Am 9. Februar 2024 12:32:18 UTC schrieb "Philippe Mathieu-Daudé"
<[email protected]>:
>Hi,
>
>Various issues related to implementing dynamic machines have
>been documented in [1].
>
>We are trying to understand what means "a qdev is realized".
>One explanation was "the device is guest visible"; however
>many devices are realized before being mapped, thus are not
>"guest visible". Some devices map / wire their IRQs before
>being realized (such ISA devices). There is a need for devices
>to be "automatically" mapped/wired (see [2]) such CLI-created
>devices, but this apply generically to dynamic machines.
>
>Currently the device creation steps are expected to roughly be:
>
> (external use) (QDev core) (Device Impl)
> ~~~~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~~~
>
> INIT enter
> ----->
> +----------------------+
> | Allocate state |
> +----------------------+
> ----->
> +---------------------+
> | INIT children |
> | |
> | Alias children
> properties
> | |
> | Expose properties |
> INIT exit +---------------------+
> <-----------------------------------
> +----------------+
> | set properties |
> | |
> | set ClkIn |
> +----------------+ REALIZE enter
> ---------------------------------->
> +----------------------+
> | Use config properties|
> | |
> | Realize children |
> | |
> | Init GPIOs/IRQs |
> | |
> | Init MemoryRegions |
> +----------------------+
> REALIZE exit
> <-----------------------------------
> ---- "realized" / "guest visible"
>+-----------------+
>| Explicit wiring:|
>| IRQs |
>| I/O / Mem |
>| ClkOut |
>+-----------------+ RESET enter
> --------------------------------->
> +----------------------+
> | Reset default values |
> +----------------------+
>
>But as mentioned, various devices "wire" parts before they exit
>the "realize" step.
>In order to clarify, I'm trying to enforce what can be done
>*before* and *after* realization.
>
>*after* a device is expected to be stable (no more configurable)
>and fully usable.
>
>To be able to use internal/auto wiring (such ISA devices) and
>keep the current external/explicit wiring, I propose to add an
>extra "internal wiring" step, happening after the REALIZE step
>as:
>
> (external use) (QDev core) (Device Impl)
> ~~~~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~~~
>
> INIT enter
> ----->
> +----------------------+
> | Allocate state |
> +----------------------+
> ----->
> +---------------------+
> | INIT children |
> | |
> | Alias children
> properties
> | |
> | Expose properties |
> INIT exit +---------------------+
> <-----------------------------------
> +----------------+
> | set properties |
> | |
> | set ClkIn |
> +----------------+ REALIZE enter
> ---------------------------------->
> +----------------------+
> | Use config properties|
> | |
> | Realize children |
> | |
> | Init GPIOs/IRQs |
> | |
> | Init MemoryRegions |
> +----------------------+
> REALIZE exit <---
> +----------------------+
> | Internal auto wiring |
> | IRQs | (i.e. ISA bus)
> | I/O / Mem |
> | ClkOut |
> +----------------------+
> <---
> ---- "realized"
>+-----------------+
>| External wiring:|
>| IRQs |
>| I/O / Mem |
>| ClkOut |
>+-----------------+ RESET enter
>---- "guest visible"
> --------------------------------->
> +----------------------+
> | Reset default values |
> +----------------------+
>
>The "realized" point is not changed. "guest visible" concept only
>occurs *after* wiring, just before the reset phase.
>
>This series introduces the DeviceClass::wire handler within qdev
>core realization code, and convert devices using the implicit
>wiring to using that explicit handler.
>
>QDev API assertions patches will be posted later as another series.
>
>Thoughts?
Hi Phil,
What exactly are you trying to achieve? I assume that the goal is to make
devices user-creatable which currently can't due to current limitations. For
this I'm presenting an idea:
I'd probably not force the new discipline onto devices which aren't
user-creatable -- that seems like too much gymnastics to me for no gain.
Instead, I'd act on the suggestion of having an interface TYPE_USER_CREATABLE
and put the wire method there.
This wire method would be responsible for performing *all* duties between the
states "realized" and "guest-visible". Based on the parameters it is given by
the user, it would find all resources it needs to wire up the device. It finds
these resources by e.g. scanning the QOM tree. A good practice would be to pass
the root node for the scanning as parameter, e.g. the current machine.
It could happen that a user-crrated device depends on another user-created
device. To make this work, the part of QEMU that deals with user-created
devices would perform a fixed point iteration over all not yet finished
user-created devices. When there is no progress within one iteration, QEMU
aborts with an error message. This error message could come from the `wire`
invocation of any -- or even better -- all not yet finished devices.
To demonstrate the approach, one could probably pick just one device, add the
interface and hack up something in the core handling of user-created devices.
Does this idea sound reasonable? Is this what you're ultimately after?
Best regards,
Bernhatd
>
>Regards,
>
>Phil.
>
>[1] https://lore.kernel.org/all/[email protected]/
>[2]
>https://lore.kernel.org/qemu-devel/[email protected]/
>
>Cc: Eduardo Habkost <[email protected]>
>Cc: Markus Armbruster <[email protected]>' --cc '
>Cc: Edgar E. Iglesias <[email protected]>'
>Cc: Richard Henderson <[email protected]>
>Cc: Mark Cave-Ayland <[email protected]>
>Cc: Thomas Huth <[email protected]>
>Cc: Paolo Bonzini <[email protected]>
>Cc: Alexander Graf <[email protected]>
>Cc: Bernhard Beschow <[email protected]>
>Cc: Stefan Hajnoczi <[email protected]>
>Cc: Peter Maydell <[email protected]>
>Cc: Cédric Le Goater <[email protected]>
>Cc: Luc Michel <[email protected]>
>Cc: Zhao Liu <[email protected]>
>Cc: Gustavo Romero <[email protected]>
>
>Philippe Mathieu-Daudé (7):
> hw/qdev: Introduce DeviceClass::[un]wire() handlers
> hw/input/pckbd: Connect i8042 GPIOs once mouse/keyboard are realized
> hw/ide/cmd646: Configure IDE bus IRQs after realization
> hw/ide/sii3112: Configure IDE bus IRQs after realization
> hw/ide/via: Configure IDE bus IRQs after realization
> hw/intc/mips_gic: Initialize IRQ array once device is realized
> hw/misc/mac_via: Have VIA1 child access parent IRQ once realized
>
> include/hw/qdev-core.h | 8 +++++++-
> hw/core/qdev.c | 21 ++++++++++++++++++++-
> hw/ide/cmd646.c | 12 +++++++++++-
> hw/ide/sii3112.c | 10 ++++++++++
> hw/ide/via.c | 10 ++++++++++
> hw/input/pckbd.c | 38 +++++++++++++++++++++++++++-----------
> hw/intc/mips_gic.c | 11 +++++++++--
> hw/misc/mac_via.c | 9 ++++++++-
> 8 files changed, 102 insertions(+), 17 deletions(-)
>
