-board temperature value to the Firmware so that it
	  may be integrated into the Enclosure Data.
	- Added SkyHawk SATA cards to device list. 2020S changes now to 
	  2020ZCR, and we add 2020SA.

aachba.c:
	- PERCRAID RAID-5 is superfluous, changed to to PERC RAID-5.

Version: 1.1.5-2328

linit.c + aacraid.h:
	- Migrate towards using CONFIG_COMPAT instead of __x86_64__

rx.c + rkt.c:
	- Added support to pick up an Adapter Blink code. ADPmp64499.

linit.c:
	- Report the Adapter Blink code to the console log. ADPmp64499.

build:
	- Correctly built the x86_64 SLES8 and ul1 driver disk. Side effects
	  discovered also fix problems with ia32 SLES8 install. ADPmp64499.

Version: 1.1.5-2329

linit.c + aacraid.h:
	- Report an AifExeFirmwarePanic AIF message to applications when the
	  adapter is in a blinkled state.

aachba.c + commsup.c: Brad House <brad@mainstreetsoftworks.com>
	- use shost_for_each_device instead of list_for_each_entry.

linit.c + aachba.c:
	- xscale (arm) systems can not have highmem_io set as virtual/phys
	  handling does not recognize the page/offset addressing.

rkt.c + rx.c:
	- The Mailbox[7] in none BBS systems is not active until shortly
	  before the Firmware kernel is booted. The Outbound Message register
	  is always active and contains the same bringup conditions. We must
	  look at the OMR during the *_init wait.

Version: 1.1.5-2330

rkt.c + rx.c + sa.c:
	- Set the time by using get_seconds (epoch January 1 1970) instead
	  of jiffies/HZ (epoch machine startup). get_seconds is provided
	  for kernels < 2.6.

Version: 1.1.5-2331

rkt.c:
	- Mailbox[7] becomes momentarily inaccessible right after PATUWAIT
	  on the Callisto, lets loop on OMR only. Do not know if this
	  problem exists on other systems.

Version: 1.1.5-2332

aachba.c + linit.c:
	- Issue CT_COMMIT_CONFIG before issuing the VM_NameServe. This is
	  for systems that do not have a BIOS to perform this step.

Version: 1.1.5-2333

aacraid.h:
	- SAS requires the ability to handle as many as 32 Adapters in a
	  system, increased the manifest that limits the number of Adapters.
	- Testing has shown that allowing 33MB I/O can starve a machine, so
	  we are limiting the maximum I/O size to 4MB (to match other drivers
	  that permit large I/O).

linit.c:
	- Make sure that the driver does not register more than
	  AAC_MAXIMUM_ADAPTERS instances.
	- Set the queue depth to each device as divided up from AAC_MAX_IO_FIB

commctrl.c: Chris Wright <chrisw@osdl.org>
	- aac_send_raw_srb added check for bounding of fibsize value.

all: Mark Haverkamp <markh@osdl.org> & Mark Salyzyn <Mark_Salyzyn@adaptec.com>
	- merge 2.6 driver changes into tree to synchronize.

Version: 1.1.5-2334

aacraid.h+linit.c+commctrl.c+comminit.c:
	- Added sg_tablesize and max_fib_size to adapter structure and
	  negotiate these plus Scsi_Host sg_tablesize, can_queue and
	  max_sectors based on the adapter capabilities.

aachba.c:
	- Added aac_raw_io command
	- Recognize that read_callback is identical to write_callback, which
	  is in turn identical to raw_io's need for a callback. Renamed to
	  one callback function io_callback.

rx.c+rkt.c+sa.c:
	- Moved initialization around to permit New Command Interface probes
	- dropped irq_mask and associated functions.
	- moved acknowledgement of F/W commands *before* processing so that
	  we get re-interrupted if a new command is added to the produced
	  index while we are processing.

linit.c+aachba.c:
	- Do not print `bad0' for the serial number

linit.c:
	- this_id = 32, because it gets in the way of Container 16 being
	  processed.

aachba.c:
	- scsi_add_timer call issued just before completion routine called
	  since error recovery handler is there *just* to detect card
	  failure and not to affect command processing.

build:
	- Added 2.4.19.SuSE-343 kernel in support of ul1-sles8-ia32 install,
	  which adds yet another installation floppy to the list.

Version: 1.1.5-2335

linit.c+all:
	- Revert temporarily to 1.1.4-2177, Changed ASR-2020S to ASR-2020ZCR,
	  and ASR-2020S Terminator to ASR-2025ZCR.

Version: 1.1.4-2336

linit.c+all:
	- Revert temporarily to 1.1.4-2322, Changed ASR-2020S to ASR-2020ZCR,
	  and ASR-2020S Terminator to ASR-2025ZCR.

Version: 1.1.4-2337

all:
	- Revert back to 1.1.5 code base.

commsup.c:
	- Fix Irq Moderation code. A Misnomer, since this is really a PCI
	  utilization moderation, interrupts are not recurring on F/W.

comminit.c:
	- Turn on Irq Moderation feature (Tentatively 30% reduction in Host
	  CPU utilization)

Version: 1.1.5-2337

aacraid.h+commsup.c+dpcsup.c+comminit.c+rx.c:
	- Added support for the new comm interface.

linit.c:
	- Added debug information to proc output

Version: 1.1.5-2338

commsup.c: Mark Haverkamp <markh@osdl.org>
	- Added scsi/scsi_device.h, scsi/scsi_driver.h to include file set
	- set removable to a value of 1, not to TRUE.

aachba.c: Mark_Salyzyn@adaptec.com
	- Switch to using max_fib_size rather than FIB_DATA_SIZE_IN_BYTES,
	  this permits SAS management applications to send ioctl FIBs larger
	  than 512 bytes in size to adapters that accept larger FIBs.
	- Added support for SAI_READ_CAPACITY_16, READ_12, WRITE_12, READ_16
	  and WRITE_16 commands.
	- Played `tricks' with device_blocked and queue_depth fields in the
	  scsi_device structure to meter the outstanding commands down when
	  large sequential activity is detected.

aacraid.h: Mark_Salyzyn@adaptec.com
	- Remove unused definition of FIB_DATA_SIZE_IN_BYTES.

linit.c:
	- Setting the maximum number of I/O requests/device to a maximum of
	  256 would act in the SCSI layer as only allocating to permit 1 I/O
	  for each device.

Version: 1.1.5-2339

build: Mark_Salyzyn@adaptec.com
	- Added support for 2.6.4-52 SuSE 9.1 Pro install
	- Added support for multiple architectures for 2.4.21-15.EL RHEL3 QU2
	  install.

aacraid.h+aachba.c+linit.c: Mark Haverkamp <markh@osdl.org>
	- Define 2131 as FSACTL_GET_CONTAINERS

commctrl.c: Adam Manthei <amanthei@redhat.com>, Mark_Salyzyn@adaptec.com
	- change all printk() to dprintk(()) as this is a user initiated
	  call for aac_send_rw_srb & aac_get_pci_info.

rx.c+rkt.c: Adam Manthei <amanthei@redhat.com>, Mark_Salyzyn@adaptec.com
	- use pci_alloc_consistent/pci_free_consistent instead of an
	  unchecked combination of kmalloc(,_GFP_DMA)/pci_map_single/
	  pci_unmap_single/kfree.

Version: 1.1.5-2340

linit.c+commctrl.c: Mark Haverkamp <markh@osdl.org>
	- adjust to reflect linux-scsi submission results

aachba.c: Mark_Salyzyn@adaptec.com
	- remove print for unhandled commands into a debug print. The
	  unhandled commands are reported to the caller as unhandled, let
	  the caller deal with this.

rx.c+rkt.c+sa.c: maximilian attems <janitor@sternwelten.at>
	- upon failure of the init routine, make sure that the registered
	  interupt handler is deregistered.

commsup.c:
	- fib_adapter_complete is supposed to free the hw_fib and that is it,
	  it tried to talk to hardware and caused a lockup.

Version: 1.1.5-2341

build:
	- use aacraid.ko for 2.6 releases

Version: 1.1.5-2342

aachba.c: Mark_Salyzyn@adaptec.com
	- added support for a module parameter 'commit=1' to enable COMMIT
	  CONFIG to be issued to the adapter.
	- added support for a module parameter 'coalescethreshold=16' which
	  sets the maximum block size to consider for pushing back to the
	  scsi_merge layer.
	- added support for a module parameter 'acbsize=8192' which sets the
	  suggested fib size to override the suggestion from Firmare.
	- dropped call to scsi_add_timer, as it causes a panic. It was placed
	  in the source to resolve a command completion race condition.

Version: 1.1.5-2343

install.sh: Mark_Salyzyn@adaptec.com
	- globbing issue caused more whiny complaints about a missing
	  installation into the initrd.
	- fixed some issued surrounding using the script for SuSE module
	  update.

linit.c: Mark_Salyzyn@adaptec.com
	- if the driver does not discover any targets, report failure.
	- drop kernel_version hack to support build

build: Mark_Salyzyn@adaptec.com
	- Use vermagic instead of kernel_version to pick up matching kernel.
	- when innoculating 2.6 tree builds, one needs a *full* compile in
	  order to generate the struct_module member.
	- use module.ko for 2.6 kernels.

Version: 1.1.5-2344

build: Mark_Salyzyn@adaptec.com
	- floppy linux/suse/${ARCH}-${VERS}/modules/${PRODUCT}.o needs to be
	  a ${PRODUCT}.ko in the 2.6 based installations.
	- Placed module in both scsi and scsi/${PRODUCT} directories as it
	  appears that the post-install is not functioning properly.

aachba.c: Mark_Salyzyn@adaptec.com
	- Checked if the lba exceeds 32 bit block address for systems that
	  can not support it. raw_io_64 enables 64 bit block addresses.
	- Redid math for u64 >> 32 as it appears the xscale 64 bit library
	  is flawed.

Version: 1.1.5-2345

aachba.c: Mark_Salyzyn@adaptec.com
	- Overrides to force 8KB fibs needs to be reverted to defaults.

Version: 1.1.5-2346

build: Mark_Salyzyn@adaptec.com
	- Added 2.4.21-15.0.2.EL kernel
	- Added 2.6.5-7.97 kernel to the build (SLES9 Gold)

rx.c+rkt.c: Mark_Salyzyn@adaptec.com
	- Mailbox7 continues to be a consternation regarding reliable
	  adapter recovery scenarios; switched to using OMRx[0].

Version: 1.1.5-2347

aachba.c: Mark_Salyzyn@adaptec.com
	- (u64)=((u8)<<24) does not give expected results, sign extension
	  occurs. Replace with (u64)=((u64)(u8)<<24)

Version: 1.1.5-2348

install.sh: Mark_Salyzyn@adaptec.com
	- initrd is blocked from incorporating our product if there is
	  something in /lib/modules/${OS}-${CONFIG}/update/${PRODUCT}.o,
	  so remove the file.

Version: 1.1.5-2349

aachba.c+aacraid.h:
	- define commit_config FIB command
	- define get_container_count FIB command.

aachba.c+aacraid.h+commsup.c+linit.c
	- fsa_dev becomes a dynamic structure to accommodate a variable
	  maximum_num_containers.

build:
	- Added 2.4.21-231 kernel to build system.

linit.c:
	- Turned on debug printing of scsi timeouts for xscale only.

Version: 1.1.5-2350

rkt.c:
	- Limit can_queue to 246 for rocket

build:
	- Added 2.4.19-306 kernel to build system.

aachba.c:
	- Removed an innocuous (obnoxious?) DEBUG printk

2004-07-15: Mark Salyzyn <Mark_Salyzyn@adaptec.com>

Version: 1.1.5-2351

build:
	- Added 2.4.9-31 to the build system

module.equiv:
	- Added 2.4.9-e.41, 2.4.9-e.43, 2.4.21-17.EL & 2.4.21-15.0.3.EL kernels

build:
	- Dropped 2.4.21-231 from build

2004-07-16: Mark Salyzyn <Mark_Salyzyn@adaptec.com>

Version: 1.1.5-2352

build:
	- Added 2.6.3-7mdk to the build system

2004-07-20: Mark Salyzyn <Mark_Salyzyn@adaptec.com>

Version: 1.1.5-2353 (7t Build w/o SLES9, SuSE9.1 & SuSE9 errata 231)
Version: 1.1.5-2354 (7t Build w/o SLES9 & SuSE9 errata 231)
Version: 1.1.5-2355 (BigQual refresh)

install.sh:
	- If missing, add a reference to the module to the kernel's module.dep
	  file (affects drivers that are *not* shipped with the OS; HostRAID
	  and some dpt_i2o)

aachba.c:
	- for __arm__ build, the default FIB size is selected by F/W and not
	  overridden to 8192 bytes.

Version: 1.1.5-2356 (Jupiter)

aacraid.h+comminit.c+rkt.c+commsup.c+linit.c: Ken Sandars <Ken_Sandars@adaptec.com> + Mark Salyzyn
	- Added AAC_NUM_MGT_FIB, and ensured can_queue represents the
	  maximum number of I/O commands allowed and not be confused as the
	  maximum number of FIB commands permitted into the Adapter. Thus
	  host->can_queue is the maximum number of I/O commands, AAC_NUM_MGT_FIB
	  is the maximum number of ioctl commands (set to 8 rather than legacy
	  of 64) and init->MaxIoCommands sent back to the adapter is the total
	  number of FIBs.

Version: 1.1.5-2357 (Jupiter+BigQual)

commctrl.c: Mark Salyzyn
	- Added support for issuing FIBs that are larger than the negotiated
	  size for the management interface.

linit.c: Mark Salyzyn
	- Added ASR-2240S, ASR-4005SAS, ASR-4000SAS, ASR-4800SAS, ASR-4805SAS
	  and AAR-2610SA to the product list.

install.sh: Mark Salyzyn
	- Fixed problems with using the RH & SuSE modules disk as an update
	  media, most of which was the selection of the extraction name from
	  the modules.cgz file or acquiring the appropriate update.tar.gz file.

build: Mark Salyzyn
	- set 700 for update.sh in the modules disks to recognize at least
	  that the RH modules disk works as an update media.

aachba.c: Mark Salyzyn
	- Dropped to 8K for the coalesce threshold for xscale builds.

Version: 1.1.5-2358 (BigQual+7t)

aachba.c+commctrl.c+aacraid.h+comminit.c+commsup.c+dpcsup.c+linit.c:
	- Merged 2.6.8-rc2-bk9 differences into the driver (COSMETIC)

compat.h
	- Added definition for __user for kernels less than 2.5 (COSMETIC)

linit.c:
	- The aac_get_next_adapter_fib_ioctl for 64 bit implementations under
	  2.6 maladdressed the transfer.

Version: 1.1.5-2359 (BigQual+SPOCK+7t)

commctrl.c:
	- Added support for CODE_STREAM_IDENTIFIER, accessed via a new
	  ioctl FSACTL_GET_VERSION_MATCHING
	- Added support for FSACTL_SEND_LARGE_FIB

aacraid.h:
	- Added definition for struct VersionMatch
	- Added definition for FSACTL_GET_VERSION_MATCHING
	- Added definition for FSACTL_SEND_LARGE_FIB

install.sh:
	- if the modules.dep file does not exist, then ensure no complaints
	  are made about mv not being able to access the file.
	- If an entry is missing in modules.dep, construct it correctly.

aachba.c:
	- Remove any leading spaces from the Container Name. Ensure that
	  if there is no container name remaining, to leave the existing
	  one alone.

build:
	- Added support for 2.4.18-e.43
	- Added support for 2.4.18-e.47
	- Added support for 2.4.9-e.48
	- Added support for 2.4.9-e.49 (RHAS 2.1 QU5)
	- Added support for 2.4.21-15.0.4.EL

rx.c+rkt.c:
	- When responding to AIFs, use DoorBellAdapterNormRespReady instead of
	  DoorBellAdapterNormCmdReady. The code appeared to work without
	  undue side effects because Firmware would clear off the queues
	  when new AIFs are delivered.

commsup.c:
	- If busy, defer scan action to next AIF. Half hearted attempt to
	  improve the reliability of this unsupported feature.

aacraid.h+linit.c+comminit.c+rx.c+rkt.c+sa.c:
	- Remove references to Mailbox7 accesses for synchronous commands.

aacraid.h+aachba.c:
	- Turned on support for Compatibility ID testing. Only enabled if
	  the build environment defines CODE_STREAM_IDENTIFIER.
	- Fortify the adapter information to include supplemental information
	  as well as the GetBusInfo in support of SAS programatic limits.

linit.c+aachba.c:
	- Use the newly acquired supplement information vendor and product
	  fields to override the cardtype selection.

Version: 1.1.5-2360
Version: 1.1.5-2361 (Branch off 1.1.5-2340 in RHEL3 QU3 with aac_info fix in 1.1.5-2364)

linit.c:
	- register a reboot notifier to flush adapter
	- faked AIF needs to call fib_init() later (ADPmp70525)

commctrl.c:
	- Since kfree has the possibility of switching in some esoteric
	  variants of the kernel, and since the BKL is held during ioctl
	  calls, we are unlocking the fib lock around these system calls.

Version: 1.1.5-2362

build:
	- Added support for 2.4.21-17.EL (RHEL3 QU3 beta)
	- Added support for 2.4.21-20.EL (RHEL3 QU3)
	- Added support for 2.6.7-1.451.2.3 (RHEL4 alpha 4)

linit.c:
	- ASR4000 series entries were flawed in the indexes. Added an
	  additional 8i entry.

Version: 1.1.5-2363

linit.c:
	- aac_info is flawed and causes periodic panics in certain systems.

build:
	- Allow the build system to operate as background build
	- em64t/ia32e binaries did not show in the rpm

Version: 1.1.5-2364 [BigQual, Pratt, Jupiter Dual & 7t]

linit.c:
	- AAR-2610SA has had a subproduct change from 0x103C/0x0295 to
	  0x103C/0x3227.
	- Some adapters have internal limits of 34 SG elements as a result
	  of a bug of not splitting requests up when cache is disabled when
	  sending them to the CHIM. Created a quirk for the adapters that
	  have this limit.

Version: 1.1.5-2365

aachba.c:
	- Neglected to add maximum_num_physical check to the srb handler.
	- leading space on inquiry code was flawed as a result of a typographic
	  error (replace ! with *)

linit.c:
	- ASR-4005SAS for IBM is designated as an 8i (ADPmp71521)
	- Added check for late completion of command during timeout
	- called aac_get_fw_debug_buffer() after init

fwdebug.c+fwdebug.h
	- Added firmware debug print handlers

build:
	- RH floppy disk was limited to 864KB, let it open up to 1.4MB. The
	  risk is that 864KB was about all the ramdisk could handle when
	  extracting, so we will no longer get a report of disk overrun as
	  a warning to existing OS releases. We do not know what will overload
	  the ramdisk during install. (ADPmp72476)
	- Product ID list for aacraid is broken in the build due to changes
	  resulting from incorporating the 2.6 tree.
	- em32t binaries did not show in the rpm fix broke SuSE releases that
	  utilize the 2.4.19 kernel (ADPmp73104)

commsup.c:
	- Added support for sending the time every 30 minutes.

Version: 1.1.5-2366

commctrl.c:
	- Fixed 64 bit version of the SCB ioctl call as it was not translating
	  the 32 bit scatter-gather correctly for scatter gather elements
	  beyond the first one. Do not believe this issue presented a problem
	  for any Adaptec management products due to their needs as they
	  limited their SG to only one entry.

build:
	- Added 2.4.19-238
	- Added 2.4.19-241
	- Added 2.4.19-248
	- Added 2.4.19-251

Version: 1.1.5-2367

linit.c:
	- Added Themisto discovery to driver
	- Added AAC_QUIRK_MASTER and AAC_QUIRK_SLAVE to deal with controller
	  pairs.
	- Changed Prowler "ESD SO-DIMM PCI-X SATA ZCR" to "ASR-2026ZCR".
	- Return FAILED when hba_reset completed, the ten second bus settling
	  delay is counter productive.

aacraid.h+linit.c: Christoph Hellwig <hch@lst.de>
	- drop casting surrounding iomap and ioremap and use the __iomem
	  type enhancement.

csmi.c+csmi.h:
	- Added CSMI ioctl support.

install.sh:
	- log both successful and failed installations and limit complaints
	  about installation to a minimum by not repeating similar failure
	  messages.

aachba.c:
	- vtune reports that 2% of the time dealing with read calls was
	  devoted to get_sd_devname(). Optimized codepath.

