[PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Cong Wang]

Signed-off-by: Cong Wang <amwang@redhat.com>
---
 drivers/ata/libata-sff.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 4cadfa2..8469cb5 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -720,13 +720,13 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
 
 		/* FIXME: use a bounce buffer */
 		local_irq_save(flags);
-		buf = kmap_atomic(page, KM_IRQ0);
+		buf = kmap_atomic(page);
 
 		/* do the actual data transfer */
 		ap->ops->sff_data_xfer(qc->dev, buf + offset, qc->sect_size,
 				       do_write);
 
-		kunmap_atomic(buf, KM_IRQ0);
+		kunmap_atomic(buf);
 		local_irq_restore(flags);
 	} else {
 		buf = page_address(page);
@@ -865,13 +865,13 @@ next_sg:
 
 		/* FIXME: use bounce buffer */
 		local_irq_save(flags);
-		buf = kmap_atomic(page, KM_IRQ0);
+		buf = kmap_atomic(page);
 
 		/* do the actual data transfer */
 		consumed = ap->ops->sff_data_xfer(dev,  buf + offset,
 								count, rw);
 
-		kunmap_atomic(buf, KM_IRQ0);
+		kunmap_atomic(buf);
 		local_irq_restore(flags);
 	} else {
 		buf = page_address(page);
-- 
1.7.4.4

[PATCH 15/62] ide: remove the second argument of k[un]map_atomic()

[Cong Wang]

Signed-off-by: Cong Wang <amwang@redhat.com>
---
 drivers/ide/ide-taskfile.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 5bc2839..729428e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -253,7 +253,7 @@ void ide_pio_bytes(ide_drive_t *drive, struct ide_cmd *cmd,
 		if (page_is_high)
 			local_irq_save(flags);
 
-		buf = kmap_atomic(page, KM_BIO_SRC_IRQ) + offset;
+		buf = kmap_atomic(page) + offset;
 
 		cmd->nleft -= nr_bytes;
 		cmd->cursg_ofs += nr_bytes;
@@ -269,7 +269,7 @@ void ide_pio_bytes(ide_drive_t *drive, struct ide_cmd *cmd,
 		else
 			hwif->tp_ops->input_data(drive, cmd, buf, nr_bytes);
 
-		kunmap_atomic(buf, KM_BIO_SRC_IRQ);
+		kunmap_atomic(buf);
 
 		if (page_is_high)
 			local_irq_restore(flags);
-- 
1.7.4.4

Re: [PATCH 15/62] ide: remove the second argument of k[un]map_atomic()

[David Miller]

From: Cong Wang <amwang@redhat.com>
Date: Sun, 27 Nov 2011 13:26:55 +0800

> 
> Signed-off-by: Cong Wang <amwang@redhat.com>

Acked-by: David S. Miller <davem@davemloft.net>

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Jeff Garzik]

On 11/27/2011 12:26 AM, Cong Wang wrote:
> Signed-off-by: Cong Wang<amwang@redhat.com>
> ---
>   drivers/ata/libata-sff.c |    8 ++++----
>   1 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
> index 4cadfa2..8469cb5 100644
> --- a/drivers/ata/libata-sff.c
> +++ b/drivers/ata/libata-sff.c
> @@ -720,13 +720,13 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
>
>   		/* FIXME: use a bounce buffer */
>   		local_irq_save(flags);
> -		buf = kmap_atomic(page, KM_IRQ0);
> +		buf = kmap_atomic(page);
>
>   		/* do the actual data transfer */
>   		ap->ops->sff_data_xfer(qc->dev, buf + offset, qc->sect_size,
>   				       do_write);
>
> -		kunmap_atomic(buf, KM_IRQ0);
> +		kunmap_atomic(buf);
>   		local_irq_restore(flags);
>   	} else {
>   		buf = page_address(page);
> @@ -865,13 +865,13 @@ next_sg:
>
>   		/* FIXME: use bounce buffer */
>   		local_irq_save(flags);
> -		buf = kmap_atomic(page, KM_IRQ0);
> +		buf = kmap_atomic(page);
>
>   		/* do the actual data transfer */
>   		consumed = ap->ops->sff_data_xfer(dev,  buf + offset,
>   								count, rw);
>
> -		kunmap_atomic(buf, KM_IRQ0);
> +		kunmap_atomic(buf);
>   		local_irq_restore(flags);
>   	} else {

Acked-by: Jeff Garzik <jgarzik@redhat.com>

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Sergei Shtylyov]

Hello.

