Ask New 2.4 Maintainer Marcelo Tosatti Anything 308
Linus keeps hinting (declaring, even) that he's nearly ready to work full-time on the 2.5 development branch of his kernel, and hand the 2.4 kernel off to Marcelo Tosatti. Marcelo's graciously agreeed to answer questions (you might want to read some of his mailing list contributions first), so here's your chance to ask him what he'll do in the famous footsteps of Linus and Alan Cox, and how he got there. Please only put one question per post; we'll pass along the top-rated comments to Marcelo for his answers, and hear back from him shortly.
My Question (Score:4, Interesting)
Re:My Question (Score:2)
Once again a valid question modded as "Troll." Moderators that moderate based on personal vendettas rather than proper moderation protocol should be taken out back and shot. If you moderated the parent as "-1, Troll" you sir are an idiot. Go read the moderators guide and make sure you know the facts before you moderate based on your inexperience. Allan Cox did choose to keep certain security fixes to the kernel out of the US change log due to potential conflicts with the DMCA. He made the decision. Now other people want to question it. Wise up moderators.
background (Score:5, Interesting)
XFS (Score:1)
Re:XFS (Score:1)
I think XFS is more mature than Reiserfs which is already include.
PS : ext3 seems just to be included
Coup (Score:3, Offtopic)
Re:Coup (Score:2, Funny)
2.4 and 2.5 (Score:5, Interesting)
Re:If I can get the damn memory bugs fixed... (Score:2)
Re:If I can get the damn memory bugs fixed... (Score:2)
2.3.1 -> This is the first upgrade to the third re-working of the second major build of FOLK.
It's not intended to be the same as the Kernel patch number, as I don't intend to release one FOLK patch per kernel, any more than Alan Cox did. The reason I use 3 digits to his one, is that something as complex as FOLK can experience major changes in behaviour with only minor changes in the code. It's far more useful for people to be able to see the degree of change, and what sort of change it is, than to just get a single revision number.
Ask Marcelo Tosatti Anything? (Score:1, Funny)
Re:Ask Marcelo Tosatti Anything? (Score:2, Insightful)
Money (Score:4, Interesting)
Re:Money (Score:3, Interesting)
Hit by a bus (Score:5, Interesting)
I'd hate to stray from the status quo where standard interview questions are concerned, so in keeping with it, I ask:
What's it like knowing that, if (God forbid) Linus and Alan were hit by a bus, you might be "The Man"?
Hey, someone had to ask.
Re:Hit by a bus (Score:1)
Re:Hit by a bus (Score:2)
Re:Hit by a bus (Score:2, Funny)
Hmmmm. What sort of bus architecture would it take to run down Linus and Alan both?
First, it has to be pretty long-range. A modified FC bus, I suppose, with the protocol altered to support transatlantic distances and repeaters. The mother of all differential SCSI, shall we say.
It would also have to have a fairly fat pipe - one can't imagine serious injury resulting from a collision at a mere 10mbit/sec, right? PCI 33/32 packs a decent punch, enough to knock you down, but I doubt it could kill you. The FC bus would have to hang off something faster like AGP4x or VME.
Then we have routing ... you know there's no direct fibre line from Swansea to Santa Clara, so the bus protocol would have to support bridging of some sort at the least. I guess FC fits the bill here too.
........oh, that kind of bus? Ummmm, never mind..
... (Score:1, Troll)
Hardware to support in 2.4? (Score:4, Interesting)
Expectations (Score:5, Interesting)
Age a question (Score:5, Interesting)
Preemptile (Score:4, Interesting)
Re:Preemptile (Score:2)
The real question is what area is the kernel headed into? Or, is it possible that a compile-time option can be set to chose from the two different schemes?
security patches in the Changelogs? (Score:4, Interesting)
Threads (Score:5, Interesting)
Any chance of Alan Cox's multi-threaded post-mortem debugging patch which dumps multiple core files for each lightweight process (LWP) making it into the kernel? How about support for post mortem debugging of multithreaded core files in general (right now there is zero support).
Any rants about threading as a general topic would work.
