[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191112091521.GX4131@hirez.programming.kicks-ass.net>
Date: Tue, 12 Nov 2019 10:15:21 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: LKML <linux-kernel@...r.kernel.org>, x86@...nel.org,
Linus Torvalds <torvalds@...uxfoundation.org>,
Andy Lutomirski <luto@...nel.org>,
Stephen Hemminger <stephen@...workplumber.org>,
Willy Tarreau <w@....eu>, Juergen Gross <jgross@...e.com>,
Sean Christopherson <sean.j.christopherson@...el.com>,
"H. Peter Anvin" <hpa@...or.com>
Subject: Re: [patch V2 11/16] x86/ioperm: Share I/O bitmap if identical
On Mon, Nov 11, 2019 at 11:03:25PM +0100, Thomas Gleixner wrote:
> @@ -59,8 +71,26 @@ long ksys_ioperm(unsigned long from, uns
> return -ENOMEM;
>
> memset(iobm->bits, 0xff, sizeof(iobm->bits));
> + refcount_set(&iobm->refcnt, 1);
> + }
> +
> + /*
> + * If the bitmap is not shared, then nothing can take a refcount as
> + * current can obviously not fork at the same time. If it's shared
> + * duplicate it and drop the refcount on the original one.
> + */
> + if (refcount_read(&iobm->refcnt) > 1) {
> + iobm = kmemdup(iobm, sizeof(*iobm), GFP_KERNEL);
> + if (!iobm)
> + return -ENOMEM;
> + io_bitmap_exit();
refcount_set(&iobm->refcnd, 1);
> }
>
> + /* Set the tasks io_bitmap pointer (might be the same) */
> + t->io_bitmap = iobm;
> + /* Mark it active for context switching and exit to user mode */
> + set_thread_flag(TIF_IO_BITMAP);
> +
Powered by blists - more mailing lists