6 ms·

I find it interesting that Linus seems to have originally come across this particular bug report as a user; he, too, was having a problem with Flash player: ht

by mjs 16y ago

I find it interesting that Linus seems to have originally come across this particular bug report as a user; he, too, was having a problem with Flash player: https://bugzilla.redhat.com/show_bug.cgi?id=638477#c8 https://bugzilla.redhat.com/show_bug.cgi?id=638477#c8 "I see this as well. Sounds like clipping or some really bad sample rate frequency conversion." (It's only later that it becomes clear that this is due to a glibc "regression".)

acqq 16y ago
For the long history of how some glibc developers are enjoying to make more problems to the users see also: http://www.tuxradar.com/content/debian-ditches-glibc-eglibc http://www.tuxradar.com/content/debian-ditches-glibc-eglibc I consider all this a real attitude problem.
- tptacek 16y ago
  The first bit of evidence linked to by that article is glibc's maintainer complaining that he's being asked to improve the strfry() function, after a detailed technical analysis. If you don't know what strfry() is, that response does look unreasonable. But he's right! strfry() is, literally, a joke function.
  - simoncion 16y ago
    strfry's man page gives no indication that it's a joke function. Contrast with memfrob(3).
    - tptacek 16y ago
      From the man page: The function below addresses the perennial programming quandary: “How do I take good data in string form and painlessly turn it into garbage?”. To make the argument that strfry isn't a joke, you might find a real program that uses it productively. I'd be interested in seeing if one exists.
      - simoncion 16y ago
        Hm. My man page doesn't have that passage: NAME strfry - randomize a string [snip] DESCRIPTION The strfry() function randomizes the contents of string by using rand(3) to randomly swap characters in the string. The result is an anagram of string. [snip] COLOPHON This page is part of release 3.32 of the Linux man-pages project. [snip] What version of the man-pages package do you have installed?
        tptacek 16y ago
        Sorry, mine came from gnu.org and probably isn't really the man page (I misread a Google result). In any case, I'm not wrong about strfry being a gag function, and it looks (if you read upthread) that my hunch was right that nobody uses it in real code.
        simoncion 16y ago
        It is a joke function. I really wish that the supplied documentation let potential users in on the joke. :) (Oh, pardon. I didn't catch that you might have thought that I thought that strfry was anything but a joke. My take on the situation is that the documentation needs patched, rather than the strfry code. :) )
  - viraptor 16y ago
    Yet, it's there. I'd understand the reaction completely if it was an "It's broken. Fix it." bug report. But this one is "It's broken, this is how, this is data confirming it, this is a patch and that's how it's fixing the issue." The work is done - review, apply, release. Or if it's considered such a joke that it's not worth fixing, just remove it completely.
    - tptacek 16y ago
      It's only broken if you don't get the joke. Asking someone to review and merge a patch for strfry() is, if not unreasonable, then at least something that's reasonable to say "no" to.
      - Someone 16y ago
        http://www.google.com/codesearch?hl=en&start=30&sa=N&as_case=y&q=strfry http://www.google.com/codesearch?hl=en&start=30&sa=N... Seems to say it isn't used. I also contest that it _is_ broken. I cannot find a man page that promises that it uses a uniform distribution; one could even argue that such a function should not use an uniform distribution. For example, randomizing the non-uniformity of the distribution depending on the phase of the moon would, IMO, be a good idea for this function.
      - mukyu 16y ago
        He didn't just say no though. When presented with a model bug report (includes good description, test cases, and even a patch) he felt the need to change the function, but instead of using the supplied patch he rewrites it himself and commits it without testing it. It is still broken. When this is pointed out he gets angry and complains about people wasting his time, when it was he that decided to waste his own time rewriting it his own way and he was the one that still got it wrong. Everyone would have been better off if he had taken the sensible path of "read bug report", "confirm bug", "test patch", "commit patch" instead of arguing and rewriting things out of spite.

by mjs 16y ago

acqq 16y ago

For the long history of how some glibc developers are enjoying to make more problems to the users see also: http://www.tuxradar.com/content/debian-ditches-glibc-eglibc http://www.tuxradar.com/content/debian-ditches-glibc-eglibc I consider all this a real attitude problem.

tptacek 16y ago
The first bit of evidence linked to by that article is glibc's maintainer complaining that he's being asked to improve the strfry() function, after a detailed technical analysis. If you don't know what strfry() is, that response does look unreasonable. But he's right! strfry() is, literally, a joke function.
- simoncion 16y ago
  strfry's man page gives no indication that it's a joke function. Contrast with memfrob(3).
  - tptacek 16y ago
    From the man page: The function below addresses the perennial programming quandary: “How do I take good data in string form and painlessly turn it into garbage?”. To make the argument that strfry isn't a joke, you might find a real program that uses it productively. I'd be interested in seeing if one exists.
    - simoncion 16y ago
      Hm. My man page doesn't have that passage: NAME strfry - randomize a string [snip] DESCRIPTION The strfry() function randomizes the contents of string by using rand(3) to randomly swap characters in the string. The result is an anagram of string. [snip] COLOPHON This page is part of release 3.32 of the Linux man-pages project. [snip] What version of the man-pages package do you have installed?
      - tptacek 16y ago
        Sorry, mine came from gnu.org and probably isn't really the man page (I misread a Google result). In any case, I'm not wrong about strfry being a gag function, and it looks (if you read upthread) that my hunch was right that nobody uses it in real code.
        simoncion 16y ago
        It is a joke function. I really wish that the supplied documentation let potential users in on the joke. :) (Oh, pardon. I didn't catch that you might have thought that I thought that strfry was anything but a joke. My take on the situation is that the documentation needs patched, rather than the strfry code. :) )
- viraptor 16y ago
  Yet, it's there. I'd understand the reaction completely if it was an "It's broken. Fix it." bug report. But this one is "It's broken, this is how, this is data confirming it, this is a patch and that's how it's fixing the issue." The work is done - review, apply, release. Or if it's considered such a joke that it's not worth fixing, just remove it completely.
  - tptacek 16y ago
    It's only broken if you don't get the joke. Asking someone to review and merge a patch for strfry() is, if not unreasonable, then at least something that's reasonable to say "no" to.
    - Someone 16y ago
      http://www.google.com/codesearch?hl=en&start=30&sa=N&as_case=y&q=strfry http://www.google.com/codesearch?hl=en&start=30&sa=N... Seems to say it isn't used. I also contest that it _is_ broken. I cannot find a man page that promises that it uses a uniform distribution; one could even argue that such a function should not use an uniform distribution. For example, randomizing the non-uniformity of the distribution depending on the phase of the moon would, IMO, be a good idea for this function.
    - mukyu 16y ago
      He didn't just say no though. When presented with a model bug report (includes good description, test cases, and even a patch) he felt the need to change the function, but instead of using the supplied patch he rewrites it himself and commits it without testing it. It is still broken. When this is pointed out he gets angry and complains about people wasting his time, when it was he that decided to waste his own time rewriting it his own way and he was the one that still got it wrong. Everyone would have been better off if he had taken the sensible path of "read bug report", "confirm bug", "test patch", "commit patch" instead of arguing and rewriting things out of spite.