Version: 1.1.5-2368

build:
	- Added 2.4.27 debian smp kernel to build
	- Added 2.6.7-1.451.2.3 RHEL4 alpha 4 to the build
	- Added 2.6.8-1.602 RHEL4 beta 1 to the build
	- Added 2.6.5-7.109.5 SuSE 9.1 errata to the build

csmi.h:
	- Changed from CSMI_8_BYTE_ALLIGNED to CSMI_8_BYTE_ALIGNED

csmi.c:
	- failure on return from copy_to_user is ignored, but (void) is
	  not the way to ignore it based on compiler warnings.
	- scb size is set to 16 always.
	- scb flags is set to 0 always.

linit.c+compat.h:
	- scsi_sleep() renamed to ssleep() in 2.6.9 kernel

commctrl.c:
	- 32 bit application issuing a raw srb in a 64 bit address space
	  is not handled correctly for the fibsize calculation.
	- Limit the number of scatter gather entries to 32, and zero
	  out the references to those sg entries.

Version: 1.1.5-2369

compat.h:
	- 2.6.8-1.602 RHEL4 beta 1 kernel looks more like 2.6.9, so needed to
	  define scsi_sleep for kernels 2.6.8 and up rather than 2.6.9 and up.

linit.c: Chris Trown <ctrown@uoregon.edu>
	- Added an include for linux/delay.h to pick up the definition of
	  ssleep().
	- Added `other' Themisto ID to list.

csmi.h:
	- bumped to 0.83 version

csmi.c:
	- Acquired slot number from Supplementary Adapter Info SlotNumber
	  field
	- Added support to determine the actual bus type (SAS, SATA, Other).

build:
	- Added support for 2.4.21-22.EL (RHEL3 errata)
	- Added support for 2.6.5-7.111 (SLES9 NLD)
	- Added support for 2.4.21-243 (SuSE 9.x errata)

aachba.c:
	- valid must clear if VM_Nameserve fails on scan
	- setinqstr has the possibility of overflowing the inquiry structure,
	  the results are a damaged stack.

commsup.c:
	- do a probe_container before issuing scan as the adapter can take
	  some time to write DDF data (not SCSI).

aacraid.h:
	- added definition of probe_container.

Version: 1.1.5-2370

aacraid.h+commsup.c+aachba.c+commctrl.c+comminit.c+linit.c+sa.c: Adrian Bunk <bunk@stusta.de>
	- Make some needlessly global code static

linit.c:
	- Added Intruder (AAR-2420SA, AAR-2620SA & AAR-2820SA)

aachba.c+commsup.c:
	- Enable a 50 second timeout on the first asynchronous command to
	  permit the driver to error out and report the hardware failure.

linit.c:
	- Fixed a comment completion problem for Intruder additions.

build:
	- Added support for 2.6.8-24 (SuSE 9.2)

Version: 1.1.5-2371

aachba.c: Jens Axboe
	- Use a busy status return with scsi_done rather than -1 to
	  signal OS to try again later for aac_read and aac_write to
	  meet with acceptable coding standards.

aachba.c+linit.c: Mark Salyzyn & Christoph Hellwig <hch@infradead.org>
	- Moved AAC_EXTENDED_TIMEOUT to set sdev->timeout instead of
	  inline for every command in the 2.6 variant.

linit.c:
	- There is a subsystem device id clash between SATAHAWK and INTRUDER
	  and is giving the BIOS group some grief. Therefore the subsystem
	  device ID for intruder is changed to 029B, 029C and 029D for 8, 6,
	  and 4 port versions respectively.
	- Added FSACTL_GET_VERSION_MATCHING and FSACTL_SEND_LARGE_FIB ioctls to
	  list of supported 32 bit ioctl calls.

build:
	- enhanced README.txt to also provide a brief description of the
	  binary file.
	- Added support for a RHEL3 i686 ISO image as well.
	- Added support for 2.6.5-7.109.12 (SLES9/SuSE9.1 SP1 B2 i386)
	- Added support for 2.6.5-7.109.13 (SLES9/SuSE9.1 SP1 B2 x86_64)
	- Added support for 2.6.5-7.115 (SLES9/SuSE9.1 SP1 B3)
	- Added support for 2.6.5-1.358 (RH FC2)
	- Added support for 2.6.9-1.667 (RH FC3)
	- Added support for 2.4.21-260 (SuSE 9 errata)
	- Added support for 2.4.21-261 (SuSE 8 errata)

csmi.c:
	- return code for CSMIGetSATASignature is under the control of the
	  firmware and should not be blindly set to SUCCESS if the command
	  succeeded.


Version: 1.1.5-2372

build:
	- Added support for 2.6.9-1.648_EL (RHEL4 beta 2)

linit.c:
	- trim space from Model

install.sh:
	- Add /etc/modprobe.conf as another modules configuration file.
	- If the module in the initrd does not match, but has the same
	  `size' values, then report a `possibly stripped' warning message.

Version: 1.1.5-2373

build:
	- Use "Avon Park SIT" for this build for the version identifier

commsup.c:
	- enable scsi-scan-single (ADPmp75534 & ADPmp69336)

Version: 1.1.5-2374

build:
	- Added support for 2.6.5-7.111.5 (SLES9 NLD errata)
	- Added support for 2.6.5-7.128 (SLES9 SP1 B4)
	- Added support for 2.6.5-7.134 (SLES9 SP1 RC)
	- Added support for 2.6.9-1.906_EL (RHEL4 RC)

commsup.c:
	- disable scsi-scan-single for 2.6 kernels, proc_scsi no longer
	  exported

install.sh
	- Missed a double quote in the scripting to reduce size sensitivity.

Version: 1.1.5-2375 (Avon Park SIT)

csmi.c:
	- Paramters.uFlags is a bit field, not a state.
	- cdbLength needs to be hard coded to 14 from 16 (beware, 2TB warning).
	- Set the srbFlags based on data direction or if there is any data to
	  send at all (HOST_FLAGS_DATA_IN/DATA_OUT/NO_DATA_TRANSFER).

csmi.h:
	- Added definition for HOST_FLAGS_DATA_IN/DATA_OUT/NO_DATA_TRANSFER.

Version: 1.1.5-2376 (This is a code stream identifier)

linit.c:
	- Added ICP Lancer products ICP9024R0 and ICP9014R0.
	- Added include of asm/ioctl32.h
	- Report ServeRAID for driver name if IBM 8i.

aacraid.h
	- Added definition for IOP_RESET synchronous command

rkt.c+rx.c+linit.c+aacraid.h+commsup.c+aachba.c+csmi.c+comminit.c+commctrl.c+compat.h
	- Merged code and style changes in 2.6.10-rc3-bk14 into codebase.

aachba.c:
	- set the scsi device as removeable during the Read Capacity call.
	  (ADPmp76369)

pci.ids:
	- Submitted patch to pci-ids@ucw.cz to update the vital product list
	  to match the products in the linit.c file (ADPmp77082)

build:
	- Added support for 2.4.21-20.0.1.EL (RHEL3 errata)
	- Added support for 2.4.21-27.EL (RHEL3 QU4)
	- Added support for 2.4.21-27.0.1.EL (RHEL3 errata)

Version: 1.1.5-2377 (Avon Park SIT)

linit.c:
	- Dropped the maximum number of commands down to 16 per target if on
	  an x86_64 machine with more than 4GB of memory when built in the
	  2.4.* environment.

Version: 1.1.5-2378 (CERC Test)

linit.c:
	- Dropped the maximum number of commands down to 10 per target for
	  this test.

Version: 1.1.5-2379 (CERC Test)

build:
	- Added support for 2.6.9-5.EL (RHEL4 RC)
	- Added support for 2.6.5-7.139 (SLES9 SP1)
	- Added support for 2.4.9-e.57 (RHAS QU6)
	- Added support for 2.4.9-e.59 (RHAS QU6 errata 59)

commsup.c:
	- Added kernel ifdef's to handle scsi_add_target and
	  scsi_remove_target calls.

aachba.c+aacraid.h:
	- Added AAC_DEBUG_INSTRUMENT_DAC_CERC to disable 64 bit scatter gather
	  for only the CERC SR2 product.

Version: 1.1.5-2380 (This is a code stream identifier)

aachba.c+comminit.c:
	- Added numacb insmod parameter.

aachba.c+aacraid.h
	- Remove AAC_DEBUG_INSTRUMENT_DAC_CERC code.

build:
	- Error in incorporating the RHAS2.1 QU6 kernel (2.4.9-e.57) on to the
	  driver disk (ADPmp78010)
	- Same problem with RHEL4 RC (2.6.9-5.EL) (ADPmp69861)

Version: 1.1.5-2381 (This is a code stream identifier)

build:
	- Added support for 2.4.21-273 (SLES 8 errata 273)
	- Added support for 2.6.5-7.111.30 (SLES 9 errata 30)
	- Added support for 2.6.8-24.11 (SuSE 9.2 errata 11)
	- Added support for 2.4.19.SuSE-256 (SLES8 x86_64 errata 256)
	- Added support for 2.4.21-9.EL (RHEL3 QU1)
	- Added support for 2.6.8.1-12mdk (Mandrake 10.1)
	- Added support for 2.6.9-1.11_FC2 (FC2)
	- Added support for 2.6.10-1.9_FC2 (FC2)
	- Updated dkms to v2.0.5

commsup.c+aacraid.h
	- Changed Plug-n-Play state machine to be per-array rather than
	  per-adapter (ADPmp77096)

Version: 1.1.5-2382 (This is a code stream identifier)

build:
	- Added support for 2.4.21-276 (SLES8/UL1 SP4 B2)
	- Added support for 2.6.10-1.8_FC2 (FC2)
	- Added support for 2.6.10-1.12_FC2 (FC2)
	- Added support for 2.6.9-1.681_FC3-2.6 (FC3)
	- Added support for 2.6.9-1.724_FC3-2.6 (FC3)
	- Added support for 2.6.10-1.737_FC3-2.6 (FC3)
	- Added support for 2.6.10-1.741_FC3-2.6 (FC3)
	- Added support for 2.6.10-1.760_FC3-2.6 (FC3)

linit.c:
	- vmware specifically utilizes the file->private_data. They will be
	  correcting the problem in future releases thus the fix will work
	  in both environments being obnoxious^H^H^H^H^H^H^H^H^Hinnocuous in
	  the later releases.

aachba.c:
	- vmware has problems with the coalescing code.

commctrl.c:
	- used sizeof(struct sgmap *) instead of sizeof(struct sgmap) and
	  misrepresented the size of the srb command truncating the scatter
	  gather size from the incoming data in 64 bit architectures with
	  more than 4G of memory populated in the system. (ADPmp78515,
	  ADPmp78128, ADPmp76236 & ADPmp78228)

Version: 1.1.5-2383 (Avon Park SIT)

Makefile:
	- Self detect the various SCSI_HAS_* flags rather than depending on
	  the build system to generate them.

aacraid.h+aachba.c
	- Added VM_NameServe64
	- Added capacity64 field to end of the mnt information.
	- Do not respond to SERVICE_ACTION_IN when card not capable of 64 bit
	  lba.

commctrl.c:
	- The srbcmd in 64 bit environments with more than 4GB of memory
	  are utilizing the sgentry64 elements for the scatter gather, thus
	  the counts were not parsed correctly when copying the data back
	  to the user. (ADPmp78846)

build:
	- Removed support for linux-2.4.21-1.1931.2.349.2.2.ent.RH
	- Removed support for linux-2.4.21-1.1931.2.393.ent.RH
	- Removed support for linux-2.4.21-1.1931.2.399.ent.RH
	- Removed support for debug configurations
	- Split AS2.1 summit/enterprise from up/smp
	- pcitable for aacraid driver disk is missing " after the ICP cards

Version: 1.1.5-2384 (Avon Park SIT)

aacraid.h+rkt.c+rx.c+sa.c:
	- use aac_io_cpu_to_le* and aac_io_le*_to_cpu to deal with perceived
	  discrepancies in write* and read* I/O handlers.

commsup.c+commctrl.c:
	- header.Size and header.SenderSize is a 16 bit field, erroneously
	  handled by 32 bit swap handlers.

commsup.c+comminit.c+commctrl.c+aachba.c+dpcsup.c:
	- missing swap handlers for various packets.

aachba.c:
	- When 'first' command times out, return error immediately, do not
	  fall through.

csmi.c+aachba.c+linit.c:
	- monitor/kernel/build information from adapter needs to be swapped
	  in BE architectures.

aachba.c:
	- Revert 64 bit LBA code

Version: 1.1.5-2385 (Avon Park SIT)
Version: 1.1.5-2386 (Avon Park SIT, revert to 1.1.5-2383, plus one managment change)

aachba.c+linit.c+compat.h: Tobias Klauser <tklauser@nuerscht.ch>
	- Use the DMA_{64,32}BIT_MASK constants

linit.c:
	- scsi_host_alloc calls scsi_register which prints a report in some
	  versions of the 2.4 kernel. The aac_info function is not ready to
	  be called at that time as the hostdata has not been set up, so we
	  report a 'default' name of "AAC" (ADPmp78060).

aacraid.h+aachba.c:
	- Adding any u64 to a structure will cause, in some cases of the
	  compiler, 8 byte alignment constraints which can reshape the
	  structure at each element even before the u64 definition. Changed to
	  using a pair of u32's for capacity64 element in the mount structure.
	  (ADPmp79142)

Version: 1.1.5-2387 (This is a code stream identifier)

build:
	- Added support for 2.4.21-277 (SLES8/UL1 SP4 RC1)

aacraid.h+rx.c+rkt.c+sa.c+linit.c+Makefile:
	- Added support for dump_poll. Currently RHEL4 and RHFC support this
	  new interface. (ADPmp79442)

Version: 1.1.5-2388 (This is a code stream identifier)

compat.h+aachba.c+aacraid.h+commctrl.c+comminit.c+commsup.c+dpcsup.c+linit.c:
	- Merged code and style changes in 2.6.11-rc5-bk3 into codebase.

linit.c:
	- Added support for shost_attrs & sysfs.

linit.c+csmi.c:
	- Dropped reference to Red Hat in printouts (ADPmp79559 & ADPmp79382)

build:
	- Strip kernel environment check for sourceball that is part of the
	  DKMS packaging, broke DKMS build environment (ADPmp79708)

Version: 1.1.5-2389 (This is a code stream identifier)

build:
	- Switch from 3.3 compiler to 3.4 for RHEL4 ia64 and FC3 all archs
	  & all errata.
	- Added support for 2.4.18-19.7.x (RH7.3 errata 19)

linit.c: Domen Puncer <domen@coderock.org>
	- Change from pci_module_init to pci_register_driver.

linit.c:
	- Loop for determining unique_id failed and generates a pattern of
	  0, 1, 0, 0, ... (ADPmp79694)
	- Added ICP9047MA and ICP9087MA to product probe. Added AvonPark Lite.

linit.c+aachba.c:
	- An SG list of 1K,4K,...,4K,64K+4K-1K got produced, the math for the
	  maximum I/O should be (sg_tablesize * 8) + 112 instead.

rkt.c+rx.c:
	- Adapter Panic handler has AAC_IO_USES_CPU_ORDER misspelled.

commsup.c:
	- time stamp is an u32, not an unsigned long.

csmi.c:
	- Send LoopCount rather than bufferOffset to param[4] of GetRAIDConfig
	  command. (ADPmp77631, ADPmp79282)

Version: 1.1.5-2390 (This is a code stream identifier)

build:
	- Added support for 2.4.21-278 (SLES8/UL1 SP4 RC3)
	- Removed support for 2.4.21-276
	- Removed support for 2.4.21-277
	- Removed support for 2.4.18-24.7.x (RH7.3 errata 24)
	- Removed support for 2.4.18-26.7.x (RH7.3 errata 26)
	- Removed support for 2.6.10-1.741_FC3-2.6 (FC3 errata 741)
	- Dropped all the sles9-sp1 betas from the packaging.
	- strip date stamps, then join.file the patches before commiting them
	  to the archive.

modules.equiv
	- Declared 2.4.21-277 to be identical to 2.4.21-278
	- Declared 2.4.18-24.7 to be identical to 2.4.18-19.7
	- Declared 2.4.18-26.7 to be identical to 2.4.18-19.7
	- Declared 2.6.10-1.737_FC3-2.6 to be identical to 2.6.10-1.741_FC3-2.6

linit.c:
	- return code from pci_register_driver() is not of the same form
	  as pci_module_init. Only negative return values should be reported
	  as an error.

install.sh:
	- Added /etc/grub.conf to list of grub files
	- redirect error on boot configuration file awk script for cases when
	  boot configuration file is not found.

Version: 1.1.5-2391 (This is a code stream identifier)
Version: 1.1.5-2392 (Branch off 1.1.5-2372 with pci_unregister_driver if aac_count drops to zero).

build:
	- Added support for 2.6.10-1.14_FC2-2.6 (RH FC2 Errata 14)
	- Added support for 2.6.10-1.770_FC2 (RH FC2 Errata 770)
	- Added support for 2.6.10-1.766_FC3-2.6 (RH FC3 Errata 766)
	- Added support for 2.6.10-1.770_FC3-2.6 (RH FC3 Errata 770)
	- Removed support for 2.6.10-1.9_FC2 (FC2) and placed in
	  module.equiv
	- Removed support for 2.4.18-e.47 (RHAS 2.1 IA64) and placed in
	  module.equiv
	- Added support for 2.4.18-e.52 (RHAS 2.1 IA64) in module.equiv
	- Added support for 2.4.18-e.54 (RHAS 2.1 IA64) in module.equiv
	- Generate dkms package with build number, but add a Branch Type
	  of "dkms" into the version information reported by the driver.
	- Generate source package with a Branch Type of "custom" into the
	  version information reported by the driver.
	- build rpm packages as 'noarch' and not as default of the build
	  system 'i386'.

csmi.c:
	- uMaxDrivesPerSet needs to be acquired from the Firmware,
	  rather than using the driver physical limits (ADPmp80188)

aachba.c:
	- Added some new container types (RAID5D, RAID5D0, RAID1E, RAID6
	  and RAID60) to default array naming list.
	- Changed over to new format of VM_NameServe64 (changed before
	  customer release of Firmware that utilized interim format).

linit.c:
	- Added Hurricane ASR-4810SAS
	- Added sensitivity to AAC_DRIVER_BRANCH in order to propagate
	  driver source with keys as to their history.

linit.c: (Mark Haverkamp <markh@osdl.org>
	- Restructured sys handler to match standards expectations.
	  ADPmp80589

install.sh:
	- Do not compare result in initrd to any backup orig drivers
	  that may have been left in the /lib/modules tree.
	- Added support for elilo.efi

commsup.c+comminit.c:
	- pci_alloc_consistent acquired GFP_DMA arena pool. This has
	  been shown as a problem on 2.4 based kernels on em64t machines
	  with > 4GB of memory which typically exhaust the DMA pool. So,
	  prior to making the call, we will acquire GFP_ATOMIC
	  memory first and check if it is 31BIT limited, and instead use
	  that memory rather than resorting to the precious DMA pool. The
	  other workarounds are to limit the memory to 4GB, set the
	  memsize to 4GB, or to tell the SW IOMMU to reduce it's memory
	  requirements (swiotlb=12288).

