linux-ide.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range
@ 2008-10-27 13:39 Atsushi Nemoto
  2008-10-27 14:07 ` Sergei Shtylyov
  0 siblings, 1 reply; 5+ messages in thread
From: Atsushi Nemoto @ 2008-10-27 13:39 UTC (permalink / raw)
  To: linux-mips; +Cc: linux-ide, Bartlomiej Zolnierkiewicz, ralf, sshtylyov

SHWT value is used as address valid to -CSx assertion and -CSx to -DIOx
assertion setup time, and contrarywise, -DIOx to -CSx release and -CSx
release to address invalid hold time, so it actualy applies 4 times and
so constitutes -DIOx recovery time.  Check requirement of the recovery
time and cycle time.  Also check SHWT maximum value.

Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
---
 drivers/ide/tx4938ide.c |    9 ++++++++-
 1 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c
index fa660f9..7e4820e 100644
--- a/drivers/ide/tx4938ide.c
+++ b/drivers/ide/tx4938ide.c
@@ -39,10 +39,17 @@ static void tx4938ide_tune_ebusc(unsigned int ebus_ch,
 	/* Address-valid to DIOR/DIOW setup */
 	shwt = DIV_ROUND_UP(t->setup, cycle);
 
+	/* -DIOx recovery time (SHWT * 4) and cycle time requirement */
+	while (shwt * cycle * 4 + t->act8b < t->cycle)
+		shwt++;
+	if (shwt > 7) {
+		pr_warning("tx4938ide: SHWT violation (%d)\n", shwt);
+		shwt = 7;
+	}
 	pr_debug("tx4938ide: ebus %d, bus cycle %dns, WT %d, SHWT %d\n",
 		 ebus_ch, cycle, wt, shwt);
 
-	__raw_writeq((cr & ~(0x3f007ull)) | (wt << 12) | shwt,
+	__raw_writeq((cr & ~0x3f007ull) | (wt << 12) | shwt,
 		     &tx4938_ebuscptr->cr[ebus_ch]);
 }
 
-- 
1.5.6.3

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range
  2008-10-27 13:39 [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range Atsushi Nemoto
@ 2008-10-27 14:07 ` Sergei Shtylyov
  2008-10-27 14:52   ` Atsushi Nemoto
  0 siblings, 1 reply; 5+ messages in thread
From: Sergei Shtylyov @ 2008-10-27 14:07 UTC (permalink / raw)
  To: Atsushi Nemoto; +Cc: linux-mips, linux-ide, Bartlomiej Zolnierkiewicz, ralf

Hello.

Atsushi Nemoto wrote:

> SHWT value is used as address valid to -CSx assertion and -CSx to -DIOx
> assertion setup time, and contrarywise, -DIOx to -CSx release and -CSx
> release to address invalid hold time, so it actualy applies 4 times and
> so constitutes -DIOx recovery time.  Check requirement of the recovery
> time and cycle time.  Also check SHWT maximum value.

> Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
> Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>

> diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c
> index fa660f9..7e4820e 100644
> --- a/drivers/ide/tx4938ide.c
> +++ b/drivers/ide/tx4938ide.c
> @@ -39,10 +39,17 @@ static void tx4938ide_tune_ebusc(unsigned int ebus_ch,
>  	/* Address-valid to DIOR/DIOW setup */
>  	shwt = DIV_ROUND_UP(t->setup, cycle);
>  
> +	/* -DIOx recovery time (SHWT * 4) and cycle time requirement */
> +	while (shwt * cycle * 4 + t->act8b < t->cycle)

    This can lead to adding an extra setup cycle due to use of "non-quantized" 
active time, here's more precise expression for the cycle time:

    (shwt * 4 + wt + (wt ? 2 : 3)) * cycle

or even this, more obsure:

    (shwt * 4 + wt + !wt + 2) * cycle

MBR, Sergei

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range
  2008-10-27 14:07 ` Sergei Shtylyov
@ 2008-10-27 14:52   ` Atsushi Nemoto
  2008-10-27 23:12     ` Sergei Shtylyov
  0 siblings, 1 reply; 5+ messages in thread
From: Atsushi Nemoto @ 2008-10-27 14:52 UTC (permalink / raw)
  To: sshtylyov; +Cc: linux-mips, linux-ide, bzolnier, ralf

On Mon, 27 Oct 2008 17:07:05 +0300, Sergei Shtylyov <sshtylyov@ru.mvista.com> wrote:
> > +	/* -DIOx recovery time (SHWT * 4) and cycle time requirement */
> > +	while (shwt * cycle * 4 + t->act8b < t->cycle)
> 
>     This can lead to adding an extra setup cycle due to use of "non-quantized" 
> active time, here's more precise expression for the cycle time:
> 
>     (shwt * 4 + wt + (wt ? 2 : 3)) * cycle
> 
> or even this, more obsure:
> 
>     (shwt * 4 + wt + !wt + 2) * cycle

Hm, indeed.  I prefer former one.  Revised.  Thanks!


