| Message ID | CAPYw7P7syZhb98oDAV53KGANqcZODxVNiEX+objtObTH-zG6ww@mail.gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel] avfilter/buffersrc: switch to activate | expand |
Paul B Mahol (12023-11-04): > From 31a6a78ebc3a3f8785ec7c8e5ffd4257c7eadec3 Mon Sep 17 00:00:00 2001 > From: Paul B Mahol <onemda@gmail.com> > Date: Fri, 27 Oct 2023 14:26:50 +0200 > Subject: [PATCH] avfilter/buffersrc: switch to activate > > Fixes error when caller keeps adding frames into filtergraph > that reached EOF by other means, for example EOF is signalled > by other filter in filtergraph or by buffersink. Filters with no more than one input and one output do not need to implement activate. If it seems they need, that indicates a bug in the framework. Then that bug needs to be found and fixed. This patch just adds code to work around it, it is not acceptable.
On Sat, Nov 4, 2023 at 8:07 PM Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-04): > > From 31a6a78ebc3a3f8785ec7c8e5ffd4257c7eadec3 Mon Sep 17 00:00:00 2001 > > From: Paul B Mahol <onemda@gmail.com> > > Date: Fri, 27 Oct 2023 14:26:50 +0200 > > Subject: [PATCH] avfilter/buffersrc: switch to activate > > > > Fixes error when caller keeps adding frames into filtergraph > > that reached EOF by other means, for example EOF is signalled > > by other filter in filtergraph or by buffersink. > > Filters with no more than one input and one output do not need to > implement activate. If it seems they need, that indicates a bug in the > framework. Then that bug needs to be found and fixed. This patch just > adds code to work around it, it is not acceptable. > Sorry to hear your deep feelings and failed realizations for this topic and in FFmpeg in general. Certainly you are not, never was, and never will be central absolute authority to say what is acceptable and what is not to do in FFmpeg code. Now on technical side of story: buffersrc can be closed by calling av_buffersrc_close(). And currently in master that is the only way how buffersrc can end processing of sending more frames to filter chain. There is possibility for filters in filtergraph to signal EOF back down (backward) to buffersrc filter in filtergraph chain, but that same thing is ignored by buffersrc filter. This patch address this issue/bug. Now why switch to .activate() is proper solution for buffersrc filter? buffersrc filter uses request_frame(), and that one is used to count number of failed requests when no frame is available for next filters in filtergraph. Inside request_frame() one can check only for EOF from inlinks, but as buffersrc filter is source filter it have no inlinks. Other function filter_frame() is not applicable to source filter types. Also you forgot fact that buffersrc filter have 0 inputs and 1 output, so to correct your claim, following holds: Any filter with only one input and only one output of same media type is not forced to use .activate(). Above may not hold if such 1->1 filter does modify timeline of frames. (Drops/inserts frames/samples, like obvious example is fps filter) Similar applies to buffersink filter, with 1 input and 0 outputs, by your standards such filter do not need .activate(), which is obviously false. For more insight into this subject and groundbreaking theory of filtering, please do consult: 912969a33e313c57c906e87a7e2367b78a2160f4 lavfi/buffersink: move to the new design. Note: the OOM happens even without split filter usage, which means its not fault of .activate() in split filter code. > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
Paul B Mahol (12023-11-06):
> Gonna apply soon.
No you are not.
Gonna apply soon.
On Mon, Nov 6, 2023 at 1:50 PM Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-06): > > Gonna apply soon. > > No you are not. > While I appreciate the brevity of this email, I think this kind of communication is non constructive. If the mailing list is the only and universal voice of the project, perhaps it could be used a bit better. Nicolas, have you thought of providing a patch to the bug Paul is describing?
Vittorio Giovara (12023-11-06): > Nicolas, have you thought of providing a patch to the bug Paul is > describing? Why do you not ask Paul if he have thought of giving me time to have a chance to do so?
On Mon, Nov 6, 2023 at 3:07 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > Nicolas, have you thought of providing a patch to the bug Paul is > > describing? > > Why do you not ask Paul if he have thought of giving me time to have a > chance to do so? > You gave no indication you were willing to, and I thought he gave you time on the other thread for a related bug and you used it for mind games. Anyway, usually code trumps words, so the course I'd suggest is agreeing on a feasible time window, and if you can get a better patch out, then submit it, otherwise Paul's can go in. After all it can always be reverted in case a better solution comes along.
Vittorio Giovara (12023-11-06): > You gave no indication you were willing to, and I thought he gave you time Bullshit. Please do not intervene in a thread when you do not know what is going on at all. Paul has been repeatedly rude and threatened to push before review, yet it is towards me you address your criticism. > on the other thread for a related bug and you used it for mind games. > Anyway, usually code trumps words, so the course I'd suggest is agreeing on > a feasible time window, and if you can get a better patch out, then submit > it, otherwise Paul's can go in. No. I will review this issue when I have the time and Paul will wait. This is how it works. Invalid patches do not go in just because the maintainer is busy or sick.