Version: 1.1.5-2393 (This is a code stream identifier)

linit.c+commctrl.c:
	- Added AAC_DEBUG_INSTRUMENT_IOCTL_SENDFIB

linit.c: Ming Zhang <mingz@ele.uri.edu> & Mark Salyzyn
	- Set max_cmd_len to 16 to permit SAI_READ_CAPACITY_16 call to
	  get through the scsi layer to the driver.

linit.c:
	- Added calls to csmi_register_ioctl32_conversion() and
	  csmi_unregister_ioctl32_conversion()

csmi.c:
	- Added csmi_register_ioctl32_conversion(void) and
	  csmi_unregister_ioctl32_conversion(void)

install.sh:
	- notice that we are doing an ELILO configuration on ia64

aachba.c:
	- Removed sundry debugging prints

build:
	- Added MODULES_CONF_OBSOLETES_ONLY[0]="${PRODUCT}" to dkms.conf
	- Added support for 2.6.5-7.147 (SLES9 SP1 errata 147) to
	  modules.equiv
	- Added support for 2.6.5-7.151 (SUSE9.1 SP1 errata 151) to
	  module.equiv
	- Added support for 2.4.21-278-x86_64 (SLES8 SP4 64 bit)

Makefile:
	- Use TOPDIR instead of AAC_ROOT

Version: 1.1.5-2394 (This is a code stream identifier)

build:
	- Added support for 2.4.20-133 (SuSE 8.2 errata 133)
	- Added support for 2.4.21-286 (SuSE 9.0 errata 286) to module.equiv
	- Added support for 2.6.8-24.14 (SuSE 9.2 errata 14)
	- Added support for 2.6.9-5.0.3.EL (RHEL4 errata 3)
	- Added support for 2.6.9-6.37.EL-2.6 (RHEL4 U1 beta)
	- Added support for 2.6.10-1.771_FC2 (FC2 errata 771)
	- Added support for 2.6.11-1.14_FC3 (FC3 errata 773)
	- Added support for 2.4.21-31.EL (RHEL3 QU5 beta)

aachba.c+linit.c: Tobias Klauser <tklauser@nuerscht.ch> & Domen Puncer <domen@coderock.org>
	- added include for linux/dma-mapping.h

linit.c: Konstantin Khorenko <khorenko@sw.ru>
	- aac_info and aac_detect need to be active on in-kernel
	  versions of the driver.

linit.c:
	- aac_show_flags is a newline separated list
	- Added SAI_READ_CAPACITY_16 to list of possible flags in
	  aac_show_flags
	- aac_get_adapter_info status needs to be placed in the 'error'
	  variable to unload correctly and deinit rather than unmap
	  (ADPmp83209).

commsup.c:
	- allocate 8 more NewFibs than noticed to deal with high AIF
	  traffic.
	- Added support for wait=-2 to do a 'silent' timeout of a
	  command.
	- Increased the timeout for wait<0 to 3 minutes from 50 seconds
	  due to paranoia (ADPmp83209)

comminit.c:
	- issue the adapter shutdown with a wait=-2 value (ADPmp78635)

aacraid.h+rx.c+rkt.c+sa.c+linit.c:
	- Added a disable_interrupt method to prevent future adapter
	  interrupts when shut down. Call this method before free_irq(),
	  preferably before dealocating structures. (ADPmp83209)

commsup.c+comminit.c:
	- revert out the GFP_KERNEL kmalloc call to see if it reports an
	  address <2GB instead of using pci_alloc_consistent. Fix this
	  another day. (ADPmp83209)

Version: 1.1.5-2395 (This is a code stream identifier)

build:
	- Added support for 2.6.5-7.162 (SLES9 sp2 beta1)

aachba.c+linit.c+aacraid.h+commsup.c+comminit.c+dpcsup.c:
	- Merge differences in the 2.6.12-rc2 kernel.org branch of the
	  driver.

linit.c:
	- Added ICP9085LI & ICP5085AU
	- Modified support for ASR4810SAS

aacraid.h+commsup.c:
	- If ROMB heritage adapter, enable printf

Version: 1.1.5-2396 (Avon Park SIT)

build:
	- Added support for 2.6.11.4-20a (SUSE 9.3)
	- Added support for 2.6.9-5.0.3.EL-2.6 (CentOS4) to driver disks
	- Added support for 2.6.9-5.0.5.EL (RHEL4 Errata 5) to
	  modules.equiv
	- Added support for 2.4.21-27.0.2.EL-2.4 (CentOS3) to driver
	  disks

linit.c+aacraid.h:
	- Merge differences in the 2.6.12-rc2 kernel.org branch of the
	  driver.

linit.c:
	- str() did not do the right thing, needed to nest macros.

aachba.c+aacraid.h+commctrl.c+comminit.c+commsup.c+dpcsup.c+linit.c: Mark Haverkamp <markh@osdl.org>
	- Remove sparce warnings

readme.txt(dkms):
	- Added strings for 'ips' driver
	- Added documentation on how to make a 'suse' driver disk

Version: 1.1.5-2397 (This is a code stream identifier)

build:
	- sort distributions when code has to cut the products into
	  pieces.

commctrl.c:
	- byte_count is converted from le32 to cpu, then again when used
	  in the following line. Dropped the second le32_to_cpu call. No
	  side effects in an LE machine.

linit.c:
	- MODULE_VERSION limited to 2.6.4 and higher

Version: 1.1.5-2398 (Avon Park SIT)
Version: 1.1.5-2399 (This is a code stream identifier)

build:
	- Added linked equivalent entries for multi-OS driver disk
	  images in driverdisks tarball
	- dkms package versioning is ${VERSION}.${REVISION}.${BUILD} as
	  required by DKMS and changed the tarball package name to
	  ${VERSION}.${REVISION}-${BUILD} to match the other build
	  product names. Adjusted dkms documentation to match.
	- Added support for 2.4.21-27.0.4.EL-2.4 (RHEL3 QU4 Errata 4) to
	  modules.equiv

install.sh
	- RHEL3, RHEL4, FC2 and FC3 all can add a -2.4 or -2.6 into the
	  kernel name that does not match the /lib/modules/kernel names.

Version: 1.1.2-2400 - 1.1.2-lk2 + LARGE_FIB patch
Version: 1.1.5-2400 (Enzo)

build:
	- Added support for 2.6.5-7.104 (SLES9 errata) to module.equiv
	- Added support for 2.6.5-7.108 (SLES9 errata)
	- Added support for 2.4.21-169 (SLES8 errata)
	- Added support for 2.4.21-190 (SLES8 errata) to module.equiv
	- Added support for 2.4.21-192 (SuSE 9 errata) to module.equiv
	- Added support for 2.4.21-196 (SLES8 errata) to module.equiv
	- Added support for 2.4.21-198 (SLES8 errata) to module.equiv
	- Added support for 2.4.21-199 (SuSE 9 errata)
	- Added support for 2.4.21-202 (SuSE 9 errata) to module.equiv
	- Added support for 2.4.21-207 (SLES8 errata) to module.equiv
	- Added support for 2.4.21-215 (SLES8 errata) to module.equiv
	- Added support for 2.4.21-226 (SuSE 9 errata) to module.equiv
	- Added support for 2.4.21-238 (SuSE 9 errata) to module.equiv
	- Added support for 2.4.21-280 (SLES8 errata) to module.equiv
	- Added support for 2.6.8-24.3 (SuSE 9.2 errata 3)
	- Added support for 2.4.9-e.62 (RHAS2.1 QU7)
	- Switched support for 2.4.19-340 to module.equiv
	- 2.6.5-7.162 had the wrong .config entries, CONFIG_RELEASE="0"
	  instead of "7.162"

Version: 1.1.5-2400 (This is a code stream identifier)

aachba.c+linit.c: James Bottomley <James.Bottomley@SteelEye.com> & Mark Haverkamp <markh@osdl.org>
	- Use blk_queue_max_segment_size() instead of 2.4 kernel fix where
	  we split up SG elements if they exceed 64KB for the older adapters.

build:
	- Added support for 2.4.21-32.EL (RHEL3 QU5)

Version: 1.1.5-2401 (Enzo)

aachba.c:
	- Removed aac_spin* macros, they are no longer needed and reduce the
	  differences between the kernel.org and Adaptec branch of the driver.

linit.c: Mark Haverkamp <markh@osdl.org>
	- Dropped aac_host_version from /sys entry names (/sys/modules/aacraid
	  is a duplicate of this information).
	- Dropped aac_ from /sys entry names
	- Dropped supporting wrapping text from /sys entry contents

aachba.c+aacraid.h+commctrl.c+comminit.c+dpcsup.c+rkt.c+sa.c+linit.c:
	- Back merged cosmetic changes from 2.6.12-rc4-git2

linit.c:
	- Move DRIVER_FULL_VERSION down so that parameters in aacraid.h
	  can be picked up if necessary.

install.sh:
	- Added support for --mail option.

build:
	- Added report in README for unusual build flags.
	- SL9.3 build was broken, needed an rpm-version file placed in the
	  TOPDIR.

Version: 1.1.5-2401 (This is a code stream identifier)

build:
	- Added support for 2.4.21-32.0.1.EL-2.4 (RHEL3 QU5 errata 1)
	- Added support for 2.6.11-1.27_FC3-2.6 (FC3 errata 27)
	- Alterred build of 2.6.11.4-20a (SUSE 9.3)
	- Alterred build of 2.6.9-6.37.EL (RHEL4 U1 beta)

install.sh:
	- SLES9, SL9.3 (and probably others) switched from mk_initrd to
	  mkinitrd using 'SuSE' parameters. Added a check to see if the
	  /sbin/mkinitrd had "i).*initrd" and "k).*kernel" contained in
	  the script to differentiate.

commsup.c:
	- When an array is clearing, we issue a scan when we should be
	  issuing a delete in the 2.6 stream (ADPmp85455).
	- When the array clear has completed, we have discovered a means to
	  add the device for kernels between 2.6.3 and 2.6.10 (ADPmp85455).

commsup.s+aacraid.h:
	- We added additional code to determine when the container zero
	  task starts, and issue a container delete when the container
	  change event occurs. We keep the original container zero detector
	  as it is also used to detect a container create. (ADPmp85455)

Version: 1.1.5-2402 (This is a code stream identifier)

commctrl.c:
	- send raw srb has some return code -1, that should be replaced
	  with ENOMEM and ENXIO to be more descriptive. -1 is
	  equivalent to indicating EPERM.
	- used the outgoing swapped value for sg count in 32 bit path
	  rather than the user value (ADPmp85466)

aacraid.h+linit.c:
	- Added AAC_DEBUG_INSTRUMENT_PENDING which displays the number
	  of commands outstanding to the adapter in both the /proc and
	  /sys interface points.

compat.h:
	- Recognize that scsi_remove_host should restore the 'aac'
	  value (ADPmp85994).

linit.c+Makefile:
	- If scsi_in_detect is present in hosts.c, then we need to
	  adjust the code that 'hacks' 2.4 functionality. This affects
	  RHEL3 QU2 and up only (ADPmp85994+). This one was a mind-
	  bender where they broke support for allocation of adapter
	  scsi resources before the registration call as a scsi module.

compat.h+Makefile:
	- If scsi_scan_host is exported in scsi_syms.c, then we need to
	  use it to improve the timeliness of delivery.

aachba.c:
	- aac_driver_version is now a character string, compatibility
	  report needs to be adjusted.

install.sh:
	- Added support to recognize other versions of em64t
	  processors, Genuine Intel processors. (ADPmp86028)

Version: 1.1.5-2403 (This is a code stream identifier)

aacraid.h:
	- Remove AAC_DEBUG_INSTRUMENT_PENDING

Version: 1.1.5-2403 (Enzo)

linit.c:
	- monitor_version and bios_version should be hba_monitor_version
	  and hba_bios_version respectively.

build:
	- Added support for 2.6.9-11.EL-2.6 (RHEL4 U1)
	- Removed support for 2.6.7-1.451.2.3 (RHEL4 alpha)
	- Removed support for 2.6.8-1.602 (RHEL4 beta)
	- Removed support for 2.6.8-1.648_EL (RHEL4 beta)
	- Removed support for 2.6.9-1.906_EL (RHEL4 beta)
	- Removed support for 2.6.9-6.37.EL_2.6 (RHEL4 U1 beta)

aachba.c+aacraid.h+csmi.c+linit.c:
	- Merged in 2.6.12-rc6-git8

commctrl.c:
	- cdb was not transfered when sending raw srb commands.

Version: 1.1.5-2404 (Enzo)

build:
	- RHEL3 U5 2.4.21-32.EL was erroneously versioned as 2.4.21-32.EL-2.4

Version: 1.1.5-2405 (Enzo)

compat.h+csmi.c+linit.c:
	- Incorrect expectations for list_for_each_entry usage, the loop
	  'entry' variable is undefined when outside the loop. Only
	  experienced side effect is a failure to load/unload the driver
	  module in the 2.4 series of the kernels (ADPmp85994). CSMI is
	  affected, but no reported problems.

compat.h:
	- 2.4.21-17.EL+ has scsi_scan_host(), but does not have the
	  corresponding scsi_remove_host() call! We construct one that
	  issued "scsi remove-single-device" calls to proc
	  (ADPmp85994).

aachba.c+linit.c:
	- move probe_container out of the queuecommand path under the
	  2.4 based timesys kernels. They can not handle a 'blocking'
	  function, especially one that requires an interrupt response
	  before returning to the caller.

build:
	- CHANGELOG inclusion in source is now parameterized
	  (--nochangelog).  Subsequent Enzo or AvonPark builds will
	  set this flag.

Version: 1.1.5-2404 (This is a code stream identifier)

build:
	- Added support for 2.6.5-7.187 (SLES9 sp2 rc3)
	- Removed driver disk support for 2.6.5-7.162 (SLES9 sp2 beta1)
	- Added support for 2.6.11-1.35_FC3_2.6 (FC3 errata)
	- Added support for 2.6.11-1.1369_FC4 (FC4 Gold)
	- Added support for GCC4 cross/native compile suite.
	- set ARCH= for all cases, just in case build machine is hosted
	  on a non-i386 environment.
	- To support a SLES9 build machine, we need to set LC_ALL and
	  LANG.

aachba.c:
	- probe_container call was not ifdef'd out properly for
	  CONFIG_TIMESYS.

aachba.c+commsup.c+linit.c+csmi.c+aacraid.h
	- merge scsi-misc-2.6.git cosmetics
	- remove aac_atoi() and replace with simple_strtol()

install.sh:
	- set LC_ALL to C.

Version: 1.1.5-2405 (This is a code stream identifier)

build:
	- Fixed permission issues with creation of iso driver disks that
	  affected the SuSE portion of the image.

comminit.c+aachba.c+linit.c:
	- Use NEW_COMM flag to switch between 64K scatter gather element
	  limits or not. The SGBCOUNT condition has outgrown it's
	  usefullness.
	- Dropped SGBCOUNT to 4096 from 8192 as we managed to cause the
	  2810 to drop out as the WD SD drives can not handle such large
	  requests.

aachba.c:
	- Adjusted white space to match better with scsi-misc-2.6.git

commctrl.c:
	- In the 64 bit data path of the raw srb handler, the scatter
	  gather data was transferred to the user side rather than the
	  fib side of the equation (ADPmp87218)

Version: 1.1.5-2406 (This is a code stream identifier)
Version: 1.1.5-2406 (Enzo)

aacraid.h: Martin Drab <drab@kepler.fjfi.cvut.cz>
	- Dropped AAC_MAX_32BIT_SGBCOUNT to 512, some field
	  instability showed up on the adapters.

build:
	- Added support for 2.6.5-7.155.29 (SuSE 9.1 errata)
	- Added support for 2.6.5-7.191 (SLES9 SP2)
	- Added support for 2.6.12-1.1387_FC4-2.6 (FC4 errata)
	- Added support for 2.6.12-1.1390_FC4-2.6 (FC4 errata)

Version: 1.1.5-2407 (Enzo)
Version: 1.1.5-2407-1 FSC branch with first two fixes (scsi_bios_ptable
	NULL, and hot action break from loop) of 1.1.5-2410.

build:
	- Added object module checking to build script to ensure that
	  fiasco resulting from a full build done in a dirty environment
	  does not produce a fubar result. More changes or policy to
	  come to control the prominence of the warning message(s)
	- When confronted with a kernel that is not configured as part
	  of the build, make note, declare an error, but continue.
	- Added 2.6.5-7.191 (SLES9 sp2) to the driver disk.
	- Removed driver disk support for 2.6.5-7.162 (SLES9 sp2 beta1)
	- Removed driver disk support for 2.6.5-7.187 (SLES9 sp2 rc3)
	- Fixed SL9.3 driver disk creation, neglected to add -20a to
	  kernel names to be added to this driver disk.
	- Added 2.6.12-1.1372_FC3 (FC3 errata)
	- Added 2.6.12-1.1398_FC4 (FC4 errata)
	- Added 2.6.11.4-21.7 (SL9.3 errata)
	- Added 2.6.9-7.2.EL (RHEL4 hot fix)
	- Removing 2.4.19-kdb

linit.c+README+readme.txt.orig+pciids.sourceforge.net
	- Added IBM ServeRAID 8k/8k-l4 and 8k/8k-l8 product IDs.
	- Added ICP9067MA product IDs

linit.c: (Christoph Hellwig <hch*lst.de>
	- Remove scsi_cmnd->state

linit.c+rx.c+rkt.c+sa.c:
	- Move thread start out of the adapter initialization routines
	  and certainly after adapter startup so that we don't get any
	  time updates before the adapter is considered ready. Currently
	  the thread blocks this from occurring by not setting the time
	  until a second or more has elapsed. This is not considered a
	  reliable fence.

linit.c:
	- ICP9085LI & ICP5085BR used the rkt interface instead of the rx
	  interface (ADPmp87781)
	- 8k and ASR4810SAS used the rx interface instead of the rkt
	  interface.

Version: 1.1.5-2407 (This is a code stream identifier)
Version: 1.1.5-2407 (AvonPark)
Version: 1.1.5-2408 (Enzo)

aachba.c+comminit.c: Mark_Salyzyn@adaptec.com & Nagendra_Tomar@adaptec.com
	- replace sgmap with sgentry when determining the maximum sg
	  entries to fit in a FIB.

aachba.c: James Bottomley <James.Bottomley@SteelEye.com>
	- Incorrect used of scsicmd->timeout - jiffies, use
	  scsicmd->timeout_per_command instead.

commctrl.c:
	- remove unnecessary unlocks surrounding kfree calls.
	- kmalloc call has arguments reversed, result was to
	  allocate a 224 byte structure which currently is 'big
	  enough' for most command requirements for the srb command.
	- If a LARGE_FIB is larger than the negotiated size and larger
	  than 2K, then return EINVAL

install.sh:
	- Added support to permit /lib/modules/2.4.21-231-SMP
	  and /lib/modules/2.4.21-231 to be handled by the installer.

build:
	- Added 2.6.5-7.193 (SL9.1 update)
	- Added 2.6.8-24.17 (SL9.2 update)

linit.c:
	- switch to using aac_shutdown instead of aac_reboot_event.
	- return -ENODEV if the adapter count is 0.
	- No need to pre-populate device structures with probe_container
	  calls before issuing the scan, as this is done by the
	  aac_get_containers() call.
	- replaced using 'serial_number' with command->request
	  and command->request->rq_status != RQ_INACTIVE

rx.c+rkt.c+linit.c+commsup.c
	- Started using IOP_RESET.

commsup.c:
	- Added a check for fsa_dev as it can be in an uninitialized state
	  at the time aac_handle_aif() is called.

aachba.c+commsup.c
	- Added update_interval and check_interval for time sync and
	  check health to make them tunable.

sa.c:
	- sa_writew(,,cpu_to_le16(0xffff)) should be sa_writew(,,0xffff)

linit.c+aachba.c:
	- Support no_uld_attach in lieu of the traditional blocking of
	  Inquiry responses for kernels that support this flag (2.6.10+).
	  enforce data protect for RAID components. This also affects
	  back ports, which currently includes SLES9 SP2 and SLES9 updates.

aachba.c: Mark Haverkamp <markh@osdl.org>
	- Fix aacraid probe breakage in scsi-block-2.6

Version: 1.1.5-2407 (IBM Jaguar SIT)

linit.c+aachba.c+commsup.c:
	- Drop references to cmd->owner & cmd->state and duplicate the
	  functionality in cmd->SCp.phase
	- split out from aac_check_health() the aac_reset_adapter()
	  routine and add a "reset_host" sys attribute to call it.
	  This same entry point returns adapter status.

linit.c:
	- In the proc handler,
		echo reset_host >/proc/scsi/aacraid/?
	  will trigger a reset of the host adapter for legacy kernels
	  using the new aac_reset_adapter() handler.

commsup.c:
	- call scsi_block_request()/scsi_unblock_request() instead of
	  relying purely on the in_reset command queue kickback. Also
	  released the host_lock early so that we do not have tasks
	  blocked (we did manage to get a CPU? hang complaint on
	  weekend testing.
	- 32 bit math errors when dealing with check_interval (ADPmp90870).

linit.c+Makefile:
	- Add support for disk dump feature in RHEL3.