From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Subject: [PATCH] tx4938ide: Check minimum cycle time and SHWT range (v2)

SHWT value is used as address valid to -CSx assertion and -CSx to -DIOx
assertion setup time, and contrarywise, -DIOx to -CSx release and -CSx
release to address invalid hold time, so it actualy applies 4 times and
so constitutes -DIOx recovery time.  Check requirement of the recovery
time and cycle time.  Also check SHWT maximum value.

Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
---
 drivers/ide/tx4938ide.c |    9 ++++++++-
 1 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c
index a0d10a9..796289c 100644
--- a/drivers/ide/tx4938ide.c
+++ b/drivers/ide/tx4938ide.c
@@ -39,10 +39,17 @@ static void tx4938ide_tune_ebusc(unsigned int ebus_ch,
 	/* Address-valid to DIOR/DIOW setup */
 	shwt = DIV_ROUND_UP(t->setup, cycle);
 
+	/* -DIOx recovery time (SHWT * 4) and cycle time requirement */
+	while ((shwt * 4 + wt + (wt ? 2 : 3)) * cycle < t->cycle)
+		shwt++;
+	if (shwt > 7) {
+		pr_warning("tx4938ide: SHWT violation (%d)\n", shwt);
+		shwt = 7;
+	}
 	pr_debug("tx4938ide: ebus %d, bus cycle %dns, WT %d, SHWT %d\n",
 		 ebus_ch, cycle, wt, shwt);
 
-	__raw_writeq((cr & ~(0x3f007ull)) | (wt << 12) | shwt,
+	__raw_writeq((cr & ~0x3f007ull) | (wt << 12) | shwt,
 		     &tx4938_ebuscptr->cr[ebus_ch]);
 }
 
-- 
1.5.6.3



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range
  2008-10-27 14:52   ` Atsushi Nemoto
@ 2008-10-27 23:12     ` Sergei Shtylyov
  2008-10-29 19:12       ` Bartlomiej Zolnierkiewicz
  0 siblings, 1 reply; 5+ messages in thread
From: Sergei Shtylyov @ 2008-10-27 23:12 UTC (permalink / raw)
  To: Atsushi Nemoto; +Cc: linux-mips, linux-ide, bzolnier, ralf

Hello.

Atsushi Nemoto wrote:

> From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
> Subject: [PATCH] tx4938ide: Check minimum cycle time and SHWT range (v2)
>
> SHWT value is used as address valid to -CSx assertion and -CSx to -DIOx
> assertion setup time, and contrarywise, -DIOx to -CSx release and -CSx
> release to address invalid hold time, so it actualy applies 4 times and
> so constitutes -DIOx recovery time.  Check requirement of the recovery
> time and cycle time.  Also check SHWT maximum value.
>
> Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
> Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
>   

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>

> diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c
> index a0d10a9..796289c 100644
> --- a/drivers/ide/tx4938ide.c
> +++ b/drivers/ide/tx4938ide.c
> @@ -39,10 +39,17 @@ static void tx4938ide_tune_ebusc(unsigned int ebus_ch,
>  	/* Address-valid to DIOR/DIOW setup */
>  	shwt = DIV_ROUND_UP(t->setup, cycle);
>  
> +	/* -DIOx recovery time (SHWT * 4) and cycle time requirement */
> +	while ((shwt * 4 + wt + (wt ? 2 : 3)) * cycle < t->cycle)
>   

   Besides, it's worth making 'wt' variable signed, otherwise you're 
risking underflow when subtracting 2 above...

WBR, Sergei



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range
  2008-10-27 23:12     ` Sergei Shtylyov
@ 2008-10-29 19:12       ` Bartlomiej Zolnierkiewicz
  0 siblings, 0 replies; 5+ messages in thread
From: Bartlomiej Zolnierkiewicz @ 2008-10-29 19:12 UTC (permalink / raw)
  To: Sergei Shtylyov; +Cc: Atsushi Nemoto, linux-mips, linux-ide, ralf

On Tuesday 28 October 2008, Sergei Shtylyov wrote:
> Hello.
> 
> Atsushi Nemoto wrote:
> 
> > From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
> > Subject: [PATCH] tx4938ide: Check minimum cycle time and SHWT range (v2)
> >
> > SHWT value is used as address valid to -CSx assertion and -CSx to -DIOx
> > assertion setup time, and contrarywise, -DIOx to -CSx release and -CSx
> > release to address invalid hold time, so it actualy applies 4 times and
> > so constitutes -DIOx recovery time.  Check requirement of the recovery
> > time and cycle time.  Also check SHWT maximum value.
> >
> > Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
> > Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
> >   
> 
> Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>

applied

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2008-10-29 19:28 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-10-27 13:39 [PATCH 1/2] tx4938ide: Check minimum cycle time and SHWT range Atsushi Nemoto
2008-10-27 14:07 ` Sergei Shtylyov
2008-10-27 14:52   ` Atsushi Nemoto
2008-10-27 23:12     ` Sergei Shtylyov
2008-10-29 19:12       ` Bartlomiej Zolnierkiewicz

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).