Change 1760 by gibbs@bitkeeper-linux-2.4 on 2002/12/12 14:30:40	aic79xx_osm.h:	aic7xxx_osm.h:		Complete adjustment for host template changes.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#90 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.h#109 editChange 1759 by gibbs@bitkeeper-linux-2.5 on 2002/12/12 14:25:19	aic79xx_osm.h:		Remove a stray reference to aic79xx_host.h.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#89 editChange 1754 by gibbs@bitkeeper-linux-2.4 on 2002/12/12 11:45:59	aic79xx_host.h:	aic7xxx_host.h:	aic79xx_osm.c:	aic7xxx_osm.c:		Eliminate separate Linux host template files and move		all host template entry ponts to one section of the Linux		osm.c file.		aic79xx_osm.c:	aic7xxx_osm.c:		Add support for larger disks under 2.5.X.		Changes prodded by: Christoph Hellwig <hch@sgi.com>Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_host.h#14 delete... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#92 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c#18 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_host.h#16 delete... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.c#160 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c#41 editChange 1751 by scottl@scottl-template on 2002/12/10 21:41:30	       Update driver version to 6.2.23Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.h#108 edit... //depot/linux_mod_devel/scsi.aic7xxx/rpm/aic7xxx.spec#9 edit... //depot/linux_mod_devel/scsi.aic7xxx/rpm/install.sh#12 editChange 1750 by gibbs@bitkeeper-linux-2.4 on 2002/12/10 20:51:23	aic79xx_host.h:	aic7xxx_host.h:		Set single_sg_okay for RH AS 2.1 too.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_host.h#13 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_host.h#15 editChange 1749 by gibbs@bitkeeper-linux-2.4 on 2002/12/10 20:45:35	aic79xx_host.h:	aic7xxx_host.h:		Use CONFIG_HIGHIO as the gating define for whether		the highmem_io host template field is present.  Also		assume that if the kernel version is less than 2.4.18,		then we must be under RedHat and need to call this field		can_dma_32.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_host.h#12 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_host.h#14 editChange 1748 by scottl@scottl-template on 2002/12/10 19:54:55	       Update driver version to 1.3.0.ALPHA3Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#88 edit... //depot/linux_mod_devel/scsi.aic79xx/rpm/aic79xx.spec#18 edit... //depot/linux_mod_devel/scsi.aic79xx/rpm/install.sh#20 editChange 1747 by gibbs@bitkeeper-linux-2.4 on 2002/12/10 18:49:21	aic79xx_host.h:	aic7xxx_host.h:		Enable the highmem_io option on kernels with the		new highmem bouncing code.		aic79xx_osm.c:	aic7xxx_osm.c:		Fix compilation error in the non-debug enabled case.		aic7xxx_osm.c:		Remove leftover #if 0'd code from DV merge.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_host.h#11 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#91 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_host.h#13 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.c#159 editChange 1746 by gibbs@bitkeeper-linux-2.5 on 2002/12/10 12:05:15	aic79xx_osm.c:	aic7xxx_osm.c:		Disable the shutdown hook for 2.5.X since it is called		to early to be useful.  Some other strategy will need to		be found.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#90 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.c#158 editChange 1745 by gibbs@bitkeeper-linux-2.5 on 2002/12/09 16:29:02	aic79xx_host.h:	aic79xx_osm.c:	aic79xx_osm_pci.c	aic7xxx_host.h:	aic7xxx_osm.c:	aic7xxx_osm_pci.c:		Conform to latest 2.5.X API changes.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_host.h#10 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#89 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c#17 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_host.h#12 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.c#157 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c#40 editChange 1743 by gibbs@bitkeeper-linux-2.4 on 2002/12/09 15:16:31	aic79xx.c:		Limit the syncrate after all option conformance		changes have taken place in ahd_devlimited_syncrate.		