Re: [PATCH] powerpc/entry: Fix double accounting of user time on interrupt entry

[Aboorva Devarajan <aboorvad@linux.ibm.com>]

On Wed, 2026-09-02 at 11:01 +0530, Mukesh Kumar Chaurasiya wrote:

Hi Mukesh,

> On Wed, Sep 02, 2026 at 10:36:28AM +0530, Aboorva Devarajan wrote:
> > Since the switch to generic entry, an interrupt taken from user mode
> > accounts user time twice: once in arch_interrupt_enter_prepare() and
> > again in arch_enter_from_user_mode(), which irqentry_enter() invokes
> > for the same interrupt:
> > 
> >   arch_interrupt_enter_prepare()
> >       account_cpu_user_entry()
> >   irqentry_enter()
> >     arch_enter_from_user_mode()
> >       account_cpu_user_entry()
> > 
> > account_cpu_user_entry() accumulates the time spent in user mode
> > since the last return to user space, so the second call charges the
> > same interval again.
> > 
> > With CONFIG_VIRT_CPU_ACCOUNTING_NATIVE=y this roughly doubles the
> > reported user time of any workload that takes interrupts. On a
> > pseries LPAR, ps/top show ~200% CPU for a single-threaded CPU-bound
> > loop, and time(1) reports user time about twice the elapsed time.
> > 
> > Remove the accounting from arch_interrupt_enter_prepare() and rely on
> > arch_enter_from_user_mode(), which runs for both syscalls and
> > interrupts. The duplicate account_stolen_time() call is removed the
> > same way.
> > 
> > Fixes: bee25f97ad24 ("powerpc: Enable GENERIC_ENTRY feature")
> > Signed-off-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
> > ---
> > Verified on a pseries LPAR (CONFIG_VIRT_CPU_ACCOUNTING_NATIVE=y),
> > 7.3.0-rc1, single-threaded CPU-bound loop:
> > 
> > Before:
> > 
> >   $ python3 -c 'while True: pass' &
> >   $ sleep 3; ps -p $! -o pid,etime,time,pcpu
> >       PID     ELAPSED     TIME %CPU
> >      4980       00:03 00:00:06  210
> > 
> > After:
> > 
> >   $ python3 -c 'while True: pass' &
> >   $ sleep 3; ps -p $! -o pid,etime,time,pcpu
> >       PID     ELAPSED     TIME %CPU
> >      4951       00:03 00:00:03  105
> > 
> >  arch/powerpc/include/asm/entry-common.h | 6 ++++--
> >  1 file changed, 4 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/powerpc/include/asm/entry-common.h b/arch/powerpc/include/asm/entry-common.h
> > index c5adb5006361..984294e62568 100644
> > --- a/arch/powerpc/include/asm/entry-common.h
> > +++ b/arch/powerpc/include/asm/entry-common.h
> > @@ -222,8 +222,6 @@ static inline void arch_interrupt_enter_prepare(struct pt_regs *regs)
> >  
> >  	if (user_mode(regs)) {
> >  		kuap_lock();
> > -		account_cpu_user_entry();
> > -		account_stolen_time();
> Yeah, that's dual accounting the time.
> 
> Thanks for the fix.
> >  	} else {
> >  		kuap_save_and_lock(regs);
> >  		/*
> > @@ -426,6 +424,10 @@ static __always_inline void arch_enter_from_user_mode(struct pt_regs *regs)
> >  #endif
> >  	kuap_assert_locked();
> >  	booke_restore_dbcr0();
> > +	/*
> > +	 * User and stolen time is accounted here for every entry from
> > +	 * user mode. The interrupt prepare hooks must not account again.
> > +	 */
> I think this comment is very specific to the current problem. We already
> have a commit history to explain this, I guess we can remove this
> comment.

Agreed, I will drop it in v2.

> 
> With that,
> 
> Reviewed-by: Mukesh Kumar Chaurasiya (IBM) <mkchauras@gmail.com>

Thanks.

> 
> Thanks,
> Mukesh
> >  	account_cpu_user_entry();
> >  	account_stolen_time();
> >  
> > 
> > base-commit: fb442a6673ff1046bf67754957d95880fdb394b5
> > -- 
> > 2.54.0
> >