ror code with "timed out", and update it with something more
appropriate before returning when other errors are seen?
though this is an internal helper, and callers may not tell the
situations apart in the first place, so this might be a minor nit
virtually yours
Gerhard Sittig
--
DENX
7;s R/B pin until the
operation has completed, to then fetch the STATUS byte to
determine the execution's result. Does this sound plausible?
For NOR, do you poll for the "busy" condition to deassert, and
check for success then?
virtually yours
Gerhard Sittig
--
DENX Software Enginee
[ removed cscope-devel from Cc:, non-subscriber mails get blocked anyway ]
On Mon, 2014-04-07 at 14:42 +0200, Gerhard Sittig wrote:
>
> On Mon, 2014-04-07 at 06:42 -0400, Neil Horman wrote:
> >
> > On Thu, Apr 03, 2014 at 03:16:15PM +0200, Y
[ ignore this if you are busy :) ]
On Mon, 2014-04-07 at 20:26 +0200, Sebastian Andrzej Siewior wrote:
>
> On 04/07/2014 02:26 PM, Gerhard Sittig wrote:
> > On Mon, 2014-04-07 at 12:13 +0200, Sebastian Andrzej Siewior wrote:
> >>
> >> d->mask contains exact the s
tries. Would it be useful to not emit the warnings then? Or
to silent those warnings when the user knows it's perfectly legal
to skip those filesytem entries? Like what you can do with the
ctags(1) command and its --links option.
virtually yours
Gerhard Sittig
--
DENX Software Engineering Gm
t; - level |= BIT(bit);
> - dwapb_toggle_trigger(gpio, bit);
> + level |= mask;
> + dwapb_toggle_trigger(gpio, d->hwirq);
these introduce another pointer dereference, unless 'bit' was
assigned from a pointer dereference (as is shown
e example has many more properties than the binding describes,
the additional items aren't even optional -- you might want to
refer to a few more common or general bindings
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, O
rst two have 29.
The commit message probably should not discuss which git tree a
driver currently resides in. This is just a temporary detail
which no longer applies to the mainline source base where your
patch will end up in.
virtually yours
Gerhard Sittig
--
DENX Software Engineering
r require parentheses. So either of
"sizeof(x)" as well as "sizeof x" are legal with regard to the C
language. It's just that the community prefers "sizeof(x)" for
the improved readability.
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH,
v7 as it is.
Am I supposed to provide ACKs at all, given that I'm a reviewer
and fellow developer. I always understood that Acked-By is
something more formal, where one maintainer signals to another
maintainer that it's OK to go though a different tree.
virtually yours
Gerhard Sittig
rq)
> + goto skip_irq;
> +
> + altera_gc->domain = irq_domain_add_linear(node,
> + altera_gc->mmchip.gc.ngpio, &altera_gpio_irq_ops, altera_gc);
> +
> + if (!altera_gc->domain) {
> + ret = -ENODEV;
> +
e
situation where you need such an extension (or something similar,
but you really should look into the ll_ops thing).
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, German
e from "divider value" to
"bit pattern to read/write". In the latter case, the clock
control module is rather special, and may not be easily get
mapped to the common primitives. Unless the ll_ops can implement
the required special handling.
virtually yours
Gerhard Sittig
--
On Tue, Feb 18, 2014 at 09:09 +0900, Mark Brown wrote:
>
> On Sun, Feb 16, 2014 at 03:25:12PM +0100, Gerhard Sittig wrote:
> > On Sat, Feb 15, 2014 at 08:09 +0200, Ivan T. Ivanov wrote:
>
> > - the total length of the SPI transfer cannot be empty (which I'd
> >
er->len / w_size;
> + /* No partial transfers accepted */
> + if (!n_words || xfer->len % xfer->bits_per_word)
> + return -EINVAL;
> +
> if (xfer->speed_hz && master->min_speed_hz &&
> xfe
certainly would be
rather wrong an assumption. Am I missing/misunderstanding
something?
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax:
er, the change appears to
be balanced, and replacing 'goto free_master' with immediate
return looks appropriate.
Can't comment on the correctness of removing the spi_master_put()
call when switching from spi_alloc_master() to
devm_spi_alloc_master(). This gets discussed in the other
sub
ransfers as well. :)
BTW am I trying to be strict about the above implicit assumption
of "dual/quad" meaning the number of data lines. To not confuse
them with dual data rate transfers, which are orthogonal to the
number of data lines in use, and are available in chips as well
(see the S25
ole case), but still would
be wrong or incomplete in the open-collector case.
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142
mean that
you cannot _disable_ broken features? Or does it mean that you
_must_ specify the non-broken features and thus break backwards
compatibility? Or are these properties not boolean (they are not
used in the example either, unfortunately), and the binding text
would need an update for clarity? W
23s08, mcp23s17, mcp23008 and mcp23017
> are
> + * supported.
> + * For the I2C versions of the chips (mcp23008 and mcp23017) generation of
> + * interrupts is also supported.
> */
Adjust these as well to reflect whether it's hardware or software
which limits the feature.
virtua
mon gate
> code somehow?
> This situation happens from time to time and I haven't seen a solution
> for this.
You may want to lookup the following message:
Date: Tue, 23 Jul 2013 15:14:06 +0200
From: Gerhard Sittig
To: linuxppc-...@lists.ozlabs.org, Anatolij Gustschin ,
y 1,
and to clear up where such an assumption is wrongly made.
If the GPIO subsystem's API wants to guarantee values of 0 and 1
(which I think it doesn't), then I feel the adjustment should be
done in the gpio_get_value() routines (in all its public
variants, or a common routine which all
On Tue, Dec 31, 2013 at 10:01 -0800, Randy Dunlap wrote:
>
> On 12/31/13 05:48, Gerhard Sittig wrote:
> > On Tue, Dec 31, 2013 at 11:51 +0100, Krzysztof Hałasa wrote:
> >>
> >> drivers/crypto/ixp4xx_crypto.c: In function 'ixp_module_init':
> >> d
to=no
>
> to your .muttrc file.
