Re: Request (2nd) for unicore32 architecture codes to merge into linux-next
From: Thomas Gleixner <hidden>
Date: 2011-02-06 22:06:36
Also in:
linux-arch, lkml
From: Thomas Gleixner <hidden>
Date: 2011-02-06 22:06:36
Also in:
linux-arch, lkml
On Sat, 22 Jan 2011, Guan Xuetao wrote:
Hi, I want to merge unicore32 repo into linux-next tree, the position is (unicore32 branch):
Sorry for late reply.
arch/unicore32/kernel/time.c | 148 +++++
static struct clock_event_device ckevt_puv3_osmr0 ...
#ifdef CONFIG_ARCH_FPGA
.shift = 18, /* correct shift val: 16, but warn_on_slowpath */
#else
.shift = 30,
#endif
....
ckevt_puv3_osmr0.mult =
div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC, ckevt_puv3_osmr0.shift);
Please use the function which calculates the optimizes mult/shift
values for you depending on the clock frequency.
clockevents_calc_mult_shift() is your friend.
arch/unicore32/kernel/irq.c | 426 +++++++++++++
Please convert your irq_chips to the new function pointers:
.mask -> .irq_mask
...
The new functions replace the "unsigned int irq" argument with
"struct irq_data *d".
So conversion for your chips is simply
s/unsigned int irq/struct irq_data *d/
and
s/irq/d->irq/ in the implementation.
Then your arch should select: GENERIC_HARDIRQS_NO_DEPRECATED
Otherwise I'm happy with your irq and time related patches.
Thanks,
tglx