Re: Makefile in /usr/src/lib/libdisk WARNS change? PLEASE READ!

On Tuesday, April 22, 2003, at 10:02 PM, Matthew Emmerton wrote:

>>> Hi,
>>>
>>> I posted the libdisk patch that was heavily ignored two days ago :).
>>> Without it I
>>> couldn't build libdisk and subsequently world.  If I lower the WARNS
>>> value
>>> in the libdisk directory Makefiel the source builds without warning 
>>> and
>>> world
>>> completes.
>>>
>>> The problem is rules.c having a switch/case that handles only 2 of 7
>>> cases for the
>>> platform value.  Adding a simple "default:" [as my patch did] will
>>> repair this problem.
>>>
>>> I should tell you that I clean-installed FBSD 5.0-RELEASE on Saturday
>>> 4/19/2003
>>> and started playing with CURRENT again the very next day at which 
>>> point
>>> I realized
>>> world wouldn't build.  I didn't do much else so I can only assume 
>>> that
>>> somehow this
>>> could happen to anyone else starting from RELEASE-5 today and
>>> attempting to build
>>> world.
>>>
>>> I have the latest version of the rules.c file [1.29]. I checked 
>>> CVSWEB
>>> and it has
>>> the exact same code that fails for me in HEAD so I know its not my
>>> source.  The
>>> only data involved in the switch is an enum type from libdisk.h 
>>> which I
>>> also
>>> have the latest version of [1.55].
>>>
>>> The error is
>>> /usr/src/lib/libdisk/rules.c:264: warning: enumeration value `p_any'
>>> not handled in switch
>>> /usr/src/lib/libdisk/rules.c:264: warning: enumeration value 
>>> `p_alpha'
>>> not handled in switch
>>> /usr/src/lib/libdisk/rules.c:264: warning: enumeration value
>>> `p_sparc64' not handled in switch
>>> /usr/src/lib/libdisk/rules.c:264: warning: enumeration value `p_ia64'
>>> not handled in switch
>>> /usr/src/lib/libdisk/rules.c:264: warning: enumeration value `p_ppc'
>>> not handled in switch
>>> *** Error code 1.
>>>
>>> I know this isn't a bug in the VM system.  Its not solving a kernel
>>> panic or
>>> some other serious commit but it does pose a barrier to those who 
>>> want
>>> to test CURRENT.
>>>
>>> Should I file a problem report?
>>>
>>> Dave Leimbach
>>
>> I know how you feel, boss. I just started a thread about how my 
>> computer
>> dies if I run out of memory, and all I got was a "me too" from another
>> fellow user. No developers jumped in. It seems like unless you have a 
>> LOR
> or
>> a laptop, you get no attention. And no, don't bother filing a bug 
>> report
>> because it will stagnate further.
>
> Please ignore the above comment.  The poster does not have the best
> interests of FreeBSD in mind.
> Remember that most (if not all) FreeBSD contributors are unpaid 
> volunteers,
> and their day jobs take precendence.
> It just happens that the people looking after LOR and ACPI issues have 
> more
> time on their hands than others.
>
> I've cc'd the committer of the last change to libdisk.h; this should 
> capture
> their attention and get this looked into more expediently.
>

I sent to PHK who dealt with rules.c so please don't annoy him more 
than I might
just have with my slightly desperate sounding email :).

I didn't want to get dramatic about this since its such a tiny issue... 
I just didn't
want to fork "LeimyBSD" over this simple fix :).

I am not worried about it... None of the FreeBSD developers owe me a 
goddamned thing.
In fact the only reason I report the bug is because I want to make sure 
it doesn't
happen to someone else :).

Dave


> --
> Matt Emmerton
>