It is obvious that we are most likely going to miss 2.2.8.........
The set_geometry_intr is fixed for promise cards.
Chipset tinkering on done if the BIOS does nothing.
hde: set_geometry_intr: status=0x01 { Error }
hde: set_geometry_intr: error=0x04 { DriveStatusError }
or
hde: lost interrupt
hde: set_geometry_intr: status=0x00 { }
or
hda: set_geometry_intr: status=0x51 { DriveReady SeekComplete Error }
hda: set_geometry_intr: error=0x04 { DriveStatusError }
Should be corrected now.
Promise Ultra/33 is verified that one can force the third card into
Ultra DMA mode by force setting the UDMA_speed_bit. Change ::
#define PDC202XX_FORCE_BURST_BIT 0
to
#define PDC202XX_FORCE_BURST_BIT 1
See the top of v0.25 ./drivers/block/pdc202xx.c
This assumes a BIOS of v1.26 or newer.
Do not set or enable ::
#define PDC202XX_FORCE_MASTER_MODE 0
This is used by FastTrak and needs more setups in the "extra" space to
make it fully functional. One report for the Ultra/66 showed that it set
the card in to master_mode from pci_mode.
Generic ATA-66 is stubbed out since the rules on paper are not in
practice. That is to say if the adapter or ribbon are not present, then
the device will not report bit 13 of word93 as set.
The original drive list are back and default for ide-dma.c.
#define IDE_DMA_NEW_LISTINGS 0
The test the new idea for refining the the white/black lists to require
the model plus firmware change ::
#define IDE_DMA_NEW_LISTINGS 1
Note that in all of the past test patches, the modular nature of the
driver breaks. This should be corrected in theory.......
If you still get kernel dumps loading ide-modules with the above option
enabled, please forward the dump.
Both ./drivers/block/hpt343.c and ./drivers/block/ide-pci.c
have new code to allow for IDE Express HPT345 (with stamp HPT363) to be
detected and distingished from it identically reporting cousin HPT343.
There is a BIOS hook problem in the HPT345 card for checking partition ID
codes. This is will not boot Linux..........
NOTE that ./drivers/block/ide-pci.c will update the bootable flag for
HPT345 to that of "OFFBOARD" from "NEVERBOARD"
"set_multmode" has been addressed.
hda: set_multmode: status=0x51 { DriveReady SeekComplete Error }
hda: set_multmode: error=0x04 { DriveStatusError }
The IBM disks that report are now handled 16383,15,63.
GOOD/BAD diskcheck is now handled as part of ide_dmaproc, so that chipset
code can get cleaner access to the test.
CMD643 simplex is handled by ide_dma.c.
All code that was previously declared in "pci_init" are moved to
"pcibios_scan_buglist" in ./arch/i386/kernel/bios32.c.
It may be necessary to duplicate the ide-pci-cards in the other arch's
that can address these animals.
Since ali15x3.c, via82c586.c, or piix.c do not have (U)DMA setup code
started or ready for testing, if your BIOS fails to perform its task
correctly, there is not help yet.
./drivers/block/piix.c is nearing completion for chipset tuning and
correction, but it is still extremely pre-alpha; however, the default
tuneproc is functional and will correct chipset code.
To add ide4/5 devices call "sh ./drivers/block/MAKEDEV-IDE45"
To add ide6/7 devices call "sh ./drivers/block/MAKEDEV-IDE67"
./drivers/block/Config.in is returned to working order for
"Builtin PowerMac IDE support"
Some of the ./Documentation/Configure.help is updated, but not that there
are some fields that are present with at info. This was done to remind me
to get is added quickly.
If there are no real pains reported late monday or early tuesday I will
attempt to submit this to Linus..........
If there are any problems, I expect us to begin entry in 2.3.X and then
move it backwards to 2.2.X.
Andre Hedrick
The Linux IDE guy
Candidate for pre-patch-2.2.8-series or pre-patch-2.2.9-series
http://www.dyer.vanderbilt.edu/server/udma/
2.2.7.uniform-ide-6.19.golf.patch.gz
http://www.dyer.vanderbilt.edu/server/udma/2.2.7.uniform-ide-6.19.patch.gz
http://www.dyer.vanderbilt.edu/server/udma/ide.2.0.37pre11+pat7.gz
http://www.dyer.vanderbilt.edu/server/udma/hdparm-3.5i.patch.gz
APC UPS Daemon Support Center.
http://www.brisse.dk/site/apcupsd/
GPLed source on April 7, 1999
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/