On Mon, Nov 6, 2023 at 3:28 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > You gave no indication you were willing to, and I thought he gave you > time > > Bullshit. Please do not intervene in a thread when you do not know what > is going on at all. Paul has been repeatedly rude and threatened to push > before review, yet it is towards me you address your criticism. > Please as a non-native English speaker to another non-native English speaker, you may miss that this comes off extremely aggressive and offensive especially in text form. And this is a public mailing list, everybody is allowed to intervene in every thread -- if you don't like it, take your discussion offline. There is no need for some language or behavior, while I can understand being mad at Paul, the community, or the world, please make a minimum of effort to keep communication polite, and efficient, as stated multiple times. Belligerence will take you nowhere. > > on the other thread for a related bug and you used it for mind games. > > Anyway, usually code trumps words, so the course I'd suggest is agreeing > on > > a feasible time window, and if you can get a better patch out, then > submit > > it, otherwise Paul's can go in. > > No. I will review this issue when I have the time and Paul will wait. > This is how it works. Invalid patches do not go in just because the > maintainer is busy or sick. > That is not how it works, you cannot indefinitely veto a patch just because you think it is invalid. Your concerns in the first email were addressed in Paul's reply. In the other thread, 3 people verified the presence of the bug while you insisted it wasn't there, while also confirming the patch is valid. If you can provide further counterpoints to Paul's explanation, please share them instead of silence, if you have anything substantial (code wise) to add please do so, otherwise let the patch in. This bickering on the list cannot continue.
Vittorio Giovara (12023-11-06): > world, please make a minimum of effort to keep communication polite Please realize that politeness resides not only in which words we use, like “bullshit”, but also in what we say, even if it is with very mild words. For example: > while you insisted it wasn't there Pretending I said something when I said precisely the opposite is extremely rude from you. Or: > That is not how it works, you cannot indefinitely veto a patch Pretending that I try to veto a patch when I am just demanding the time to properly review it is extremely rude from you.
On Mon, Nov 6, 2023 at 3:50 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > world, please make a minimum of effort to keep communication polite > > Please realize that politeness resides not only in which words we use, > like “bullshit”, but also in what we say, even if it is with very mild > words. For example: > > > while you insisted it wasn't there > > Pretending I said something when I said precisely the opposite is > extremely rude from you. > > Or: > > > That is not how it works, you cannot indefinitely veto a patch > > Pretending that I try to veto a patch when I am just demanding the time > to properly review it is extremely rude from you. > Nice try but pointing out your logical fallacies is not being rude, but rather is just showing you how easily your message can be misinterpreted, due to the aggressive and non-constructive way of communication you insist on using. And yes, asking for time when you yourself do not provide a timeline is equivalent to vetoing a patch (indefinitely too, since you yourself say you "will review this issue when [you] have the time", which could very well be never). I feel like, once again, we've exhausted this topic, and before delving into what is rude and what is time, let's get back to the technical merits of the patch. Please justify your objections to the proposed solution, provide proof that this patch is invalid, or send code that improves upon this. Thank you
Vittorio Giovara (12023-11-06): > Nice try but pointing out your logical fallacies is not being rude, but > rather is just showing you how easily your message can be misinterpreted, You have done nothing of the sort. You have just misrepresented my words, showing that you decided to barge into a conversation without knowing its details. > And yes, asking for time when you yourself do not provide a > timeline is equivalent to vetoing a patch (indefinitely too, since you > yourself say you "will review this issue when [you] have the time", which > could very well be never). And now you are accusing me of outright dishonesty. Better and better. > I feel like, once again, we've exhausted this topic You have exhausted because you barged in without knowing the first thing about it. > Please justify your objections to the proposed solution, I have already done so, but again you neglected to read it: if there is a bug, it is in the framework code.
On Mon, Nov 6, 2023 at 4:46 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > Nice try but pointing out your logical fallacies is not being rude, but > > rather is just showing you how easily your message can be misinterpreted, > > You have done nothing of the sort. You have just misrepresented my > words, showing that you decided to barge into a conversation without > knowing its details. > No, you barged out and started calling fair arguments "bullshit" while I'm trying to have a polite discussion. I'm not misrepresenting anybody's words, just making sure you understand how belittling and prone to misinterpretation your comments are due to harsh style of communication. I think I'm advising you to improve that through the lines in order to have better success in getting your point across. > And yes, asking for time when you yourself do not provide a > > timeline is equivalent to vetoing a patch (indefinitely too, since you > > yourself say you "will review this issue when [you] have the time", which > > could very well be never). > > And now you are accusing me of outright dishonesty. Better and better. > I'm not a lawyer, this is not a trial, nobody's accusing you of anything. I'm just pointing out the facts that lead to *a* reasonable conclusion, you could prove me wrong by providing a sound technical discussion but you insist in this syllogistic battle, hence the long off topic, and the continued bickering. > > I feel like, once again, we've exhausted this topic > > You have exhausted because you barged in without knowing the first thing > about it. > No, we've exhausted the topic because you insist on playing the "who is right" game with someone who has no stake in the matter. Right now, if I acted like you do, I'd start a long paragraph saying something like "oh this is a process of intentions" or "so you're accusing me of barging in, this rude/dishonest/$adjective", but instead I'm patiently trying to bring the discussion back to the technical points raised previously. Can we focus on them please? > > Please justify your objections to the proposed solution, > > I have already done so, but again you neglected to read it: if there is > a bug, it is in the framework code. > Paul already explained why it is not, in the email starting with "Sorry to hear your deep feelings and failed realizations for this topic and in FFmpeg in general." You seem to have neglected to read it and reply to the points made there.
