1 2Information you need to know about netdev 3----------------------------------------- 4 5Q: What is netdev? 6 7A: It is a mailing list for all network-related Linux stuff. This includes 8 anything found under net/ (i.e. core code like IPv6) and drivers/net 9 (i.e. hardware specific drivers) in the Linux source tree. 10 11 Note that some subsystems (e.g. wireless drivers) which have a high volume 12 of traffic have their own specific mailing lists. 13 14 The netdev list is managed (like many other Linux mailing lists) through 15 VGER ( http://vger.kernel.org/ ) and archives can be found below: 16 17 http://marc.info/?l=linux-netdev 18 http://www.spinics.net/lists/netdev/ 19 20 Aside from subsystems like that mentioned above, all network-related Linux 21 development (i.e. RFC, review, comments, etc.) takes place on netdev. 22 23Q: How do the changes posted to netdev make their way into Linux? 24 25A: There are always two trees (git repositories) in play. Both are driven 26 by David Miller, the main network maintainer. There is the "net" tree, 27 and the "net-next" tree. As you can probably guess from the names, the 28 net tree is for fixes to existing code already in the mainline tree from 29 Linus, and net-next is where the new code goes for the future release. 30 You can find the trees here: 31 32 http://git.kernel.org/?p=linux/kernel/git/davem/net.git 33 http://git.kernel.org/?p=linux/kernel/git/davem/net-next.git 34 35Q: How often do changes from these trees make it to the mainline Linus tree? 36 37A: To understand this, you need to know a bit of background information 38 on the cadence of Linux development. Each new release starts off with 39 a two week "merge window" where the main maintainers feed their new 40 stuff to Linus for merging into the mainline tree. After the two weeks, 41 the merge window is closed, and it is called/tagged "-rc1". No new 42 features get mainlined after this -- only fixes to the rc1 content 43 are expected. After roughly a week of collecting fixes to the rc1 44 content, rc2 is released. This repeats on a roughly weekly basis 45 until rc7 (typically; sometimes rc6 if things are quiet, or rc8 if 46 things are in a state of churn), and a week after the last vX.Y-rcN 47 was done, the official "vX.Y" is released. 48 49 Relating that to netdev: At the beginning of the 2-week merge window, 50 the net-next tree will be closed - no new changes/features. The 51 accumulated new content of the past ~10 weeks will be passed onto 52 mainline/Linus via a pull request for vX.Y -- at the same time, 53 the "net" tree will start accumulating fixes for this pulled content 54 relating to vX.Y 55 56 An announcement indicating when net-next has been closed is usually 57 sent to netdev, but knowing the above, you can predict that in advance. 58 59 IMPORTANT: Do not send new net-next content to netdev during the 60 period during which net-next tree is closed. 61 62 Shortly after the two weeks have passed (and vX.Y-rc1 is released), the 63 tree for net-next reopens to collect content for the next (vX.Y+1) release. 64 65 If you aren't subscribed to netdev and/or are simply unsure if net-next 66 has re-opened yet, simply check the net-next git repository link above for 67 any new networking-related commits. 68 69 The "net" tree continues to collect fixes for the vX.Y content, and 70 is fed back to Linus at regular (~weekly) intervals. Meaning that the 71 focus for "net" is on stabilization and bugfixes. 72 73 Finally, the vX.Y gets released, and the whole cycle starts over. 74 75Q: So where are we now in this cycle? 76 77A: Load the mainline (Linus) page here: 78 79 http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git 80 81 and note the top of the "tags" section. If it is rc1, it is early 82 in the dev cycle. If it was tagged rc7 a week ago, then a release 83 is probably imminent. 84 85Q: How do I indicate which tree (net vs. net-next) my patch should be in? 86 87A: Firstly, think whether you have a bug fix or new "next-like" content. 88 Then once decided, assuming that you use git, use the prefix flag, i.e. 89 90 git format-patch --subject-prefix='PATCH net-next' start..finish 91 92 Use "net" instead of "net-next" (always lower case) in the above for 93 bug-fix net content. If you don't use git, then note the only magic in 94 the above is just the subject text of the outgoing e-mail, and you can 95 manually change it yourself with whatever MUA you are comfortable with. 96 97Q: I sent a patch and I'm wondering what happened to it. How can I tell 98 whether it got merged? 99 100A: Start by looking at the main patchworks queue for netdev: 101 102 http://patchwork.ozlabs.org/project/netdev/list/ 103 104 The "State" field will tell you exactly where things are at with 105 your patch. 106 107Q: The above only says "Under Review". How can I find out more? 108 109A: Generally speaking, the patches get triaged quickly (in less than 48h). 110 So be patient. Asking the maintainer for status updates on your 111 patch is a good way to ensure your patch is ignored or pushed to 112 the bottom of the priority list. 113 114Q: How can I tell what patches are queued up for backporting to the 115 various stable releases? 