Control: tag -1 + moreinfo
09.03.2013 00:30, Helge Deller wrote:
> On 03/08/2013 01:16 PM, Michael Tokarev wrote:
[]
>> This should not be necessary on modern kernels anymore, since kernel
>> interface allows either short or long packets. Why do you wan this
>> change?
>
> What exactly do you mean with "modern kernel"?
> My patch fixes automounter (autofs5-5.0.7-3) with upstream Linux kernel
> 3.7, 3.8 and 3.9-rc1 on the parisc architecture (32bit userspace and
> 64bit Linux kernel). This is what I actually built and tested successfully.
It does not matter anymore since 3.4 kernel and all previous 3.x stable
series as well, since upstream commit 64f371bc3107e69efce563a3d0f0e6880de0d537
which has been accepted to all stable releases.
This is what fixed the root cause and basically made the size of the packet
irrelevant as long as it is the same or larger in the kernel.
Whole if..arch, whole this function and adopting struct size difference
are not needed anymore in userspace and can be removed from automountd,
Ian kept these "for documentation and historical reasons" and in order
to not break stuff with old (before that commit) kernel.
> I know there has been various approaches to work around those problems
> ("32/64bit alignment strangenesses"), but as far as I can see, all those
> have been rejected or at least reverted again upstream.
>
> Of course I may be wrong, but maybe you can give me a hint?
The above commit is what made it all work. Note that systemd also implements
autofs functionality internally, but does not have this arch-specific hack,
it just use sizeof(kernel struct).
As Ian replied to your patch on autofs mailinglist, this is only needed for
old kernels, but he applies your patch anyway. I'd go for removing whole
hack entirely, maybe after a bit more waiting (say, when all major distributions
will move on to the fixed kernel already), because whole thing is just
disgusting and hurt eyes. This hack, with or without your change, is
definitely not needed on Debian.
So, unless there's some actual breakage going on, I'll close this bugreport.
And if there is, it must be something else, because since the above mentioned
commit, it all Just Works, and if there's a problem it is elsewhere.
Thanks,
/mjt
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