Vittorio Giovara (12023-11-06):
> I'm not misrepresenting anybody's words,
Me: “I never doubted that Paul's patches do make the bug go away in your
test case.”
You: “you insisted it wasn't there”
This is precisely misrepresenting my words.
So stop lying and present excuses.
On Mon, Nov 6, 2023 at 5:49 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > I'm not misrepresenting anybody's words, > > Me: “I never doubted that Paul's patches do make the bug go away in your > test case.” > > You: “you insisted it wasn't there” > > This is precisely misrepresenting my words. > > So stop lying and present excuses > Stop twisting people's words and implying things they never said :) And thank you for ignoring my request to stay on topic, this really shows the shortcomings of the mailing list and how important live communication and IRC texting is. Paul, feel free to push at any time.
Vittorio Giovara (12023-11-06): > Stop twisting people's words and implying things they never said :) This is not you presenting excuses. > And thank you for ignoring my request to stay on topic There is no topic to stay on: I will review this bug when time permits, and you do not know anything about lavfi's scheduling so you cannot help.
On Mon, Nov 6, 2023 at 6:05 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > Stop twisting people's words and implying things they never said :) > > This is not you presenting excuses. > Correct, it's you trying to win arguments with aggressive communication > And thank you for ignoring my request to stay on topic > > There is no topic to stay on: I will review this bug when time permits, > and you do not know anything about lavfi's scheduling so you cannot > help. > Projecting much?
Vittorio Giovara (12023-11-06):
> Correct, it's you trying to win arguments with aggressive communication
No. You came into an argument without knowing the first thing about it
and you decided to side with the guy who wants to violate the project's
policy about review. Your behavior in this thread was toxic from your
very first post.
On Tue, Nov 7, 2023 at 2:44 AM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-06): > > Correct, it's you trying to win arguments with aggressive communication > > No. You came into an argument without knowing the first thing about it > and you decided to side with the guy who wants to violate the project's > policy about review. > Oh you like policies now? Interesting :) Your behavior in this thread was toxic from your very first post. I'm sorry but it was your behavior that has been toxic, rude, and frankly unacceptable. It seems like you pick on everybody who dares to confront you, and don't care about taking an entire discussion offtopic just for the sake of winning an argument. To better explain the case at hand, there is no developer policy that lets you request an indefinite amount of review time with a single "No, you are not" message. A good way to ask for an extension would be a message like "Please let me have a few days to review this and debug it. I know last time I used the same request only to test you and spread more toxicity around me, but this time I really mean it". ( https://ffmpeg.org/pipermail/ffmpeg-devel/2023-October/316297.html)... Hm I wonder why your request is being ignored now. And no, I am not siding with anybody, skipping over your personal attack about my libavfilter knowledge, I literally have no stake in the matter. I do recognize toxic behavior tho, and I think it should stop at a certain point, as the list deserves better than reading this email exchange. If you would like to provide more technical feedback to the patch (or bring references to points already made), please go ahead, otherwise just drop this vendetta you seem to have against everybody in your path. Please and thank you, I believe in you.
Vittorio Giovara (12023-11-07): > To better explain the case at hand, there is no developer policy that lets > you request an indefinite amount of review time with a single "No, you are > not" message. There is a policy that maintainers review patches and patches are not pushed when somebody wants to review. > A good way to ask for an extension would be a message like > "Please let me have a few days to review this and debug it. I know last > time I used the same request only to test you and spread more toxicity > around me, but this time I really mean it". ( > https://ffmpeg.org/pipermail/ffmpeg-devel/2023-October/316297.html)... Hm I > wonder why your request is being ignored now. Thanks for pointing this e-mail, anybody reading this honestly will realize Paul's “I did proper analysis already. Pushed.” was the problem in the first place, even more so since it was purely trolling.
Will apply in next 48h. Libavfilter code is not so complex to need so big time interval for review from experienced C developer.
Paul B Mahol (12023-11-07): > Will apply in next 48h. So, you are again threatening to violate the project's policy. > Libavfilter code is not so complex to need so big time interval for review > from experienced C developer. First, the issue is not only analyzing the code but first finding time to do so. Realize that some of us have a job distinct from FFmpeg. And a life. And that does not make us second class citizens of the project. Second, yes, the scheduling of libavfilter is quite complex, and the fact that you believe it is not shows how little qualified you are to judge your own patch.
