[dm-devel] Latest dm-userspace kernel code
Dan Smith
danms at us.ibm.com
Sat Sep 30 02:39:46 UTC 2006
FT> - I think that you can use kernel threads for map_worker since
FT> dmu_map says that we might reorder I/Os.
Yes, I think you are probably right. I disabled the kernel thread at
one point as a sanity check, but I think it might be ok to turn it
back on. I'll do some testing.
FT> - What is the purpose of dmu_request->deps? I guess that it is for
FT> preventing I/O being reordered. if so, again, you don't need to do
FT> that.
Well, the problem is that a block copy would overwrite other requests.
I think that while a copy is happening, you have to be careful to not
write into that block so that your writes are not lost.
Am I over-thinking it?
FT> - Would be better to move non-transport functions (map_worker,
FT> copy_block, etc) to dm-userspace.c?
Yes. You did a much better job of isolating those functions than I
did :) ... I plan to revisit that soon.
FT> - How does the current kernel/user interface support multiple
FT> destination feature?
I'm not sure what you mean here. Can you elaborate?
FT> - Needs some cosmetic fixes (removing needless whitespaces and
FT> braces, etc)
Indeed. I made sure to mention in my post that it was not in "final
form" for this reason :) I'll try to clean it up a little more before
I post it again.
Thanks for the feedback!
--
Dan Smith
IBM Linux Technology Center
Open Hypervisor Team
email: danms at us.ibm.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 190 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/dm-devel/attachments/20060929/0919f862/attachment.sig>
More information about the dm-devel
mailing list