116 117A: Normally Greg Kroah-Hartman collects stable commits himself, but 118 for networking, Dave collects up patches he deems critical for the 119 networking subsystem, and then hands them off to Greg. 120 121 There is a patchworks queue that you can see here: 122 http://patchwork.ozlabs.org/bundle/davem/stable/?state=* 123 124 It contains the patches which Dave has selected, but not yet handed 125 off to Greg. If Greg already has the patch, then it will be here: 126 http://git.kernel.org/cgit/linux/kernel/git/stable/stable-queue.git 127 128 A quick way to find whether the patch is in this stable-queue is 129 to simply clone the repo, and then git grep the mainline commit ID, e.g. 130 131 stable-queue$ git grep -l 284041ef21fdf2e 132 releases/3.0.84/ipv6-fix-possible-crashes-in-ip6_cork_release.patch 133 releases/3.4.51/ipv6-fix-possible-crashes-in-ip6_cork_release.patch 134 releases/3.9.8/ipv6-fix-possible-crashes-in-ip6_cork_release.patch 135 stable/stable-queue$ 136 137Q: I see a network patch and I think it should be backported to stable. 138 Should I request it via "stable@vger.kernel.org" like the references in 139 the kernel's Documentation/stable_kernel_rules.txt file say? 140 141A: No, not for networking. Check the stable queues as per above 1st to see 142 if it is already queued. If not, then send a mail to netdev, listing 143 the upstream commit ID and why you think it should be a stable candidate. 144 145 Before you jump to go do the above, do note that the normal stable rules 146 in Documentation/stable_kernel_rules.txt still apply. So you need to 147 explicitly indicate why it is a critical fix and exactly what users are 148 impacted. In addition, you need to convince yourself that you _really_ 149 think it has been overlooked, vs. having been considered and rejected. 150 151 Generally speaking, the longer it has had a chance to "soak" in mainline, 152 the better the odds that it is an OK candidate for stable. So scrambling 153 to request a commit be added the day after it appears should be avoided. 154 155Q: I have created a network patch and I think it should be backported to 156 stable. Should I add a "Cc: stable@vger.kernel.org" like the references 157 in the kernel's Documentation/ directory say? 158 159A: No. See above answer. In short, if you think it really belongs in 160 stable, then ensure you write a decent commit log that describes who 161 gets impacted by the bugfix and how it manifests itself, and when the 162 bug was introduced. If you do that properly, then the commit will 163 get handled appropriately and most likely get put in the patchworks 164 stable queue if it really warrants it. 165 166 If you think there is some valid information relating to it being in 167 stable that does _not_ belong in the commit log, then use the three 168 dash marker line as described in Documentation/SubmittingPatches to 169 temporarily embed that information into the patch that you send. 170 171Q: Someone said that the comment style and coding convention is different 172 for the networking content. Is this true? 173 174A: Yes, in a largely trivial way. Instead of this: 175 176 /* 177 * foobar blah blah blah 178 * another line of text 179 */ 180 181 it is requested that you make it look like this: 182 183 /* foobar blah blah blah 184 * another line of text 185 */ 186 187Q: I am working in existing code that has the former comment style and not the 188 latter. Should I submit new code in the former style or the latter? 189 190A: Make it the latter style, so that eventually all code in the domain of 191 netdev is of this format. 192 193Q: I found a bug that might have possible security implications or similar. 194 Should I mail the main netdev maintainer off-list? 195 196A: No. The current netdev maintainer has consistently requested that people 197 use the mailing lists and not reach out directly. If you aren't OK with 198 that, then perhaps consider mailing "security@kernel.org" or reading about 199 http://oss-security.openwall.org/wiki/mailing-lists/distros 200 as possible alternative mechanisms. 201 202Q: What level of testing is expected before I submit my change? 203 204A: If your changes are against net-next, the expectation is that you 205 have tested by layering your changes on top of net-next. Ideally you 206 will have done run-time testing specific to your change, but at a 207 minimum, your changes should survive an "allyesconfig" and an 208 "allmodconfig" build without new warnings or failures. 209 210Q: Any other tips to help ensure my net/net-next patch gets OK'd? 211 212A: Attention to detail. Re-read your own work as if you were the 213 reviewer. You can start with using checkpatch.pl, perhaps even 214 with the "--strict" flag. But do not be mindlessly robotic in 215 doing so. If your change is a bug-fix, make sure your commit log 216 indicates the end-user visible symptom, the underlying reason as 217 to why it happens, and then if necessary, explain why the fix proposed 218 is the best way to get things done. Don't mangle whitespace, and as 219 is common, don't mis-indent function arguments that span multiple lines. 220 If it is your first patch, mail it to yourself so you can test apply 221 it to an unpatched tree to confirm infrastructure didn't mangle it. 222 223 Finally, go back and read Documentation/SubmittingPatches to be 224 sure you are not repeating some common mistake documented there. 225