Hi, On Tue, Nov 7, 2023, 16:50 Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-07): > > Will apply in next 48h. > > So, you are again threatening to violate the project's policy. > > > Libavfilter code is not so complex to need so big time interval for > review > > from experienced C developer. > > First, the issue is not only analyzing the code but first finding time > to do so. Realize that some of us have a job distinct from FFmpeg. And a > life. And that does not make us second class citizens of the project. > > Second, yes, the scheduling of libavfilter is quite complex, and the > fact that you believe it is not shows how little qualified you are to > judge your own patch. > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". > This is really unfair to Paul to let this drag on for so long. If you are genuinely this busy, is there really no one else on the list who you can delegate to review this piece? -- Sean McGovern >
Sean McGovern (12023-11-07): > This is really unfair to Paul to let this drag on for so long. What does fair have to do with this? If there is a bug, it has been there for years, it can be there for a few more months without any drawback. Plus, Paul has the patch that makes the bug disappear in his test case, he can use it if he want. Some reviews have dragged for weeks before somebody even looked at the patch. It has been just three days since Paul deigned giving some kind of analysis. > If you are genuinely this busy, is there really no one else on the list who > you can delegate to review this piece? This is a public mailing-list, have you seen somebody volunteering? The scheduling in libavfilter is tricky, and I am not aware of anybody else having manifested interest about the subtle details.
I have now been able to set up a reasonable test case with limit memoryuse 600M. Next step is to simplify the test case. (I cannot understand how you can pretend you analyzed the bug when your test case contains unrelated filters like hqdn3d or setsar.)
Paul B Mahol (12023-11-08): > I did. I simplified it right from start. But I kept it for myself as It is > not really relevant. It is very relevant for reproducing the problem and reviewing the patch. Please show what you have, it will save time. > With this pace of your debugging it can be expected that universe will > froze first before you manage to came to any conclusion about the > "framework". Then work on something else. There is no urgency here.
On Wed, Nov 8, 2023 at 5:53 PM Nicolas George <george@nsup.org> wrote: > I have now been able to set up a reasonable test case with limit > memoryuse 600M. Next step is to simplify the test case. > > (I cannot understand how you can pretend you analyzed the bug when your > test case contains unrelated filters like hqdn3d or setsar.) > I did. I simplified it right from start. But I kept it for myself as It is not really relevant. With this pace of your debugging it can be expected that universe will froze first before you manage to came to any conclusion about the "framework". > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
On Wed, Nov 8, 2023 at 7:22 PM Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-08): > > I did. I simplified it right from start. But I kept it for myself as It > is > > not really relevant. > > It is very relevant for reproducing the problem and reviewing the patch. > Please show what you have, it will save time. > > > With this pace of your debugging it can be expected that universe will > > froze first before you manage to came to any conclusion about the > > "framework". > > Then work on something else. There is no urgency here. > I waited enough, applied, now for 100% true statement. > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
Paul B Mahol (12023-11-11):
> I waited enough, applied, now for 100% true statement.
Your teacher called, you forgot your pacifier in class yesterday.
I am still expecting you to share the test cases you have so that the
bug can be properly fixed.
On Sat, Nov 11, 2023 at 1:27 PM Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-11): > > I waited enough, applied, now for 100% true statement. > > Your teacher called, you forgot your pacifier in class yesterday > No, bad Nicolas, stop this language, it's unacceptable and unbecoming. You can ask for more time in a kinder manner. No, don't reply to me with insults or excuses, just take a walk outside and reflect on your mistakes.
Vittorio Giovara (12023-11-11): > No, bad Nicolas, stop this language, it's unacceptable and unbecoming. You > can ask for more time in a kinder manner. > No, don't reply to me with insults or excuses, just take a walk outside and > reflect on your mistakes. Address your reproaches to the person who started the “unacceptable and unbecoming” behavior. You just shown that you are biassed against me.
On Sat, Nov 11, 2023 at 1:42 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-11): > > No, bad Nicolas, stop this language, it's unacceptable and unbecoming. > You > > can ask for more time in a kinder manner. > > No, don't reply to me with insults or excuses, just take a walk outside > and > > reflect on your mistakes. > > Address your reproaches to the person who started the “unacceptable and > unbecoming” behavior. You just shown that you are biassed against me. > I said don't reply to me! I'm not biased, reread what you wrote and think if "pacifier" belongs on a technical mailing list. The other person's behavior can be right or wrong depending on the interpretation, but using the language you insist on using makes you wrong right away. Please just stop, you cannot have a bad day on the ml every day, it's bad for your mental health.
Vittorio Giovara (12023-11-11): > I said don't reply to me! Fortunately you have no authority over me. > I'm not biased, You are. In this message again. > reread what you wrote and think > if "pacifier" belongs on a technical mailing list. I agree. > The other person's behavior can be right or wrong depending on the > interpretation, but using the language you insist on using makes you wrong > right away. “I don't know what the other did, what you did is obviously wrong” is the discourse of adults empowering bullies everywhere by punishing the victims who try to defend themselves. > Please just stop, you cannot have a bad day on the ml every day, it's bad > for your mental health. Oh, so “pacifier” is inappropriate but this is? I hope your are only engaging in a comedy act.
