On 12/22/19 10:39 AM, Marc Lehmann wrote: > (Note that I have a conversation with Jens in private, as per his request, > but since he replied to this publicly, so do I) > >>> https://bugzilla.kernel.org/show_bug.cgi?id=204081 that bug >>> https://bugzilla.kernel.org/show_bug.cgi?id=204065 oops bug >>> https://bugzilla.kernel.org/show_bug.cgi?id=204085 linux aio potential >>> security bug (not iouring) >> >> Well, none of these have me on the CC list, and they are filed against >> the aio subsystem. There's no way I'd see these, and I'm pretty sure the >> person listed for aio has long since moved elsewhere. >> >> For bug reports, always at least CC the person, otherwise you have no >> guarantee they are seen! > > I thought the same, which is why, as I wrote, I _mailed_ *you* > _directly_. Not as CC:, but a brand-new mail specific to iouring where I > didn't even make the assumption that I hit a bug (because, maybe it works > as designed).
I've never received an email from you! I just double checked both this email and [email protected], and I've never received an email. When did you send it? Subject? To whom? I guess this is the core of the issue! > Keep in mind I was documenting what I did in a reply to Benjamin, which > was about the whole issue, which affects both linux aio and io_uring, and > I clearly distinguish between the two everywhere, I think. I'm just talking about the bugzilla, they were all filed under aio, which meant I would never see them unless someone added me to the CC, I pointed it out to me on the side. >> I'm not subscribed to linux-kernel, I need to be CC'ed on emails to see >> them. Otherwise I couldn't do anything but read emails, and I do have >> other things to do. > > Sure, and I didn't imply you should, or did I? Well sort of, you're implying you did what you could to try and reach me, and I'm trying to explain why most of those didn't work so well as they would never reach me. The private email is still a puzzle, hopefully we can get to the bottom of that. I'm not trying to place blame here, I'm trying to explain why assuming I was well informed about any of this was false as I never got any of them. > I wrote that, among other things, I wrote to you personally, so you > shouldn't be that surprised. Why didn't you quote that part? It's > literally in the first sentence of my reply? Why do you keep complaining > about not being cc'ied on bug reports when you got sent your own personal > copy, just for you? > > Also, keep in mind you do not have to be so awfully defensive here, I am > just documenting what I did. Not being defensive, I'm explaining why I never saw any of this. > A simple (fictitious!) "I am a busy person and I don't even remember such > a mail, what the heck" is totally defensible in my eyes. _I_ can't even > reply to all personal emails I get. In fact, I am probably much worse at > replying or even readong mail then you are. And I would have, had I gotten an email. > In fact, I don't care why you didn't reply, as long as you (I assume) have > a valid reason. I am well aware that you are not my personal slave. > > If you wish, we can just drop this topic - I have you now on mail, why > would I care why I never saw a reply. We can also drag this out to any > detail you want, I might even be able to have an old enough log file > showing who eventually accepted that mail from our mailserver, if need be. Yes let's drop this topic, it's counter productive. As I explained higher up, I never got a single email on this topic. And I'd like to figure out why, in terms of when you sent it and to what email. Because that's the only thing I don't get. >>> I understand we are all busy, and bugzilla.kernel.org sems to work a lot >>> like the usenet wizard groups (saome areas are just blackholes, such as >>> btrfs and aio/iouring, others are quite active, such as ext4 :), but on >>> the other hand, Jens really should not be that _surprised_ :) >> >> You may think this is fair, but it's definitely not. > > You are attacking the messenger. > >> First of all, you bundle aio/iouring, they are totally not the same >> thing. > > I think with "you" you mean bugzila.kernel.org (which doesn't have a > category for io_uring specifically). Do not attack the messenger. I don't want to use bugzilla if I can avoid it. Most folks have been filing issues on the liburing github, which is fine with me. Or sending email, which is my preferred way of being contacted. Filing under a different category is kind of pointless. >> bug report for libevent and call it libev/libevent and expect you to >> find it. > > Neither do I, why do you bring this up? Because it's pretty similar to filing a bugzilla for a different entity? > I took your e-mail address from the maintainers file for io_uring, which > is the same address as you are using now. OK good, so you used this one. I'm using gmail hosting for that, and I only find the emails from today and yesterday. Nothing on libev either. Maybe you can understand why I'm puzzled, and why it's literally the first thing I've heard about this? >> I've replied to ALL bug reports I've received, and taken care of them. > > I have no clue what you received, and I