On Tue, Jul 17, 2012 at 8:40 PM, Linus Walleij [off-list ref] wrote:

On Tue, Jul 17, 2012 at 12:39 PM, Sedat Dilek [off-list ref] wrote:

quoted

On Tue, Jul 17, 2012 at 12:21 PM, Sedat Dilek [off-list ref] wrote:

quoted

...looks like AB3100 is the *only* driver still using rand_initialize_irq()...

drivers/mfd/ab3100-core.c:939:  rand_initialize_irq(client->irq);

...but why has random.h still this...?

include/linux/random.h:51:extern void rand_initialize_irq(int irq);

Just thinking "loud"...

- Sedat -

Is this attached patch OK, Ted?

You're just deleting a line from the AB3100 driver, so I guess you're
regressing the system since that IRQ will no longer contribute
to the entropy pool?

How do we simultaneously make sure that this IRQ is still
contributing to the entropy?

For the sake of completeness:
* Ted explained in his answer in [1].
* Just FYI: [2] has these changes merged into the original commit.

- Sedat -

[1] http://marc.info/?l=linux-next&m=134256445707799&w=2
[2] http://git.kernel.org/?p=linux/kernel/git/tytso/random.git;a=commitdiff;h=627574db8b21c51217c683909458500916e64963

Yours,
Linus Walleij