Re: [Qemu-devel] [qemu-s390x] [PATCH v2 2/8] s390x/css: IO instr handler ending control

[Thomas Huth <thuth@redhat.com>]

On 17.10.2017 13:10, Halil Pasic wrote:
> 
> 
> On 10/12/2017 01:44 PM, Halil Pasic wrote:
>>
>>
>> On 10/12/2017 08:58 AM, Thomas Huth wrote:
>>> On 10.10.2017 13:41, Halil Pasic wrote:
> [..]
>>> So yes, please don't do a "typedef unsigned int IOInstEnding" either. I
>>> think the best match for QEMU would be a
>>>
>>> typedef enum IOInstEnding {
>>>     CC_...,
>>>     CC_...,
>>>     CC_...,
>>>     CC_...
>>> } IOInstEnding;
>>>
>>
>> I also prefer this over #define NAME val.
>>
> 
> @Conny @Thomas
> 
> I'm almost done with v3, but I've realized we did not agree on the
> names for the enum constants, so before posting something to ugly
> again, I would like to inquire your opinion.
> 
> My current version of the enum is the following (but I can easily change
> to whatever you like with sed):
> 
> +/*
> + * IO instructions conclude according this. Currently we have only
> + * cc codes. Valid values are 0,1,2,3 and the generic semantic for IO instructions
> + * is described briefly. For more details consult the PoP.
> + */
> +typedef enum IOInstEnding {
> +    IOINST_CC_0 = 0, /* produced expected result */
> +    IOINST_CC_1 = 1, /* status conditions were present, or alternate result */
> +    IOINST_CC_2 = 2, /* ineffective, busy with previously initiated function */
> +    IOINST_CC_3 = 3  /* ineffective, not operational */
> +} IOInstEnding;
> +
> 
> Alternatives I had in mind are IOINST_CC_0_EXPECTED, IOINST_CC_1_STATUS_PRESENT, 
> IOINST_CC_2_BUSY, IOINST_CC_3_NOT_OPERATIONAL or the same without the numerical
> code (e.g. just IONIST_CC_EXPECTED).

FWIW, I'd prefer your last suggestion (e.g. IOINST_CC_EXPECTED).

 Thomas