LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH 2.4] decrypt/update ide help entries
@ 2004-05-17 17:05 O.Sezer
       [not found] ` <200405172020.36892.bzolnier@elka.pw.edu.pl>
  0 siblings, 1 reply; 7+ messages in thread
From: O.Sezer @ 2004-05-17 17:05 UTC (permalink / raw)
  To: marcelo.tosatti; +Cc: B.Zolnierkiewicz, linux-kernel

[-- Attachment #1: Type: text/plain, Size: 122 bytes --]

This has been in Alan's tree for ages, why not merge
in mainline? Patch below, happily stolen from -ac/-pac.

Özkan Sezer

[-- Attachment #2: ide_help_update.patch --]
[-- Type: text/plain, Size: 3750 bytes --]

--- 27p2/Documentation/Configure.help~
+++ 27p2/Documentation/Configure.help
@@ -1309,20 +1309,23 @@
 
   If unsure, say N.
 
-PROMISE PDC20246/PDC20262/PDC20265/PDC20267/PDC20268 support
+PROMISE PDC20246/PDC20262/PDC20265/PDC20267 support
 CONFIG_BLK_DEV_PDC202XX_OLD
-  Promise Ultra33 or PDC20246
-  Promise Ultra66 or PDC20262
-  Promise Ultra100 or PDC20265/PDC20267/PDC20268
+  Promise Ultra 33 [PDC20246]
+  Promise Ultra 66 [PDC20262]
+  Promise FastTrak 66 [PDC20263]
+  Promise MB Ultra 100 [PDC20265]
+  Promise Ultra 100 [PDC20267]
 
   This driver adds up to 4 more EIDE devices sharing a single
-  interrupt. This add-on card is a bootable PCI UDMA controller. Since
+  interrupt. This device is a bootable PCI UDMA controller. Since
   multiple cards can be installed and there are BIOS ROM problems that
-  happen if the BIOS revisions of all installed cards (three-max) do
+  happen if the BIOS revisions of all installed cards (max of three) do
   not match, the driver attempts to do dynamic tuning of the chipset
-  at boot-time for max-speed.  Ultra33 BIOS 1.25 or newer is required
+  at boot-time for max speed.  Ultra33 BIOS 1.25 or newer is required
   for more than one card. This card may require that you say Y to
-  "Special UDMA Feature".
+  "Override-Enable UDMA for Promise Contr." (or "Special UDMA Feature")
+  to force UDMA mode for connected UDMA capable disk drives.
 
   If you say Y here, you need to say Y to "Use DMA by default when
   available" as well.
@@ -1355,7 +1358,30 @@
 
   If unsure, say N.
 
-Special UDMA Feature
+PROMISE PDC202{68|69|70|71|75|76|77} support
+CONFIG_BLK_DEV_PDC202XX_NEW
+  Promise Ultra 100 TX2 [PDC20268]
+  Promise Ultra 133 PTX2 [PDC20269]
+  Promise FastTrak LP/TX2/TX4 [PDC20270]
+  Promise FastTrak TX2000 [PDC20271]
+  Promise MB Ultra 133 [PDC20275]
+  Promise MB FastTrak 133 [PDC20276]
+  Promise FastTrak 133 [PDC20277]
+
+  This driver adds up to 4 more EIDE devices sharing a single
+  interrupt. This device is a bootable PCI UDMA controller. Since
+  multiple cards can be installed and there are BIOS ROM problems that
+  happen if the BIOS revisions of all installed cards (max of five) do
+  not match, the driver attempts to do dynamic tuning of the chipset
+  at boot-time for max speed.  Ultra33 BIOS 1.25 or newer is required
+  for more than one card. 
+
+  If you say Y here, you need to say Y to "Use DMA by default when
+  available" as well.
+
+  If unsure, say N.
+
+Override-Enable UDMA for Promise Controllers
 CONFIG_PDC202XX_BURST
   This option causes the pdc202xx driver to enable UDMA modes on the
   PDC202xx even when the PDC202xx BIOS has not done so.
@@ -1365,14 +1391,24 @@
   used successfully on a PDC20265/Ultra100, allowing use of UDMA modes
   when the PDC20265 BIOS has been disabled (for faster boot up).
 
-  Please read the comments at the top of
-  <file:drivers/ide/pci/pdc202xx_old.c>.
-
   If unsure, say N.
 