Version: 1.1.5-2408 (IBM Jaguar SIT)

aachba.c:
	- use aac->max_fib_size rather than sizeof(struct hw_fib) in
	  the BUG statement because if the Adapter negotiates a larger
	  fib, you *will* get a panic. This is not an issue for any
	  current driver from Adaptec or from the scsi-misc-2.6 tree,
	  but since the raw I/O code did not propagate to 2.6.13 yet, it
	  is a serious issue there.

install.sh:
	- Did not accommodate module.o & module.ko entries in the
	  RH driver disk format (ADPmp90502)
	- One can not trust a break in a case statement (ADPmp90870)

build:
	- Add 2.6.11.4-21.8 (SL9.3 update)
	- Add support for CentOS 4.1
	- Add support for the generation of an RHEL4, FC3, FC4, CentOS
	  CDROM driver disk.

commsup.c:
	- The AifDen* packets should result in immediate driver action
	  (ADPmp90701)

Version: 1.1.5-2408 (This is a code stream identifier)

commsup.c:
	- user submitted IOP_RESET should perform a VM_CloseAll (flush
	  cache) before submitting command. Unwind if IOP_RESET is not
	  supported.

aacraid.h+aachba.c:
	- If the SlotNumber in the supplement information is unknown,
	  then set the slot number from the Linux concept of slot
	  number and inform Adapter of the setting (ADPmp90861).
	  Should be set by the BIOS normally, but not all Linux systems
	  have a BIOS!

aachba.c+commsup.c:
	- Merged current scsi-misc-2.6 cosmetic changes.

build:
	- x86_64 SuSE dud template is not being transferred.

Version: 1.1.5-2409 (This is a code stream identifier)
Version: 1.1.5-2409 (IBM Jaguar SIT)

linit.c: Martin Wilck <martin.wilck@fujitsu-siemens.com> & Mark_Salyzyn@adaptec.com
	- scsi_bios_ptable can return a NULL pointer, if memory is
	  exhausted or if someone created a zero sized array.
	  ADPmp91471

commsup.c: Martin Wilck <martin.wilck@fujitsu-siemens.com> & Mark_Salyzyn@adaptec.com
	- break of device loop once one is discovered so that we are
	  not presented with a borken device list following one of
	  the hot actions. ADPmp91471

aachba.c: Ken Sandars <Ken_Sandars@adaptec.com>
	- scsicmd->cmnd[9] << 32 should be scsicmd->cmnd[5] << 32

aachba.c:
	- Ifdef SlotNumber setting, decided that BIOS and Management
	  needs to get their act together.

commctrl.c:
	- If the aac_command_thread appears to have died, resurrect it!
	  ADPmp91208.

build:
	- Add 2.6.0-test5_2 (TL 10 Desktop boot)
	- Add 2.6.0-6 (TL 10 Desktop install)
	- Add 2.6.0-24 (TL 10 Desktop update)
	- Add 2.6.8-1 (TL 10 Server)
	- Add 2.6.8-5 (TL 10 Server update)

Version: 1.1.5-2410 (This is a code stream identifier)
Version: 1.1.5-2410 (IBM Jaguar SIT)

build:
	- Add 2.4.21-37.EL (RHEL3 U6)
	- Removed CODE_STREAM_IDENTIFIER from channel build
	- Fixed 2.4.21-37.EL (RHEL3 U6) so that it would land on driver
	  disk
	- Add 2.4.22-1.2115.nptl (FC)
	- Add 2.4.22-1.2199.nptl (FC errata)

aachba.c+commsup.c+linit.c:
	- Merge scsi-rc-fixes-2.6

commctrl.c+Makefile:
	- Can not use find_task_by_pid universally, not all kernels
	  export the symbol for use by the modules ADPmp91208 &
	  ADPmp91941.

aachba.c: Torry Hoffman <torry@lockdownnetworks.com>
	- ifdef out variables i and count as they are unused.

Version: 1.1.5-2411 ()
Version: 1.1.5-2411 (Avon Park SIT)
Version: 1.1.5-2411 (IBM Jaguar SIT)

build:
	- README.txt now reports the 2.4.19 sub-version on SuSE
	  kernels incorporated into the driver disks.
	- Add 2.6.9-22.EL (RHEL4 U2)
	- Add 2.6.5-7.224 (SuSE SP3 beta2)

linit.c:
	- 2.4.2 and 2.4.4 kernels dropped off of build due to
	  usage of snprintf. Continued pedantic build checking
	  of the driver in multiple kernels only reason for
	  repair of this code.

aacraid.h: Juan D Ch <jchimienti@most.com.ar>
	- Dropped AAC_MAX_32BIT_SGBCOUNT to 256, some field
	  instability showed up on the adapters.

commsup.c: John Hanson <John_Hanson@adaptec.com>
	- When we requeue a container for later processing, lets move on
	  down the list and check if another could be done in the
	  current cycle ADPmp91038.

Version: 1.1.5-2412 ()
Version: 1.1.5-2412-IBM () ADPmp92508

commctrl.c:
	- Only resurrect the aac_command_thread if the resources are
	  available ADPmp92417

commsup.c: Jesper Juhl <jesper.juhl@gmail.com>
	- Removed all if (ptr) kfree(ptr) style calls.

Version: 1.1.5-2413-IBM ()

build:
	- add SCSI_HAS_DUMP* checking in the build scripts to complement
	  those picked up by the Makefile. This is useful for the ips
	  and dpt_i2o drivers.
	- Added 2.6.11.4-21.9 SuSE 9.3 errata kernel.
	- Various SuSE Pro releases dropped the bigsmp driver because
	  the build script specified *-bigsmp-ia32 instead of
	  *-bigsmp-i586

Version: 1.1.5-2413 ()

build:
	- Added 2.6.13-3 SuSE 10.0 pro
	- Added 2.6.13-9 SuSE 10.0 pro

compat.h:
	- include linux/dma-mapping.h ahead of the MASK definitions to
	  ensure that we do not enter into a situation where the
	  definitions differ from what the driver utilizes. Purely a
	  warning issues as the MASK does not differ at this time.

compat.h+aachba.c: Jeff Garzik <jgarzik@pobox.com>
	- replace scsicmd->device->id and scsicmd->device->channel with
	  scmd_id(scsicmd) & scmd_channel(scsicmd) respectively.
	- replace sdev->id with sdev_id(sdev);

commctrl.c: Mark_Salyzyn@adaptec.com & Keith_Conner@adaptec.com
	- aacraid thread restart needs to occur with the fib_lock
	  released.

aachba.c+aacraid.h+commctrl.c+comminit.c+commsup.c+csmi.c: Mark_Salyzyn@adaptec.com & Olivier Kaloudoff <kalou@kalou.net>
	- changed to aac_fib_init defined and used in network code.

aachba.c+commctlr.c+comminit.c:
	- changed over from pci_map_single to pci_map_page.

aachba.c:
	- highmem_io was not set when using raw_io.

commsup.c: Christoph Hellwig <hch@lst.de>
	- replace Scsi_Device with scsi_device in comments to more
	  accurately reflect the naming of the structure.

commsup.c:
	- Merged changes from scsi-misc-2.6

Version: 1.1.5-2414 ()

commsup.c: John_Hanson@adaptec.com
	- perform the aac_probe_container after busy check, but before
	  scan request to the OS.

comminit.c+aachba.c+linit.c+dpcsup.c compat.h commsup.c aacraid.h
	- Merged in changes from VMWARE

comminit.c+aachba.c+linit.c+dpcsup.c+commsup.c+aacraid.h+csmi.c+commctrl.c
	- fib_adapter_complete+fib_init+fib_alloc+fib_send+
	  fib_complete+fib_free+fib_map_free+fib_setup+probe_container
	  all have a prefix of aac_ to prevent name space collision when
	  the driver is compiled into the kernel.

aachba.c+linit.c+rx.c+rkt.c+README
	- Merged changes from scsi-misc-2.6

aachba.c: Ken Sandars <ken_sandars@adaptec.com>, John Hanson <John_Hanson@adaptec.com> & Mark_Salyzyn@adaptec.com
	- SAI_READ_CAPACITY_16 compliance with sbc2r16

build:
	- Added 2.6.13-15 SuSE 10.0 pro

Version: 1.1.5-2415 ()

build:
	- Added 2.6.12-1.1378_FC3-2.6
	- Added 2.6.12-1.1380_FC3-2.6
	- Added 2.6.12-1.1381_FC3-2.6
	- Added 2.6.12-1.1456_FC4-2.6
	- Added 2.6.13-1.1526_FC4-2.6
	- Added 2.6.13-1.1532_FC4-2.6
	- Added 2.6.9-22.8.EL

aacraid.h+linit.c+commsup.c:
	- Dell CERC card performs poorly when more than 17 sg elements
	  are passed into the CHIM.

Version: 1.1.5-2415-IBM ()

aachba.c+commsup.c:
	- revert change introduced to move aac_probe_container after
	  busy check and instead decide not to take targets offline if
	  they happen to be in the aac_probe_container state.

build:
	- Updated DKMS to 2.0.8

aachba.c+commctlr.c+comminit.c+commsup.c+compat.h:
	- pci_map_page was rejected by the community, code pulled.

linit.c:
	- scsi_bios_ptable return check is not in coding style.
	- set skip_ms_page_8 to soften read cache page warning.

linit.c+aachba.c+compat.h: Mark_Salyzyn@adaptec.com & Christoph Hellwig <hsh@infradead.org>
	- Used CONTAINER_CHANNEL when matching the virtual channel 0
	- use sdev_channel & sdev_id

build:
	- Add 2.4.9-e.62 to modules.equiv
	- Add 2.4.9-e.65 to modules.equiv
	- Add 2.6.14-1.1637_FC4-2.6 (FC4 errata)
	- Add 2.6.14-1.1644_FC4-2.6 (FC4 errata)
	- Add 2.6.9-22.0.1.EL-2.6 (RHEL4 errata)
	- Corrected configuration of 2.6.5-7.224 (SuSE SP3 beta2)
	- Add 2.6.5-7.236 (SuSE SP3 RC1)
	- Add 2.6.5-7.243 (SuSE SP3 RC3)
	- Add 2.6.5-7.244 (SLES9 SP3)
	- Add 2.6.13-15.7 (SuSE 10 errata update)
	- Add 2.6.5-7.224 as a module.equiv of 2.6.5-7.243
	- Add 2.6.5-7.236 as a module.equiv of 2.6.5-7.243
	- Add 2.6.5-7.193 as a module.equiv of 2.6.5-7.191

aachba.c+rx.c+rkt.c+sa.c:
	- Added support for an insmod parameter startup_timeout. The
	  default value is set to 3 minutes for channel, 9 minutes for
	  external.

Version: 1.1.5-2416 ()

linit.c:
	- set skip_ms_page_3f to soften write enable warning.

aachba.c+commsup.c:
	- Added support for an insmod parameter aif_timeout. The default
	  value is set to 2 minutes for channel, 9 minutes for external.

Version: 1.1.5-2416-IBM ()

build:
	- Removed 2.6.5-7.224 (SLES9 sp3 b2)
	- Removed 2.6.5-7.236 (SLES9 sp3 rc1)
	- Removed driver disk support for 2.6.5-7.243 (SLES9 sp3 rc3)
	- Added 2.6.9-22.EL i586 configuration (CentOS 4.2 i386 desktop)
	- Added 2.6.9-27.EL (RHEL4 U3 rc)

compat.h+linit.c: M.Gehre <M.Gehre@gmx.de>
	- Added DMA_31BIT_MASK

Version: 1.1.5-2417 ()

Makefile: Martin Wilck <martin.wilck@fujitsu-siemens.com> & Mark_Salyzyn@adaptec.com
	- SCSI_HAS_SCSI_SCAN_HOST should check for the presence of
	  scsi_scan_host in the file.

build:
	- Moved ia64 and ppc architectures out of RHEL4 and CentOS disks
	  and moved them into RHEL4-enterprise and CentOS-enterprise.
	- Added support for reporting architectures in README.txt.
	- Split sles9-sp's into their own sets, they were mixing with
	  u11-sp's before.
	- Adjusted Driver Disk ISO images, moved 2.6 RH kernels into the
	  SuSE set, moving 2.4 RH kernels on their own. Added sl10. Have
	  *not* dealt with sles9-sp's as they each require their own
	  ISOs and will wait until the fan hits ...
	- Added 2.6.14-1.1653_FC4 (FC4 errata)
	- Added 2.6.14-1.1656_FC4 (FC4 errata)
	- Added 2.6.11.4-21.10 (SuSE 9.3 errata)
	- Added 2.6.9-22.25.EL

commctrl.c:
	- check the in_reset flag before attempting to restart the AIF
	  thread.

linit.c:
	- Added /sys/class/scsi_host/host?/max_channel
	- Added /sys/class/scsi_host/host?/max_id
	- max_channel is '0' based, not 1 as it is currently treated.
	- for kernels from 2.6.7 to 2.6.9 use inq_periph_qual as an
	  equivalent to no_uld_attach.
	- for kernels from 2.6.0 to 2.6.7 use inquiry[0] >> 5 as an
	  equivalent to no_uld_attach.
	- for kernels before 2.6.0 use type >> 5 as an equivalent to
	  no_uld_attach.

aachba.c:
	- Added support for the expose_physicals insmod flag. This flag
	  sets nondasd & turns off r/w protection for the physicals.
	- for kernels from 2.6.7 to 2.6.9 use inq_periph_qual as an
	  equivalent to no_uld_attach.
	- for kernels from 2.6.0 to 2.6.7 use inquiry[0] >> 5 as an
	  equivalent to no_uld_attach.
	- for kernels before 2.6.0 read and write protect the drives.

Version: 1.1.5-2418 ()

build:
	- Add 2.6.9-32.EL (RHEL4 U3 beta)
	- Add 2.6.13-15.8 (SuSE 10 errata update)
	- Add 2.4.21-4.17.EL (RHEL3 Gold 64 bit)
	- Add 2.4.21-37.18AX (Miracle Linux 3)
	- Add 2.6.5-7.252 (SLES9 update)
	- Add 2.6.11.4-21.11 (SuSE 9.3 update)
	- Add 2.6.15-git12-6 (SLES 10 beta 1)

aachba.c:
	- Data protect WRITE_LONG, WRITE_VERIFY, MODE_SELECT and
	  LOG_SELECT.

commsup.c:
	- added lock_kernel/unlock_kernel around call to write_proc.

aacraid.h+comminit.c+commsup.c+linit.c: Christoph Hellwig <hch@lst.de>
	- use kthread_API

commctrl.c:
	- use kthread_API

aachba.c: Takashi Iwai <tiwai@suse.de>
	- Fix the comparison of sizeof()

Version: 1.1.5-2419 ()

build:
	- Add 2.4.21-37.EL (RHEL3 U7 beta)
	- Add 2.6.9-22.0.2.EL-2.6 (RHEL4 errata)

linit.c+commsup.c+aachba.c+comminit.c:
	- Merge cosmetic changes from scsi-misc-2.6

linit.c+commsup.c+comminit.c+commctrl.c+aacraid.h
	- kthread_API only invoked past 2.6.5-7.162 ADPmp95987

commsup.c:
	- Bumped up the AIF timeout to 30 seconds, the adapter could
	  introduce delays in processing when in lockdrive. ADPmp94961?

Version: 1.1.5-2420 ()

build:
	- Add 2.6.9-34.EL (RHEL4 U3)

Version: 1.1.5-2419-IBM ()

build:
	- Add 2.4.21-40.EL (RHEL3 U7)

commsup.c+commctrl.c+linit.c:
	- Cosmetic changes merged from scsi-misc-2.6

build:
	- Split RHEL4 & CentOS driver disks into ia32 and x86_64 parts
	- Dropped RHEL3 beta kernels 2.4.21-17.EL, 2.4.21-31.EL,
	  and 2.4.21-38.EL from the RHEL3 driver disks.

Version: 1.1.5-2420-HP ()
Version: 1.1.5-2420-IBM ()

build:
	- Add 2.6.15-1.2054_FC5 (FC5 Gold)
	- Add 2.4.21-37.ELvmnix (VMWare 3.1)
	- Add 2.4.9-e.65 (RHAS2.1 errata 65)
	- Add 2.4.9-e.68 (RHAS2.1 errata 68)
	- Add 2.4.9-e.68 to modules.equiv
	- Dropped 2.4.9-e.62 products (already in modules.equiv)

linit.c:
	- 32 bit applications in 64 bit space are not handled correctly
	  in kernels above 2.6.10 for acquisition of AIFs ADPmp95861 &
	  ADPmp95859.

aacraid.h+linit.c+Makefile: Nobuhiro Tachino <ntachino@redhat.com>
	- one can not ssleep or msleep during disk dump, so we call our
	  own versions to differentiate and use mdelay instead during.

aacraid.h+aachba.c+commctrl.c+commsup.c+linit.c:
	- Merge in scsi-misc-2.6

Version: 1.1.5-2421

build:
	- Add 2.6.16-rc5-git9-2 (SLES10 rc7)

comminit.c: Andreas Mohr <andi@rhlx01.fht-esslingen.de>
	- spelling correction.

aacraid.h+linit.c+Makefile:
	- Added support for diskdump under 2.6.16-rc5-git9-2.SuSE kernel

Version: 1.1.5-2422

build:
	- Add 2.6.15-1.2054_FC5 (FC5)
	- Fix 32 bit compile of 2.6.15-git12-6 (SLES10 beta 1)
	- Fix 32 bit compile of 2.6.16-rc5-git9-2 (SLES10 beta 7)

commsup.c+linit.c:
	- moved locking around to fix problems with user initiated reset

aachba.c+commsup.c+linit.c:
	- Added a parameter to aac_get_config_status to request that the
	  CT_COMMIT_CONFIG be issued. This is to ensure that the
	  BlinkLED recovery always does a commit config. This change may
	  morph as we get feedback from the external products.

commsup.c:
	- block scsi commands & wait for commands to flush before
	  issuing the VM_CloseAll in the user initiated reset handler.

install.sh:
	- 'sort +2' is deprecated in FC5. Replace with 'sort -k 2'.

Makefile:
	- Dropped support for picking up linux/dumplib.h, the library
	  did not appear to be present during install ADPmp96886. This
	  reverts a change provided by RedHat by Nobuhiro Tachino
	  <ntachino@redhat.com>

aachba.c+commsup.c+aacraid.h:
	- Dropped fib queue list, is not being used.

Version: 1.1.5-2423

README+linit.c:
	- Hurricane8/4810SAS becomes Huricane44/3800SAS

aachba.c:
	- enabled aacraid_setup to function from the kernel command
	  line.

Version: 1.1.5-2424

aachba.c+commsup.c: Alexy Dobriyan <adobriyan@gmail.com> and Eric Sesterhenn <smakebyte@qmx.de>
	- BUG_ON conversion

linit.c:
	- clear_user returns 0 on success, not number of bytes zero'd
	  [ADPmp95861]

Version: 1.1.5-2425

build:
	- Add 2.4.21-40.6.EL (RHEL3 U8 beta)
	- Add 2.4.21-41.EL (RHEL3 U8 beta)
	- Add 2.4.21-43.EL (RHEL3 U8 beta)
	- Add 2.6.15-1.1830_FC4 (FC4 errata)
	- Add 2.6.15-1.1831_FC4 (FC4 errata)
	- Add 2.6.15-1.1833_FC4 (FC4 errata)
	- Add 2.6.16-1.2069_FC4 (FC4 errata)
	- Add 2.6.16-1.2096_FC4 (FC4 errata)
	- Add 2.6.16-1.2096_FC5 (FC5 errata)
	- Add 2.6.16-1.2107_FC4 (FC4 errata)
	- Add 2.6.16-1.2108_FC4 (FC4 errata)
	- Add 2.6.16-1.2111_FC5 (FC5 errata)
	- Add 2.6.9-34.19.EL (RHEL4 U4 beta)
	- Add 2.6.9-36.1.EL (RHEL4 U4 beta)
	- Add 2.6.9-36.EL (RHEL4 U4 beta)

linit.c: Rafael_Ayala@adaptec.com & Mark_Salyzyn@adaptec.com
	- diskdump_mode is not accessible in 2.4.21-27.EL (RHEL3 QU4)

Version: 1.1.5-2426

aacraid.h:
	- debugging was turned on for the 1.1.5-2426 build :-(

Version: 1.1.5-2427

aachba.c:
	- null check of aacraid variable in 2.4 ADPmp97630

commsup.c:
	- clear out count field in PAUSE_IO request.

