linux-kernel - 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?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 29 Mar 2020 19:55:38 -0700
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Toralf Förster <toralf.foerster@....de>,
        Linux Kernel <linux-kernel@...r.kernel.org>
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?

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