-Special FastTrak Feature
+Use FastTrak RAID capable device as plain IDE controller
 CONFIG_PDC202XX_FORCE
-  For FastTrak enable overriding BIOS.
+  Setting this option causes the kernel to use your Promise IDE disk
+  controller as an ordinary IDE controller, rather than as a FastTrak
+  RAID controller. RAID is a system for using multiple physical disks
+  as one virtual disk.
+
+  You need to say Y here if you have a PDC20276 IDE interface but either
+  you do not have a RAID disk array, or you wish to use the Linux
+  internal RAID software (/dev/mdX).
+
+  You need to say N here if you wish to use your Promise controller to
+  control a FastTrak RAID disk array, and you you must also say Y to
+  CONFIG_BLK_DEV_ATARAID_PDC.
+
+  If unsure, say Y.
 
 SiS5513 chipset support
 CONFIG_BLK_DEV_SIS5513


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries
       [not found] ` <200405172020.36892.bzolnier@elka.pw.edu.pl>
@ 2004-05-18 13:42   ` O.Sezer
  2004-05-18 14:09     ` Bartlomiej Zolnierkiewicz
  0 siblings, 1 reply; 7+ messages in thread
From: O.Sezer @ 2004-05-18 13:42 UTC (permalink / raw)
  To: Bartlomiej Zolnierkiewicz; +Cc: marcelo.tosatti, linux-kernel

[-- Attachment #1: Type: text/plain, Size: 1554 bytes --]

Bartlomiej Zolnierkiewicz wrote:
> This patch was disccussed long time ago and nobody cared to correct it.

Hmm.. too many errors, careless duplicate entries.. ...
too many trust in trusted trees.. Me embarassed ;)

> +  "Override-Enable UDMA for Promise Contr." (or "Special UDMA Feature")
> +  to force UDMA mode for connected UDMA capable disk drives.
> 
>  It is about forcing burst UDMA transfers not UDMA mode.

Fixed

> +PROMISE PDC202{68|69|70|71|75|76|77} support
> +CONFIG_BLK_DEV_PDC202XX_NEW
[...]
> This is just copied from CONFIG_BLK_DEV_PDC202XX_OLD
> ('Ultra33') and probably is incorrect for newer Promise controllers.

Removed the old copied one wrote something generic

> +  You need to say Y here if you have a PDC20276 IDE interface but either
> +  you do not have a RAID disk array, or you wish to use the Linux
> +  internal RAID software (/dev/mdX).
> 
> This is needed not only for PDC20276.

I think this time I took the correct chipset names upon reading
pdc202XX_old.h and pdc202XX_new.h. Please check.

> +  You need to say N here if you wish to use your Promise controller to
> +  control a FastTrak RAID disk array, and you you must also say Y to
> +  CONFIG_BLK_DEV_ATARAID_PDC.
> 
> This is incorrect.
> 
> You must say Y to this option and to CONFIG_BLK_DEV_ATARAID_PDC.

Whoops, sorry. Fixed.

> If you want to correct Promise IDE help entries, do it for 2.6 first.
> 

Don't know much about 2.6; if you can review this one, I can make
similar changes for 2.6 (in case options didn't change much).

Thanks,
Özkan Sezer

[-- Attachment #2: ide_help_update2.patch --]
[-- Type: text/plain, Size: 5323 bytes --]

--- 27p2/Documentation/Configure.help.orig
+++ 27p2/Documentation/Configure.help
@@ -1309,20 +1309,22 @@
 
   If unsure, say N.
 
-PROMISE PDC20246/PDC20262/PDC20265/PDC20267/PDC20268 support
+PROMISE PDC20246/PDC20262/PDC20265/PDC20267 support
 CONFIG_BLK_DEV_PDC202XX_OLD
-  Promise Ultra33 or PDC20246
-  Promise Ultra66 or PDC20262
-  Promise Ultra100 or PDC20265/PDC20267/PDC20268
+  Promise Ultra 33 [PDC20246]
+  Promise Ultra 66 [PDC20262]
+  Promise FastTrak 66 [PDC20263]
+  Promise MB Ultra 100 [PDC20265]
+  Promise Ultra 100 [PDC20267]
 
   This driver adds up to 4 more EIDE devices sharing a single
-  interrupt. This add-on card is a bootable PCI UDMA controller. Since
+  interrupt. These devices are bootable PCI UDMA controllers. Since
   multiple cards can be installed and there are BIOS ROM problems that
-  happen if the BIOS revisions of all installed cards (three-max) do
+  happen if the BIOS revisions of all installed cards (max of three) do
   not match, the driver attempts to do dynamic tuning of the chipset
   at boot-time for max-speed.  Ultra33 BIOS 1.25 or newer is required
   for more than one card. This card may require that you say Y to
-  "Special UDMA Feature".
+  "Force Burst UDMA transfers" (old name: "Special UDMA Feature").
 
