Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: why does "mem auto-init: clearing system memory may take some
 time..." takes about 14 sec under 5.4.17 but only few msec under 5.5.1?
To:     =?UTF-8?Q?Toralf_F=c3=b6rster?= <toralf.foerster@gmx.de>,
        Linux Kernel <linux-kernel@vger.kernel.org>
References: <c7a23250-6b23-2a9e-88c8-924c34fb9139@gmx.de>
From:   Randy Dunlap <rdunlap@infradead.org>
Message-ID: <a8e02d42-a16a-c665-6a55-1c43818fa48c@infradead.org>
Date:   Sun, 29 Mar 2020 19:55:38 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.6.0
MIME-Version: 1.0
In-Reply-To: <c7a23250-6b23-2a9e-88c8-924c34fb9139@gmx.de>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2/1/20 9:52 AM, Toralf Förster wrote:
> I do wonder (and I'm surprised) about increased boot speed according to dmesg for thi9s hardend Gentoo Linux system:
> Linux mr-fox 5.5.1 #1 SMP Sat Feb 1 18:25:21 CET 2020 x86_64 Intel(R) Xeon(R) CPU E5-1650 v3 @ 3.50GHz GenuineIntel GNU/Linux
> 
> Under 5.4.17 I do have usually 14 sec to wait:
> 
> [    0.497171] mem auto-init: stack:byref_all, heap alloc:on, heap free:on
> [    0.497171] mem auto-init: clearing system memory may take some time...
> [    0.501549] Calgary: detecting Calgary via BIOS EBDA area
> [    0.501550] Calgary: Unable to locate Rio Grande table in EBDA - bailing!
> [   14.083235] Memory: 131812972K/134101088K available (12291K kernel code, 1400K rwdata, 1912K rodata, 1272K init, 2040K bss, 2288116K reserved, 0K cma-reserved)
> [   14.083342] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=12, Nodes=1
> [   14.083350] Kernel/User page tables isolation: enabled
> [   14.083403] rcu: Hierarchical RCU implementation.
> [   14.083404] rcu: RCU calculated value of scheduler-enlistment delay is 25 jiffies.
> [   14.083414] NR_IRQS: 4352, nr_irqs: 928, preallocated irqs: 16
> 
> but with 5.5.1 there's no wait?:
> 
> [    0.000000] mem auto-init: stack:byref_all, heap alloc:on, heap free:on
> [    0.000000] mem auto-init: clearing system memory may take some time...
> [    0.000000] Memory: 131812876K/134101088K available (12291K kernel code, 1412K rwdata, 1936K rodata, 1260K init, 2012K bss, 2288212K reserved, 0K cma-reserved)
> [    0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=12, Nodes=1
> [    0.000000] Kernel/User page tables isolation: enabled
> [    0.000000] rcu: Hierarchical RCU implementation.
> [    0.000000] rcu: RCU calculated value of scheduler-enlistment delay is 25 jiffies.
> [    0.000000] NR_IRQS: 4352, nr_irqs: 928, preallocated irqs: 16
> [    0.000000] random: get_random_bytes called from start_kernel+0x4a8/0x659 with crng_init=0
> 
> Do I miss something? (The kernel config was derived via "make oldconfig").

Hi > Toralf,

Curious/odd.

Do you have the kernel .config files for 5.4.17 and 5.5.1?
Looks like it would have to be a config difference AFAICT.

-- 
~Randy