Re: [PATCH] percpu_counter: Fix __percpu_counter_sum()

[Mingming Cao]

å 2008-12-08äç 17:12 -0500ïTheodore Tsoåéï
> On Sun, Dec 07, 2008 at 08:52:50PM -0800, Andrew Morton wrote:
> > 
> > The first patch which was added (pre-2.6.27) was "percpu_counter: new
> > function percpu_counter_sum_and_set".  This added the broken-by-design
> > percpu_counter_sum_and_set() function, **and used it in ext4**.
> > 
> 
> Mea culpa, I was the one who reviewed Mingming's patch, and missed
> this.  Part of the problem was that percpu_counter.c isn't well
> documented, and I so saw the spinlock, but didn't realize it only
> protected reference counter, and not the per-cpu array.  I should have
> read through code more thoroughly before approving the patch.
> 
> I suppose if we wanted we could add a rw spinlock which mediates
> access to a "foreign" cpu counter (i.e., percpu_counter_add gets a
> shared lock, and percpu_counter_set needs an exclusive lock) but it's
> probably not worth it.
> 
> Actually, if all popular architectures had a hardware-implemented
> atomic_t, I wonder how much ext4 really needs the percpu counter,
> especially given ext4's multiblock allocator;

Delayed allocation will makes multiple block allocation possible for
buffered IO.  

However, we still need to check the percpu counter on write_begin() time
for every single possible block allocation (this is to make sure fs is
not overbooked), unless write_begin() could cluster the write requests
and maps multiple blocks in a single shot. So in reality in ext4 the
free blocks percpu_counter check and the s_dirty_blocks (percpu counter
too, for delayed blocks) only takes 1 block at a time:(


Mingming

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/