Date:   Fri, 12 May 2017 14:58:53 +0200
From:   "Niklas Söderlund" <niklas.soderlund@...natech.se>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Sergei Shtylyov <sergei.shtylyov@...entembedded.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Linux-Renesas <linux-renesas-soc@...r.kernel.org>
Subject: Re: [PATCH] ravb: add wake-on-lan support via magic packet

Hi Geert,

Thanks for testing.

On 2017-05-12 14:47:53 +0200, Geert Uytterhoeven wrote:
> Hi Niklas,
> 
> On Fri, May 12, 2017 at 12:27 AM, Niklas Söderlund
> <niklas.soderlund+renesas@...natech.se> wrote:
> > WoL is enabled in the suspend callback by setting MagicPacket detection
> > and disabling all interrupts expect MagicPacket. In the resume path the
> > driver needs to reset the hardware to rearm the WoL logic, this prevents
> > the driver from simply restoring the registers and to take advantage of
> > that ravb was not suspended to reduce resume time. To reset the
> > hardware the driver closes the device, sets it in reset mode and reopens
> > the device just like it would do in a normal suspend/resume scenario
> > without WoL enabled, but it both closes and opens the device in the
> > resume callback since the device needs to be reset for WoL to work.
> >
> > One quirk needed for WoL is that the module clock needs to be prevented
> > from being switched off by Runtime PM. To keep the clock alive the
> > suspend callback need to call clk_enable() directly to increase the
> > usage count of the clock. Then when Runtime PM decreases the clock usage
> > count it won't reach 0 and be switched off.
> >
> > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@...natech.se>
> 
> Thanks for your patch!
> 
> Wake-on-LAN now works for me on Salvator-X (both H3 and M3).
> 
> However, after a few cycles, Ethernet stopped working, because CMA ran out of
> memory:
> 
>     cma: cma_alloc: alloc failed, req-size: 6 pages, ret: -4
>     dpm_run_callback(): ravb_resume+0x0/0x148 returns -12
>     PM: Device e6800000.ethernet failed to resume: error -12
> 
> Are we still missing some ravb patches in net-next, or is this a different
> issue?

Interesting, I did a 100 loop test on H3 based on v4.11 without any 
issues. What are you using as a base for your test and I will test if I 
can reproduce it.

> 
> When using PSCI suspend/resume, and Wake-on-LAN cannot work due to
> PSCI firmware issues, Ethernet fails to come up afterwards:
> 
>     ravb e6800000.ethernet eth0: failed to switch device to config mode
>     ravb e6800000.ethernet eth0: device will be stopped after h/w
> processes are done.
>     ravb e6800000.ethernet eth0: failed to switch device to config mode
>     dpm_run_callback(): ravb_resume+0x0/0x148 returns -110
>     PM: Device e6800000.ethernet failed to resume: error -110
> 
> Your resume routine cannot assume RAVB is in a sane mode, as it will
> have been reset if PSCI suspend was used.

Ouch, yes this is true thanks for reporting will look in to it.

The problem is that in the resume handler if WoL is enabled it will try 
to close the device before reinitializing it from reset state. If WoL is 
not enabled the device will be closed at suspend time so no need to 
close it before restoring operation from reset in the resume handler.

> 
> Gr{oetje,eeting}s,
> 
>                         Geert
> 
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@...ux-m68k.org
> 
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
>                                 -- Linus Torvalds

-- 
Regards,
Niklas Söderlund

Powered by blists - more mailing lists