On Sat, Nov 11, 2023 at 2:04 PM Nicolas George <george@nsup.org> wrote: > Vittorio Giovara (12023-11-11): > > I said don't reply to me! > > Fortunately you have no authority over me. > I said that to avoid ridiculing yourself any more > > I'm not biased, > > You are. In this message again. > shameless accusation, trying to discredit someone who's tired of this toxic language > > The other person's behavior can be right or wrong depending on the > > interpretation, but using the language you insist on using makes you > wrong > > right away. > > “I don't know what the other did, what you did is obviously wrong” is > the discourse of adults empowering bullies everywhere by punishing the > victims who try to defend themselves. > Nice victimization, but "i'm behaving poorly because the other person is a meanie" is a childish argument > > Please just stop, you cannot have a bad day on the ml every day, it's bad > > for your mental health. > > Oh, so “pacifier” is inappropriate but this is? I hope your are only > engaging in a comedy act. I am genuinely concerned for whatever reason is pushing you to use such an alienating and toxic language, yes.
On Sat, Nov 11, 2023 at 7:27 PM Nicolas George <george@nsup.org> wrote: > Paul B Mahol (12023-11-11): > > I waited enough, applied, now for 100% true statement. > > Your teacher called, you forgot your pacifier in class yesterday. > > I am still expecting you to share the test cases you have so that the > bug can be properly fixed. > So all this time you play arrogant moves and expect something out of nothing but just posting cryptic text. The case can be simplified by just using single filter, "trim=.." with same parameters as in script. > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
Paul B Mahol (12023-11-12): > So all this time you play arrogant moves and expect something out of > nothing but just posting cryptic text. “Out of nothing”? What does that even mean? It is a collaboration: you do not share because I want it but because it helps your own goal. > The case can be simplified by just using single filter, "trim=.." with same > parameters as in script. Please show the exact command so that we do not talk at cross-purposes.
Vittorio Giovara (12023-11-11): > Nice victimization, but "i'm behaving poorly because the other person is a > meanie" is a childish argument Mock all your want, the fact that you attacked me even though I was just replying and not even told Paul to stop his childish games proves you are biassed beyond all credibility. > I am genuinely concerned for whatever reason is pushing you to use such an > alienating and toxic language, yes. Yeah, right. You mocking somebody's mental health was way more inappropriate than anything that I or Paul ever wrote.
On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: > Vittorio Giovara (12023-11-11): > > Nice victimization, but "i'm behaving poorly because the other person is a > > meanie" is a childish argument > > Mock all your want, the fact that you attacked me even though I was just > replying and not even told Paul to stop his childish games proves you > are biassed beyond all credibility. > > > I am genuinely concerned for whatever reason is pushing you to use such an > > alienating and toxic language, yes. > > Yeah, right. You mocking somebody's mental health was way more > inappropriate than anything that I or Paul ever wrote. Vittorio, Nicolas, Paul. guys i love you all but can you rest this thread until we have a new community committee ? thanks! PS: if there are technical questions and you want, there is a technical committe that could look at it. (At least IMHO the TC is fully functional, thats just my personal oppinion iam not speaking for the TC here) [...]
On Sun, Nov 12, 2023 at 5:37 PM Michael Niedermayer <michael@niedermayer.cc> wrote: > On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: > > Vittorio Giovara (12023-11-11): > > > Nice victimization, but "i'm behaving poorly because the other person > is a > > > meanie" is a childish argument > > > > Mock all your want, the fact that you attacked me even though I was just > > replying and not even told Paul to stop his childish games proves you > > are biassed beyond all credibility. > > > > > I am genuinely concerned for whatever reason is pushing you to use > such an > > > alienating and toxic language, yes. > > > > Yeah, right. You mocking somebody's mental health was way more > > inappropriate than anything that I or Paul ever wrote. > > Vittorio, Nicolas, Paul. > > guys i love you all but can you rest this thread until we have a new > community committee ? > > thanks! > > PS: if there are technical questions and you want, there is a technical > committe that could look at it. > (At least IMHO the TC is fully functional, thats just my personal oppinion > iam not speaking for the TC here) > You are blocking my patch. You should be ashamed. > > [...] > > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > The bravest are surely those who have the clearest vision > of what is before them, glory and danger alike, and yet > notwithstanding go out to meet it. -- Thucydides > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
