Re: [PATCH] gdb: fix auxv caching

[John Baldwin <jhb@FreeBSD.org>]

On 10/11/22 1:34 PM, Pedro Alves wrote:
> 
> On 2022-10-11 9:31 p.m., Pedro Alves wrote:
>> On 2022-10-09 1:39 a.m., Simon Marchi via Gdb-patches wrote:
>>>
>>>> I think this approach is fine.  Having two variants of target_read_auxv is
>>>> a bit verbose, and I'm not sure it's abundantly clear to a new person when
>>>> to use one vs the other.  That said, these are used rarely, so probably
>>>> will intuit the right thing by looking at existing uses.  I agree with the
>>>> idea that the auxv reads during gdbarch_core_read_description should
>>>> effectively all be "raw" and uncached.
>>>
>>> The second one is perhaps not essential, call sites could call
>>>
>>>    target_read_alloc (ops, TARGET_OBJECT_AUXV, NULL)
>>>
>>> themselves.  But I find it convenient to have this little wrapper.
>>>
>>> As to how to know which overload to call, perhaps that can be improved
>>> with better documentation and comments.  I'm not sure what to add
>>> though, the problem is so fresh in my mind that it's obvious to me.  So
>>> I'm open to suggestions.
>>
>> Maybe call the one that works with the cache, target_read_auxv_cached ?
>>
> 
> Or the other one target_read_auxv_raw, of course.

I think "_raw" would be better here as using the cached values should be the
default.

>> Two overloads that do different things is a sign that they shouldn't be
>> overloads, to me.
>>


-- 
John Baldwin