Time for renumbering? (Score:1, Interesting)
Wouldn't it make more sense to take 2.4.14 (or 2.4.15 whenever it comes out) and call that 2.5.0, while moving the unstable kernels over to 2.6? While it would break the even/odd numbering paradigm, that seems a better option than having a subversion which starts out unstable and partway through morphs into a stable form.
Re:Time for renumbering? (Score:2)
As long as people use the latest 2.4, it will be the most stable. (Unless Marcelo screws up.)
Stability vs Features (Score:5, Interesting)
For example, do new or updated device drivers rank more highly than VM updates?
-- Azaroth
How do you plan to handle the big companies? (Score:5, Interesting)
How are you going to deal with the submissions from people like IBM and SGI who are going to want to make significant changes to 2.4?
vi or emacs? (Score:1, Offtopic)
Re:vi or emacs? (Score:2)
emacs or vi?
RMS reminder (Score:2)
its GNUEmacs or vi?
VM and EXT3 (Score:1, Redundant)
Re:VM and EXT3 (Score:1)
Sound drivers (Score:5, Informative)
Not having ALSA in 2.4.x means no good sound support in the stable kernel for another year of two. Do you plan to integrate ALSA into the 2.4 branch? If yes, will it happen after it's done on the 2.5 branch?
Re:Sound drivers (Score:2)
Of course - the people who use Linux on their server doesn't give a damn about the sound OSS vs. ALSA, and others who do - can simply install the RPM and get over it. I can hardly belive that ANYONE will seriously consider replacing OSS with ALSA at this stage.
What would you rather do: (Score:2, Funny)
2) Date Daisy Fuentes [daisyfuentes.com] (or any one person of your choice)
3) Get to play around with a bat and various people from Microsoft
Re:What would you rather do: (Score:2)
Ginger or Maryann?
Nurse Chapel or Counselor Troi?
Princess Leia or Queen Amidala?
Do you use a distribution? (Score:4, Interesting)
Re:Do you use a distribution? (Score:5, Interesting)
What operating systems and platforms do you personally use and which ones do you also use (and why)?
Do you run a common environment on all your machines (in as far as possible) or do you run different things in different places and which environments do you prefer for what?
What development tools do you use (especially for the kernel), would you do anything differebtly for the kernel (like make it compile with other compilers) if you could (or will you) and would you like to (or will you) place the Linux stable kernel into CVS or another version control system?
Do you feel any personal preferences for anything might actually be in anyway reflected in your work as the stable kernel maintainer?
Do you have a good lawyer? Are you planing on travelling to the U.S.A. (for all I know you live there, excuse my ignorance:-)? Have you experience dealing with politicians, business leaders and large groups and do you see this as a part of your job description?
When you stop maintaing the stable kernel, what would you like people to be saying about your reign?
Linux or BSD
Alan Cox and politics (Score:5, Interesting)
But is the Linux kernel the right forum for politics? Do you feel that it's a bad idea to involve the kernel in politics -- a slippery slope in which the software itself becomes a political pawn? Or would you say that the kernel -- and all software -- has already become a political pawn, and Cox's actions are entirely justified given that free software's existence is under increasing threat?
Human interest (Score:4, Interesting)
The Deal with Marcelo, Alan and Linus? (Score:4, Interesting)
What is going on here?
It seems that Alan has stopped doing his -AC series...
Linus is finishing up 2.4.X after making some deep changes to the VM (Adrea's new VM), and thankfully adding EXT3, but form what I have been reading, 2.4.15 is the end of the stable kernel series (with no XFS or JFS support which is upsetting).
Andrea has a plethora of experimental/tweaked patches in ftp.kernel.org/pub/linux/kernel/people/andrea. What is interesting/worrisome to me is that Marcelo's directory there is empty.
Marcelo must be a great guy, I've seen a lot of banter on newsgroups with him, and a few chages in the stable changelog here and there, but to me it looks like a lot of talk. Alan used to release (past tense, it seems it has stopped) AC patches on a near daily basis, to me, Marcelo is vapor.
Is the maintainer's jobs simple to make small changes to the kernel when errata is found? Or is it, as Alan has done, to integrate and merge a LOT of stuff to produce a useful and robust hybrid kernel and then suggest that maybe the unbroken things should be merged in.