On Sun, Nov 12, 2023 at 09:19:57PM +0100, Paul B Mahol wrote: > On Sun, Nov 12, 2023 at 5:37 PM Michael Niedermayer <michael@niedermayer.cc> > wrote: > > > On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: > > > Vittorio Giovara (12023-11-11): > > > > Nice victimization, but "i'm behaving poorly because the other person > > is a > > > > meanie" is a childish argument > > > > > > Mock all your want, the fact that you attacked me even though I was just > > > replying and not even told Paul to stop his childish games proves you > > > are biassed beyond all credibility. > > > > > > > I am genuinely concerned for whatever reason is pushing you to use > > such an > > > > alienating and toxic language, yes. > > > > > > Yeah, right. You mocking somebody's mental health was way more > > > inappropriate than anything that I or Paul ever wrote. > > > > Vittorio, Nicolas, Paul. > > > > guys i love you all but can you rest this thread until we have a new > > community committee ? > > > > thanks! > > > > PS: if there are technical questions and you want, there is a technical > > committe that could look at it. > > (At least IMHO the TC is fully functional, thats just my personal oppinion > > iam not speaking for the TC here) > > > > > You are blocking my patch. > You should be ashamed. dear patch iam ashamed but iam not actually blocking you. Maybe you can forgive me if i provide this short ode to you Oh, Paul's Patch, you sly little thing, A fix for libavfilter, oh joy it brings. In the world of code, where bugs do dwell, You swoop in gracefully, breaking their spell. With your lines of logic, so neat and clean, You dance through the bugs like a code routine. Oh, Paul's Patch, you're a hero in disguise, Saving us from errors, what a pleasant surprise! You sneak into the code, like a ninja in the night, Fixing issues left and right, oh what a sight. Libavfilter trembles, but not in fear, For with Paul's Patch, victory is near. Oh, patch so fine, with your bug-crushing might, You turn errors into pixels of pure delight. Paul's Patch, the troubleshooter supreme, In the open source world, you reign supreme. Programmers cheer, and users applaud, For with Paul's Patch, we're never flawed. So here's to you, oh mighty fixer-upper, In the kingdom of code, you're the top-drawer. To Paul's Patch, a toast we raise, In the realm of software, you set ablaze. For every bug quivered, and every glitch shivered, Thanks to Paul's Patch, we code delivered! thx [...]
On Mon, Nov 13, 2023, 11:59 Michael Niedermayer <michael@niedermayer.cc> wrote: > On Sun, Nov 12, 2023 at 09:19:57PM +0100, Paul B Mahol wrote: > > On Sun, Nov 12, 2023 at 5:37 PM Michael Niedermayer < > michael@niedermayer.cc> > > wrote: > > > > > On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: > > > > Vittorio Giovara (12023-11-11): > > > > > Nice victimization, but "i'm behaving poorly because the other > person > > > is a > > > > > meanie" is a childish argument > > > > > > > > Mock all your want, the fact that you attacked me even though I was > just > > > > replying and not even told Paul to stop his childish games proves you > > > > are biassed beyond all credibility. > > > > > > > > > I am genuinely concerned for whatever reason is pushing you to use > > > such an > > > > > alienating and toxic language, yes. > > > > > > > > Yeah, right. You mocking somebody's mental health was way more > > > > inappropriate than anything that I or Paul ever wrote. > > > > > > Vittorio, Nicolas, Paul. > > > > > > guys i love you all but can you rest this thread until we have a new > > > community committee ? > > > > > > thanks! > > > > > > PS: if there are technical questions and you want, there is a technical > > > committe that could look at it. > > > (At least IMHO the TC is fully functional, thats just my personal > oppinion > > > iam not speaking for the TC here) > > > > > > > > > You are blocking my patch. > > You should be ashamed. > > dear patch iam ashamed but iam not actually blocking you. > Maybe you can forgive me if i provide this short ode to you > > Oh, Paul's Patch, you sly little thing, > A fix for libavfilter, oh joy it brings. > In the world of code, where bugs do dwell, > You swoop in gracefully, breaking their spell. > > With your lines of logic, so neat and clean, > You dance through the bugs like a code routine. > Oh, Paul's Patch, you're a hero in disguise, > Saving us from errors, what a pleasant surprise! > > You sneak into the code, like a ninja in the night, > Fixing issues left and right, oh what a sight. > Libavfilter trembles, but not in fear, > For with Paul's Patch, victory is near. > > Oh, patch so fine, with your bug-crushing might, > You turn errors into pixels of pure delight. > Paul's Patch, the troubleshooter supreme, > In the open source world, you reign supreme. > > Programmers cheer, and users applaud, > For with Paul's Patch, we're never flawed. > So here's to you, oh mighty fixer-upper, > In the kingdom of code, you're the top-drawer. > > To Paul's Patch, a toast we raise, > In the realm of software, you set ablaze. > For every bug quivered, and every glitch shivered, > Thanks to Paul's Patch, we code delivered! > > thx > > [...] > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > Many things microsoft did are stupid, but not doing something just because > microsoft did it is even more stupid. If everything ms did were stupid they > would be bankrupt already. > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". > Made me laugh, +1. -- Sean McGovern >
