Messages in this thread | | | Subject | Re: [PATCH RFC 1/1] arm64: Use PSCI calls for CPU stop when hotplug is supported | From | Scott Branden <> | Date | Wed, 23 Jan 2019 09:33:16 -0800 |
| |
Hi Sudeep,
On 2019-01-23 9:21 a.m., Sudeep Holla wrote: > On Wed, Jan 23, 2019 at 09:05:26AM -0800, Scott Branden wrote: >> Hi Mark, >> >> Hopefully I can shed some light on the use case inline. >> >> On 2019-01-23 8:48 a.m., Mark Rutland wrote: >>> On Mon, Jan 21, 2019 at 11:30:02AM +0530, Pramod Kumar wrote: >>>> On Mon, Jan 21, 2019 at 11:28 AM Pramod Kumar <pramod.kumar@broadcom.com> >>>> wrote: >>>> >>>> Need comes from a specific use case where one Accelerator card(SoC) is >>>> plugged in a sever over a PCIe interface. This Card gets supply from a >>>> battery, which could provide very less power for a very small time, in case >>>> of any power loss. Once Card switches to battery, this has to reduce its >>>> power consumption to its lowest point and back-up the DDR contents asap >>>> before battery gets fully drained off. >>> In this example is Linux running on the server, or on the accelerator? >> Accelerator >>> What precisely are you trying to back up from DDR, and why? >> Data in DDR is being written to disk at this time (disk is connected to >> accelerator) >>> What is responsible for backing up that contents? >> A low power M-class processor and DMA engine which continues necessary >> operations to transfer DDR memory to disk. >> >> The high power processors on the accelerator running linux needed to be >> halted ASAP on this power loss event and M0 take over. Graceful shutdown of >> linux and other peripherals is unnecessary (and we don't have the power >> necessary to do so). >> > It may be unnecessary for your use-case, but not recommended. No choice - we don't have the time/power for a graceful shutdown. > >>>> Since battery can provide limited power for a very short time hence need to >>>> transition to lowest power. As per the transition process , CPUs power >>>> domain has to be off but before that it needs to flush out its content to >>>> system memory(L3) so that content could be backed-up by a MCU, a controller >>>> consuming very less power. Since we can not afford plugging-out every >>>> individual CPUs in sequence hence uses ipi_cpu_stop for all other CPUs >>>> which ultimately switch to ATF to flush out all the CPUs caches and comes >>>> out of coherency domain so that its power rails could be switched-off. >>> If you're stopping CPUs from completely arbitrary states, what is the >>> benefit of saving the RAM contents? >> Some of the RAM contains data that was in the process of being written to >> disk by the accelerator. >> >> This data must be saved to disk and the high power CPUs consume too much >> power to continue performing this operation. >> > Why will suspend to ram or idle not work ? It will power off the secondaries > which this patch is trying to achieve, but in more sane way so that no > data/state is lost/corrupted as I stated earlier. We need to take over control of the disk write operations. There is no time/power available to leave linux running. > >>> CPUs might be running with IRQs disabled for an arbitrarily long time, >> In an embedded linux system we control everything running. >> > By which I assume you have patches to do all sorts of things to make this > work and this patch standalone is of no use :)
I believe other patch is in a standalone driver to be upstreamed.
Remainder of code is on a standalone M0 processor not running linux.
> > I don't like this as it's not scalable to big systems as this is in the > same code path as system off/reset. Many things are not used by big systems and vice versa. What do you suggest is done otherwise? > > -- > Regards, > Sudeep
Regards,
Scott
| |