* [PULLv2] vhost: cleanups and fixes
@ 2013-05-02 20:11 Michael S. Tsirkin
2013-05-02 20:26 ` Linus Torvalds
0 siblings, 1 reply; 3+ messages in thread
From: Michael S. Tsirkin @ 2013-05-02 20:11 UTC (permalink / raw)
To: Linus Torvalds; +Cc: kvm, virtualization, netdev, linux-kernel, asias, mst, nab
Hi Linus,
This is exactly same code as my previous pull sent earlier, the diffstat
was wrong however, I did a merge with master which fixed that. And I
forgot to copy you so you likely
did not see it at all :(
Here comes.
The following changes since commit 5a148af66932c31814e263366094b5812210b501:
Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc (2013-05-02 10:16:16 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git tags/for_linus2
for you to fetch changes up to ef1f2e2ae31069a3484f77f00ea6ef92031d611a:
Merge branch 'vhost' into 'master' (2013-05-02 22:40:01 +0300)
----------------------------------------------------------------
vhost: cleanups and fixes for 3.10
Here are vhost cleanups and fixes by Asias He and myself.
They affect both vhost-net and vhost-scsi devices.
They also *depend* on both net-next and target-pending,
where the net and target commits these changes depend on
are already merged.
So merging through the common vhost tree.
Note2: I had to make a merge from vhost into master otherwise
request-pull generates a wrong diffstat.
It's harmless since git merge will figure it all out
easily.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
----------------------------------------------------------------
Asias He (3):
vhost: Allow device specific fields per vq
tcm_vhost: Wait for pending requests in vhost_scsi_flush()
vhost: move vhost-net zerocopy fields to net.c
Michael S. Tsirkin (9):
Merge branch 'for-next-merge' of git://git.kernel.org/.../nab/target-pending into vhost-net-next
tcm_vhost: document inflight ref-counting use
vhost: move per-vq net specific fields out to net
tcm_vhost: remove virtio-net.h dependency
vhost: fix error handling in RESET_OWNER ioctl
vhost: src file renames
tcm_vhost: header split up
vhost_scsi: module rename
Merge branch 'vhost' into 'master'
drivers/vhost/Kconfig | 10 +-
drivers/vhost/Kconfig.tcm | 6 -
drivers/vhost/Makefile | 3 +-
drivers/vhost/net.c | 255 +++++++++++++++++++++++++++-------
drivers/vhost/{tcm_vhost.c => scsi.c} | 255 +++++++++++++++++++++++++++++++---
drivers/vhost/tcm_vhost.h | 128 -----------------
drivers/vhost/test.c | 9 +-
drivers/vhost/vhost.c | 153 +++++++-------------
drivers/vhost/vhost.h | 32 +----
include/uapi/linux/vhost.h | 28 ++++
10 files changed, 533 insertions(+), 346 deletions(-)
delete mode 100644 drivers/vhost/Kconfig.tcm
rename drivers/vhost/{tcm_vhost.c => scsi.c} (86%)
delete mode 100644 drivers/vhost/tcm_vhost.h
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PULLv2] vhost: cleanups and fixes
2013-05-02 20:11 [PULLv2] vhost: cleanups and fixes Michael S. Tsirkin
@ 2013-05-02 20:26 ` Linus Torvalds
2013-05-02 20:32 ` Linus Torvalds
0 siblings, 1 reply; 3+ messages in thread
From: Linus Torvalds @ 2013-05-02 20:26 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: KVM list, virtualization, Network Development,
Linux Kernel Mailing List, Asias He, Nicholas A. Bellinger
On Thu, May 2, 2013 at 1:11 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> Hi Linus,
> This is exactly same code as my previous pull sent earlier, the diffstat
> was wrong however, I did a merge with master which fixed that.
See my other email. THIS IS NOT A VALID REASON TO DO A BACK-MERGE!
Extraneous merge commits that have no actual reason for development
are bad. They make it much harder to see what the actual development
thread was, because they mix up different branches. In fact, the whole
problem with the diffstat started with *another* extraneous merge, so
the "fix" was a result of a previous merge. This is exactly the kind
of reason why we want merges to be done by upstream developers, not
"back-merges" by downstream.
Now that merge is in your development tree, and if/when you start
doing development on top of that, the same kind of problem will
continue. Together with making it much harder to see in the history
what _your_ development stream was, because now it's no longer linear
and mixes in random points of the merge window.
Don't do this. Why do I have to write one of these "DON'T DO
BACK-MERGES" every single merge window? I really wish people listened,
instead of every single developer deciding that "it's ok if _I_ do it,
I'm sure Linus is complaining about all those *other* morons out
there".
When I say "don't do back-merges", I mean *you*. Yes, you. Not
somebody else. You.
Linus
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PULLv2] vhost: cleanups and fixes
2013-05-02 20:26 ` Linus Torvalds
@ 2013-05-02 20:32 ` Linus Torvalds
0 siblings, 0 replies; 3+ messages in thread
From: Linus Torvalds @ 2013-05-02 20:32 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: KVM list, Network Development, Linux Kernel Mailing List,
Asias He, Nicholas A. Bellinger
I pulled the previous tag, which didn't have the merge.
And what the heck is up with "virtualization@lists.osdl.org"? That
thing doesn't work. "osdl.org"? Did we suddenly take a delorean back a
decade? Nonworking list cc removed.
Linus
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2013-05-02 20:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-05-02 20:11 [PULLv2] vhost: cleanups and fixes Michael S. Tsirkin
2013-05-02 20:26 ` Linus Torvalds
2013-05-02 20:32 ` Linus Torvalds
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).