Changes in options may change the final syncrate we		accept.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.c#143 editChange 1741 by gibbs@bitkeeper-linux-2.4 on 2002/12/09 14:11:44	Makefile:		Turn off -g.Affected files ...... //depot/linux-aic79xx-2.4.0/drivers/scsi/aic7xxx/Makefile#10 editChange 1740 by scottl@scottl-template on 2002/12/09 10:15:50	       Update driver version to 6.2.22Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.h#107 edit... //depot/linux_mod_devel/scsi.aic7xxx/rpm/aic7xxx.spec#8 edit... //depot/linux_mod_devel/scsi.aic7xxx/rpm/install.sh#11 editChange 1737 by scottl@scottl-template on 2002/12/06 16:14:24	       Update driver version to 1.3.0.ALPHA2Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#87 edit... //depot/linux_mod_devel/scsi.aic79xx/rpm/aic79xx.spec#17 edit... //depot/linux_mod_devel/scsi.aic79xx/rpm/install.sh#19 editChange 1736 by gibbs@overdrive on 2002/12/06 16:04:51	Regenerate Linux firmware.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_reg.h#51 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_reg_print.c#32 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_seq.h#52 editChange 1735 by gibbs@bitkeeper-linux-2.4 on 2002/12/06 16:03:54	aic79xx.h:		Use consistent names for AH?_ASYNC_XFER_PERIOD.		aic79xx.c:	aic79xx_proc.c:		Print out IU, QAS, and RTI features when showing		transfer options.		aic79xx.c:		Save and restore the NEGOADDR address when setting		new transfer settings.  The sequencer performs lookups		in the negotiation table too and it expects NEGOADDR		to remain consistent across pause/unpause sessions.			Consistently use "offset" instead of "period" to determine		if we are running sync or not.			Add a SHOW_MESSAGES diagnostic for when we assert ATN		during message processing.		aic79xx_osm.c:	aic7xxx_osm.c:		In ah?_devlimited_syncrate enforce that we can never		attempt DT transfers if we are narrow. 			Remove the ALLOW_MEMIO config option.			Modify ah?_linux_filter_inquiry to assume that the		inquiry buffer is large enough and has been setup		correctly (i.e. memset to 0 prior to command issue)		so that range checks are unecessary.			Remove comments about the need to perform short inquiry		requests.  The code does this now.			Split out the bottom half handling of DV inquiry results		to make the code clearer.			Re-arrange the DV state machine so that:			o REB descriptor fetching is performed ASYNC so we		  don't confuse a failure with the need to fallback.		o Devices using Level 2 DV don't bother with Level 1		  DV.		o REB collisions are retried after a delay randomized		  by our ID on the bus.		o Level 2 DV is only attempted on devices that support		  DT transfers.			Fix ah?_linux_fallback to always favor the fastest		speed during fallback and to favor sync over async speeds.			Don't let a target reponding to a PPR message with 0		options set fool us into falling back to far.  We still		must try to negotiate without using PPR to get by some		expanders.					Use the SCB_SILENT flag on DV commands.		with 0 PPR options set			Fix large inquiry size calculation.  It was off by one.		We need to add 5 and not 4.		aic79xx_osm_pci.c:		Tell the user that our failure of the MMAP test is non		fatal.  We're just dropping back to PIO.		aiclib.c:	aiclib.h:		Add flags for a randomized delay in response to an error.			Include table entries for REB collisions.			Add the proper response codes for sense codes indicating		inquiry data has changed.