Re: A Transformation of our Global Context

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 10/06/2016, Gregory Farnum wrote:
> So you want everything to stay with/go back to having a global member
> for the ProcessContext?

Precisely. I want a ProcessContext to handle dout and the like in
libcommon (since it doesn't vary by cluster I don't think there's any
reason to restrict it to libglobal/daemon/whatever) and let everything
else be passed in.

> I'm surprised it needs to find its way into so many random things;
> most objects have references to their parent that let them go back to
> the MDSRank/OSD/whatever main class there is.

A lot of them do have a link back to the main class but a lot
don't. A lot of 'light weight' data classes in the MDS for example
don't have a link back to anything but still need g_ceph_context to
tell time or dout something. A lot of callbacks do that, too. other
places they link back but the cct is protected (which I just switched
to public in my version, but it isn't ideal.)

After having tried to do it I think having a fairly minimal (process
specific) part that's global in libcommon and the rest that Isn't
Global Anywhere is much less disruptive.

-- 
Senior Software Engineer           Red Hat Storage, Ann Arbor, MI, US
IRC: Aemerson@{RedHat, OFTC, Freenode}
0x80F7544B90EDBFB9 E707 86BA 0C1B 62CC 152C  7C12 80F7 544B 90ED BFB9
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [CEPH Users]     [Ceph Large]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux