ata(4) says : To see the devices' current access modes, use the command line: sysctl hw.atamodes which results in the modes of the devices being displayed as a string like this: hw.atamodes: dma,pio,---,pio,dma,---,dma,---, (--- = no device) This means that ata0-master is in DMA mode, ata0-slave is in PIO mode, and so forth. You can set the mode with sysctl and a string like the above, for example: sysctl hw.atamodes=pio,pio,---,dma,pio,---,dma,---, The new modes are set as soon as the sysctl command returns. while it should refer to atacontrol(8) to change ata modes since the sysctl hw.atamodes doesn't exists anymore. How-To-Repeat: man 4 ata
On Wed, Jun 12, 2002 at 10:29:59PM +0200, Cyrille Lefevre wrote: > while it should refer to atacontrol(8) to change ata modes > since the sysctl hw.atamodes doesn't exists anymore. I think the ata(4) man page has missed a few MFCs. The following diff would sync it with -current (leaving out the bits which relate to -current only device stuff). Soren - would you like me to commit this when the code freeze is over, or would you like to do it on your next MFC? David. --- src/share/man/man4/ata.4:1.3.2.14 Sun Feb 10 07:24:08 2002 +++ src/share/man/man4/ata.4 Fri Nov 2 12:22:17 2001 @@ -72,11 +74,11 @@ .It Va hw.ata.atapi_dma set to 1 for DMA access, 0 for PIO (default is PIO). .It Va hw.ata.wc -set to 1 to enable Write Caching, 0 to disable (default is enabled) -(WARNING might cause data loss on power failures) +set to 1 to enable Write Caching, 0 to disable (default is disabled). +(WARNING: might cause data loss on power failures.) .It Va hw.ata.tags -set to 1 to enable Tagged Queuing support, 0 to disable (default is disabled) -(only IBM DPTA and DTLA drives support that) +set to 1 to enable Tagged Queuing support, 0 to disable (default is disabled). +(Only IBM DPTA and DTLA drives support that.) .El .Sh DESCRIPTION This driver provides access to disk drives, ATAPI CD-ROM and DVD drives, @@ -86,9 +88,9 @@ .Pp The currently supported controllers with their maximum speed include: .Pp -.Bl -tag -width "xxxxxxxxxxxxxxxxxxxxxxxxxxxxx" -compact +.Bl -tag -width "Promise Ultra/Fasttrak-100 TX2/TX4" -compact .It Acerlabs Aladdin -Ultra DMA 33 (UDMA2), 33 MB/sec +Ultra DMA 100 (UDMA5), 100 MB/sec (depending on model, max stated at boot) .It AMD 756 Ultra DMA 66 (UDMA4), 66 MB/sec .It AMD 766 @@ -101,10 +103,14 @@ Ultra DMA 100 (UDMA5), 100 MB/sec .It Cypress 82C693 DMA 2 (WDMA2), 16 MB/sec +.It Cyrix 5530 +Ultra DMA 33 (UDMA2), 33 MB/sec .It HighPoint HPT366 Ultra DMA 66 (UDMA4), 66 MB/sec .It HighPoint HPT370 Ultra DMA 100 (UDMA5), 100 MB/sec +.It HighPoint HPT372 +Ultra DMA 133 (UDMA6), 133 MB/sec .It Intel PIIX DMA 2 (WDMA2), 16 MB/sec .It Intel PIIX3 @@ -125,57 +131,41 @@ Ultra DMA 66 (UDMA4), 66 MB/sec .It Promise Ultra/Fasttrak-100 Ultra DMA 100 (UDMA5), 100 MB/sec +.It Promise Ultra/Fasttrak-100 TX2/TX4 +Ultra DMA 100 (UDMA5), 100 MB/sec .It ServerWorks ROSB4 Ultra DMA 33 (UDMA2), 33 MB/sec .It SiS 5591 Ultra DMA 33 (UDMA2), 33 MB/sec -.It Cyrix 5530 -Ultra DMA 33 (UDMA2), 33 MB/sec .It VIA 82C586 Ultra DMA 33 (UDMA2), 33 MB/sec +.It VIA 82C596 +Ultra DMA 66 (UDMA4), 66 MB/sec (depending on model, max stated at boot) .It VIA 82C686a Ultra DMA 66 (UDMA4), 66 MB/sec .It VIA 82C686b Ultra DMA 100 (UDMA5), 100 MB/sec .El .Pp -All unknown chipsets can be supported at the maximum speed of 16 MB/sec. +All unknown chipsets are supported at the maximum speed of 16 MB/sec. .Pp The .Nm driver also allows for changes to the transfer mode of the devices -at a later time when the system is up and running. +at a later time when the system is up and running, see +.Xr atacontrol 8 . .Pp The driver attempts to set the maximum performance transfer mode on your disk drives by selecting the highest possible DMA mode. ATAPI devices are left in PIO mode because DMA problems are common despite the device specifications. -You can always try to set DMA mode on an ATAPI device using the sysctl -method described here, +You can always try to set DMA mode on an ATAPI device using +.Xr