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.c#142 edit... //depot/aic7xxx/aic7xxx/aic79xx.h#76 edit... //depot/aic7xxx/aic7xxx/aic7xxx.c#103 edit... //depot/aic7xxx/aic7xxx/aic7xxx.h#66 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#88 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#86 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c#16 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_proc.c#9 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.c#156 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm.h#106 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c#39 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aiclib.c#4 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aiclib.h#3 editChange 1734 by gibbs@bitkeeper-linux-2.4 on 2002/12/06 15:49:37	Config.in:		With the advent of our command line option and the		new memory mapped register test, there is no need		for a kernel compile time option to allow memory		mapped I/O.  Remove it.Affected files ...... //depot/linux-aic79xx-2.4.0/drivers/scsi/aic7xxx/Config.in#10 editChange 1733 by gibbs@bitkeeper-linux-2.4 on 2002/12/06 15:31:31	aic79xx.seq:		Correct ignore wide residue processing check for		a wide negotiation being in effect.  We must be		in the SCSI register window in order to access the		negotiation table.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.seq#74 editChange 1732 by gibbs@aslan on 2002/12/06 15:14:01	aicasm_gram.y:		Add two missing ';'s that for some reason yacc/bison		never complained about as a syntax error.Affected files ...... //depot/aic7xxx/aic7xxx/aicasm/aicasm_gram.y#27 editChange 1730 by cde@haywire on 2002/12/05 17:55:31	Add AHD_DV_STATE_INQ_SHORT_ASYNC to probe inquiry buff len	<enter description here>Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.c#87 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_osm.h#85 editChange 1729 by gibbs@aslan on 2002/12/05 17:23:45	aic7xxx.h:		AHD -> AHC.Affected files ...... //depot/aic7xxx/aic7xxx/aic7xxx.h#65 editChange 1728 by gibbs@aslan on 2002/12/05 17:21:45	aic79xx.c:	aic79xx.h:	aic7xxx.c:	aic7xxx.h:		Cleanup usage of the SCB_SILENT flag by using		a new macro SCB_IS_SILENT(scb).			Apply the silent treatment to outgoing LQ CRC		errors.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.c#141 edit... //depot/aic7xxx/aic7xxx/aic79xx.h#75 edit... //depot/aic7xxx/aic7xxx/aic7xxx.c#102 edit... //depot/aic7xxx/aic7xxx/aic7xxx.h#64 editChange 1727 by gibbs@aslan on 2002/12/05 17:01:15	aic79xx.c	aic7xxx.c:	aic79xx.h:	aic7xxx.h:		Implement the SCB_SILENT flag.  This is useful for		hushing up the driver during DV or other operations		that we expect to cause transmission errors.  The		messages will still print if the SHOW_MASKED_ERRORS		debug option is enabled.		aic7xxx_inline.h:	aic7xxx_pci.c:		Implement ahc_[in|out][w|l|q].  This removes the need		for manual 'or and shift" type operations through out		the driver.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.c#140 edit... //depot/aic7xxx/aic7xxx/aic79xx.h#74 edit... //depot/aic7xxx/aic7xxx/aic7xxx.c#101 edit... //depot/aic7xxx/aic7xxx/aic7xxx.h#63 edit... //depot/aic7xxx/aic7xxx/aic7xxx_inline.h#39 edit... //depot/aic7xxx/aic7xxx/aic7xxx_pci.c#54 editChange 1726 by gibbs@overdrive on 2002/12/04 17:33:18	Regenerate Linux firmwareAffected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_reg.h#20 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_reg_print.c#7 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic7xxx_seq.h#21 editChange 1725 by gibbs@aslan on 2002/12/04 17:32:26	aic7xxx.reg:		Add a definition for MAX_OFFSET.Affected files ...... //depot/aic7xxx/aic7xxx/aic7xxx.reg#36 editChange 1724 by gibbs@overdrive on 2002/12/04 16:35:11	Regenerate Linux firmware.Affected files ...... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_reg.h#50 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_reg_print.c#31 edit... //depot/aic7xxx/linux/drivers/scsi/aic7xxx/aic79xx_seq.h#51 editChange 1723 by gibbs@aslan on 2002/12/04 16:34:34	aic79xx.seq:		The sequencer downloading code assumes that all jump		labels are acurate in relation to a fully compiled		sequencer program (all patches downloaded).  Correct		a few occurances of a relative jump across a macro		that ended up jumping us into the last instruction		of the macro.Affected files ...... //depot/aic7xxx/aic7xxx/aic79xx.seq#73 edit