> 在 2023年11月14日，上午1:45，Sean McGovern <gseanmcg@gmail.com> 写道： > > On Mon, Nov 13, 2023, 11:59 Michael Niedermayer <michael@niedermayer.cc> > wrote: > >>> On Sun, Nov 12, 2023 at 09:19:57PM +0100, Paul B Mahol wrote: >>> On Sun, Nov 12, 2023 at 5:37 PM Michael Niedermayer < >> michael@niedermayer.cc> >>> wrote: >>> >>>> On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: >>>>> Vittorio Giovara (12023-11-11): >>>>>> Nice victimization, but "i'm behaving poorly because the other >> person >>>> is a >>>>>> meanie" is a childish argument >>>>> >>>>> Mock all your want, the fact that you attacked me even though I was >> just >>>>> replying and not even told Paul to stop his childish games proves you >>>>> are biassed beyond all credibility. >>>>> >>>>>> I am genuinely concerned for whatever reason is pushing you to use >>>> such an >>>>>> alienating and toxic language, yes. >>>>> >>>>> Yeah, right. You mocking somebody's mental health was way more >>>>> inappropriate than anything that I or Paul ever wrote. >>>> >>>> Vittorio, Nicolas, Paul. >>>> >>>> guys i love you all but can you rest this thread until we have a new >>>> community committee ? >>>> >>>> thanks! >>>> >>>> PS: if there are technical questions and you want, there is a technical >>>> committe that could look at it. >>>> (At least IMHO the TC is fully functional, thats just my personal >> oppinion >>>> iam not speaking for the TC here) >>>> >>> >>> >>> You are blocking my patch. >>> You should be ashamed. >> >> dear patch iam ashamed but iam not actually blocking you. >> Maybe you can forgive me if i provide this short ode to you >> >> Oh, Paul's Patch, you sly little thing, >> A fix for libavfilter, oh joy it brings. >> In the world of code, where bugs do dwell, >> You swoop in gracefully, breaking their spell. >> >> With your lines of logic, so neat and clean, >> You dance through the bugs like a code routine. >> Oh, Paul's Patch, you're a hero in disguise, >> Saving us from errors, what a pleasant surprise! >> >> You sneak into the code, like a ninja in the night, >> Fixing issues left and right, oh what a sight. >> Libavfilter trembles, but not in fear, >> For with Paul's Patch, victory is near. >> >> Oh, patch so fine, with your bug-crushing might, >> You turn errors into pixels of pure delight. >> Paul's Patch, the troubleshooter supreme, >> In the open source world, you reign supreme. >> >> Programmers cheer, and users applaud, >> For with Paul's Patch, we're never flawed. >> So here's to you, oh mighty fixer-upper, >> In the kingdom of code, you're the top-drawer. >> >> To Paul's Patch, a toast we raise, >> In the realm of software, you set ablaze. >> For every bug quivered, and every glitch shivered, >> Thanks to Paul's Patch, we code delivered! >> >> thx >> >> [...] >> -- >> Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB >> >> Many things microsoft did are stupid, but not doing something just because >> microsoft did it is even more stupid. If everything ms did were stupid they >> would be bankrupt already. >> _______________________________________________ >> ffmpeg-devel mailing list >> ffmpeg-devel@ffmpeg.org >> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >> >> To unsubscribe, visit link above, or email >> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >> > > Made me laugh, +1. More poems and no drama, We can be half of a Shakespeare. > > -- Sean McGovern > >> > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with s
On Mon, Nov 13, 2023 at 6:00 PM Michael Niedermayer <michael@niedermayer.cc> wrote: > On Sun, Nov 12, 2023 at 09:19:57PM +0100, Paul B Mahol wrote: > > On Sun, Nov 12, 2023 at 5:37 PM Michael Niedermayer < > michael@niedermayer.cc> > > wrote: > > > > > On Sun, Nov 12, 2023 at 11:02:49AM +0100, Nicolas George wrote: > > > > Vittorio Giovara (12023-11-11): > > > > > Nice victimization, but "i'm behaving poorly because the other > person > > > is a > > > > > meanie" is a childish argument > > > > > > > > Mock all your want, the fact that you attacked me even though I was > just > > > > replying and not even told Paul to stop his childish games proves you > > > > are biassed beyond all credibility. > > > > > > > > > I am genuinely concerned for whatever reason is pushing you to use > > > such an > > > > > alienating and toxic language, yes. > > > > > > > > Yeah, right. You mocking somebody's mental health was way more > > > > inappropriate than anything that I or Paul ever wrote. > > > > > > Vittorio, Nicolas, Paul. > > > > > > guys i love you all but can you rest this thread until we have a new > > > community committee ? > > > > > > thanks! > > > > > > PS: if there are technical questions and you want, there is a technical > > > committe that could look at it. > > > (At least IMHO the TC is fully functional, thats just my personal > oppinion > > > iam not speaking for the TC here) > > > > > > > > > You are blocking my patch. > > You should be ashamed. > > dear patch iam ashamed but iam not actually blocking you. > Maybe you can forgive me if i provide this short ode to you > > Oh, Paul's Patch, you sly little thing, > A fix for libavfilter, oh joy it brings. > In the world of code, where bugs do dwell, > You swoop in gracefully, breaking their spell. > > With your lines of logic, so neat and clean, > You dance through the bugs like a code routine. > Oh, Paul's Patch, you're a hero in disguise, > Saving us from errors, what a pleasant surprise! > > You sneak into the code, like a ninja in the night, > Fixing issues left and right, oh what a sight. > Libavfilter trembles, but not in fear, > For with Paul's Patch, victory is near. > > Oh, patch so fine, with your bug-crushing might, > You turn errors into pixels of pure delight. > Paul's Patch, the troubleshooter supreme, > In the open source world, you reign supreme. > > Programmers cheer, and users applaud, > For with Paul's Patch, we're never flawed. > So here's to you, oh mighty fixer-upper, > In the kingdom of code, you're the top-drawer. > > To Paul's Patch, a toast we raise, > In the realm of software, you set ablaze. > For every bug quivered, and every glitch shivered, > Thanks to Paul's Patch, we code delivered! > If this is ChatGPT, I'm even more sad. > > thx > > [...] > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > Many things microsoft did are stupid, but not doing something just because > microsoft did it is even more stupid. If everything ms did were stupid they > would be bankrupt already. > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
Michael Niedermayer (12023-11-13):
> dear patch iam ashamed but iam not actually blocking you.