atacontrol 8 , but be aware that your hardware might .Em not support it and can .Em hang the system. -.Pp -To see the devices' current access modes, use the command line: -.Pp -.Dl sysctl hw.atamodes -.Pp -which results in the modes of the devices being displayed as a string -like this: -.Pp -.Dl hw.atamodes: dma,pio,---,pio,dma,---,dma,---, (--- = no device) -.Pp -This means that ata0-master is in DMA mode, -ata0-slave is in PIO mode, -and so forth. -You can set the mode with sysctl and a string like the above, -for example: -.Pp -.Dl sysctl hw.atamodes=pio,pio,---,dma,pio,---,dma,---, -.Pp -The new modes are set as soon as the sysctl command returns. .Sh FILES .Bl -tag -width "/sys/i386/conf/GENERIC " -compact .It Pa /dev/ad* @@ -219,6 +209,9 @@ unless they are run at the non-UDMA4 device's lower speed. The driver has been designed to handle that kind of setup but lots of older devices do not like this. +.Sh SEE ALSO +.Xr atacontrol 8 , +.Xr burncd 8 .Sh HISTORY The .Nm
On Sat, Jun 15, 2002 at 01:47:55AM +0100, David Malone wrote: > -.Pp > -To see the devices' current access modes, use the command line: > -.Pp > -.Dl sysctl hw.atamodes > -.Pp > -which results in the modes of the devices being displayed as a string > -like this: > -.Pp > -.Dl hw.atamodes: dma,pio,---,pio,dma,---,dma,---, (--- = no device) > -.Pp > -This means that ata0-master is in DMA mode, > -ata0-slave is in PIO mode, > -and so forth. > -You can set the mode with sysctl and a string like the above, > -for example: > -.Pp > -.Dl sysctl hw.atamodes=pio,pio,---,dma,pio,---,dma,---, > -.Pp > -The new modes are set as soon as the sysctl command returns. > .Sh FILES > .Bl -tag -width "/sys/i386/conf/GENERIC " -compact > .It Pa /dev/ad* why do you get rid of this sample section, that I've translated to atacontrol(8) equivalents, which may help many people having troubles w/ ATA modes ? an alternative would be to move it to a true EXAMPLES section in atacontrol(8) ? Cyrille. -- Cyrille Lefevre mailto:cyrille.lefevre@laposte.net
> why do you get rid of this sample section, that I've translated to > atacontrol(8) equivalents, which may help many people having troubles > w/ ATA modes ? I was really just suggesting that we sync up ata(4) with -current. > an alternative would be to move it to a true EXAMPLES section in > atacontrol(8) ? I think adding them to the examples section of atacontrol would probably be a better idea, though Soren might have a better informed opinion. David.
It seems David Malone wrote: > On Wed, Jun 12, 2002 at 10:29:59PM +0200, Cyrille Lefevre wrote: > > while it should refer to atacontrol(8) to change ata modes > > since the sysctl hw.atamodes doesn't exists anymore. > > I think the ata(4) man page has missed a few MFCs. The following > diff would sync it with -current (leaving out the bits which relate > to -current only device stuff). > > Soren - would you like me to commit this when the code freeze is > over, or would you like to do it on your next MFC? Uhm I hashed out the changes with the RE@ and the agreement was made that they should commit it, apparently that newer happend :( -Søren
If memory serves me right, Soeren Schmidt wrote: > > Soren - would you like me to commit this when the code freeze is > > over, or would you like to do it on your next MFC? > > Uhm I hashed out the changes with the RE@ and the agreement was > made that they should commit it, apparently that newer happend :( That's not the way I remember things. :-p On 30 May, I sent you a diff which would sync up ata(4) in RELENG_4 with its counterpart in -CURRENT. Later that day, you sent me back a complete manpage, with the comment: "Actually this is more like it, but I [haven't] even gotten to commit it to -current yet :)" To me, this meant "hold off, I'm need to commit a new manpage to -CURRENT". I was waiting for your commit so I could do a MFC. Sorry we had "crossed wires" on this one. Since the code freeze is over, RE has no objections to updates of the ata(4) manpage. Bruce.
State Changed From-To: open->closed Committed! Thanks!