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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161210014804.GA3023@lerouge>
Date:   Sat, 10 Dec 2016 02:48:06 +0100
From:   Frederic Weisbecker <fweisbec@...il.com>
To:     LKML <linux-kernel@...r.kernel.org>
Cc:     Martin Schwidefsky <schwidefsky@...ibm.com>,
        Tony Luck <tony.luck@...el.com>,
        Wanpeng Li <wanpeng.li@...mail.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Michael Ellerman <mpe@...erman.id.au>,
        Heiko Carstens <heiko.carstens@...ibm.com>,
        Benjamin Herrenschmidt <benh@...nel.crashing.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Paul Mackerras <paulus@...ba.org>,
        Ingo Molnar <mingo@...nel.org>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Rik van Riel <riel@...hat.com>,
        Stanislaw Gruszka <sgruszka@...hat.com>
Subject: Re: [PATCH 09/10] s390/cputime: delayed accounting of system time

On Tue, Dec 06, 2016 at 03:32:22AM +0100, Frederic Weisbecker wrote:
> From: Martin Schwidefsky <schwidefsky@...ibm.com>
> 
> The account_system_time() function is called with a cputime that
> occurred while running in the kernel. The function detects which
> context the CPU is currently running in and accounts the time to
> the correct bucket. This forces the arch code to account the
> cputime for hardirq and softirq immediately.
> 
> Such accounting function can be costly and perform unwelcome divisions
> and multiplications, among others.
> 
> The arch code can delay the accounting for system time. For s390
> the accounting is done once per timer tick and for each task switch.
> 
> Signed-off-by: Martin Schwidefsky <schwidefsky@...ibm.com>
> Cc: Benjamin Herrenschmidt <benh@...nel.crashing.org>
> Cc: Paul Mackerras <paulus@...ba.org>
> Cc: Michael Ellerman <mpe@...erman.id.au>
> Cc: Heiko Carstens <heiko.carstens@...ibm.com>
> Cc: Martin Schwidefsky <schwidefsky@...ibm.com>
> Cc: Tony Luck <tony.luck@...el.com>
> Cc: Fenghua Yu <fenghua.yu@...el.com>
> Cc: Peter Zijlstra <peterz@...radead.org>
> Cc: Rik van Riel <riel@...hat.com>
> Cc: Thomas Gleixner <tglx@...utronix.de>
> Cc: Ingo Molnar <mingo@...nel.org>
> Cc: Stanislaw Gruszka <sgruszka@...hat.com>
> Cc: Wanpeng Li <wanpeng.li@...mail.com>
> [rebase against latest cputime tree, massaged changelog accordingly]
> Signed-off-by: Frederic Weisbecker <fweisbec@...il.com>

Looking at this patch again, I think I need to do another pass on it.
Comments below:

>  /*
>   * Update process times based on virtual cpu times stored by entry.S
>   * to the lowcore fields user_timer, system_timer & steal_clock.
>   */
>  static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
>  {
> -	u64 timer, clock, user, system, steal;
> -	u64 user_scaled, system_scaled;
> +	u64 timer, clock, user, guest, system, hardirq, softirq, steal;
>  
>  	timer = S390_lowcore.last_update_timer;
>  	clock = S390_lowcore.last_update_clock;
> @@ -110,36 +119,57 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
>  #endif
>  		: "=m" (S390_lowcore.last_update_timer),
>  		  "=m" (S390_lowcore.last_update_clock));
> -	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
> -	S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock;
> +	clock = S390_lowcore.last_update_clock - clock;
> +	timer -= S390_lowcore.last_update_timer;
> +
> +	if ((tsk->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
> +		S390_lowcore.guest_timer += timer;
> +	else if (hardirq_count() - hardirq_offset)
> +		S390_lowcore.hardirq_timer += timer;

We should get rid of the hardirq_offset argument, it doesn't really make sense
anymore. Also it makes the accounting buggy now. It's called from the tick
through account_user_time() with hardirq_offset=1, so the irq time is incorrectly
accumulated as system time. Guest time may be incorrect too.

In fact it may have been buggy even before this patchset because vtime_account_user()
isn't only called from the tick but also from task switch, and hardirq_offset remains 1
for those two cases. Not good.

