[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZGzTZWq/hAYCE3DA@pc636>
Date: Tue, 23 May 2023 16:53:25 +0200
From: Uladzislau Rezki <urezki@...il.com>
To: Christoph Hellwig <hch@...radead.org>
Cc: "Uladzislau Rezki (Sony)" <urezki@...il.com>, linux-mm@...ck.org,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, Baoquan He <bhe@...hat.com>,
Lorenzo Stoakes <lstoakes@...il.com>,
Matthew Wilcox <willy@...radead.org>,
"Liam R . Howlett" <Liam.Howlett@...cle.com>,
Dave Chinner <david@...morbit.com>,
"Paul E . McKenney" <paulmck@...nel.org>,
Joel Fernandes <joel@...lfernandes.org>,
Oleksiy Avramchenko <oleksiy.avramchenko@...y.com>
Subject: Re: [PATCH 4/9] mm: vmalloc: Add a per-CPU-zone infrastructure
On Mon, May 22, 2023 at 11:08:38PM -0700, Christoph Hellwig wrote:
> On Mon, May 22, 2023 at 01:08:44PM +0200, Uladzislau Rezki (Sony) wrote:
> > +#define fbl(z, i, m) z->fbl[i].m
> > +#define fbl_root(z, i) fbl(z, i, root)
> > +#define fbl_head(z, i) fbl(z, i, head)
> > +
> > +#define fbl_lock(z, i) spin_lock(&fbl(z, i, lock))
> > +#define fbl_unlock(z, i) spin_unlock(&fbl(z, i, lock))
>
> Even if it is just temporary, I don't think adding these wrappers
> make much sense.
>
If open-coded, it looks like:
spin_lock(&z->fbl[BUSY].lock);
fbl_lock(z, BUSY);
the reason of adding such helpers is to make the name shorter.
> > +struct cpu_vmap_zone {
> > + /*
> > + * FREE, BUSY, LAZY bookkeeping data of this CPU zone.
> > + */
> > + struct {
> > + struct rb_root root;
> > + struct list_head head;
> > + spinlock_t lock;
> > + } fbl[NFBL];
>
> Maybe replace NFBL with something longer and more descriptive?
>
> But also in general it feels like this should be folded into a patch
> doing real work. As-is it doesn't look very useful.
>
I split it for better understanding for review. But i can fold it.
--
Uladzislau Rezki
Powered by blists - more mailing lists