On 27-11-2011 9:26, Cong Wang wrote:

> Signed-off-by: Cong Wang<amwang@redhat.com>
> ---
>   drivers/ata/libata-sff.c |    8 ++++----
>   1 files changed, 4 insertions(+), 4 deletions(-)

    IMO, you should do this in a single patch, not piecemeal to keep the 
kernel bisectable.

WBR, Sergei

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Cong Wang]

于 2011年11月28日 18:01, Sergei Shtylyov 写道:
> Hello.
>
> On 27-11-2011 9:26, Cong Wang wrote:
>
>> Signed-off-by: Cong Wang<amwang@redhat.com>
>> ---
>> drivers/ata/libata-sff.c | 8 ++++----
>> 1 files changed, 4 insertions(+), 4 deletions(-)
>
> IMO, you should do this in a single patch, not piecemeal to keep the kernel bisectable.
>

Hi,

Please see patch 01/62,

https://lkml.org/lkml/2011/11/28/35

So even without it, the kernel can still compile, just with some warnings.

Thanks.

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Cong Wang]

于 2011年11月28日 19:42, Cong Wang 写道:
> 于 2011年11月28日 18:01, Sergei Shtylyov 写道:
>> Hello.
>>
>> On 27-11-2011 9:26, Cong Wang wrote:
>>
>>> Signed-off-by: Cong Wang<amwang@redhat.com>
>>> ---
>>> drivers/ata/libata-sff.c | 8 ++++----
>>> 1 files changed, 4 insertions(+), 4 deletions(-)
>>
>> IMO, you should do this in a single patch, not piecemeal to keep the kernel bisectable.
>>
>
> Hi,
>
> Please see patch 01/62,
>
> https://lkml.org/lkml/2011/11/28/35
>
> So even without it, the kernel can still compile, just with some warnings.
>

Ah, no, sorry that I replied too quickly. We have to have at least 01/62
to compile.

Actually I did make the patches as a whole, but Andrew Morton prefers to
split them, so I make V2 of them.

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[James Bottomley]

On Mon, 2011-11-28 at 19:49 +0800, Cong Wang wrote:
> 于 2011年11月28日 19:42, Cong Wang 写道:
> > 于 2011年11月28日 18:01, Sergei Shtylyov 写道:
> >> Hello.
> >>
> >> On 27-11-2011 9:26, Cong Wang wrote:
> >>
> >>> Signed-off-by: Cong Wang<amwang@redhat.com>
> >>> ---
> >>> drivers/ata/libata-sff.c | 8 ++++----
> >>> 1 files changed, 4 insertions(+), 4 deletions(-)
> >>
> >> IMO, you should do this in a single patch, not piecemeal to keep the kernel bisectable.
> >>
> >
> > Hi,
> >
> > Please see patch 01/62,
> >
> > https://lkml.org/lkml/2011/11/28/35
> >
> > So even without it, the kernel can still compile, just with some warnings.
> >
> 
> Ah, no, sorry that I replied too quickly. We have to have at least 01/62
> to compile.
> 
> Actually I did make the patches as a whole, but Andrew Morton prefers to
> split them, so I make V2 of them.

I'd vote for as a whole as well, but if there's some legitimate reason
why not, then take them through Jiri's trivial tree; it's pretty trivial
anyway and at least going that way would keep them bisectable.

James

Re: [PATCH 10/62] ata: remove the second argument of k[un]map_atomic()

[Cong Wang]

于 2011年11月29日 03:00, James Bottomley 写道:
>
> I'd vote for as a whole as well, but if there's some legitimate reason
> why not, then take them through Jiri's trivial tree; it's pretty trivial
> anyway and at least going that way would keep them bisectable.
>

Ok, I just made a branch which keeps these patches as a single one,

git://github.com/congwang/linux.git #kmap_atomic-single

As the single patch is 233K, so I don't send it to mailing list.

Thanks.