   If you say Y here, you need to say Y to "Use DMA by default when
   available" as well.
@@ -1342,22 +1344,18 @@
   Promise MB FastTrak 133 [PDC20276]
   Promise FastTrak 133 [PDC20277]
 
-  This driver adds up to 4 more EIDE devices sharing a single
-  interrupt. This device is a bootable PCI UDMA controller. Since
-  multiple cards can be installed and there are BIOS ROM problems that
-  happen if the BIOS revisions of all installed cards (max of five) do
-  not match, the driver attempts to do dynamic tuning of the chipset
-  at boot-time for max speed.  Ultra33 BIOS 1.25 or newer is required
-  for more than one card.
+  This driver adds support for the newer Udma100/133 Promise chipsets
+  listed above. These chipsets are not driven by the pdc202XX_old,
+  but by this pdc202XX_new driver.
 
   If you say Y here, you need to say Y to "Use DMA by default when
   available" as well.
 
   If unsure, say N.
 
-Special UDMA Feature
+Force Burst UDMA transfers
 CONFIG_PDC202XX_BURST
-  This option causes the pdc202xx driver to enable UDMA modes on the
+  This option causes "pdc202xx_old" driver to enable UDMA modes on the
   PDC202xx even when the PDC202xx BIOS has not done so.
 
   It was originally designed for the PDC20246/Ultra33, whose BIOS will
@@ -1365,14 +1363,33 @@
   used successfully on a PDC20265/Ultra100, allowing use of UDMA modes
   when the PDC20265 BIOS has been disabled (for faster boot up).
 
-  Please read the comments at the top of
-  <file:drivers/ide/pci/pdc202xx_old.c>.
+  (Please read the comments in <file:drivers/ide/pci/pdc202xx_old.c>
+   about this option.)
 
   If unsure, say N.
 
-Special FastTrak Feature
+Use FastTrak RAID capable device as plain IDE controller
 CONFIG_PDC202XX_FORCE
-  For FastTrak enable overriding BIOS.
+  This option has effect on Promise chipsets PDC20270 and PDC20276
+  while using the pdc202xx_new driver.
+
+  Setting this option causes the kernel to use your Promise IDE disk
+  controller as an ordinary IDE controller, rather than as a FastTrak
+  RAID controller (RAID is a system for using multiple physical disks
+  as one virtual disk).
+
+  You need to say Y here if you have one of the above mentioned IDE
+  interfaces,  but either you do not have a RAID disk array,  or you
+  wish to use the Linux internal RAID software (/dev/mdX).
+
+  If you wish to use your Promise controller to control a FastTrak
+  RAID disk array, you need to say Y here AND you you must also say Y
+  to CONFIG_BLK_DEV_ATARAID_PDC.
+
+  This option also has effect on the chipsets run by the pdc202xx_old
+  driver ( PDC202{46|62|63|65|67} ).
+
+  If unsure, say Y.
 
