Re: [Minios-devel] [UNIKRAFT PATCHv4 35/43] plat/common: Implement CPU suspend for arm64

[Wei Chen <Wei.Chen@xxxxxxx>]

Hi Julien,

> -----Original Message-----
> From: Julien Grall <julien.grall@xxxxxxx>
> Sent: 2018年7月16日 21:48
> To: Wei Chen <Wei.Chen@xxxxxxx>; minios-devel@xxxxxxxxxxxxxxxxxxxx;
> simon.kuenzer@xxxxxxxxx
> Cc: Kaly Xin <Kaly.Xin@xxxxxxx>; nd <nd@xxxxxxx>
> Subject: Re: [Minios-devel] [UNIKRAFT PATCHv4 35/43] plat/common: Implement
> CPU suspend for arm64
> 
> Hi Wei,
> 
> On 06/07/18 10:03, Wei Chen wrote:
> > Using PSCI despatch function to call PSCI_CPU_SUSPEND
> > function to halt CPU.
> >
> > Signed-off-by: Wei Chen <Wei.Chen@xxxxxxx>
> > ---
> >   plat/common/arm/cpu_native.c        | 44 +++++++++++++++++++++++++++++
> >   plat/common/include/arm/arm64/cpu.h |  2 ++
> >   2 files changed, 46 insertions(+)
> >   create mode 100644 plat/common/arm/cpu_native.c
> >
> > diff --git a/plat/common/arm/cpu_native.c b/plat/common/arm/cpu_native.c
> > new file mode 100644
> > index 0000000..07548bf
> > --- /dev/null
> > +++ b/plat/common/arm/cpu_native.c
> > @@ -0,0 +1,44 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause */
> > +/*
> > + * Authors: Wei Chen <wei.chen@xxxxxxx>
> > + *
> > + * Copyright (c) 2018, Arm Ltd., All rights reserved.
> > + *
> > + * Redistribution and use in source and binary forms, with or without
> > + * modification, are permitted provided that the following conditions
> > + * are met:
> > + *
> > + * 1. Redistributions of source code must retain the above copyright
> > + *    notice, this list of conditions and the following disclaimer.
> > + * 2. Redistributions in binary form must reproduce the above copyright
> > + *    notice, this list of conditions and the following disclaimer in the
> > + *    documentation and/or other materials provided with the distribution.
> > + * 3. Neither the name of the copyright holder nor the names of its
> > + *    contributors may be used to endorse or promote products derived from
> > + *    this software without specific prior written permission.
> > + *
> > + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
> IS"
> > + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
> THE
> > + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
> PURPOSE
> > + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS
> BE
> > + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
> > + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
> > + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
> > + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
> > + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
> > + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
> THE
> > + * POSSIBILITY OF SUCH DAMAGE.
> > + *
> > + * THIS HEADER MAY NOT BE EXTRACTED OR MODIFIED IN ANY WAY.
> > + */
> > +
> > +#include <arm/cpu.h>
> > +#include <arm/cpu_defs.h>
> > +
> > +void halt(void)
> 
> Looking at the usage of halt, I think PSCI CPU suspend is too expensive
> for brief idling.
> 
> I am not entirely sure of the semantic expected, but it looks like "wfi"
> would be more suitable here.

Does timer interrupt can wake up wfi? If yes, I think I can't use wfi here.

> 
> > +{
> > +   if (psci_method == PSCI_METHOD_HVC)
> > +           psci_hvc_call(PSCI_FNID_CPU_SUSPEND);
> > +   else if (psci_method == PSCI_METHOD_SMC)
> > +           psci_smc_call(PSCI_FNID_CPU_SUSPEND);
> 
> Regardless my comment above, I don't think this code follow the PSCI
> spec. CPU_SUSPEND takes another parameter that is left unknown. Also,
> 

Yes, I have ignored other parameters. I will use another patch series to
improve the PSCI library. Let's record it in TODO list.

> Furthermore, you may want to introduce a function pointer and detect at
> initialization which call you want to use rather than checking in every
> function.
> 

That's sensible. I will use such a function pointer.

> Lastly, in case of non-PSCI platform function will just do nothing. This
> looks slightly strange.
> 

Yes, because in the code context, while cpu_suspend/cpu_halt could not take
effect, the code will enter dead loop.

> Cheers,
> 
> --
> Julien Grall
_______________________________________________
Minios-devel mailing list
Minios-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/minios-devel