Thank you for telling me, I was not aware. Had no MFT related
setting in my config, learned from the manual that $followup_to
defaults to yes, have turned it off now. Other mutt users may
want to check as well. Happy new year! :)
virtually yours
Gerhard Sitti
int num = ARRAY_SIZE(ixp4xx_algos);
> - int i, err ;
> + int i, err;
>
> pdev = platform_device_register_full(&ixp_dev_info);
> if (IS_ERR(pdev))
> return PTR_ERR(pdev);
>
> - dev = &pdev->dev;
> -
> spin_lock_init(&desc_lock);
>
On Thu, Dec 26, 2013 at 11:31 -0800, Stephen Boyd wrote:
>
> On 12/24, Gerhard Sittig wrote:
> > On Mon, Dec 23, 2013 at 17:12 -0800, Stephen Boyd wrote:
> > >
> > > The clock framework already has support for simple gate clocks
> > > but if drivers want t
ill
break when yours gets added later (regardless of the number of
review iterations it will see).
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-6
t the patch: Do you expect callers to remember whether
a clock gate is backed by mmio or by regmap access, to call a
different set of routines? Should this not be hidden behind the
API and be transparent after clock registration?
I'd suggest to fold regmap support into Tero Kristo's
On Thu, Dec 19, 2013 at 21:24 +0100, Gerhard Sittig wrote:
>
> Wasn't the idea to extend the set of register accessor routines
> in such that memory mapped I/O as well as
> regmap style becomes possible? This is what I understood from
> past iterations of discussing this app
mation is just too useful to hide
behind another option that requires manual enabling. :)
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0
thogonal to the API which wraps register access.
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: off...@denx
you may have to re-create the patch
on top of something official. :) And you may want to mark this
patch as "trivial".
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzel
he patch nor its description at hand.
And you might improve the description's body. Something like
"the 'direction' member of 'struct dma_slave_config' is of data
type 'enum dma_transfer_direction', so update the 'struct
dma_slave_config' kerneldoc
+ irqd_to_hwirq(d));
> + byt_irq_unmask(d);
> + return 0;
> +}
Just a thought: If failure to lock is non-fatal, should the
message be a warning then?
I do agree that failure to lock the GPIO should be non-fatal, as
there was debate and still might be concerns abo
fix a trivial copy'n'paste error in the regmap kerneldoc, s/write/read/
for the regmap_read(), regmap_raw_read() and regmap_bulk_read() routines
Signed-off-by: Gerhard Sittig
---
drivers/base/regmap/regmap.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
ng the
series, and thus would hinder bisection.
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: off...@denx.
a was asking is where to put
code for the component which is neither a strict subset of any
subsystem. Please correct me if I'm wrong.
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-8
er that
should be easy to fix. (Strictly speaking it's not the data
gathering that breaks, but the already broken network driver
which accesses the hardware without acquiring the clock, and not
finds the clock disabled.)
virtually yours
Gerhard Sittig
--
DENX Software Engineering GmbH, MD: Wol
On Wed, Jul 31, 2013 at 11:20 +0400, Alexander Popov wrote:
>
> From: Gerhard Sittig
>
> Concentrate the test and the specific code for MPC8308
> in the 'if' branch and handle MPC512x in the 'else' branch.
>
> This modification only reorders
return -EINVAL;
> +}
> +
and here
I think it's unfortunate to attribute the "will access
peripheral" to the channel instead of the transfer job, and to
set the flag from within the device control callback, and to
nevery clear the flag (what will happen if a channel
On Fri, Jul 12, 2013 at 14:42 +0400, Alexander Popov wrote:
>
> 2013/7/10 Gerhard Sittig :
> > On Wed, Jul 10, 2013 at 14:21 +0400, Alexander Popov wrote:
> >>
> >> +
> >> +struct mpc512x_lpbfifo_request {
> >> + unsigned int cs;
> >>
On Fri, Jul 12, 2013 at 14:14 +0200, Gerhard Sittig wrote:
>
> We shall combine the parts which currently are floating around.
> So that others aren't supposed to pick up pieces but instead can
> use a series and get something consistent.
>
> Here are the parts that I
s only useful after implementing the support (here: slave
S/G in the DMA engine).
On Fri, Jul 12, 2013 at 14:15 +0400, Alexander Popov wrote:
>
> 2013/7/10 Gerhard Sittig :
>
> >> @@ -256,7 +256,9 @@ static void mpc_dma_execute(struct mpc_dma_chan *mchan)
> >>
> >&g
r_res;
> + }
> +
> + if (!request_mem_region(r->start, resource_size(r), DRV_NAME)) {
> + e = -EBUSY;
> + goto err_res;
> + }
> +
> + lpbfifo.irq = irq_of_parse_and_map(pdev->dev.of_node, 0);
> + if (!lpbfifo.irq) {
> + e = -ENODEV;
> +
default:
> + return -ENOSYS;
> + }
> +
> + return -EINVAL;
> +}
> +
> static int mpc_dma_probe(struct platform_device *op)
> {
> struct device_node *dn = op->dev.of_node;
> @@ -725,9 +855,12 @@ static int mpc_dma_probe(struct platform_devic
y
lack free resources for this very driver extension.
Alternatively I might send an RFC series since the current state
isn't good enough for v1. Just ask if you'd like to test or
review what I have come up with so far (it builds but wasn't
run-tested).
virtually yours
Gerhard Sitti
47 matches
Mail list logo
