[PATCH] fixup! dmaengine: pxa_dma: add debug information

[PATCH] fixup! dmaengine: pxa_dma: add debug information

[Robert Jarzmik]

This fixes the following error:
drivers/dma/pxa_dma.c: In function ‘dbg_show_requester_chan’:
drivers/dma/pxa_dma.c:192:2: error: void value not ignored as it ought to be
  pos += seq_printf(s, "DMA channel %d requester :\n", phy->idx);
  ^
drivers/dma/pxa_dma.c:197:8: error: void value not ignored as it ought to be
        !!(drcmr & DRCMR_MAPVLD));
        ^
scripts/Makefile.build:258: recipe for target 'drivers/dma/pxa_dma.o' failed

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
 drivers/dma/pxa_dma.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c
index ddcbbf5..be5b524 100644
--- a/drivers/dma/pxa_dma.c
+++ b/drivers/dma/pxa_dma.c
@@ -184,19 +184,18 @@ static unsigned int pxad_drcmr(unsigned int line)
 
 static int dbg_show_requester_chan(struct seq_file *s, void *p)
 {
-	int pos = 0;
 	struct pxad_phy *phy = s->private;
 	int i;
 	u32 drcmr;
 
-	pos += seq_printf(s, "DMA channel %d requester :\n", phy->idx);
+	seq_printf(s, "DMA channel %d requester :\n", phy->idx);
 	for (i = 0; i < 70; i++) {
 		drcmr = readl_relaxed(phy->base + pxad_drcmr(i));
 		if ((drcmr & DRCMR_CHLNUM) == phy->idx)
-			pos += seq_printf(s, "\tRequester %d (MAPVLD=%d)\n", i,
-					  !!(drcmr & DRCMR_MAPVLD));
+			seq_printf(s, "\tRequester %d (MAPVLD=%d)\n", i,
+				   !!(drcmr & DRCMR_MAPVLD));
 	}
-	return pos;
+	return 0;
 }
 
 static inline int dbg_burst_from_dcmd(u32 dcmd)
-- 
2.1.4

Re: [PATCH] fixup! dmaengine: pxa_dma: add debug information

[Arnd Bergmann]

On Wednesday 27 May 2015 23:23:40 Robert Jarzmik wrote:
> This fixes the following error:
> drivers/dma/pxa_dma.c: In function ‘dbg_show_requester_chan’:
> drivers/dma/pxa_dma.c:192:2: error: void value not ignored as it ought to be
>   pos += seq_printf(s, "DMA channel %d requester :\n", phy->idx);
>   ^
> drivers/dma/pxa_dma.c:197:8: error: void value not ignored as it ought to be
>         !!(drcmr & DRCMR_MAPVLD));
>         ^
> scripts/Makefile.build:258: recipe for target 'drivers/dma/pxa_dma.o' failed
> 
> Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
> 

Acked-by: Arnd Bergmann <arnd@arndb.de>

I ran into the problem yesterday and came up with the same fix, but you
submitted it first, thanks!

	Arnd

Re: [PATCH] fixup! dmaengine: pxa_dma: add debug information

[Vinod Koul]

On Wed, May 27, 2015 at 11:23:40PM +0200, Robert Jarzmik wrote:
> This fixes the following error:
> drivers/dma/pxa_dma.c: In function ‘dbg_show_requester_chan’:
> drivers/dma/pxa_dma.c:192:2: error: void value not ignored as it ought to be
>   pos += seq_printf(s, "DMA channel %d requester :\n", phy->idx);
>   ^
> drivers/dma/pxa_dma.c:197:8: error: void value not ignored as it ought to be
>         !!(drcmr & DRCMR_MAPVLD));
>         ^
> scripts/Makefile.build:258: recipe for target 'drivers/dma/pxa_dma.o' failed
Applied now, but not pleased that you sent broken code. Sending stuff that
compiles is basic step

-- 
~Vinod

Re: [PATCH] fixup! dmaengine: pxa_dma: add debug information

[robert.jarzmik]

----- Mail original -----
De: "Vinod Koul" <vinod.koul@intel.com>
À: "Robert Jarzmik" <robert.jarzmik@free.fr>

On Wed, May 27, 2015 at 11:23:40PM +0200, Robert Jarzmik wrote:
> This fixes the following error:
> drivers/dma/pxa_dma.c: In function ‘dbg_show_requester_chan’:
> drivers/dma/pxa_dma.c:192:2: error: void value not ignored as it ought to be
>   pos += seq_printf(s, "DMA channel %d requester :\n", phy->idx);
> Applied now.
Thanks.

> but not pleased that you sent broken code. Sending stuff that
> compiles is basic step
You should know me better by now. I told you I compiled *and* tested.

And with a a bit closer look check this commit, especially the date :
commit 0bac33653b06a5f80a5c04e275eb33db9493b85a
Author: Joe Perches <joe@perches.com>
Date:   Sat May 23 12:00:38 2015 +1000

    fs/seq_file: convert int seq_vprint/seq_printf/etc... returns to void

Cheers.

--
Robert

Re: [PATCH] fixup! dmaengine: pxa_dma: add debug information

[Joe Perches]

On Fri, 2015-05-29 at 11:29 +0200, robert.jarzmik@free.fr wrote:
> ----- Mail original -----
> De: "Vinod Koul" <vinod.koul@intel.com>
> On Wed, May 27, 2015 at 11:23:40PM +0200, Robert Jarzmik wrote:
> > This fixes the following error:
[]
> And with a a bit closer look check this commit, especially the date :
> commit 0bac33653b06a5f80a5c04e275eb33db9493b85a
> Author: Joe Perches <joe@perches.com>
> Date:   Sat May 23 12:00:38 2015 +1000
> 
>     fs/seq_file: convert int seq_vprint/seq_printf/etc... returns to void

Hello Robert.

The return value of that dbg_show_requester_chan function
was incorrect and the increment of pos was an example of
why the seq_printf return value was changed to void.

It was a pretty common error and easy to understand given
the expectation that seq_printf should work like printf.

It's a pity it bit you. but this type of patch collision
should be a bit more difficult to reoccur in the future.

cheers, Joe

Re: [PATCH] fixup! dmaengine: pxa_dma: add debug information

[Robert Jarzmik]

Joe Perches <joe@perches.com> writes:

> Hello Robert.
>
> The return value of that dbg_show_requester_chan function
> was incorrect and the increment of pos was an example of
> why the seq_printf return value was changed to void.
>
> It was a pretty common error and easy to understand given
> the expectation that seq_printf should work like printf.
Sure.

> It's a pity it bit you. but this type of patch collision
> should be a bit more difficult to reoccur in the future.

I got what I deserved. I knew it was coming, I've seen it in other drivers of
mine, but my left brain didn't tell the other half to crosscheck in this
patch. Anyway, it's not a big fixup, so everything is fine now.

Cheers.


-- 
Robert