What was blocking this patch is that the bug had not been properly
analyzed and a review process was in progress, slowed down by Paul's own
refusal of providing the info his has (most likely because that info
does not exist).
But Paul pushed anyway, violating deliberately the project's policy. Out
of courtesy to everybody else I will not revert immediately as I should,
but I will revert with extreme prejudice if I find a simpler fix.
Regards,
On Mon, Nov 13, 2023 at 10:32 PM Nicolas George <george@nsup.org> wrote: > Michael Niedermayer (12023-11-13): > > dear patch iam ashamed but iam not actually blocking you. > > What was blocking this patch is that the bug had not been properly > analyzed and a review process was in progress, slowed down by Paul's own > refusal of providing the info his has (most likely because that info > does not exist). > What info that does not exist? Just use trim filter in ffmpeg script already provided, here goes that info, now exist also on this mailing list. > But Paul pushed anyway, violating deliberately the project's policy. Out > of courtesy to everybody else I will not revert immediately as I should, > but I will revert with extreme prejudice if I find a simpler fix. > > Regards, > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
From 31a6a78ebc3a3f8785ec7c8e5ffd4257c7eadec3 Mon Sep 17 00:00:00 2001 From: Paul B Mahol <onemda@gmail.com> Date: Fri, 27 Oct 2023 14:26:50 +0200 Subject: [PATCH] avfilter/buffersrc: switch to activate Fixes error when caller keeps adding frames into filtergraph that reached EOF by other means, for example EOF is signalled by other filter in filtergraph or by buffersink. Signed-off-by: Paul B Mahol <onemda@gmail.com> --- libavfilter/buffersrc.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git a/libavfilter/buffersrc.c b/libavfilter/buffersrc.c index 453fc0fd5c..b0a905d455 100644 --- a/libavfilter/buffersrc.c +++ b/libavfilter/buffersrc.c @@ -36,6 +36,7 @@ #include "audio.h" #include "avfilter.h" #include "buffersrc.h" +#include "filters.h" #include "formats.h" #include "internal.h" #include "video.h" @@ -194,7 +195,7 @@ FF_ENABLE_DEPRECATION_WARNINGS if (!frame) return av_buffersrc_close(ctx, s->last_pts, flags); if (s->eof) - return AVERROR(EINVAL); + return AVERROR_EOF; s->last_pts = frame->pts + frame->duration; @@ -484,21 +485,28 @@ static int config_props(AVFilterLink *link) return 0; } -static int request_frame(AVFilterLink *link) +static int activate(AVFilterContext *ctx) { - BufferSourceContext *c = link->src->priv; + AVFilterLink *outlink = ctx->outputs[0]; + BufferSourceContext *c = ctx->priv; - if (c->eof) - return AVERROR_EOF; + if (!c->eof && ff_outlink_get_status(outlink)) { + c->eof = 1; + return 0; + } + + if (c->eof) { + ff_outlink_set_status(outlink, AVERROR_EOF, c->last_pts); + return 0; + } c->nb_failed_requests++; - return AVERROR(EAGAIN); + return FFERROR_NOT_READY; } static const AVFilterPad avfilter_vsrc_buffer_outputs[] = { { .name = "default", .type = AVMEDIA_TYPE_VIDEO, - .request_frame = request_frame, .config_props = config_props, }, }; @@ -507,7 +515,7 @@ const AVFilter ff_vsrc_buffer = { .name = "buffer", .description = NULL_IF_CONFIG_SMALL("Buffer video frames, and make them accessible to the filterchain."), .priv_size = sizeof(BufferSourceContext), - + .activate = activate, .init = init_video, .uninit = uninit, @@ -521,7 +529,6 @@ static const AVFilterPad avfilter_asrc_abuffer_outputs[] = { { .name = "default", .type = AVMEDIA_TYPE_AUDIO, - .request_frame = request_frame, .config_props = config_props, }, }; @@ -530,7 +537,7 @@ const AVFilter ff_asrc_abuffer = { .name = "abuffer", .description = NULL_IF_CONFIG_SMALL("Buffer audio frames, and make them accessible to the filterchain."), .priv_size = sizeof(BufferSourceContext), - + .activate = activate, .init = init_audio, .uninit = uninit, -- 2.42.0