I wish Marcelo luck in this endeavor, but also wish to see loads of "maintainence," Linux really, really needs feverish active development, and there a lot of people, Like Alan and Linus, who put out quite a bit. I am hoping Marcelo will set a new precedent for uber-feverish maintenance - maybe even see XFS and JFS and other things that the distributions have to waste huge amounts of time tinkering and adding various enterprise-ish things to make the Linux kernel stand up for, as they put it, prime time.
Confusing two different issues... (Score:2)
In addition to maitaining the 2.2.x series, Alan also has his own, more experimental tree knows as the -ac tree. This is where a lot of developers submit their patches for initial testing for a few reasons: Alan can get them in his tree sooner because his tree is more experimental than the main kernel tree; Alan is a direct link to Linus so if Alan approves of your patch, he will eventually submit it to Linus (all those merging comments in the kernel change logs).
If you want to compare Marcelo's position to Alan's, your question would have to be, do you plan to start an -mt tree for the 2.6.x kernel series while maintaining the 2.4.x kernel?
Re:The Deal with Marcelo, Alan and Linus? (Score:2)
/Brian
Re:The Deal with Marcelo, Alan and Linus? (Score:2)
2.4.x is a stable series kernel, where 'x' is a release number.
2.5.x is the development unstable kernel.
2.6.0 will happen when 2.5.x hits a point of stability and feature-freeze.
Re:The Deal with Marcelo, Alan and Linus? (Score:2)
/Brian
Re:The Deal with Marcelo, Alan and Linus? (Score:2)
then became the rule
Re:The Deal with Marcelo, Alan and Linus? (Score:2)
Are you asking where the kernel number came from because it is so high? Did you look at kernel.org to see the archive of all the old kernels down to 1.x? There was a
I parse your question but it seems like you either left some essential component of your question out or I'm missing something... By all means clarify your question if you aren't getting the answers you seek.
Money (Score:4, Interesting)
CML2, cvs, kdb, crypto and more (Score:5, Interesting)
Also, would you consider moving crypto into the main tree in the near future? Debian has, Redhat will "soon"..
Would you consider using bitkeeper, cvs, or even complete changelogs with proper attribution of WHICH merges took place?
And finally, would you consider FINALLY bringing kdb into the main kernel? Linus doesnt want it, but he doesnt want it because he doesnt see the value. He didnt say he banned it..
Patches (Score:5, Interesting)
Linux in the Latin world? (Score:4, Informative)
Re:Linux in the Latin world? (Score:2)
Why you? (Score:5, Interesting)
What makes you think that you can handle maintaining the Linux Kernel? What qualifications do you have and why should we (the people) trust you with the Linux kernel.
I realize that this is actually two questions, but they more or less go together in one answer.
Interaction with distros (Score:3, Interesting)
A pretty generic question (Score:5, Interesting)
Any plans to improve documenting the kernel? (Score:5, Interesting)
doesn't this somehow defeat the point of the source being available in the first place? Basically the question I have to ask is this, "I have flirted with the thought of sending comment only patches to the kernel to further help people understand certain subtleties (e.g. why the pprev and next pointers in user_struct are not what they seem) in the source code especially CS students who are learning about the kernel in operating systems classes. If someone were to start such a program would such patches be accepted into the kernel?
Re:Any plans to improve documenting the kernel? (Score:2, Insightful)
Heh. This is very similar to the comment in the original Unix sources. The idea was that it was completely obscure until one understood something magical that couldn't be documented, at which point it didn't need commenting. A lot of really low-level stuff can be pretty obscure and mind-bending - it's just a fact of life when dealing with such things and documenting it doesn't help.
By the way, the Buddy System is a memory allocation strategy given by Donald Knuth in his book "Fundamental Algorithms". It's pretty obvious once you've seen how it works, but I'd have never thought of it independently. I would assume that understanding the code requires one to understand the algorithm first - e.g. by reading Knuth's excellent description that is unfortunately too long to stick in a comment.
Re:Any plans to improve documenting the kernel? (Score:3, Interesting)
When you write code you write for a specific standard (or compiler or interpreter) but when you write English you write for unknown. Descriptions could at some point get more lengthy than code. Would that be bad? I don't know. I'd rather have the information than not, personally. But I can see why bibles of information are kept out of the kernel.
Re:Any plans to improve documenting the kernel? (Score:2)
It's perfectly fine to have highly technical comments of the sort that you or I would never understand. But it is NOT okay to have comments that say "you really aren't expected to understand this." If you are unable to explain something to someone else, you don't really understand it.
Re:Any plans to improve documenting the kernel? (Score:2)
That's got to be the stupidest thing I've read all day. Unfortunately for the software industry it isn't orginal but in fact an unwritten policy.
Go look at the blueprints for a house. Fully documented. Look at the blueprints for nuclear reactor. Fully documented. Would YOU trust a nuclear reactor that was not documented? Software will not leave the dark ages until the developers start treating it as a professional discipline.
Re:Any plans to improve documenting the kernel? (Score:2)
Every programming language I know has syntax for including comments. Presumably these comments are for more than just satisfying license requirements to document changes within the code.
Not all of your documentation needs to be in the form of comments. In the case of the buddy system, it would have been appropriate to include a reference to external docs. But not even this was done. As a professional software developer, my documentation is expected to include specifications and commented code at the bare minimum. A code review for uncommented code? Unthinkable!
Comment your code. Comment your code. Comment your code. It's standard practice in academia, industry, and yes, even volunteer open source projects.
Are you ready....... (Score:2, Interesting)
code control (Score:5, Interesting)
Re:code control (Score:4, Insightful)
National biases. (Score:3, Interesting)
--saint
A question (Score:2, Interesting)
No, the -real- question is whether you will be treating 2.4.x as a "maintained but dead" branch, or whether you'll be actively beating the code senseless with a (plank *) twoByFour?
(Quick followup: If you -are- going to actively develop 2.4.x, will you be restricting that to merging in 2.5.x code, or will you be seeking out new patches, boldly coding where no kernel has gone before?)
Re:A question (Score:2)
/Brian
Conectiva & Kernel & Security (Score:3, Interesting)
cage match... (Score:5, Funny)
Non-x86 architectures (Score:2)
Re:Non-x86 architectures (Score:2)
/brian
When? (Score:2)
Also how do you see yourself handling issues like that in 2.4.11 and 2.4.14 where simple drivers would not compile or would not run, because of obsolete function calls in the kernel? I.E. Wait till 2.4.15 to fix this or release 2.4.15 with just that fix?
Kernel growth -- is it a problem? (Score:5, Interesting)
My question: I presume kernel bloat, both in terms of code size and resource demands, can impact heavily on maintainability. Do you see this as a significant issue for the future, how much of a problem is it, and is it something that can be easily addressed without tearing apart the code base and reorganizing it from the ground up?
/Brian
Re:Kernel growth -- is it a problem? (Score:2)
only the driver code has grown since 2.2 (Score:2)
By the main parts I mean everything except drivers. The largest parts of this are the file system and network code. The file system grew a bit when Rieserfs was include into the kernel but it's a tiny increase compared to the rest of the code.
Drivers make up more than 75% of the Linux source code. And this driver code is growing rapidly.
People talk about the fact that you can cut things out of the Linux kernel if you don't want to use them. This is true. But it's even more true to say that if you did want to use everything in the Linux kernel you would have to buy a lot of hardware.
diary (Score:5, Interesting)
Because we all like to know, if you'll actually be doing what your gonna promise now
Re:diary (Score:2)
list of changes for the common folk (Score:5, Insightful)
One thing that is really missing is a list of changes in each kernel release that is meant to be consumed by the masses. The "changelogs" that are offered up are sorely lacking for us non-kernel hackers. What I'd like to see is a prose description of the changes in each version. Something like Release notes for 2.2.18 by Alan Cox [linux.org.uk] is a step in the right direction, but some of it is even a little too technical. For example, in the above document,
means little to me and probably a lot of other people. Under what condition does this occur? The question why should I care about this change? should be answered for each entry.How do you feel about doing something like this?
