[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180221165006.GA27687@bombadil.infradead.org>
Date: Wed, 21 Feb 2018 08:50:06 -0800
From: Matthew Wilcox <willy@...radead.org>
To: Andy Lutomirski <luto@...nel.org>
Cc: Konstantin Khlebnikov <khlebnikov@...dex-team.ru>,
Dave Hansen <dave.hansen@...el.com>,
LKML <linux-kernel@...r.kernel.org>,
Christoph Hellwig <hch@...radead.org>,
Linux-MM <linux-mm@...ck.org>,
Andrew Morton <akpm@...ux-foundation.org>,
"Kirill A. Shutemov" <kirill@...temov.name>
Subject: Re: Use higher-order pages in vmalloc
On Wed, Feb 21, 2018 at 04:11:16PM +0000, Andy Lutomirski wrote:
> On Wed, Feb 21, 2018 at 3:42 PM, Matthew Wilcox <willy@...radead.org> wrote:
> > +++ b/kernel/fork.c
> > @@ -319,12 +319,12 @@ static void account_kernel_stack(struct task_struct *tsk, int account)
> > if (vm) {
> > int i;
> >
> > - BUG_ON(vm->nr_pages != THREAD_SIZE / PAGE_SIZE);
...
> > + if (j) {
> > + area->nr_pages -= (1UL << j) - 1;
>
> Is there any code that expects area->nr_pages to be the size of the
> area in pages? I don't know of any such code.
I found one and deleted it ;-)
Powered by blists - more mailing lists