Version: 1.1.5-2428

build:
	- Add 2.6.5-7.257 (SLES9 errata)
	- Moved 2.6.8-24.11 to module.equiv as alias to 2.6.8-24.21
	- Moved 2.6.8-24.14 to module.equiv as alias to 2.6.8-24.21
	- Moved 2.6.8-24.17 to module.equiv as alias to 2.6.8-24.21
	- Moved 2.6.8-24.18 to module.equiv as alias to 2.6.8-24.21
	- Moved 2.6.8-24.19 to module.equiv as alias to 2.6.8-24.21
	- Moved 2.6.8-24.20 to module.equiv as alias to 2.6.8-24.21
	- Add 2.6.8-24.21 (SuSE 9.2 errata)
	- Moved 2.6.11.4-21.7 to module.equiv as alias to 2.6.11.4-21.9
	- Moved 2.6.11.4-21.8 to module.equiv as alias to 2.6.11.4-21.9
	- Moved 2.6.11.4-21.10 to module.equiv as alias to 2.6.11.4-21.11
	- Add 2.6.11.4-21.12 (SuSE 9.3 errata) Not quite 100% match to 2.6.11.4-21.11, may work, risk not taken today.
	- Moved 2.6.13-15.8 to module.equiv as alias to 2.6.13-15.8
	- Add 2.6.13-15.10 (SuSE 10 errata)
	- Add 2.6.16.16-1.6 (SLES10/SLED10 RC2)

aachba.c+commctrl.c: Tobias Klauser <tklauser@neurscht.ch> & Andrew Morton <akpm@osdl.org>
	- use ARRAY_SIZE macro

aachba.c: Christoph Hellwig <hch@lst.de>
	- fix up request buffer reference

aachba.c: Tejun Heo <htejun@gmail.com>
	- add cpu cache flushes after kmapping and modifying a page

commsup.c+aacraid.h:
	- Added support for IOP_RESET_ALWAYS

Version: 1.1.5-2429

build:
	- Moved 2.6.16-1.2107_FC4-2.6 to module.equiv as alias to 2.6.16-1.2108_FC4-2.6
	- Moved 2.6.16-1.2069_FC4-2.6 to module.equiv as alias to 2.6.16-1.2096_FC4-2.6
	- Moved 2.6.15-1.1831_FC4-2.6 to module.equiv as alias to 2.6.15-1.1833_FC4-2.6
	- Moved 2.6.15-1.1830_FC4-2.6 to module.equiv as alias to 2.6.15-1.1833_FC4-2.6
	- Moved 2.6.14-1.1637_FC4-2.6 to module.equiv as alias to 2.6.14-1.1644_FC4-2.6
	- Moved 2.6.13-1.1526_FC4-2.6 to module.equiv as alias to 2.6.13-1.1532_FC4-2.6
	- Moved 2.6.12-1.1390_FC4-2.6 to module.equiv as alias to 2.6.12-1.1398_FC4-2.6
	- Moved 2.6.12-1.1387_FC4-2.6 to module.equiv as alias to 2.6.12-1.1398_FC4-2.6
	- Moved 2.6.12-1.1378_FC3-2.6 to module.equiv as alias to 2.6.12-1.1381_FC3-2.6
	- Moved 2.6.10-1.770_FC3-2.6 to module.equiv as alias to 2.6.10-1.760_FC3-2.6
	- Moved 2.6.10-1.766_FC3-2.6 to module.equiv as alias to 2.6.10-1.760_FC3-2.6
	- Moved 2.6.9-22.25.EL-2.6 to module.equiv as alias to 2.6.9-5.0.3.EL-2.6
	- Moved 2.6.5-7.243 (+aliases) to module.equiv as alias to 2.6.5-7.244
	- Moved 2.6.5-7.257 to module.equiv as alias to 2.6.5-7.244
	- Moved 2.4.21-17.EL to module.equiv as alias to 2.4.21-15.EL
	- Moved 2.4.21-38.EL to module.equiv as alias to 2.4.21-37.EL
	- Moved 2.4.21-41.EL to module.equiv as alias to 2.4.21-43.EL
	- Moved 2.4.21-243 to module.equiv as alias to 2.4.21-241
	- Moved 2.4.21-251 to module.equiv as alias to 2.4.21-241
	- Moved 2.4.21-260 to module.equiv as alias to 2.4.21-241
	- Moved 2.4.21-261 to module.equiv as alias to 2.4.21-241
	- Moved 2.4.21-273 to module.equiv as alias to 2.4.21-241
	- Moved 2.4.18-26.8.0 to module.equiv as alias to 2.4.18-24.8.0

commsup.c+comsup.c:
	- Allow a waiting aac_fib_send to be interruptible ADPmp97267.

Version: 1.1.5-2429-IBM ()

build:
	- Moved 2.6.8-24.23 to module.equiv as alias to 2.6.8-24.21 (+aliases)
	- Added 2.6.16.20-0.12 (SLES10 RC3)
	- Added 2.4.21-37.ELvmnix as alias to 2.4.21-27.ELvmnix. This
	  relationship has not been proven, added for expediency.
	- Added 2.4.21-37.0.2.ELvmnix, Compile issue remains.

comminit.c:
	- Merge kernel.org change to drop GART definition in 2.6.17+

Version: 1.1.5-2430 ()

build:
	- Added 2.6.16.21-0.8 (SLES10/SLED10)
	- Added VMware.mak

aachba.c+comminit.c+linit.c+compat.h+rx.c:
	- Merge changes from 2.4.21-37.0.2.ELvmnix/drivers/scsi/aacraid_esx30/

rkt.c+rx.c+sa.c+compat.h: Thomas Gleixner <tglx@linutronix.de>
	- scsi: Use the new IRQF_ constansts

Version: 1.1.5-2431 ()

build:
	- Added 2.6.9-42.EL (RHEL4 U4)
	- Added 2.4.21-47.EL (RHEL3 U8)

Version: 1.1.5-2431-IBM ()
Version: 1.1.5-2432 ()

build:
	- Added 2.6.16.13-4 (SuSE 10.1)
	- Added 2.6.16-1.2111_FC4 (FC4 errata)
	- Added 2.6.16-1.2115_FC4 (FC4 errata)
	- removed 2.6.16-1.2107_FC4, alias of 2.6.16-1.2111_FC4
	- removed 2.6.16-1.2108_FC4, alias of 2.6.16-1.2111_FC4
	- Added 2.6.17-1.2139_FC4, alias of 2.6.17-1.2142_FC4
	- Added 2.6.17-1.2141_FC4, alias of 2.6.17-1.2142_FC4
	- Added 2.6.17-1.2142_FC4 (FC4 errata)
	- Added 2.6.16-1.2122_FC5 (FC5 errata)
	- removed 2.6.16-1.2111_FC5, alias of 2.6.16-1.2122_FC5
	- Added 2.6.16-1.2133_FC5 (FC5 errata)
	- Added 2.6.17-1.2139_FC5, alias of 2.6.17-1.2157_FC5
	- Added 2.6.17-1.2145_FC5, alias of 2.6.17-1.2157_FC5
	- 2.6.17-1.2157_FC5 (FC5 errata)
	- removed 2.6.9-22.0.1.EL-2.6, alias of 2.6.9-22.0.2.EL-2.6
	  (RHEL4 errata)
	- removed 2.4.21-199 as alias to 2.4.21-278
	- removed 2.4.21-169 as alias to 2.4.21-278
	- removed 2.4.18-27.7.x as alias to 2.4.18-19.7.x
	- removed 2.4.18-27.8.0 as alias to 2.4.18-24.8.0
	- removed 2.4.9-38 as alias to 2.4.9-31
	- removed 2.6.11.4-21.11 as alias to 2.6.11.4-21.12
	- remove aacraid_patches, they do not appear to be relevant
	  anymore.
	- problem with 2.6.9-42.EL due to presence of localvers file in
	  configuration.

aachba.c: Mike Christie <michaelc@cs.wisc.edu>
	- Convert scsi to the block layer request timer code. The block
	  layer functions are exactly the ones that used to live in
	  scsi, so the conversion is rename.

rx.c+rkt.c:
	- Report the OMRx[0] status if the Adapter is in a Panic state
	  on init.
	- Merge rx and rkt code.

sa.c+rx.c+rkt.c+nark.c:
	- Added support for split bar, by adding an ioremap function.

aachba.c+commsup.c:
	- return EBUSY, instead of ENODEV, when we are in the middle of
	  an adapter reset. No consequence to existing management code.

commsup.c:
	- on adapter reset, make sure we close all outstanding ioctl
	  commands as well.
	- Added additional AIF instrumentation code.
	- Do not call scsi_remove_device as it leaves kref droppings.

install.sh
	- output of grep fell on floor, every rpm install adds yet
	  another hostadapter instance. Only side effect was growth of
	  /etc/modules.conf or /etc/modprobe.conf files.

Version: 1.1.5-2432-IBM ()

build:
	- problem with 2.6.9-42.EL, 2.6.9-36.1.EL, 2.6.9-36.EL,
	  2.6.9-34.19.EL, 2.6.9-34.EL, 2.6.9-32.EL, 2.6.9-27.EL
	  & 2.6.9-22.25.EL with 64 bit largesmp configurations stomping
	  all over the UP configurations. ADPml01331.

aachba.c+rx.c+commsup.c:
	- Merge from scsi-misc-2.6, cosmetics.

commsup.c:
	- Do not call scsi_remove_device as it leaves kref droppings.

Version: 1.1.5-2433-IBM ()
Version: 1.1.5-2433 ()

build:
	- 2.4.21-47.EL (RHEL3 U8) was misconfigured when placing
	  ia32e (em64t) products on the floppy disk.

Version: 1.1.5-2434-IBM ()

build:
	- Upgraded to VMware DDK v1.0
	- Altered VMware.inc, VMware.mak, VMware.spec and added
	  VMware.xml to merge new DDK changes in the asa7211 sample
	  source (still does not work on latest ESX release).
	- esxcfg-pciid needed to be made an optional artifact in the
	  rpm postinstall
	- use aacraid.xml rather than Vmware.xml for the pciids file.
	- fill out *all* the pciids in /etc/vmware/pciid/aacraid.xml.
	- accept vmware-hwdata version 0.99, the DDK's value of 1.00
	  must be for a followon OS
	- Added linux_module_heap.c to the source file list to fill in
	  for some of the linux-kernel-style API, ESX 3.0 deprecated
	  since the beta.
	- Add 2.6.8-24.24 (SuSE 9.2 errata)
	- Remove 2.6.8-24.23 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.11 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.14 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.17 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.18 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.19 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.20 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Changed 2.6.8-24.21 (SuSE 9.2 errata) as alias to 2.6.8-24.24
	- Add 2.6.11.4-21.13 (SuSE 9.3 errata)
	- Remove 2.6.11.4-21.12 (SuSE 9.3 errata) as alias to
	  2.6.11.4-21.13
	- Changed 2.6.11.4-21.11 (SuSE 9.3 errata) as alias to
	  2.6.11.4-21.13
	- Changed 2.6.11.4-21.10 (SuSE 9.3 errata) as alias to
	  2.6.11.4-21.13
	- Add 2.6.13-15.11 (SuSE 10.0 errata)
	- Remove 2.6.13-15.10 (SuSE 10.0 errata) as alias to
	  2.6.13-15.11
	- Add 2.6.5-7.267 (SLES9 errata) to module.equiv as alias to
	  2.6.5-7.244
	- Add 2.6.5-7.276 (SLES9 errata) to module.equiv as alias to
	  2.6.5-7.244
	- Add 2.6.17-1.2519.4.21.el5 (RHEL5 beta1)
	- Added gcc 4.1.1 to cross compiler set
	- Added adaptec.spec to adaptec.spec

Version: 1.1.5-2434 ()

README+linit.c+DUD:pcitable:
	- Changed product names to reflect dropping SAS post-script
	  when card supports SATA as well.

install.sh:
	- Added detection of /sbin/vmware-mkinitrd

build:
	- Moved add-on makes (VMware.* namely) to after the compile
	  section, but before the prebuilt section so that we can also
	  populate the rpm.
	- Added 2.4.9-e.70 to module.equiv file.
	- Removed 2.6.17-1.2157_FC5 & friends and added them as an alias
	  of 2.6.17-1.2187_FC5
	- Added 2.6.17-1.2174_FC5, alias of 2.6.17-1.2187_FC5
	- Added 2.6.9-42.0.2.EL (RHEL4 errata)
	- Added 2.6.9-34.0.1.EL as alias to 2.6.9-42.0.2.EL
	- Added 2.6.9-34.0.2.EL as alias to 2.6.9-42.0.2.EL
	- Corrected largesmp configurations for x86_64 and ppc64
	  environments for RHEL4
	- Corrected kdump configurations for FC5

VMware.mak:
	- Copy module products to OBJECT_DIR using appropriate names.

comminit.c:
	- Somehow, debugging left new_comm off, turned it back on.
	- Added some additional debugging for init struct
	  investigations, ifdef'd out so no side effects.

commsup.c:
	- Give a little extra time for the ioctl to complete under
	  resetiop.

commctrl.c:
	- return ioctl busy when controller is in reset state.

aacraid.h+rx.c+sa.c+commsup.c+comminit.c:
	- Added platform functions for interrupt enable, interrupt
	  handling, packet delivery and transport selection.

aacraid.h+aachba.c:
	- Added platform functions for bounds checking, read and write.

rx.c+rkt.c+nark.c:
	- Consolidate startup routines due to addition of transport
	  selection.

aachba.c:
	- Move probe_container into aac_get_containers

linit.c:
	- Added 9005/0288 family (split BAR)

Version: 1.1.5-2435 ()

linit.c+README:
	- Any references to ICP R0, should be replaced with ICP RO.

aachba.c:
	- adapter_write is never filled out (ADPml02706)

Version: 1.1.5-2436 ()

README:
	- Trimmed and Updated list of supported controllers.

build:
	- Added 2.6.18-1.2747.el5 (RHEL5 beta2)

aachba.c:
	- Added checking of the parameters fed to the
	  aac_internal_transfer function (ADPml02711)

install.sh:
	- Follow KMP guidelines and install driver into the /updates
	  directory before considering overriding as-shipped driver
	  modules.

Version: 1.1.5-2437 ()

aacraid.h+aachba.c:
	- Create _aac_probe_container as a asynchronous response
	  version of aac_probe_container.
	- Removed nested, no longer needed for _aac_probe_container.

linit.c+aachba.c:
	- expose_physicals is now in three states. Default behavior is
	  to expose protected DASD physical devices, 0 now blocks the
	  DASD devices and 1 forces full access to those physical
	  devices.

intall.sh:
	- some problems with placing modules.dep in updates/modules.dep
	- speedup dealing with KMP matching.

Version: 1.1.5-2438 ()

aachba.c:
	- expose_physicals=0 did not work for kernels 2.6.10 and higher.

commsup.c:
	- we do not deal with a case of a command aborted between
	  command setup and wait for completion. We also should perform
	  the BUG_ON check for done within an event lock? (ADPml01735)

aacraid.h+comminit.c+dpcsup.c+rx.c+release.sh+aachba.c:
	- Adding APRE transport code, ifdef'd with
	  INITFLAGS_APRE_SUPPORTED

aacraid.h+commctrl.c+commsup.c+csmi.c+dpcsup.c+rx.c
:
	- change hw_fib references to hw_fib_va to prevent namespace
	  collision with struct hw_fib.

install.sh:
	- corrected minor scripting problem.

Version: 1.1.5-2439 ()

rx.c+commsup.c:
	- Support USE_OTHER_METHOD for Sunrise Lake IOP reset.

rx.c+comminit.c+aacraid.h+linit.c:
	- Merge changes in 2.6.20 into driver, dropping pt_regs from
	  interrupt service routines.

rkt.c:
	- Merge in changes due to code inspection.

linit.c+aachba.c: Mark Salyzyn & Christoph Hellwig
	- Use slave configure returning ENXIO to block physicals for
	  2.6.15+

aachba.c+aacraid.h+comminit.c+linit.c+rkt.c+rx.c+sa.c:
	- Backmerge scsi-misc-2.6 tree, all cosmetic style issues.

aachba.c: Mark Salyzyn & Christoph Hellwig <hch@infradead.org>
	- Removed kernel command line parsing from 2.6 based kernels,
	  already implemented as <module>.<parm>=

install.sh:
	- Do not check for specifically AMD-K7 or AMD athlon processors
	  when determining athlon support, instead accept just AMD as
	  enough of an indicator (ADPml03707)

Version: 1.1.5-2440 ()

build:
	- Add 2.6.18-8.el5 (RHEL5 RC/GA)
	- Removed 2.6.17-1.2519.4.21.el5 (RHEL5 beta1)
	- Removed 2.6.18-1.2747.el5 (RHEL5 beta2)
	- Removed 2.6.16.20-0.12 (SLES10 RC3) (ADPml03802)

aacraid.h+linit.c+nark.c+rkt.c+rx.c: Adrian Bunk <bunk@susta.de>
	- Manually applied a cleanup patch that moves function externs
	  to aacraid.h rather then defined and referenced where needed.

install.sh:
	- Do not process Module.symvers variants until primary modules
	  and module.equiv variants are discovered and fleshed out
	  (ADPml03802)

Version: 1.1.5-2441 ()

build:
	- Removed 2.6.15-git12-6 (SLES10 beta 1)
	- Removed 2.6.16-rc5-git9-2 (SLES10 beta 7)
	- Removed 2.6.16.16-1.6 (SLES10/SLED10 RC2)

commctrl.c:
	- Added code to sense if the incoming frame has 64 bit sg.
	  (ADPml03719)

Version: 1.1.5-2441-IBM ()
Version: 1.1.5-2442 ()

build:
	- Added 2.4.9-e.71 to modules.equiv
	- Added 2.6.9-42.0.3.EL to modules.equiv
	- Added 2.6.9-42.0.8.EL to modules.equiv

linit.c+dpcsup.c+comminit.c+commctrl.c:
	- Switch from (unsigned long) or (long) casts to (ptrdiff_t)
	  casts wherever the intent is to convert a pointer to a
	  same sized integer or back.

comminit.c+rx.c+commsup.c:
	- Suppress compile warning messages, change kernel version
	  ifdefs to accomodate.

