Bug 75021

Summary: hw.ata.atapi_dma patch for ata(4) man page
Product: Documentation Reporter: Craig Rodrigues <rodrigc>
Component: Books & ArticlesAssignee: Peter Pentchev <roam>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: Latest   
Hardware: Any   
OS: Any   
Attachments:
Description Flags
file.diff none

Description Craig Rodrigues 2004-12-13 16:00:57 UTC
In ata(4), it states:

 hw.ata.atapi_dma
     set to 1 for DMA access, 0 for PIO (default is PIO).

In src/sys/dev/ata/ata-all.c, this sysctl is by default
set to 1 (on -CURRENT, and also on 5.3-RELEASE).

Perhaps this patch is in order?
Comment 1 Joel Dahl 2004-12-19 11:32:27 UTC
On Mon, 2004-12-13 at 10:50 -0500, Craig Rodrigues wrote:	
> >Description:
> 
> In ata(4), it states:
> 
>  hw.ata.atapi_dma
>      set to 1 for DMA access, 0 for PIO (default is PIO).
> 
> In src/sys/dev/ata/ata-all.c, this sysctl is by default
> set to 1 (on -CURRENT, and also on 5.3-RELEASE).
> 
> Perhaps this patch is in order?
> 	
> >Fix:
> 
> --- share/man/man4/ata.4.orig   Thu Dec  9 19:07:53 2004
> +++ share/man/man4/ata.4        Thu Dec  9 19:08:08 2004
> @@ -86,7 +86,7 @@
>  .It Va hw.ata.ata_dma
>  set to 1 for DMA access, 0 for PIO (default is DMA).
>  .It Va hw.ata.atapi_dma
> -set to 1 for DMA access, 0 for PIO (default is PIO).
> +set to 1 for DMA access, 0 for PIO (default is DMA).
>  .It Va hw.ata.wc
>  set to 1 to enable Write Caching, 0 to disable (default is enabled).
>  .Em WARNING :

There is another part in ata(4) that also requires a correction,
(assuming Craig's patch is correct):

"
ATAPI devices are set to PIO mode by default because severe DMA problems
are common even if the device capabilities indicate support.
"

ATAPI devices are set to DMA mode by default (as Craig mentions), so
someone should probably reword this part a bit.
Comment 2 Peter Pentchev freebsd_committer freebsd_triage 2004-12-21 20:02:58 UTC
State Changed
From-To: open->closed

Both the hw.ata.atapi_dma default and the discussion further down the page 
were fixed in -CURRENT.  I'll merge the changes into 5.x in a week.
Comment 3 Peter Pentchev freebsd_committer freebsd_triage 2004-12-21 20:05:21 UTC
State Changed
From-To: closed->patched

Not my day, is it... 


Comment 4 Peter Pentchev freebsd_committer freebsd_triage 2004-12-21 20:05:21 UTC
Responsible Changed
From-To: freebsd-doc->roam

I'll take care of the MFC to 5.x.
Comment 5 Peter Pentchev freebsd_committer freebsd_triage 2005-02-27 14:12:50 UTC
State Changed
From-To: patched->closed

The fixes were merged to the RELENG_5 branch.