don't know what you define as > bug, but nobody replied to the mail I sent to the io_uring maintainer > address. Thats simply a fact. I also didn't blame you for that. But let's get to the bottom of that, please. >>> https://urldefense.proofpoint.com/v2/url?u=http-3A__cvs.schmorp.de_libev_ev-5Fiouring.c&d=DwIBAg&c=5VD0RTtNlTh3ycd41b3MUw&r=cK1a7KivzZRh1fKQMjSm2A&m=YL0hyN7rczX2id-XEDVKfTKUvACqEeGAK2-3FUwHsdY&s=ZO_rZs9vCG9MPUeiSp36c4PJnR9Vwo_ApX1WtQRTy1s&e= >>> > > This quote that is supposedly from my mail is not from my mail, I think > something patches your mails. Probably not the source of the problems > here, though. It's the fb.com email, which does some weird url mangling to redirect through some kind of phishing filter. It's part of the reason why I want to use this email, not the fb.com email. >> I'll take a look, but as you probably know, timing is unfortunate with >> the holidays coming up. I'll wrapping things up tomorrow, then not much >> after that until after the new year. Just to set expectations. > > I'm not in a hurry. I already put your points onto my "TODO" list: > > http://cvs.schmorp.de/libev/ev_iouring.c > > I will _eventually_ come to them, and I thought the minimum I could do > _now_ is to document new developments, so that people reading the file > (which I know are many) do get the right impression. > > I didn't have a look at the io_uring kernel side for a few months, because > I didn't have a perspective on any future bugfixes, or rather, not even > knowing if io_uring is supposded to be generic or not and these are bugs, > or simply not supported. > > I documented what I did, and I think my assessment is very fair. That's great! Thanks. >>> throwaway oneliners), but between kernel oopses, having to use epoll, the >>> issues with io_uring under high load, the abysmal state of documentation >>> and so on, I didn't really bother to play with it much more in its current >>> state. >> >> I'll try that too. 5.1 is ancient, and the first release with io_uring. >> I would not expect too much, it's a lot more mature now. > > I think you mean 5.2, sure, and I accept your interesting definition of > ancient :) Yeah both of them. I realize they are not THAT old, but in terms of io_uring being introduced in 5.1, they are :-) Replace 5.1 with io_uring 0.1, and it makes more sense. >> Jens Axboe > > So, Jens. When I got your first mail I was very excited because I thought > there is a real way out of this dilemma. As I said, I really would like > io_uring to become the default backend for libev (and thus a great deal > of other software). You cannot fathom how excited I was when I first read > about io_uring. Well, actually, it was more like "shit, now I absolutely > _have_ to implement this, even though I have no time". > > I really want this to work. > > But this mail just attacks me for things I haven't said or done, and now > I have the feeling of a maintainer who just wants to bash people who > criticise their work, rather than making as good software solution. > > I am quite sure I do not deserve your attacks on me, as I was quite > factual. I'm sorry if they came across as attacks, it was not my intent. If you put yourself in my position, here are the facts from my end: 1) I don't get the bugzilla entries, hence I never saw them 2) I didn't get a private email. While #1 is totally expected, this one is not. And I believe this point 2 is the core of the issue! Which means that what I see is someone that hasn't done the due diligence and then is posting about issues that I would have loved to have known about and fixed. I want to get to the bottom of the personal email issue. But probably better if we just focus on moving ahead on the technical issues here, as long as there are no emails lost in the future. > For example, I told you that I mailed you *personally*. I do not > expect, and I certainly do not wish you to then chastise me for not > cc'ing you on a bug report or not telling you, or unfairly mixing > linux aio with io_uring just because I *correctly* point out that > io_uring and linux aio suffer from the same bug. > > So, we can either stop here, and leave me (and probably you) > disappointed with the state of affairs, or you get down from your > needlessly defensive posture. On the other side, I have zero desire to > throw around shit and see what sticks - you can criticise me to your > hearts content, I am used to usenet-style blunt words, but at least > try to attack me for things I actually said or did. Or actually > didn't. > > The choice is yours, I very much wish for an amiable and productive > exchange, and I sincerely wish you nice holidays. > > And, we can continue this in private, or in public, but I would like > to stick to either, and not both. I'm fine with either, I only sent you a private email to cover the a-z list you had. And to ensure that we can have some sort of stable communication. In either case, I would strongly suggest we just move on to technical issues, too much time has been wasted already on procedure. -- Jens Axboe _______________________________________________ libev mailing list [email protected] http://lists.schmorp.de/mailman/listinfo/libev