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig:
	- Added 9005:0285:108e:7aac for SUN RAID 4i4e
	- Added 9005:0285:15d9:02b5 for SMC AOC-USAS-S4i (ADPml04007)
	- Added 9005:0285:15d9:02b6 for SMC AOC-USAS-S8i (ADPml04007)
	- Added 9005:0285:15d9:02c9 for SMC AOC-USAS-S4iB (ADPml04007)
	- Added 9005:0285:15d9:02ca for SMC AOC-USAS-S8iB (ADPml04007)

dpcsup.c+linit.c+aacraid.h+commsup.c:
	- Set the FIB_CONTEXT_FLAG_TIMED_OUT flag for the commands that
	  are involved in error recovery; when the completion interrupt
	  happens, simply free the fib resources, do not complete the
	  command to the OS. (ADPml03975)

commsup.c+aachba.c:
	- Widen the net for capturing 'busy' to also include the
	  eh_active state flag if available. (ADPml03975)

linit.c:
	- For kernels from 2.6.0 to 2.6.12 we set the queue_depth of the
	  device to zero during the error recovery handler to trick the
	  LOWER layers to block commands. These kernels have a locking
	  bug where the error handler holds the host_lock, but commands
	  issued by user or plug-n-play scan requests manage to slip
	  through and hit the LOWER layers where host_lock is taken as
	  well *before* the host check for whether we are in error
	  recovery or not. Later kernels do not take the host_lock in
	  the error recovery (ADPml03975)
	- Added an aac_eh_abort handler, simply testing for TUR and
	  setting FIB_CONTEXT_FLAG_TIMED_OUT flag and return SUCCESS.
	  Return FAILED for any other command. TUR is simple enough
	  that the outstanding portions of the request in the Adapter
	  are inert. We could apply this to INQUIRY and READ CAPACITY
	  for the same reasons ... but chose not to at this time
	  (ADPml03975)

aachba.c:
	- If the device is offline when the completion handler is called,
	  free the FIB resources and return in srb_callback (ADPml03975)

Version: 1.1.5-2443 ()

linit.c:
	- For kernels from 2.6.0 to 2.6.12 we also set the
	  device_blocked flag during the error recovery handler to trick
	  the LOWER layers to block commands (ADPml03975)
	- When we wake up from a sleep, before retaking the locks,
	  innoculate the block I/O for kernels 2.6.0 through 2.6.12.
	  (ADPml03975).
	- The 2.4 calculation for queuedepth erroneously divided the
	  entire pool for every known target, this limits the number
	  of outstanding commands available for the logicals by too
	  much.
	- If we have a definition of crashdump_mode that we can live
	  with, let us call aac_poll during error recovery even if
	  we do not have the ability to enhance msleep or ssleep
	  calls (ADPml04175).
	- Simon Waters <simonw@zynet.net> discovered a panic in
	  vsnprintf within aac_procinfo. I added checking of
	  (bytes_available - total_len) to make sure it is never
	  negative in value.
	- Modified abort handler to close all device related commands.

rx.c:
	- optimized code by using likely() and unlikely() conditional
	  hints.
	- RESET_IOP, RESET_IOP_ALWAYS and RESET_IOP_ALTERNATE handling
	  adjusted. Added recognition of AAC_FEATURE_MU_RESET.

aacraid.h:
	- Added definition for AAC_FEATURE_MU_RESET

aachba.c:
	- If the device is offline when the completion handler is
	  called, free the FIB resources and return in io_callback,
	  _aac_probe_container2 and get_container_name_callback
	  (ADPml03975)
	- Added instrumentation to report if a scsi command is corrupt.
	- Adjusted the command line handling to be compliant with
	  2.6 standards (aacraid.parm=value) and add the 'dd' flag
	  recognition to set the expose_physicals flag to zero
	  (ADPml04274)
	- Added command line parsing 'back' into 2.6 version, but only
	  perform the parsing for the 'dd' parameter (ADPml04274)
	- Added a 'dud' parameter that when set, set the
	  expose_physicals flag to zero (ADPml04274)
	- When a command goes through the error handler, it gets
	  reissued to the controller, so that we have multiple FIBs
	  referencing the same command. Drop the re-issued command on
	  the floor if it is already on the Adapter (ADPml03975).

linit.c+aacraid.h:
	- turn on diskdump_mdelay, diskdump_msleep and diskdump_ssleep
	  even though we do not have DUMPLIB support. Remove calls to
	  diskdump_update, which did not export under RHEL4 until U6
	  (ADPml04175)

Version: 1.1.5-2444-IBM ()

build:
	- Added 2.6.18-1.2200.fc5-2.6 as equivalent to 2.6.18-1.2239.fc5-2.6
	- Added 2.6.18-1.2239.fc5-2.6
	- Added 2.6.18-1.2257.fc5-2.6
	- Added 2.6.18-1.2798.fc6
	- Added 2.6.18-1.2849.fc6-2.6 as equivalent to 2.6.18-1.2869.fc6-2.6
	- Added 2.6.18-1.2868.fc6-2.6 as equivalent to 2.6.18-1.2869.fc6-2.6
	- Added 2.6.18-1.2869.fc6-2.6
	- Added 2.6.19-1.2895.fc6-2.6 as equivalent to 2.6.19-1.2911.fc6-2.6
	- Added 2.6.19-1.2911.fc6-2.6
	- Moved 2.6.9-27.EL (RHEL4 U3 rc) as equivalent to 2.6.9-34.EL (RHEL4 U3)
	- Moved 2.6.9-32.EL (RHEL4 U3 beta) as equivalent to 2.6.9-34.EL (RHEL4 U3)
	- Moved 2.6.5-7.252 (SLES9 update) as equivalent to 2.6.5-7.244 (SLES9 SP3)

aachba.c:
	- Fix some compiler warnings associated with typecasting
	  constants to scsi_done.
	- if HAS_BOOT_CONFIG is enabled, default for expose_physicals
	  is zero (ADPml04274)
	- When issuing the probe_container, we need to set the device
	  as online so it passes the device validity check (ADPml04333,
	  ADPml04331 and ADPml03975)

linit.c:
	- For kernels from 2.6.0 to 2.6.12 we set the queue_depth of the
	  device to zero and the device_blocked flag during the error
	  recovery abort handler to trick the LOWER layers to block
	  commands. These kernels have a locking bug where the error
	  handler holds the host_lock, but commands issued by user or
	  plug-n-play scan requests manage to slip through and hit the
	  LOWER layers where host_lock is taken as well *before* the
	  host check for whether we are in error recovery or not.
	  Later kernels do not take the host_lock in the error recovery
	  abort handler (ADPml03975)
	- When we wake up from a printk, or possible dprintk,
	  innoculate the blocking I/O for kernels 2.6.0 through 2.6.12
	  and consider altering the locks (ADPml03975).
	- List of outstanding commands is for a device, so no need to
	  repeat the debug information outlining the id for each command.

Makefile+aacraid.h:
	- Define HAS_BOOT_CONFIG if in a limited resource driver, remove
	  optional pieces. This is one means to compress the resources
	  on the precious floppy disks. (ADPml04333)

install.sh:
	- Code to detect the existence of the /updates/ placement was
	  adjusted from an erroneous 'for' loop into a 'if' test.
	  (ADPml04333)
	- Do not complain if the modules.cgz is truncated, if it can
	  not provide the module, it will be evident later as missing.
	  (ADPml04333)

sa.c:
	- aac_sa_init drops pt_regs argument for kernels above 2.6.18

Version: 1.1.5-2445-IBM ()

build:
	- Added 2.6.11.13mdk-1-1mdk
	- Added 2.6.12.21mdk-1-1mdk

aachba.c+linit.c+release.sh:
	- Added BOOTCD ifdef to fortify initialization time firmware
	  print diagnostics for these configurations.

Version: 1.1.5-2444 ()

build:
	- Added 2.6.18-8.1.1.el5-2.6
	- Moved 2.6.9-36.EL as alias to 2.6.9-42.EL
	- Added 2.6.9-42.0.10.EL
	- Added 2.6.19-1.2288.fc5
	- Added 2.6.19-1.2288.2.1.fc5 as alias to 2.6.19-1.2288.2.4.fc5
	- Added 2.6.19-1.2288.2.4.fc5
	- Added 2.6.19-1.2911.6.4.fc6 as alias to 2.6.19-1.2911.6.5.fc6
	- Added 2.6.19-1.2911.6.5.fc6
	- Added 2.6.20-1.2925.fc6

aachba.c:
	- Added display of TSID.

install.sh:
	- for ia32e configuration, any Intel processor running currently
	  on a 64 bit kernel will do (Son of ADPml04333 and ADPml103707)

Version: 1.1.5-2446-IBM ()

build:
	- Moved FC ia64 and ppc binaries on driver disks into an
	  FC-enterprise disk to free up mainstream space since fc6
	  release.
	- Moved 2.6.5-7.134 (SLES9 SP1 RC) as alias to 2.6.5-7.139
	  (SLES9 SP1)

Version: 1.1.5-2445 ()

build:
	- Added support for RedHat modules.alias file. Base design
	  on ddiskit-0.9.8. Maybe one day we could use RedHat's
	  scripting merged into our meta-build.

linit.c+aachba.c+rx.c+dpcsup.c+commsup.c+comminit.c+commctrl.c+aacraid.h:
	- Adjust code to reflect a back-port of scsi-misc-2.6
	  March 21 2006 snapshot. All changes are cosmetic.

Makefile+aacraid.h:
	- Added HAS_SECTOR_T and check for definition of sector_div
	  macro. If sector_t is 64 bit, then make sure that the return
	  value maxes out rather than wraps.
	- Added kdump config detection, turns off SCSI_HAS_DUMP.

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig+Linux:pci.ids:
	- Changed 9005:0285:108e:7aac to SUN STK RAID REM
	- Added 9005:0285:108e:0286 for SUN SG-XPCIESAS-R-IN
	- Added 9005:0285:108e:0287 for SUN SG-XPCIESAS-R-EX
	- Changed 9005:0285:15d9:02c9 for SMC AOC-USAS-S4iR
	- Changed 9005:0285:15d9:02ca for SMC AOC-USAS-S8iR

rx.c+aachba.c:
	- For 2.6.19 and above, use reset_devices flag, for others
	  read the command line and pick up the reset_devices kernel
	  flag. Reset the adapter as part of initialization.
	- Check if the interrupts have been enabled, this is a
	  fine clue that another driver preceded this one. Useful
	  for hot transitions from one operating system to the
	  next (kexec and kdump are but two examples). This is a
	  substitute for the forced reset_devices flag.

linit.c:
	- Added kernel version or UTS_RELEASE to the Firmware Print

linit.c+csmi.c+aachba.c:
	- Changed serial number printout from %x to %06X format.

sa.c:
	- added aac_sa_restart_adapter method stub.

rx.c: Mark Salyzyn <aacraid@adaptec.com> & Duane Cox <dcox@conxxus.com>
	- Set adapter_sync_cmd before permitting adapter reset
	  commands.
	- If we restart the adapter in the context of the driver,
	  extend startup timeouts less than 5 minutes to 5 minutes.

Makefile:
	- Also look in drivers/scsi/scsi_scan.c for scsi_scan_host()

linit.c:
	- Added driver_version to sysfs
	- added support for 12 digit serial number.
	- exported aac_show_serial_number for aachba and csmi.c
	- Lock mishandling when debugging off (ADPml03975).

aachba.c+csmi.c:
	- Call aac_show_serial_number to acquire serial number value.

commctrl.c+commsup.c+linit.c:
	- Turn off ioctl and aif support if using the BOOT
	  configuration to save space.

aachba.c:
	- dropped definition of min(a,b) because it is in compat.h
	- ifdef on 2.6.21 to saved_command_line as a pointer
	  Steve Fox <drfickle@linux.vnet.ibm.com>
	- Drop check for HIDDEN as this is not compatible with older
	  hardware.

fwdebug.c:
	- if in_atomic(), just time out immediately rather than wait
	  for the Firmware to catch up

Version: 1.1.5-2446 ()

build:
	- Added 2.6.9-55.EL (RHEL4 U5)
	- Added to all legacy RH releases the RHEL_VERSION and
	  RHEL_UPDATE that correspond with this releases.

aachba.c+comminit.c+commsup.c+dpcsup.c:
	- replace kmalloc/memset pairs with kzalloc where possible.

fwdebug.c:
	- use vsnprintf to protect the on-stack firmware buffer.
	- Add buffering to guarantee delivery of messages issued
	  during interrupt service routines. Easier said than done,
	  and includes references to architectural specific assembler
	  fragment helpers to help maintain atomic operations.
	- Fortified the conditions where we time out immediately
	  rather than wait for the Firmware to catch up. These
	  include host locks, interrupt contexts and when error
	  recovery actions are currently active. We only allow a
	  sleep, effectively, at the [aacraid] background thread.
	- Switch to using readb,readl,writeb,writel to communicate
	  with the Firmware print buffer, enforce ordering by doing a
	  readback shortly before issuing the count.

commsup.c:
	- polled fwprint in adapter kernel thread.

aacraid.h+commsup.c+linit.c:
	- Added the ability to force an IOP_RESET_ALWAYS from user
	  initiated sources. 'reset_host!' to the proc node, or
	  and '!' written to the sys reset_host node.

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig+Linux:pci.ids:
	- Changed 9005:0285:108e:0286 for SUN STK RAID INT
	- Changed 9005:0285:108e:0287 for SUN STK RAID EXT

install.sh:
	- Added support to decode xen and xen Linux executables,
	  file's magic was not up to the task.

grub.awk:
	- Added support to detect a xen boot configuration and lift
	  out the Linux kernel and initrd references.

linit.c:
	- Added support for recognizing the RHEL_VERSION and
	  RHEL_UPDATE when reporting to the fwprint
	- aac_queuedepth for 2.4 kernels did not check if the
	  incoming scsi_device was a nul or not (ADPml05512)

aacraid.h+linit.c:
	- Given the apparent instability of long term use of
	  fwprintf, I suggest keeping it off for releases except
	  on the bootable CD where things are more controlled. To
	  aid in that, have added fwprintf next to the dprintk
	  flag output in /proc and /sys.

aachba.c:
	- Turn off physicals access by default on the Bootable CD,
	  allow it to be changed by insmod paramters unlike the
	  BOOT config which was motivated by code space issues.

rx.c+commctrl.c:
	- Added more AAC_DEBUG_INSTRUMENT_IOCTL_SENDFIB
	  instrumentation (ADPml04695)

Version: 1.1.5-2447-IBM ()

rx.c+sa.c+aacraid.h: Mark Salyzyn & Rainer Malitzke <malitzke@metronets.com>
	- Initialize adapter_deliver with aac_rx_deliver_producer
	  (http://bugzilla.kernel.org/show_bug.cgi?id=8469)

csmi.c:
	- linux/ioctl32.h has been deprecated in 2.6.21

aachba.c:
	- Corrected issue with AAC_DEBUG_INSTRUMENT_SG_PROBE
	- Respond with DID_BUS_BUSY to SRB_STATUS_BUSY.

aachba.c: Mark Salyzyn & James Bottomley <James.Bottomley@SteelEye.com>
	- fix panic on short Inquiry

aacraid.h+rx.c:
	- MU_RESET was moved to Options2

aacraid.h+linit.c+aachba.c:
	- Added support for FUA, use IO_SUREWITE for now.
	- Added mode page 8 in order to support WCE and FUA
	- enabled page 8 and page 3f from scsi layers

rx.c+aachba.c:
	- issue commit config if the adapter was reset during
	  initialization.

Version: 1.1.5-2447 ()

build:
	- corrected configuration files for 2.6.5-1.358 (FC2)
	  namely adding the x86_64 & x86_64-smp configurations.

aachba.c:
	- SCSI SYNCHRONIZE observes the range fields and only
	  blocks on writes remaining in the command queue.

compat.h+aachba.c: Fujita Tomonori <fujita.tomonori@lab.ntt.co.jp>
	- remove the non-use-sg case
	- convert to use the data buffer accessors

csmi.c+dpcsup.c+sa.c+compat.h:
	- removed/moved include linux/pci.h

linit.c:
	- Add change scsi queue depth
	- Increased resource cleanup of aac_shutdown

linit.c+commsup.c:
	- After 60 seconds of additional waiting in reset handler,
	  if still have outstanding commands, issue a blind
	  reset to the SRL adapters.

aacraid.h+linit.c+commsup.c+aachba.c:
	- Added support for options ignore reset (SRL).

Version: 1.1.5-2448 ()
Version: 1.1.5-2448-IBM ()

build:
	- Add 2.4.21-50.EL (RHEL3 U9)
	- Drop 2.4.21-9.EL i686 (RHEL3 U1) due to RHEL3 floppy overload
	- Add 2.6.16.46-0.12 (SLES10/SLED10 SP1)

rx.c:
	- Only set commit if in default state should the adapter
	  be reset during initialization.

linit.c:
	- check if thread running before attempting to shut it down.

linit.c: Brian King <brking@linux.vnet.ibm.com>
	- driver_version does not make sense given presence of
	  /sys/module/aacraid/version in later kernels.

linit.c:
	- Kernels that use the reboot notifier can not perform a
	  clean (interrupts disabled) shutdown of the controller
	  because sd_shutdown is called afterwards.  This makes
	  kexec always issue an Adapter Reset in transition.
	- SCSI_HAS_SCSI_IN_DETECTION needs to be considered 'off'
	  for any vmkernel builds.

Version: 1.1.5-2449-IBM ()
Version: 1.1.5-2449 ()

build:
	- Fixup 2.6.16.46-0.12 (SLES10/SLED10 SP1) xen, kdump and xenpae
	  configurations.
	- Fixup 2.6.16.13-4 (SuSE 10.1) x86_64 confifurations

aachba.c:
	- aac_build_sgraw move dev assignment into sub-block.
	- Test unit ready behind an already outstanding long command
	  to a physical device, assume sequential and report back
	  NEEDS_RETRY (ADPml05517)
	- Merged code from 2.6.16.46-0.12 (setinqserial). Why did SuSE
	  not submit this to kernel.org? (ADPml06710)

aachba.c+aacraid.h+commsup.c: Andrew Morton <akpm@linux-foundation.org>
	- rename check_reset to aacraid_check_reset
	- rename aacraid_check_reset to aac_check_reset

linit.c:
	- Added sysfs uart node, returns YES if supported, NO if
	  firmware prints are disabled. Printing reports driver
	  timestamp.
	- Added proc 'uart=' ability as well, same drill except
	  that one checks the flags fields for fwprintf to
	  determine if the feature has a chance.
	- bug in proc writes, if they are malformed at all, they
	  never return (!) returning 0 on writes is bad m'kay.

linit.c: Alan Cox <alan@lxorguk.ukuu.org.uk>
	- Check to make sure ioctl caller is a root user.

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig+Linux:pci.ids:
	- Added 9005:0285:9005:02ce for 51245
	- Added 9005:0285:9005:02cf for 51645
	- Added 9005:0285:9005:02d0 for 52445
	- Added 9005:0285:9005:02d1 for 5405

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig
	- Added 9005:0285:108e:7aae for SUN STK RAID EM

DUD:pcitable:
	- Changed 9005:0285:108e:0286 to SUN STK RAID INT
	- Changed 9005:0285:108e:0287 to SUN STK RAID EXT

rx.c:
	- If the Adapter responds 'Use IRCSR to reset' or 0x3803000F
	  then make sure we transition to the alternate code.
	  (ADPml06338)

aachba.c+aacraid.h+linit.c+fwdebug.c:
	- Drop most of the fwdebug instrumentation (ADPml06620)

Version: 1.1.5-2450-IBM ()

linit.c:
	- removed warning about seconds being unused if the
	  firmware prints are disabled.

linit.c: Alan Cox <alan@lxorguk.ukuu.org.uk>
	- Check to make sure compat ioctl caller is a root user.

aachba.c:
	- Linux standard codify SLES10 setinqserial code.
	- Added retrieval of Container UID as serial number, so
	  that VPD does not change through a morph (APDml06710)
	- corrected locking around search for outstanding commands
	  to physical targets. (ADPml05517)

aachba.c+commsup.c:
	- Merge in cosmetics from scsi-misc-2.6 July 18th 2007
	  snapshot

aachba.c: Ellen Gardiner <gardiner@us.ibm.com>
	- place flush_kernel_dcache_page() within an ifdef of
	  ARCH_HAS_FLUSH_ANON_PAGE

release.sh+aacraid.h+fwdebug.c:
	- filter out AAC_CSMI, AAC_PRINTF_ENABLE, GCC warning filters
	  and fwprintf

install.sh:
	- change from 'echo <files> | grep -v orig' to
	  'ls <files> | grep -v orig' form to correctly list the
	  files checked to ensure that the initrd contents are
	  correct.
	- If /etc/modprobe.conf.local exists s references by
	  /etc/modprobe.conf and there is currently no references
	  to aacraid in /etc/modprobe.conf, then use the local
	  configuration file to add the references instead.

rx.c: Tim Ng <tim_ng@adaptec.com>
	- drain OutboundQueue when performing a reset of the adapter

Version: 1.1.5-2450 ()

build:
	- SLES10 (2.6.16.21-0.8 & 2.6.16.46-0.12) utilized the gcc 4.0.2
	  cross compiler for i586 configurations, it was supposed to use
	  gcc 4.1.1 (ADPml06710)

Version: 1.1.5-2451 ()

Linux:pci.ids+DKSM:readme.txt.orig:
	- Added 9005:0285:108e:7aae for SUN STK RAID EM

install.sh
	- A module failing to install should not be considered fatal,
	  adjusted the message to suggest that the inbox driver is more
	  advanced than the package. We still error with a code from
	  the install script with the hopes that the user will
	  intelligently accept this should it be a late model operating
	  system.

aachba.c:
	- RHEL5 does not exibit the problem with CD writers (ADPml05517)
	  and thus we will ifdef the code out for the later OS releases.
	- For *released* SMC cards, support coming up with SMC vendor
	  id in setinqstr for all 'AOC*' cards.
	- default maximum num physicals should Get_BusInfo fail is set to
	  16 (ok for SCSI, kind'a limiting for SAS).

aachba.c+linit.c+aacraid.h:
	- For OEM release we support sharing the setinqstr handler. Need
	  to rationalize this differently?

linit.c+aacraid.h: inspired by Sri Dharmasanam <sdharmas@cisco.com>
	- use the pci device shutdown function to provide a
	  post-sd_shutdown notification for 2.6.0-2.6.11 kernels.
	  This will help reduce the occurrences of Bad Stripes.
	- Move registration of reboot notifier to beginning of
	  init function so that it is called last (FILO) on way
	  down.

linit.c:
	- Set default timeouts to 45 seconds for the arrays.
	- Added hba_max_channel, hba_max_physical and hba_max_array to
	  sysfs. Only available for AAC_DEBUG_INSTRUMENT_SETUP for now.
	- Make sure the outstanding command is either Async or
	  NoResponseExpected before trusting the callback_data value.
	  (ADPml08062)

linit.c: Alan Cox <alan@lxorguk.ukuu.org.uk>
	- Check to make sure config ioctl caller is a root user.

linit.c: Fujita Tomonori <tomof@acm.org>
	- Added use_sg_chaining enabled.

aacraid.h:
	- Turn off AAC_CSMI to conserve space.
	- dropped maximum transfer size to 64K for kernels less than
	  2.4.18 as we find some of them are capable of producing
	  requests with too many scatter gather elements coupled with
	  >64K element sizes that need to be split for legacy cards
	  (ADPml07201)

Version: 1.1.5-2452 ()

build:
	- Add 2.6.18-53.el5 (RHEL5.1 RC2)
	- Added 2.6.9-65.EL (RHEL4 U6 RC)
	- upgraded to DKMS 2.0.17.4 and added debian version as well.
	- Add 2.6.22-14.46 (Debian 7.10)

aacraid.h+commsup.c+dpcsup.c+linit.c:
	- Added FIB_CONTEXT_FLAG to report that there is a scsi layer
	  context associated with this FIB. (ADPml08062)
	- if check_reset=-1, then use the always reset feature, rather
	  than accepting the Firmware reporting that the reset is
	  disabled. This permits us to bypass Q/A disable to ensure
	  that the reset paths are doing what we expect (ADPml08062)

commsup.c:
	- Do not send a VM_Closeall on a blind reset.

aacraid.h+linit.c+aachba.c:
	- Add a Quirk for the ScsiPortCommand64 missing.

