trace state variables -- remote protocol

trace state variables -- remote protocol

[David Taylor]

A minor documentation nit --

I am looking into implementing trace state variables for our new GDB
stub and noticed a discrepancy between the documentation and the code.

In the documentation the QTDV message is described as having two fields

    QTDV:n:value

where 'n' is the trace state variable number and 'value' is its value.

Looking at the code in GDB, it should be:

    QTDV:n:value:builtin:name

where

    n is the trace state variable number
    value is its value
    builtin is a boolean -- is it builtin to the stub or not
    name is a string

David
--
David Taylor
dtaylor at emc dot com

Re: trace state variables -- remote protocol

[Samuel Bronson]

David Taylor <dtaylor@emc.com> writes:

> A minor documentation nit --
>
> I am looking into implementing trace state variables for our new GDB
> stub and noticed a discrepancy between the documentation and the code.
>
> In the documentation the QTDV message is described as having two fields
>
>     QTDV:n:value
>
> where 'n' is the trace state variable number and 'value' is its value.
>
> Looking at the code in GDB, it should be:
>
>     QTDV:n:value:builtin:name
>
> where
>
>     n is the trace state variable number
>     value is its value
>     builtin is a boolean -- is it builtin to the stub or not
>     name is a string

Maybe try writing a patch, submitting a bug report, or CCing Eli?
Otherwise, I'd worry that your suggestion might fall through the cracks.
(I've taken the liberty of CCing Eli.)

I would suggest CCing an area maintainer as well, but I'm not sure if
Daniel Jacobowitz is really still active.

(Note that it's not worth writing a patch yourself if you can't/don't
want to complete the annoying copyright assignment paperwork.)

-- 
Hi! I'm a .signature virus! Copy me into your ~/.signature to help me spread!