Re:list of changes for the common folk (Score:2)
I'm not trolling. I'm serious. Don't install a kernel if you don't know what you're doing. The kernel is the base of your system so you shouldn't toy around with it.
The changelogs are quite readible to the indented audience.
How often (Score:2, Interesting)
Age (Score:3, Interesting)
Do you think this will reflect badly on the image of linux? (as seen from IT decision makers)
2.4 a good switch from FreeBSD? (Score:2)
Linux 2.4 future focus (Score:2, Insightful)
What will be your main focus while maintaining 2.4, stability or backported extra functionality. It is doubtless that there will be some backports. But what will you focus on stability and speed or features?
Google/VM/Corporate influences (Score:2, Interesting)
It seems obvious from this post that Google can't use 2.4 until this gets worked out, and Google is one of Linux's big success stories...
How does that weigh in on the todo scale versus my problem with the Visor driver? What if it was Connectiva that had the problem? (Or Redhat?)
(and a quick second question: How can any human read teh 7000 posts a month to the mailing list and still get any coding done?)
XFree86 4.0 and 4.1 DRI (Score:2)
Last I checked (which was a while ago), Linux kernels included only DRI support for XFree86 4.1, whereas Alan kernels included DRI support for both (or, rather one or the other, selected at compilation time). What will be the case with future official 2.4.x kernels?
-Rob
Too many questions for Marcelo - Ask Slashdot??? (Score:4, Funny)
Where can we get the questions answered? I think that there ought to be a Linux kernel documentation project to bring the docs up to speed and answer questions like this. I have not the knowledge of the kernel for this, but such a project would be invaluable to the Linux community.
Ask him *anything*?? (Score:2, Funny)
Boot Messsages (Score:3, Interesting)
Do you plan on furthering this cause, and helping make our boot times a simpler, less complicated read?
You are an encouragement to 18 yr olds everywhere! (Score:4, Offtopic)
I must say that you are an amazing yet rare individual, to have commanded that much respect at the age of 18. Wow. I can't handle OS design now, let alone when I was 18. Would you be so kind as to share some about your upbringing, environment, faith, or other factors that were beneficial to you in making you who you are today?
Thank you,
girls running to you yet? (Score:4, Funny)
Features vs Stability (Score:3, Interesting)
Am I too late? (Score:3, Redundant)
For instance:
1. The "lousy magazine"-question: Who would you rather be stuck with on a desert island?
2. The "bitter old sysadmin"-question: You are 18, do you punk even know what Multics or VMS is?
3. The "trick"-question: what do you prefer, emacs or vi? (trick question because we all know emacs is superior).
Conectiva issues (Score:4, Interesting)
I read that you live in the same house as Rik van Riel, who wrote the first Linux 2.4 VM and who also works for Conectiva. Rik wrote in his not-that-very-often updated diary that he travelled a lot in Brazil, to help customers anywhere. That takes a lot of time. Is your job going to be different once Alan Cox passes the maintainership?
stuff&life.... (Score:2, Interesting)
I have grown to respect quite a bit you as well as some of the other folks who work in kernel development at conectiva in Brazil (Rik, Arnaldo..).
I have been wanting to try to read more and more about the linux kernel, and maybe(?) try to get involved, so i must ask how did you start out? Any pointers that i should follow to get to understand more deeply the kernel? I mean you are not in heavy kernel hacking and you are a few years younger than me... (My guess is that it wasnt in college, since by now you should be doing the infamous college entrance exams... aka vestibular =P)
Also it seems that you value your privacy, since there isnt much public information about you out there... Would you elaborate a little on how you got where you are? =)
Just one more thing... I feel i must say that all Brazilians and young people in general are very proud and happy to see where you are now! Thanks a lot and keep up the good work!question for Marcello (Score:2, Interesting)
"I've been examining the existing kernel configuration system, and I have about concluded that the best favor we could do everybody involved with it is to take it out behind the barn and shoot it through the head."
For someone as persuasive and influential as Raymond to make such a statement, it begs for a response, if not a rebuttal.
what do you think about HURD? (Score:2)
linus says: Don't do drugs. (Or "i don't care")
what is you opinion about this?
Re:I have a version numbering question (Score:2)