qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Mark McLoughlin <markmc@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 10/11] slirp: Rework external configuration interface
Date: Thu, 28 May 2009 18:23:14 +0100	[thread overview]
Message-ID: <1243531394.18588.63.camel@blaa> (raw)
In-Reply-To: <4A1EB3D4.7030804@siemens.com>

On Thu, 2009-05-28 at 17:55 +0200, Jan Kiszka wrote:
> > You're renaming "redir" and "channel" here which isn't a big deal since
> > you've only introduced them in the previous patch, but it would be
> > better to use the final names in the original patch.
> 
> Well, the purpose of the original patch was only to pull "redir" and
> "channel" as-is into the -net parameter list, not to refactor the
> interface otherwise.

It's a bisectability thing - we shouldn't introduce parameters in a
patch that we intend to rename in a subsequent patch.

e.g. if a distro cherry-picked this patch, we'd have options in the wild
that aren't available upstream.

> > More importantly, though, you've dropped the "ip" parameter which is
> in
> > the 0.10.x releases. We can't just drop existing parameters.
> 
> OK, I see the problem - though this parameter was probable rarely used
> (it was undocumented and suffered from the poor configurability). Will
> have a closer look.

Great. Probably best to re-send 7-11 with both of these things fixed up.

> > By way of meta-comment, some of these patches are much harder to
> review
> > than they need to be, because e.g. you're doing lots of cleanups
> > together with the real changes, or not breaking changes into smaller
> > chunks.
> 
> There might be a few coding style updates included, when I touch
> related
> lines. Or please give some (or the most annoying) example.

Okay:

  - In "Avoid zombie processes after fork_exec" you needlessly 
    re-arrange the code in launch_script() - it could have very 
    reasonably been an easy to review +6 hunk rather than a -24/+36 
    hunk if you'd split the cleanup out into its own patch

  - "Real fix for check_params users" could have been split into the
    revert followed by the better fix

  - In "Improve parameter error reporting", you replace a strdup() with 
    qemu_strdup(), unrelated to the goal of the patch

  - In "Reorder initialization", you change the formatting of the args 
    to the qemu_new_vlan_client() call - conflicted with my patches

  - In the same patch you've added a whole pile of braces in 
    what was net_slirp_redir() - made re-basing onto Alexander's slirp 
    stuff that bit more involved

  - You could have split up "slirp: Move smb, redir, tftp and bootp
    parameters and -net channel", maybe even made a separate patch for 
    each move

  - It's very hard to understand why each of the changes in "slirp:    
    Rework internal configuration" is needed and it's a big patch - 
    e.g. you completely re-wrote tcp_ctl(). Could that have been done 
    with a cleanup patch with no functional changes followed by the 
    actual functional changes?

  - "slirp: Rework external configuration interface" introduces several 
    new slirp options. Surely should be possible to split up into 
    smaller patches.

Cheers,
Mark.

  reply	other threads:[~2009-05-28 17:23 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-08 10:34 [Qemu-devel] [PATCH 00/11] Networking fixes and slirp enhancements Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 02/11] net: Fix and improved ordered packet delivery Jan Kiszka
2009-05-08 15:24   ` Mark McLoughlin
2009-05-08 10:34 ` [Qemu-devel] [PATCH 01/11] net: Don't deliver to disabled interfaces in qemu_sendv_packet Jan Kiszka
2009-05-08 15:20   ` Mark McLoughlin
2009-05-08 22:27     ` [Qemu-devel] "FLOSS bounty" ( FB )for running QEMU on SheevaPlug AGSCalabrese
2009-05-08 22:47       ` Marek Vasut
2009-05-08 22:58       ` Paul Brook
2009-05-08 10:34 ` [Qemu-devel] [PATCH 04/11] net: Real fix for check_params users Jan Kiszka
2009-05-19  7:57   ` Mark McLoughlin
2009-05-19  9:34     ` Jan Kiszka
2009-05-19  9:57       ` Mark McLoughlin
2009-05-28 15:04   ` Mark McLoughlin
2009-05-28 15:05     ` [Qemu-devel] [PATCH 1/3] Revert "Fix output of uninitialized strings" Mark McLoughlin
2009-05-28 15:06       ` [Qemu-devel] [PATCH 2/3] net: Real fix for check_params users Mark McLoughlin
2009-05-28 15:06         ` [Qemu-devel] [PATCH 3/3] net: fix error reporting for some net parameter checks Mark McLoughlin
2009-05-28 15:56           ` [Qemu-devel] " Jan Kiszka
2009-05-28 15:22     ` [Qemu-devel] [PATCH 04/11] net: Real fix for check_params users Kevin Wolf
2009-05-08 10:34 ` [Qemu-devel] [PATCH 03/11] slirp: Avoid zombie processes after fork_exec Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 10/11] slirp: Rework external configuration interface Jan Kiszka
2009-05-28 15:07   ` Mark McLoughlin
2009-05-28 15:55     ` Jan Kiszka
2009-05-28 17:23       ` Mark McLoughlin [this message]
2009-05-28 20:41         ` Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 07/11] Introduce get_next_param_value Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 09/11] slirp: Rework internal configuration Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 08/11] slirp: Move smb, redir, tftp and bootp parameters and -net channel Jan Kiszka
2009-05-28 15:07   ` Mark McLoughlin
2009-05-28 15:52     ` Jan Kiszka
2009-05-29 11:42     ` Paul Brook
2009-05-29 14:19       ` Jan Kiszka
2009-05-29 15:36         ` Paul Brook
2009-05-08 10:34 ` [Qemu-devel] [PATCH 05/11] net: Improve parameter error reporting Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 06/11] slirp: Reorder initialization Jan Kiszka
2009-05-08 10:34 ` [Qemu-devel] [PATCH 11/11] slirp: Bind support for host forwarding rules Jan Kiszka
2009-05-08 16:25 ` [Qemu-devel] [PATCH 00/11] Networking fixes and slirp enhancements Anthony Liguori
2009-05-08 17:01   ` Jan Kiszka
2009-05-09  7:41     ` [Qemu-devel] [PATCH 08/11 v2] slirp: Move smb, redir, tftp and bootp parameters and -net channel Jan Kiszka

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1243531394.18588.63.camel@blaa \
    --to=markmc@redhat.com \
    --cc=jan.kiszka@siemens.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).