Makefile+linit.c+VMware.mak+aacraid.h+aachba.c:
	- Added HAS_COMPILE_H
	- renamed HAS_DUMPLIB to HAS_DISKDUMPLIB_H
	- renamed HAS_DISKDUMP to HAS_DISKDUMP_H
	- renamed HAS_BOOTSETUP to HAS_BOOTSETUP_H
	- renamed HAS_DUMP_DISKDUMP to HAS_DISKDUMP_H

aacraid.h+rx.c+sa.c+Makefile: Jeff Garzik <jeff@garzik.org>
	- irq_removal: scsi driver trivial. defined
	  HAS_NEW_IRQ_HANDLER_T to aid transition.

commsup.c: Stephen Rothwell <sfr@canb.auug.org.au>
	- don't assign cpu_to_le32(constant) to u8

Version: 1.1.5-2453 ()

build:
	- Removed 2.6.9-65.EL (RHEL4 U6 RC)
	- Added 2.6.9-67.EL (RHEL4 U6)

aacraid.h+aachba.c+linit.c:
	- Add support for scsi device raid level report. This
	  is an experimental feature for the moment.
	- Add in support for caching commands (FUA and
	  SYNCHRONIZE_CACHE) to be disabled either permanently, or only
	  when the card's cache is protected.

aacraid.h+aachba.c+linit.c+commsup.c:
	- Added initial crack at support for Voodoo Lite (JBOD).

compat.h+commctrl.c+comminit.c+dpcsup.c+linit.c+aachba.c:
	- switch to uintptr_t from ptrdiff_t, inspired by scsi-misc-2.6
	  changes on November 5 2007

dpcsup.c:
	- Removed code to limit number of allocations attempted
	  for AIF slots. This is now in-line with scsi-misc-2.6.

linit.c:
	- Use existing fsa_dev info to count the number of containers
	  to make a more intelligent division of the initial queue
	  depth. JBOD drives remain as-discovered though.

linit.c: Alan Cox <lxorguk.ukuu.org.uk>
	- Address weakness in the AACRAID driver, change capability
	  checking from ADMIN to RAWIO.

aachba.c: Fujita Tomonon <tomof@acm.org>
	- replace sizeof sense_buffer with SCSI_SENSE_BUFFERSIZE

aachba.c:
	- Followup to change to SCSI_SENSE_BUFFERSIZE, switch to
	  min() macro.

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig+Linux:pci.ids:
	- Added 9005:0285:9005:02d4 for 2045
	- Added 9005:0285:9005:02d5 for 2405
	- Added 9005:0285:9005:02d6 for 2445
	- Added 9005:0285:9005:02d7 for 2805

linit.c: James Bottomley <James.Bottomley@SteelEye.com>
	- removed use of use_sg_chaining. Only on for 2.6.23-2.6.24.

Version: 1.1.5-2454 ()

linit.c+commsup.c+commctrl.c+aachba.c+aacraid.h:
	- removes pigs in space

linit.c:
	- Added SUPPORTED_JBOD to flags (sysfs & procfs) when jbod
	  adapter is detected.
	- fixed the serial number report.

aacraid.h+aachba.c:
	- Added definitions and enabled support code for AifEnAddJBOD
	  and AifEnDeleteJBOD.

commctrl.c:
	- wrapped the fib lock around the fib context.

aacraid.h+aachba.c+rx.c+linit.c+sa.c:
	- Added MSI capability.

linit.c: Fujita Tomonori <tomof@acm.org>
	- iommu sg merging: use pci_set_dma_max_seg_size

aachba.c:
	- Drop valid-data bit in sense return data.
	- optimized set_sense to minimum functionality.

aachba.c+linit.c+commsup.c:
	- problem with state of Ignore Adapter Reset.

fwdebug.c:
	- added aacraid.firmware_debug insmod parameter.

commsup.c: Marcin Slusarz <marcin.slusarz@gmail.com>
	- le*_add_cpu conversion

Version: 1.1.5-2455 ()

install.sh:
	- Return does not work in context of script, only subroutines,
	  use exit instead.
	- If a package has been relocated, and we can not find the
	  master location, use the current working directory.
	- If a make init rd command is not found, then revert the
	  backup copy of the initrd before exiting with the error.
	- Added support for update-initramfs command present in
	  Gutsy.

linit.c:
	- As proposed by Jonas Bonn <jonas@southpole.se> changed
	  struct pci_device_id table to use DECLARE_PCI_DEVICE_TABLE
	  if available.
	- use __devinitdata or __devinitconst depending on kernel
	  vintage for the legacy pci_device_id table definitions.

aacraid.h+aachba.c+linit.c:
	- Added definition and code for a ST_NOT_READY response from
	  an I/O request.
	- Added FSCS_NOT_READY bit in VM_NAMESERVE
	- Added Power Management Initflags and option report.
	- Add defintion and code for issuing START_STOP fib.

aachba.c:
	- Added additional instrumentation for
	  AAC_DEBUG_INSTRUMENT_VM_NAMESERVE (ADPml08612)

commsup.c:
	- Added a retry loop handler for queued scans (ADPml08614
	  and ADPml08615)

aachba.c+aacraid.h+commctrl.c+commsup.c+dpcsup.c+linit.c+rx.c+sa.c:
	- Synchronize to latest scsi-misc-2.6 changes on April 30 2008
	- Changed from class_device_attribute to device_attribute (why?)

README+DUD:pcitable+VMWare.xml+DKMS:readme.txt.orig+Linux:pci.ids:
	- Change 9005:0285:9005:02d4 for ASR-2045
	- Change 9005:0285:9005:02d5 for ASR-2405
	- Change 9005:0285:9005:02d6 for ASR-2445
	- Change 9005:0285:9005:02d7 for ASR-2805
	- Add all 'G' cards.

aachba.c+commsup.c+aacraid.h+linit.c+rx.c+commsup.c:
	- Merged upcoming VMWare port

linit.c: Harvey Harrison <harvey.harrison@gmail.com>
	- aac_show_serial_number (2.6.25+) should be static

build:
	- Add VMware 3.5 ddk
	- Build VMware 3.0, 3.0.1, 3.0.2 and 3.5 products ADPml09413

Version: 1.1.5-2456 ()

VMware*.spec:
	- Print out instructions on how to do the vmeare-mkinitrd.
	- Replace yes with true.

build:
	- Add SLES9.4 2.6.5-7.308
	- Add RHEL5.2 2.6.18-92.el5
	- Add SLES10.2 2.6.16.60-0.21

Version: 1.1.5-2457 ()

build:
	- Added a message telling the user to reboot in the VMware
	  package.
	- Added 2.4.21-47.0.1 (RHEL3 u8 errata 1)

commctrl.c:
	- ioctl scb passthrough needs to check the validity of the
	  individual scatter gather count fields to the maximum the
	  adapter supports. Doing so will have the side effect of
	  preventing copy_from_user() from bugging out while populating
	  the dma buffers.

aacraid.h:
	- Added Performance Modes and additional Feature Bits to
	  Supplemental Info structure.

aachba.c:
	- Do not provide VPD data for VMware builds. Messes up the SAN
	  signature.
	- Add aacraid.wwn insmod parameter to control user preference.

README:
	- Comment out G series cards.

Version: 1.1.5-2458 ()

commctrl.c:
	- if the driver is in shutdown, do not restart aacraid kernel
	  thread (cleanup)

aachba.c:
	- SLES10 SP2 uses wwn=1, SLES10 SP1 uses wwn=2 as default

Version: 1.1.5-2459 ()

build:
        - RHEL4.7 added: 2.6.9-78.el
        - RHEL5.3 added: 2.6.18-92.el5

linit.c+aachba.c:
        - aac_cache=2 as default to avoid performance problems due to flushes 
        (Novell Bugzilla #469922)
        - disable nondasd support for AAC_QUIRK_SCSI_32 controllers to avoid hang
        (RedHat Bugzilla #457552)

all sources:
        - changes for VMWare ESX4.0

README:
        - BLBU Voodoos added

Version: 1.1.5-2460 ()

2009-02-24	Achim_Leubner@adaptec.com

linit.c:
        - aac_eh_device_reset() and aac_eh_bus_reset() now return SUCCESS
          to avoid "vmkfstools -L lunreset ..." and "vmkfstools -L busreset ..."
          failures in the test suite of VMware

all sources:
        - scsi-misc-2.6 changes, patch against 2.6.29-rc7

Version: 1.1.5-2461 ()

2009-03-31	Achim_Leubner@adaptec.com

commsup.c:
        - AIF handling: If there is a JBOD drive added to the system, identify the 
          old one with scsi_device_lookup() and remove it to enable a 
          fresh scsi_add_device(); otherwise the new JBOD is not available
          until reboot

Version: 1.1.5-2463 ()

2009-04-14	Achim_Leubner@adaptec.com

aachba.c:
	- expose_physicals: Phys. drives not available for expose_physicals>0 with new
	  JBOD firmware - function aac_expose_phy_device() added to reset the appropriate
	  bit in the first byte of inquiry data 	

Version: 1.1.5-2464 ()

2009-05-18	Mahesh_Rajashekhara@adaptec.com

aachba.c:
commctrl.c:
comminit.c
commsup.c
dpcsup.c
linit.c
aacraid.h

	- Added support for building the driver against Kernel - 2.6.25

Version: 1.1.5-2466 ()

2009-06-24	Mahesh_Rajashekhara@adaptec.com

aachba.c:
commctrl.c:
csmi.c
fwdebug.c
linit.c

	- Added support for building the driver against Kernel - 2.6.29 & 2.6.30

Version: 1.1.5-2467 ()

2009-07-10	Mahesh_Rajashekhara@adaptec.com

aachba.c:
commctrl.c:
csmi.c
fwdebug.c
linit.c

	- Added support for building the driver against Kernel - 2.6.26, 2.6.27 & 2.6.28

Version: 1.1.5-2468 ()

2009-07-22	Mahesh_Rajashekhara@adaptec.com

linit.c

	- "DDTS #11875: vmware 4.0 : Shows some junk value in serial number field"
	  aac_show_serial_number - Added the fix to copy serial number into the buffer

Version: 1.1.5-24800 ()

2009-08-20	Mahesh_Rajashekhara@adaptec.com

aachba.c

	- Added support for handling ATA pass thru commands
	  When the CC bit is SET by the host in ATA pass thru CDB, driver is supposed to return DID_OK	 
	  When the CC bit is RESET by the host, driver should return DID_ERROR
	 
Version: 1.1.5-24900 ()

2009-14-10	Mahesh_Rajashekhara@adaptec.com

aachba.c

	- VMWARE ESX: Added support for creating RVOLUME 

Version: 1.1.5-25000 ()

2009-11-05	Achim_Leubner@adaptec.com

commctrl.c
	- Change to avoid XferState != 0 kernel message after a possible sync. cmd timeout after 180s

commsup.c
	- schedule() call added for ESX35 to avoid deadlock because otherwise bottom half handler does
	  not run and up() call never occurs
	- wait=-1 setting removed for ESX40 because we can block under that OS

commsup.c+dpcsup.c
	- handling after a possible sync. cmd timeout after 180s changed to avoid PSOD due to wrong
	  event_wait semaphore status

Version: 1.1.5-25100 ()

2009-11-12	Mahesh_Rajashekhara@adaptec.com

aachba.c

	- ADPml11898 SUNMR Spitfire issue
 	  Firmware  layer exposes lesser container capacity than the actual one.	
	  It exposes [Actaul size - Spitfire space(10MB)] to the OS, IO's to the 10MB should be prohibhited from the Linux driver.
	  Sensekey sets to HARDWARE_ERROR and sending the notification to the MID layer.
	  Added "expose_hidden_space" flag, by default the fix will be executed. 
	  Only if the user sets "expose_hidden_space=1", user can access beyond the Array reported size(Spitfire region).	 

commsup.c
	
	- ADPml12131 Cannot see 2 luns with a prometheus card on a J4400
	  When the Enclosure cable is Hot removed/inserted,driver doesn't inform SCSI mid layer about the device Online/Offline status 
	  leaving "stale" entries. Now with this fix in the driver, it informs the appropriate device Online/Offline status to the 
	  SCSI mid layer with the help of AIF notification from FW.

Version: 1.1.5-25200 ()

2009-11-17	Mahesh_Rajashekhara@adaptec.com

aachba.c

	- ADPml11898 SUNMR Spitfire issue
	  Removed extra condition for this fix in aac_read/aac_write function. 
	  
Version: 1.1.5-25300 ()

2009-11-30	Achim_Leubner@adaptec.com

commctrl.c
	- small fix to return correct status if sync. cmd timeout after 180s

commsup.c
	- schedule() call replaced by direct interrupt handler call due to recommendations by the VMware team
	- wait=-1 not set for ESX40

Version: 1.1.5-25400 ()

2009-12-04	Mahesh_Rajashekhara@adaptec.com

build:
        - SLES10 SP3 added: 2.6.16.60-0.54.5

Version: 1.1.5-25500 ()

2009-12-08	Achim_Leubner@adaptec.com

aacraid.h:
        - "#if (defined(SCSI_HAS_DUMP))" removed for the "aac_adapter_intr()" definition 

build:
        - RHEL4.8 added: 2.6.9-81.el

Version: 1.1.5-25600 ()

2009-12-16	Mahesh_Rajashekhara@adaptec.com

build:
        - RHEL5.4 added: 2.6.18-164.el5

Version: 1.1.5-25700 ()

2010-01-08	Mahesh_Rajashekhara@adaptec.com

aacraid.h:
        - Based on the notification from FW, the driver calls "scsi_remove_device" for the deleted array.
	  This call not only informs the scsi device status to the SCSI mid layer 
	  and also it will remove corresponding scsi device entries from the Linux sysfs.
	  Enabled "AAC_DEBUG_INSTRUMENT_AIF_DELETE" flag  in the Header file, with this flag the driver uses "scsi_remove_device"	

Version: 1.1.5-25800 ()

2010-01-14	Achim_Leubner@adaptec.com

aachba.c:
        - ESX40: Set aac_wwn=0 to avoid serious driver update problem (cannot mount root - vsd_mount failure)
aacraid.h+commctrl.c+comminit.c+commsup.c+dpcsup.c:
	- ESX40: Use cmdready wait queue instead of semaphore to wait for competion in aac_command_thread()
	- ESX40: Use kthread_stop instead of thread_die flag during shutdown to kill aac_command_thread(), older driver
	  had a deadlock here

Version: 1.1.5-25900 ()

2010-01-20	Mahesh_Rajashekhara@adaptec.com

build:
	- RHEL4.8 added: 2.6.9-89.el

aacraid.src.spec:
	- Added standalone XEN kernel support

Version: 1.1.5-26000 ()

2010-02-02	Mahesh_Rajashekhara@adaptec.com

linit.c:
	-  "cat /proc/scsi/aacraid/<Node #>" displays invalid argument
	   The return value in "aac_procinfo" was incorrect and changed to return '0'

Version: 1.1.5-26100 ()

2010-02-16	Achim_Leubner@adaptec.com

build:
	- Added support to build ESX 4.0 driver ISO image

Version: 1.1.5-26200 ()

2010-02-24	Mahesh_Rajashekhara@adaptec.com

build:
	- ESX4: Driver ISO image file is renamed to ESX4 product

Version: 1.1.5-26300 ()

2010-02-26	Mahesh_Rajashekhara@adaptec.com

linit.c:
aachba.c:
	- Fixed build issues for SLES11


Version: 1.1.6-26400 ()

2010-04-21	Achim_Leubner@adaptec.com

linit.c + frey.c + aacraid.h + Makefile + VMware-4.0.sc + VMware.inc:
        - Cardinal Frey controller support added

Version: 1.1.6-26500 ()

2010-07-28	Mahesh_Rajashekhara@adaptec.com

aacraid.h + compat.h + linit.c + build:
        - ESXi4.0 - Added fix to create device nodes(/dev/aacX)
	- ESXi4.0 - Added build support to produce seperate driver binary package

linit.c + fwdebug.c:
	- Added build support for 2.6.34 and 2.4.37 kernels

build:
	- RHEL5.5 added: 2.6.18-194.el5

Version: 1.1.6-26600 ()

2010-08-11	Mahesh_Rajashekhara@adaptec.com

build:
	- SLES11 added: 2.6.27.19-5

Version: 1.1.6-26700 ()

2010-08-24	Mahesh_Rajashekhara@adaptec.com

build:
	- Fedora Core 11 added: 2.6.29.4-167.fc11
	- Fedora Core 12 added: 2.6.31.5-127.fc12

Version: 1.1.6-26800 ()

2010-08-30	Mahesh_Rajashekhara@adaptec.com

build:
	- RHEL6 Beta2 added: 2.6.32-37.el6
	- RHEL6 Snap8 added: 2.6.32-52.el6

Version: 1.1.7-26900 ()

2010-09-04	Mahesh_Rajashekhara@adaptec.com

build:
	- SLES11 SP1 added: 2.6.32.12-0.7

Version: 1.1.7-27000 ()

2010-09-06	Achim_Leubner@pmc-sierra.com

frey.c: 
	- Removed because Marvell Frey support removed
src.c: 
	- Added to support PMC SRC controller family (Kittyhawk, Tupelo)

Version: 1.1.7-27100 ()

2010-09-17	Mahesh_Rajashekhara@adaptec.com

build:
	- SLES11 SP1: Added DUD pre and post installation scripts
	- SLES11 SP1: Updated README for SLES11 SP1 DUD installation

Version: 1.1.7-27200 ()

2010-09-23	Achim_Leubner@pmc-sierra.com

VMware-3.5.xml + VMware-4.0.xml:
	- Kittyhawk/Tupelo device IDs added for VMware

rx.c + sa.c + src.c + fwdebug.c + aacraid.h: 
	- UART trace output reworked because debug buffer is on BAR1 for PMC SRC

Version: 1.1.7-27400 ()

2010-10-19	Mahesh_Rajashekhara@adaptec.com

VMware-4.0.xml:
	- ESXi 4: Fixed driver load issue

Version: 1.1.7-27500 ()

2010-10-21	Achim_Leubner@pmc-sierra.com

commsup.c:
	- Deadlock patch (PR 414597) also added for ESX4
src.c: 
	- Wait 5 sec. after doorbell initiated FW reset to avoid wrong FW status read

Version: 1.1.7-27600 ()

2010-11-01	Achim_Leubner@pmc-sierra.com

src.c: 
	- Prep #178897: Avoid wrong doorbell bit clear in the interrupt routine
	- Prep #178892: Adjust RRQ index in the interrupt routine

Version: 1.1.7-27700 ()

2010-11-18	Achim_Leubner@pmc-sierra.com

src.c: 
	- Prep #178892: Adjust RRQ index in the interrupt routine - BUG FIXED FROM 27600 BUILD

