[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.20.1708260008051.2124@nanos>
Date: Sat, 26 Aug 2017 00:09:18 +0200 (CEST)
From: Thomas Gleixner <tglx@...utronix.de>
To: Jan Beulich <JBeulich@...e.com>
cc: mingo@...e.hu, hpa@...or.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] x86: use KERN_CONT in dump_pagetable()
On Thu, 24 Aug 2017, Jan Beulich wrote:
> The lack of newlines in preceding format strings is a clear indication
> that these were meant to be continuations of one another, and indeed
> output ends up quite a bit more compact (and readable) that way.
>
> Signed-off-by: Jan Beulich <jbeulich@...e.com>
> ---
> arch/x86/mm/fault.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
> --- 4.13-rc6/arch/x86/mm/fault.c
> +++ 4.13-rc6-x86-dumppgt-cont/arch/x86/mm/fault.c
> @@ -415,9 +415,9 @@ static void dump_pagetable(unsigned long
> goto out;
>
> pte = pte_offset_kernel(pmd, address);
> - printk("*pte = %0*Lx ", sizeof(*pte) * 2, (u64)pte_val(*pte));
> + printk(KERN_CONT "*pte = %0*Lx ", sizeof(*pte) * 2, (u64)pte_val(*pte));
Can you please make that pr_cont() while at it and change the pure
printk()s in that function to pr_info()?
Thanks,
tglx
Powered by blists - more mailing lists