 SiS5513 chipset support
 CONFIG_BLK_DEV_SIS5513
--- 27p2/drivers/ide/Config.in.orig
+++ 27p2/drivers/ide/Config.in
@@ -64,10 +64,10 @@
 	    dep_tristate '    NS87415 chipset support' CONFIG_BLK_DEV_NS87415 $CONFIG_BLK_DEV_IDEDMA_PCI
 	    dep_tristate '    OPTi 82C621 chipset enhanced support (EXPERIMENTAL)' CONFIG_BLK_DEV_OPTI621 $CONFIG_EXPERIMENTAL
 	    dep_tristate '    PROMISE PDC202{46|62|65|67} support' CONFIG_BLK_DEV_PDC202XX_OLD $CONFIG_BLK_DEV_IDEDMA_PCI
-	    dep_mbool     '      Special UDMA Feature' CONFIG_PDC202XX_BURST $CONFIG_BLK_DEV_PDC202XX_OLD $CONFIG_BLK_DEV_IDEDMA_PCI
+	    dep_mbool     '      Force Burst UDMA transfers' CONFIG_PDC202XX_BURST $CONFIG_BLK_DEV_PDC202XX_OLD $CONFIG_BLK_DEV_IDEDMA_PCI
 	    dep_tristate '    PROMISE PDC202{68|69|70|71|75|76|77} support' CONFIG_BLK_DEV_PDC202XX_NEW $CONFIG_BLK_DEV_IDEDMA_PCI
 	    if [ "$CONFIG_BLK_DEV_PDC202XX_OLD" = "y" -o "$CONFIG_BLK_DEV_PDC202XX_OLD" = "m" -o "$CONFIG_BLK_DEV_PDC202XX_NEW" = "y" -o "$CONFIG_BLK_DEV_PDC202XX_NEW" = "m" ]; then
-	        bool     '    Special FastTrak Feature' CONFIG_PDC202XX_FORCE
+	        bool     '    Use FastTrak RAID capable device as plain IDE controller' CONFIG_PDC202XX_FORCE
 	    fi
 	    dep_tristate '    RZ1000 chipset bugfix/support' CONFIG_BLK_DEV_RZ1000 $CONFIG_X86
 	    dep_tristate '    SCx200 chipset support' CONFIG_BLK_DEV_SC1200 $CONFIG_BLK_DEV_IDEDMA_PCI

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries
  2004-05-18 13:42   ` O.Sezer
@ 2004-05-18 14:09     ` Bartlomiej Zolnierkiewicz
  2004-05-18 14:18       ` O.Sezer
  0 siblings, 1 reply; 7+ messages in thread
From: Bartlomiej Zolnierkiewicz @ 2004-05-18 14:09 UTC (permalink / raw)
  To: O.Sezer; +Cc: marcelo.tosatti, linux-kernel

On Tuesday 18 of May 2004 15:42, O.Sezer wrote:
> Bartlomiej Zolnierkiewicz wrote:
> > This patch was disccussed long time ago and nobody cared to correct it.
>
> Hmm.. too many errors, careless duplicate entries.. ...
> too many trust in trusted trees.. Me embarassed ;)

hehe

> > +  "Override-Enable UDMA for Promise Contr." (or "Special UDMA Feature")
> > +  to force UDMA mode for connected UDMA capable disk drives.
> >
> >  It is about forcing burst UDMA transfers not UDMA mode.
>
> Fixed

OK

> > +PROMISE PDC202{68|69|70|71|75|76|77} support
> > +CONFIG_BLK_DEV_PDC202XX_NEW
>
> [...]
>
> > This is just copied from CONFIG_BLK_DEV_PDC202XX_OLD
> > ('Ultra33') and probably is incorrect for newer Promise controllers.
>
> Removed the old copied one wrote something generic

OK

> > +  You need to say Y here if you have a PDC20276 IDE interface but either
> > +  you do not have a RAID disk array, or you wish to use the Linux
> > +  internal RAID software (/dev/mdX).
> >
> > This is needed not only for PDC20276.
>
> I think this time I took the correct chipset names upon reading
> pdc202XX_old.h and pdc202XX_new.h. Please check.

OK

> > +  You need to say N here if you wish to use your Promise controller to
> > +  control a FastTrak RAID disk array, and you you must also say Y to
> > +  CONFIG_BLK_DEV_ATARAID_PDC.
> >
> > This is incorrect.
> >
> > You must say Y to this option and to CONFIG_BLK_DEV_ATARAID_PDC.
>
> Whoops, sorry. Fixed.

+  Setting this option causes the kernel to use your Promise IDE disk
+  controller as an ordinary IDE controller, rather than as a FastTrak
+  RAID controller (RAID is a system for using multiple physical disks
+  as one virtual disk).
+
+  You need to say Y here if you have one of the above mentioned IDE
+  interfaces,  but either you do not have a RAID disk array,  or you
+  wish to use the Linux internal RAID software (/dev/mdX).
+
+  If you wish to use your Promise controller to control a FastTrak
+  RAID disk array, you need to say Y here AND you you must also say Y
+  to CONFIG_BLK_DEV_ATARAID_PDC.

This sounds awful. ;-)

Use controller even if marked as disabled by BIOS
CONFIG_PDC202XX_FORCE

Say Y unless you want to use Promise proprietary RAID driver.