> +	else if (in_serving_softirq())
> +		S390_lowcore.softirq_timer += timer;
> +	else
> +		S390_lowcore.system_timer += timer;
>  
>  	/* Update MT utilization calculation */
>  	if (smp_cpu_mtid &&
>  	    time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
>  		update_mt_scaling();
>  
> +	/* Calculate cputime delta */
>  	user = S390_lowcore.user_timer - tsk->thread.user_timer;
> -	S390_lowcore.steal_timer -= user;
>  	tsk->thread.user_timer = S390_lowcore.user_timer;
> -
> +	guest = S390_lowcore.guest_timer - tsk->thread.guest_timer;
> +	tsk->thread.guest_timer = S390_lowcore.guest_timer;
>  	system = S390_lowcore.system_timer - tsk->thread.system_timer;
> -	S390_lowcore.steal_timer -= system;
>  	tsk->thread.system_timer = S390_lowcore.system_timer;
> +	hardirq = S390_lowcore.hardirq_timer - tsk->thread.hardirq_timer;
> +	tsk->thread.hardirq_timer = S390_lowcore.hardirq_timer;
> +	softirq = S390_lowcore.softirq_timer - tsk->thread.softirq_timer;
> +	tsk->thread.softirq_timer = S390_lowcore.softirq_timer;
> +	S390_lowcore.steal_timer +=
> +		clock - user - guest - system - hardirq - softirq;
>  
> -	user_scaled = user;
> -	system_scaled = system;
> -	/* Do MT utilization scaling */
> -	if (smp_cpu_mtid) {
> -		u64 mult = __this_cpu_read(mt_scaling_mult);
> -		u64 div = __this_cpu_read(mt_scaling_div);
> +	/* Push account value */
> +	if (user) {
> +		account_user_time(tsk, user);
> +		tsk->utimescaled += scale_vtime(user);
> +	}
>  
> -		user_scaled = (user_scaled * mult) / div;
> -		system_scaled = (system_scaled * mult) / div;
> +	if (guest) {
> +		account_guest_time(tsk, guest);
> +		tsk->utimescaled += scale_vtime(guest);
>  	}
> -	account_user_time(tsk, user);
> -	tsk->utimescaled += user_scaled;
> -	account_system_time(tsk, hardirq_offset, system);
> -	tsk->stimescaled += system_scaled;
> +
> +	if (system)
> +		account_system_index_scaled(tsk, system, scale_vtime(system),
> +					    CPUTIME_SYSTEM);
> +	if (hardirq)
> +		account_system_index_scaled(tsk, hardirq, scale_vtime(hardirq),
> +					    CPUTIME_IRQ);
> +	if (softirq)
> +		account_system_index_scaled(tsk, softirq, scale_vtime(softirq),
> +					    CPUTIME_SOFTIRQ);
>  
>  	steal = S390_lowcore.steal_timer;
>  	if ((s64) steal > 0) {
> @@ -147,16 +177,22 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
>  		account_steal_time(steal);
>  	}
>  
> -	return virt_timer_forward(user + system);
> +	return virt_timer_forward(user + guest + system + hardirq + softirq);
>  }
>  
>  void vtime_task_switch(struct task_struct *prev)
>  {
>  	do_account_vtime(prev, 0);

This call should be removed, the task switch already calls vtime_account_user().

>  	prev->thread.user_timer = S390_lowcore.user_timer;
> +	prev->thread.guest_timer = S390_lowcore.guest_timer;
>  	prev->thread.system_timer = S390_lowcore.system_timer;
> +	prev->thread.hardirq_timer = S390_lowcore.hardirq_timer;
> +	prev->thread.softirq_timer = S390_lowcore.softirq_timer;
>  	S390_lowcore.user_timer = current->thread.user_timer;
> +	S390_lowcore.guest_timer = current->thread.guest_timer;
>  	S390_lowcore.system_timer = current->thread.system_timer;
> +	S390_lowcore.hardirq_timer = current->thread.hardirq_timer;
> +	S390_lowcore.softirq_timer = current->thread.softirq_timer;
>  }

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