Version: 1.1.7-27800 ()

2010-11-30	Mahesh_Rajashekhara@adaptec.com

aachba.c: 
	- Prep #181085: Fixed write error and read-only filesystem issue

Version: 1.1.7-27900 ()

2010-12-07	Mahesh_Rajashekhara@pmc-sierra.com

build:
	- RHEL6 GA added: 2.6.32-71.el6

Version: 1.1.7-28000 ()

2011-02-16	Mahesh_Rajashekhara@pmc-sierra.com

linit.c:
	- Prep #192790: Fixed VMware certification SAS_CheckDriver test issue

Version: 1.1.7-28100 ()

2011-04-05	Mahesh_Rajashekhara@pmc-sierra.com

linit.c:
aachba.c:
VMware-5.0.mak:
VMware-5.0.sc:
VMware-5.0.xml:
	- Added support for ESXi 5.0 OS

build:
	- Fedora Core 13 added: 2.6.33.3-85.fc13
	- Fedora Core 14 added:	2.6.35.6-45.fc14
	- DKMS: Added Series 6 PCI id's into the 'pcitable' file and latest DKMS tool binary RPM v2.1.1.2

Version: 1.1.7-28200 ()

2011-05-24	Mahesh_Rajashekhara@pmc-sierra.com

linit.c:
commctrl.c:
commsup.c:
	- Fixed Ubuntu 11.04 build issues
	- Fixed VMware 5.0 SAS_CoreDump test failure issue

build:
	- Updated driver README

Version: 1.1.7-28300 ()

2011-07-01	Mahesh_Rajashekhara@pmc-sierra.com

linit.c:
	- Prep #209412: Fixed bug in ioctl interface

Version: 1.1.7-28400 ()

2011-08-22	Mahesh_Rajashekhara@pmc-sierra.com

aachba.c:
aacraid.h:
commctrl.c:
comminit.c:
commsup.c:
dpcsup.c:
VMware-5.0.mak:
	- Included HCL ServeRAID team submitted upstream driver patch(Patch: [SCSI] aacraid: fix File System going into read-only mode)
	  http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=cacb6dc3d7fea751879a225c15e48228415e6359
	- Added RDM(RAW DEVICE MAPPING) support for VMware ESX/ESXi 4.1 and ESXi5.0
	  For RDM support, set "aac_wwn = 1" RAID controller response to inquiry VPD page 0x80 to return a unique serial number
	- Added new DDK (#406165) Makefile for VMware ESXi5.0
	- Changed sync. timeout from 30s to 300s for rx and src

Version: 1.1.7-28500 ()

2011-09-09	Mahesh_Rajashekhara@pmc-sierra.com

aacraid.h:
build:
	- Fixed Prep #216490
	- RHEL5.6 added: 2.6.18-238.el5 

Version: 1.1.7-28600 ()

2011-10-10	Mahesh_Rajashekhara@pmc-sierra.com

linit.c:
commsup.c:
	- Fixed Prep #211135: Changed 'AIF_SNIFF_TIMEOUT' from 30 to 60 seconds because this timeout is too short for a config with a lot of drives. 
	  Included block request timeout 45 seconds for the kernel versions >= 2.6.27 version 
	  Initialized 'fibptr->flags' to 0 to avoid kernel derefrencing pointer issue
	  Initialized 'aac->fibs[]' to 0 completely after allocation

Version: 1.1.7-28700 ()

2011-11-08	Mahesh_Rajashekhara@pmc-sierra.com

build:
	- A warning message to be printed during the DKMS build that indicates to the user to adjust settings in conf file for DKMS auto_install support

Version: 1.1.7-28800 ()

2011-12-06	Achim_Leubner@pmc-sierra.com

aachba.c:
linit.c:
comminit.c:
commsup.c:
src.c:
aacraid.h:
	- Series 7 (Denali) support added
	- Sync. mode (for testing) added
	- Sync. mode switched on automatically if async. mode not supported (for future products - Series 7 with new async. interface, Series 8/9) 

build:
	- RHEL 6.2 RC added: 2.6.32-220.el6
	- RHEL 6.1 added: 2.6.32-131.0.15.el6
	- SLES 10 SP4 added: 2.6.16.60-0.85.1

Version: 1.1.7-28900 ()

2011-12-15	Achim_Leubner@pmc-sierra.com

commsup.c:
src.c:
aacraid.h:
linit.c:
	- Sync. mode changes to avoid tool errors (management fib handling changes)

Version: 1.2.0-29000 ()

2011-01-18	Mahesh_Rajashekhara@pmc-sierra.com

VMware-4.0.xml:
VMware-5.0.xml:
	- Added PCI device id's for Huawei80T, Series 7, 8 and 9 controllers

Version: 1.2.1-29100 ()

2012-03-12	Achim_Leubner@pmc-sierra.com

aacraid.h:
aachba.c:
comminit.c:
commsup.c:
dpcsup.c:
linit.c:
src.c:
	- Async. (performance) mode added for Series 7 with (native) S/G list, new FIB header, new Raw IO command (RawIo2)

Version: 1.2.1-29500 ()

2012-04-12	Achim_Leubner@pmc-sierra.com

aachba.c:
comminit.c:
commctrl.c:
	- Fast JBOD mode added

2012-04-17	Mahesh_Rajashekhara@pmc-sierra.com

aachba.c:
	- Added a fix for 'BUG: scheduling while atomic' issue

2012-04-30	Achim_Leubner@pmc-sierra.com

aachba.c:
dpcsup.c:
aacraid.h:
	- Bugfix Fast JBOD mode

2012-05-10      Mahesh_Rajashekhara@pmc-sierra.com

aachba.c:
commsup.c:
linit.c:
Makefile:
build:
        - Fedora 15 added: 2.6.38.6-26.rc1.fc15
        - Fedora 16 added: 3.1.0-7.fc16

Version: 1.2.1-29600 ()

2012-05-25	Mahesh_Rajashekhara@pmc-sierra.com

build:
        - RHEL 5.8 added: 2.6.18-308.el5
        - SLES 11 SP2 added: 3.0.13-0.27
        - Updated README

Version: 1.2.1-29700 ()

2012-06-21	Mahesh_Rajashekhara@pmc-sierra.com

aachba.c:
	- When we have an I/O that adjoins the previous request, we change the queue depth by calling 'scsi_adjust_queue_depth' API. 
	  It appears that latest versions of kernel doesn`t support changing the queue depth and backing up of sequential I/O`s at
	  the SCSI merge layer on a faster systems or controllers, the *lockdep* error caused due to clash between 'scsi_adjust_queue_depth' and
	  'scsi_dispatch_cmd' routines at the SCSI upper layers.

src.c:
	- Workaround: To keep OS happy and to avoid disabling IRQ line issue, when the driver is in Async mode and if driver gets an
	  extra sync. command completion response from the firmware, fix clear`s the ISR interrupt bits and return IRQ_HANDLED from 
	  the interrupt handler fixes the "Disabling IRQ #" issue.

Version: 1.2.1-29800 ()

2012-07-03	Mahesh_Rajashekhara@pmc-sierra.com

commsup.c:
	- `udelay` in aac_fib_send to wait for ShutDown command complete that block the other CPUs too long during shutdown if array is rebuilding.
	  Changed it to use `msleep` instead of `udelay`. 

Version: 1.2.1-29900 ()

2012-07-26	Mahesh_Rajashekhara@pmc-sierra.com

commsup.c:
aacraid.h:
	- Added new AIF raw device remove support

VMware-5.1.mak:
VMware-5.1.sc:
VMware-5.1.xml:
	- Added support for VMware 5.1 OS

Version: 1.2.1-30000 ()

2012-12-19	Mahesh.Rajashekhara@pmcs.com
		Vinay.Goli@pmcs.com

aachba.c:
	- SCSI dma mapping failure case handling
	  http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=0b4334473d48aa18e8448f9f718f1dcd0398c550

aacraid.h:
comminit.c:
	- 1024 command support related changes

build:
fc-post-install.sh:
install.sh:
	- New OSes support related changes for 2013.1 program

Makefile:
	- Instead of modifying OS GLOBAL config settings, creating driver specific conf file.

Version: 1.2.1-30100 ()

2013-01-16	Mahesh.Rajashekhara@pmcs.com
		Vinay.Goli@pmcs.com

build:
	- RHEL 5.9 added: 2.6.18-348.el5

aacraid.h:
comminit.c:
	- 1024 command support changes for S7 and above hardware only.

Version: 1.2.1-30200 ()

2013-02-11	Mahesh.Rajashekhara@pmcs.com
		Vinay.Goli@pmcs.com

build:
install.sh:
fc-post-install.sh:
	- Fedora 18 added: 3.6.10-4.fc18

aacraid.h:
comminit.c:
src.c:
	- Added Dual flash fw support

Version: 1.2.1-30300 ()

2013-05-13	Mahesh.Rajashekhara@pmcs.com

aachba.c:
commsup.c:
aacraid.h:
	- Fixes for Prep #309671 and Prep #309674

2013-05-13	Gilbert.Wu@pmcs.com

commctrl.c:
	- Fix for Prep #304874

Version: 1.2.1-40000 ()

2013-04-04	Achim.Leubner@pmcs.com
		Mahesh.Rajashekhara@pmcs.com

aacraid.h:
linit.c:
aachba.c:
comminit.c:
commsup.c:
src.c:
	- Added 4K sector support changes
	- Added MSI-X support changes

Version: 1.2.1-40100 ()

2013-08-12	Mahesh.Rajashekhara@pmcs.com

aacraid.h:
comminit.c:
commsup.c:
linit.c:
src.c:
	- Added MSI-X IOP RESET handling changes
	- Added fix for hang during driver re-load
	- Added support for the kernel versions >= 3.8
	- New OSes support packages:
	  RHEL 6.4 added
	  SLES 11 SP3 added
	  VMware 5.5 added
	  CentOS 6.2 added
	- Fixed Prep #319611 and Prep #319516
	- Suppress two GCC warnings

Version: 1.2.1-40200 ()

2013-10-09	Mahesh.Rajashekhara@pmcs.com

comminit.c:
aachba.c:
	- Fixed Prep #317771: DKMS driver did not install or build for the updated kernel 3.9.6-200.fc18.x86_64 in Fedora 18.x86_64
	- Fixed Prep #323488: Size of AUTOLD1 exposed in ARCCONF is different from that in OS
	- Changed QD = 512 and Maximum FIB size = 4096 for performance

Version: 1.2.1-40300 ()

2013-10-24	Mahesh.Rajashekhara@pmcs.com

aacraid.h:
dpcsup.c:
linit.c:
src.c
	- Fixes for Prep's below:
	  Prep #325216: VMware ESXi 5.1 Cert - `ErrorHandling::SAS_CoreDumpTest` failed.
	  Prep #324456: SLES 11 SP3(x64) : System hang due to CIM Server failed while running overnight IO.
	  Prep #324481: Kernel panic found while array roaming from 8885E to 81605ZQ

Version: 1.2.1-40400 ()

2013-12-02	Mahesh.Rajashekhara@pmcs.com

build:
	- RHEL 5.10 added

Version: 1.2.1-40500 ()

2014-02-14	Mahesh.Rajashekhara@pmcs.com

VMware-4.0.xml:
VMware-5.0.xml:
VMware-5.1.xml:
VMware-5.5.xml:
rx.c:
	- Updated latest PCI device IDs
	- kdump fix: http://marc.info/?l=linux-scsi&m=138960673329856&w=2

Version: 1.2.1-40600 ()

2014-03-31	Mahesh.Rajashekhara@pmcs.com

aacraid.h:
aachba.c:
	- Fix for Prep#338719: Multiple issues in creating more than 64 Simple volume in SV 240 mode

Version: 1.2.1-40700 ()

2014-05-05	Achim.Leubner@pmcs.com

aacraid.h:
comminit.c:
src.c:
	- Fix for Prep #341911: doorbell reset value now available during initialization for driver

2014-05-05	Mahesh.Rajashekhara@pmcs.com

aacraid.h:
comminit.c:
linit.c:
	- Fix for Prep #337965:  During the shutdown, if driver gets commands from the application layer, driver sends error code to the upper layers.

Version: 1.2.1-50000 ()

2014-07-02	Mahesh.Rajashekhara@pmcs.com

aachba.c:
aacraid.h:
commsup.c:
linit.c:

	- Added support for Katmai boards
	- Prep 350269: 2K/streamline continuation FIB support
	- Prep 349878: Linux driver has to populate the right value in FIB header for CF_STATUS_HEADER

Version: 1.2.1-50001 ()

2014-07-23	Achim.Leubner@pmcs.com
		Santosh.Akula@pmcs.com
		Mahesh.Rajashekhara@pmcs.com

aachba.c:
aacraid.h:
commctrl.c:
commsup.c:
linit.c:
src.c:

	- 352427 - Katmai: Force controller mode to SMART MODE
	- 351878 - Katmai: Hot plug doesnt work in linux
	- 351879 - Katmai: Add persistent mapping support
	- 349443 - change AIF vector number
	- 348570 - Error code changes
	- 350053 - Transport reset implemented
	 
Version: 1.2.1-50002 ()

2014-08-20	Achim.Leubner@pmcs.com
		Mahesh.Rajashekhara@pmcs.com
		Murthy.Bhat@pmcs.com
		Santosh.Akula@pmcs.com

aachba.c:
aacraid.h:
linit.c:
VMware-4.0.xml:
VMware-5.0.xml:
VMware-5.1.xml:
VMware-5.5.xml:
build:
install.sh:

	- Prep 334262: Do not expose SEP devices on VMware ESXi
	- Prep 352682: AATF Found: LD Name and UID not exposed to the OS
	- Prep 354136: AATF Found: Controller model not displayed in `cat /proc/scsi/aacraid/` output
	- RHEL 6.5 GA Kernel version: 2.6.32-431.el6
	- Added Lenovo 8885E support

Version: 1.2.1-50003 ()

2014-08-26	Mahesh.Rajashekhara@pmcs.com
		Achim.Leubner@pmcs.com

aacraid_build:
build:
CHANGELOG:
CheckInLog.txt:
comminit.c:
commsup.c:
linit.c:
aachba.c:
commctrl.c:
dpcsup.c:

	- RHEL 7.0 Kernel version: 3.10.0-123.el7 added
	- Fixed RHEL 5.x Xen kernel issue
	- 354406 - AATF Found: Drive Firmware version not displayed for SATA drives in ARCConf output in HBA mode
	- 355642 - Driver module load failed with ARC controller

Version: 1.2.1-50004 ()

2014-09-05	Mahesh.Rajashekhara@pmcs.com
		Achim.Leubner@pmcs.com

CheckInLog.txt:
CHANGELOG:
aachba.c:
comminit.c:
aacraid.h:
build:

	- Increase supported LUNs from 8 to 256
	- Add 2K FIB support bit
	- Add Ubuntu 14.04 driver support packages.

Version: 1.2.1-50005 ()

2014-10-07	Mahesh.Rajashekhara@pmcs.com

CHANGELOG:
CheckInLog.txt:
aacraid_key.config:
aacraid_key.priv:
aacraid_key_pub.der:
build/aacraid_kmod.spec:
build/aacraid_build:
build/aacraid_dd_skel.tar:
aacraid.spec:
build/aacraid_dd_skel/dd_sles10_32:
build/aacraid_dd_skel/dd_sles10_64:

	- Prep # 359111: Fixed SLES 10.x / 11.x driver binary package install/un-install issues.
	- Prep # 357753: Add RHEL 7.0 GA kernel driver module signing support.

Version: 1.2.1-50006 ()

2014-11-18	Rajinikanth.pandurangan@pmcs.com

aachba.c
CHANGELOG
CheckInLog.txt:
	- Fix for PREP361037 - Add support for Inquiry vpd page 0x83 for logical wwn.
	- Fix for "smartctl -i" failure in getting mode sense data.


ToDo:
	-

2014-12-19	Gilbert.Wu@pmcs.com

CHANGELOG:
CheckInLog.txt:
src.c

1. Fix the IOP_RESET does not work on S8. Sending IOP_RESET command need to wait for only 10 sec instead of 5 minutes
   in case of firmware does not response IOP_RESET command.
2. Disable interrupt before setup interrupt routine to prevent spurious interrupt.

Version: 1.2.1-50007 ()

2014-12-19	Rajinikanth.pandurangan@pmcs.com

aachba.c
CHANGELOG
CheckInLog.txt:
	- Fix for PREP361037 - Add support for Inquiry vpd page 0x83 for logical wwn.
	- Fix for "smartctl -i" failure in getting mode sense data.
ToDo:
	-
	
Version: 1.2.1-50007 ()

2015-01-13	Rajinikanth.pandurangan@pmcs.com

aachba.c
comminit.c
commsup.c
dpcsup.c
linit.c
rx.c
src.c
aachba.h
CheckInLog.txt:
 - As part of the performance improvement, there are RHEL specific changes and some generic changes.
    1. Specific for RHEL6.5/CentOS6.5 - For RHEL 6.X, now our driver exposes as lockless driver. This
    means kernel will not lock before it calls our driver IO entry point.
    2. Generic changes:
       a. Now driver sets IRQ affinity hints.  This would help kernel to assign specific MSIx vector
       to specific cpu core.
       b. Couple of spin_lock has been removed in IO path.
       c. If kernel > 2.6.37, by default kernel will not lock before it calls our IO entry.  Now our
       driver doesn't lock itself before it submits each IO request. 
ToDo:
	- There is one more spinlock which can be removed by using atomic variables.
	- Or rewrite IO path to make it compatible for NUMA.


2015-1-16 - Murthy.bhat@pmcs.com

CHANGELOG:
CheckInLog.txt:
aacraid_build:
    - Added Ubuntu 14.10 13.10 and Debian 7.6 OS support
    - Updated driver README

ToDo:
	-
	
2015-1-23 - Murthy.bhat@pmcs.com

CHANGELOG
CheckInLog.txt
build/aacraid_dd_skel.tar
build/aacraid_dd_skel/dd_fc
build/aacraid_kmod_fc.spec
build/kmodtool_aacraid_fc

        - Added new driver SPEC/build scripts for building driver binaries for
          Fedora, so that driver binaries can be built automatically on a native
          Fedora virtual machines (automated build server).
ToDo:
	-

2015-2-16 - Murthy.bhat@pmcs.com

CHANGELOG
CheckInLog.txt
build/aacraid_dd_skel.tar
build/aacraid_dd_skel/dkms
build/aacraid_dkms.spec
build/aacraid.src.spec

        - Added new driver SPEC/build scripts for building DKMS driver binaries
          and source rpm, so that driver binaries, source rpm can be built 
          automatically on an automated build server.
ToDo:
	-

2015-2-27 - Murthy.bhat@pmcs.com

CHANGELOG
CheckInLog.txt
build/aacraid_dd_skel.tar
build/README.txt
build/aacraid_dd_skel/dd_fc/README.txt

        - New master README, which explains changed methods in installing the OS on ARC 
	  array and package name changes etc
ToDo:
	-
2015-3-17 - rajinikanth.pandurangan@pmcs.com

CHANGELOG
CheckInLog.txt
build/aacraid_build
build/aacraid-kmod.spec
build/README.txt

	- Added automatic build support for RHEL7.1
ToDo:
	-
	
2015-3-19 - Murthy.bhat@pmcs.com

CHANGELOG
CheckInLog.txt
build/aacraid_build
build/aacraid_dd_skel.tar
build/README.txt
build/aacraid-kmod_fc20.spec
build/aacraid_dd_skel/dd_fc/fc-post-install.sh

    - Added Fedora 21 OS support
    - Updated driver README

ToDo:
	-

2015-3-27 - Murthy.bhat@pmcs.com

build/aacraid_kmod.spec
CHANGELOG
CheckInLog.txt:
	- Fix for PREP373778 - KMOD RPM installation not working

ToDo:
	-
2015-4-16 - rajinikanth.pandurangan@pmcs.com

linit.c
CHANGELOG
CheckInLog.txt:
	- Fix for PREP #374209 (Title: The arcconf/maxView can not see the controller after release system from Suspend state(Ubuntu 14.04.1))

ToDo:
	-

2015-4-17 - Murthy.bhat@pmcs.com

CHANGELOG
commsup.c
aacraid.h
CheckInLog.txt:
	- Fix for Prep 371737 - Container UID to Serial Number to OS is all zeros.

ToDo:
	-

