Re: permissions (and other) problems

On Tuesday, September 17, 2002, at 08:22 PM, Philip Martin wrote:

> Garrett Rooney <rooneg@electricjellyfish.net> writes:
>
>> the thing that worries me about apr_atomic_t is that if atomic ops
>> aren't available on the platform, it falls back to an implementation
>> using mutexes, which means that apr_atomic_{inc,dec,get,set} wouldn't
>> be safe to use in a signal handler.
>
> So apr_atomic_t is atomic with respect to threads, and not with
> respect to signals, then I'd just use sig_atomic_t. If we restrict
> ourselves to "Subversion" platforms, i.e. ones where neon and/or db4
> can be built, I'd hazard a guess that sig_atomic_t will work :)

sounds good enough to me.

i'll commit this stuff as soon as i figure out why my 'fix' to the pool
lifetime problem that was causing segfaults appears to have broken
'make check'...

this probably means it won't be committed until at least tomorrow,
since i'm about to go out and i doubt i'll find it in the next 5
minutes, so if anyone has any other problems with the latest patch i
posted (other than the apr_signal versus signal thing, which i've fixed
locally), you have about 24 hours to speak up ;-)

-garrett

-- 
garrett rooney                    Remember, any design flaw you're
rooneg@electricjellyfish.net      sufficiently snide about becomes
http://electricjellyfish.net/     a feature.       -- Dan Sugalski
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org