Makes a lot more sense IMHO.
BTW it needs to be split-up for old and new driver (hint, hint!).

> > If you want to correct Promise IDE help entries, do it for 2.6 first.
>
> Don't know much about 2.6; if you can review this one, I can make
> similar changes for 2.6 (in case options didn't change much).

No, they are the same (see drivers/ide/Kconfig).

Thanks,
Bartlomiej


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries
  2004-05-18 14:09     ` Bartlomiej Zolnierkiewicz
@ 2004-05-18 14:18       ` O.Sezer
  2004-05-18 14:31         ` Bartlomiej Zolnierkiewicz
  0 siblings, 1 reply; 7+ messages in thread
From: O.Sezer @ 2004-05-18 14:18 UTC (permalink / raw)
  To: Bartlomiej Zolnierkiewicz; +Cc: marcelo.tosatti, linux-kernel

Bartlomiej Zolnierkiewicz wrote:
[...]
Thanks for your patience.

> BTW it needs to be split-up for old and new driver (hint, hint!).
> 

Why? Do you think that an admin may want to enable/disable it for
Chipset-A/pdc-old and disable/enable it for chipset-B/pdc-new ?

Thanks;
Özkan Sezer


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries
  2004-05-18 14:18       ` O.Sezer
@ 2004-05-18 14:31         ` Bartlomiej Zolnierkiewicz
  0 siblings, 0 replies; 7+ messages in thread
From: Bartlomiej Zolnierkiewicz @ 2004-05-18 14:31 UTC (permalink / raw)
  To: O.Sezer; +Cc: marcelo.tosatti, linux-kernel

On Tuesday 18 of May 2004 16:18, O.Sezer wrote:
> Bartlomiej Zolnierkiewicz wrote:
> [...]
> Thanks for your patience.
>
> > BTW it needs to be split-up for old and new driver (hint, hint!).
>
> Why? Do you think that an admin may want to enable/disable it for
> Chipset-A/pdc-old and disable/enable it for chipset-B/pdc-new ?

dunno but it is easier to add help entries for *_OLD and *_NEW ;-)
or we can remove FIXME from Kconfig (2.6) instead

> Thanks;
> Özkan Sezer


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries
  2004-05-17 17:35 dwm
@ 2004-05-17 17:52 ` O.Sezer
  0 siblings, 0 replies; 7+ messages in thread
From: O.Sezer @ 2004-05-17 17:52 UTC (permalink / raw)
  To: dwm; +Cc: marcelo.tosatti, B.Zolnierkiewicz, linux-kernel

dwm@austin.ibm.com wrote:
> ...
> 
>>+  Promise MB Ultra 133 [PDC20275]
> 
> 
> This controller is UDMA5 (100) max.  At least the ones I have seen.
> 
> ++doug

http://www.promise.com/product/oem_ataasic_pdc20275_eng.htm
says the contrary:
"PROMISE PDC20275 ASIC is a PCI bus mastering ATA/ATAPI controller
chip which supports complete UDMA/133 specifications. [...] "

Anyone else knows otherwise?

Özkan


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2.4] decrypt/update ide help entries 
@ 2004-05-17 17:35 dwm
  2004-05-17 17:52 ` O.Sezer
  0 siblings, 1 reply; 7+ messages in thread
From: dwm @ 2004-05-17 17:35 UTC (permalink / raw)
  To: O.Sezer; +Cc: marcelo.tosatti, B.Zolnierkiewicz, linux-kernel


On Mon, 17 May 2004 20:05:43 +0300, "O.Sezer" wrote:
...
>This has been in Alan's tree for ages, why not merge
>in mainline? Patch below, happily stolen from -ac/-pac.
>
...
>+  Promise MB Ultra 133 [PDC20275]

This controller is UDMA5 (100) max.  At least the ones I have seen.

++doug




^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2004-05-18 14:29 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-05-17 17:05 [PATCH 2.4] decrypt/update ide help entries O.Sezer
     [not found] ` <200405172020.36892.bzolnier@elka.pw.edu.pl>
2004-05-18 13:42   ` O.Sezer
2004-05-18 14:09     ` Bartlomiej Zolnierkiewicz
2004-05-18 14:18       ` O.Sezer
2004-05-18 14:31         ` Bartlomiej Zolnierkiewicz
2004-05-17 17:35 dwm
2004-05-17 17:52 ` O.Sezer

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).