Re: [PATCH v4 08/10] hw/core/resettable: add support for warm reset

From: Damien Hedde <damien.hedde@greensocs.com>
To: "Philippe Mathieu-Daudé" <f4bug@amsat.org>, qemu-devel@nongnu.org
Cc: edgar.iglesias@xilinx.com, peter.maydell@linaro.org,
	berrange@redhat.com, ehabkost@redhat.com,
	mark.burton@greensocs.com, pbonzini@redhat.com,
	david@gibson.dropbear.id.au
Subject: Re: [PATCH v4 08/10] hw/core/resettable: add support for warm reset
Date: Mon, 11 May 2020 11:37:26 +0200	[thread overview]
Message-ID: <1c84c058-0b8b-7e70-0739-09fe3b65dcc4@greensocs.com> (raw)
In-Reply-To: <762b6429-15af-4f6e-2a31-87e3d3313f86@amsat.org>

Hi Philippe,

On 5/10/20 10:17 PM, Philippe Mathieu-Daudé wrote:
> Hi Damien,
> 
> On 8/21/19 6:33 PM, Damien Hedde wrote:
>> Add the RESET_TYPE_WARM reset type.
>> Expand the actual implementation to support several types.
>>
>>     /**
>>    * ResetType:
>> - * Types of reset.
>> + * Types of reset, values can be OR'ed together.
>>    *
>>    * + Cold: reset resulting from a power cycle of the object.
>> - *
>> - * TODO: Support has to be added to handle more types. In particular,
>> - * ResetState structure needs to be expanded.
>> + * + Warm: reset without power cycling.
>>    */
>>   typedef enum ResetType {
>> -    RESET_TYPE_COLD,
>> +    RESET_TYPE_COLD = 0x1,
>> +    RESET_TYPE_WARM = 0x2,
> 
> I'm a bit lost with the various iterations, what is the plan with warm
> reset, is this blocked due to discussion, API, something else?
> 

I removed it in the last versions of the series because it was adding
complexity. There were unsolved issues and discussions, in particular
regarding the interactions and propagation along trees between the
different types.

Damien