The problem with memory not being freed up might be bug 130157. It's particularly bad on Linux systems that do not have an mmap/munmap based allocator, but only a brk based one. The problem isn't that memory isn't being freed, but freed memory cannot be returned back to the operating system due to excessive memory fragmentation.
You really need to give a detailed set of steps to perform to see the problem. I leave Firefox open, going to lots of sites and having lots of tabs open, to sites that use JavaScript, Flash, and Java, and I don't see any obvious huge problem. After several days I can see some leaks from the leak-gauge script, and I have reported several memory leaks. However, even in the worst circumstances it seems like I would be able to use Firefox for many weeks at a time without having any serious memory use problems, even with only 1 GB of RAM.
That advice is not coming from "Mozilla." It's coming from a fellow user who used to have memory problems and is trying to help those who are still having memory problems to fix them. If you do not wish to listen to their advice, just ignore it. However, I get the distinct impression that those of you who are still complaining have no interest in fixing the problem, but would rather just keep complaining about it.
The developers have been hunting for memory leaks and fixing them. However, they can fix only leaks they know about. If you feel that going hunting for memory leaks is not of any importance, then that is your decision. Don't complain that someone else isn't doing something about it when you aren't.
I've surfed with SeaMonkey every day for several days without closing it, and the memory use still hangs around 100 MB. If you see memory use go up to 300 MB and want the problem fixed, you should describe what you're doing to cause that so the bug can be investigated. Until then, all we can say is that we simply don't see the problem you're referring to.
I still don't see how you reach the conclusion that "Firefox definitely counts as a huge memory hog" without trying the same set of steps on other browsers too see how much memory they use. Would you say that IE or Opera are memory hogs if they also use up to 100 MB of memory?
I wonder why Firefox seems to use up so much memory for some people, whilst others get away with relatively little.
That's what we all wonder. Without these people giving a clue as to how to see the problem, there's no way it can be investigated. Thankfully, it's only a very small percentage of Firefox users seeing memory problems this severe.
The memory usage was about 170mb, while Firefox 1.5 is ~140mb. (These are still mostly the same tabs as with IE 7, and that used over 400mb). 30mb increase isn't a big deal for me, but I would like to know what causes such an increase, and if it's worth it.
If you're looking at the Mem Usage column in the Windows Task Manager, perhaps some of the memory Firefox 1.5 was using got swapped out to disk, making memory usage appear lower. Compare the VM Size instead.
You don't need to take the bad with the good. Just stay away from the bad extensions and use only the good ones. It's not all extensions that leak memory; it's just a handful as far as anyone can determine. But use just one extension with a bad memory leak, and don't be surprised to see memory use skyrocket.
No, the version didn't change. It looks like a VeriSign time stamp at the end of the file and a checksum at the beginning of the file are all that changed. Firefox 2 RC3 identified itself as Firefox 2.0 because it was a true release candidate. There was no s/2 RC3/2.0/g needed.
If they discover a last minute bug in the build, it might end up not being the final release. That's what's meant by a "bad build," not some dipshit stupidly uploading the wrong build by mistake. The same thing could not happen to an update, because it has to be officially released before an update will occur. Once it's officially released, they would have to release another release with a different version number.
The product with the fewest advisories is not necessarily the safest to use. It could be that black hats have discovered a serious vulnerability in Opera and are exploiting it right now. And besides, you can't necessarily predict the future from the past, otherwise making predictions would be trivial.
I suspect much of ff's balkiness comes from IE specific coding on sites. When those sites change, maybe that will go away. In the meantime, it happens more than I (and a lot of others) like.
The results (Mem Usage/Peak Mem Usage/VM Size):
IE 7: 153/162/130
Fx 2: 96/113/85
It looks like browser users who don't like memory leaks should start complaining about IE 7. Actually, IE 7 didn't even finish the test; the Flash plug-in crashed before it was done. Otherwise, memory use probably would have climbed even higher.
Slashdot Mirror
The problem with memory not being freed up might be bug 130157. It's particularly bad on Linux systems that do not have an mmap/munmap based allocator, but only a brk based one. The problem isn't that memory isn't being freed, but freed memory cannot be returned back to the operating system due to excessive memory fragmentation.
You really need to give a detailed set of steps to perform to see the problem. I leave Firefox open, going to lots of sites and having lots of tabs open, to sites that use JavaScript, Flash, and Java, and I don't see any obvious huge problem. After several days I can see some leaks from the leak-gauge script, and I have reported several memory leaks. However, even in the worst circumstances it seems like I would be able to use Firefox for many weeks at a time without having any serious memory use problems, even with only 1 GB of RAM.
That advice is not coming from "Mozilla." It's coming from a fellow user who used to have memory problems and is trying to help those who are still having memory problems to fix them. If you do not wish to listen to their advice, just ignore it. However, I get the distinct impression that those of you who are still complaining have no interest in fixing the problem, but would rather just keep complaining about it.
You might want to reinstall your operating system. There have been several clues recently that third-party software can cause Firefox to hog memory.
If you're still convinced that it's a problem with Firefox, help to narrow down what the problem is so it can be investigated.
The developers have been hunting for memory leaks and fixing them. However, they can fix only leaks they know about. If you feel that going hunting for memory leaks is not of any importance, then that is your decision. Don't complain that someone else isn't doing something about it when you aren't.
I've surfed with SeaMonkey every day for several days without closing it, and the memory use still hangs around 100 MB. If you see memory use go up to 300 MB and want the problem fixed, you should describe what you're doing to cause that so the bug can be investigated. Until then, all we can say is that we simply don't see the problem you're referring to.
No, they just want to complain about them rather than help track them down.
If you do want to help track down memory leaks, it might be good to start a discussion on MozillaZine.
That's basically the idea they're testing on the German Wikipedia.
If that's true, you should file a bug report so the problem can be fixed.
I still don't see how you reach the conclusion that "Firefox definitely counts as a huge memory hog" without trying the same set of steps on other browsers too see how much memory they use. Would you say that IE or Opera are memory hogs if they also use up to 100 MB of memory?
Making the changes you mention should not affect memory usage, because those are the default settings. You can try setting browser.cache.memory.capacity to a low number like 4096 to save memory.
You don't need to take the bad with the good. Just stay away from the bad extensions and use only the good ones. It's not all extensions that leak memory; it's just a handful as far as anyone can determine. But use just one extension with a bad memory leak, and don't be surprised to see memory use skyrocket.
No, the version didn't change. It looks like a VeriSign time stamp at the end of the file and a checksum at the beginning of the file are all that changed. Firefox 2 RC3 identified itself as Firefox 2.0 because it was a true release candidate. There was no s/2 RC3/2.0/g needed.
If they discover a last minute bug in the build, it might end up not being the final release. That's what's meant by a "bad build," not some dipshit stupidly uploading the wrong build by mistake. The same thing could not happen to an update, because it has to be officially released before an update will occur. Once it's officially released, they would have to release another release with a different version number.
Firefox also lets you decide how much memory is used for caching. The problem is that all images on all currently displayed pages are stored uncompressed in the memory cache, even if the storage exceeds the maximum size you've set. It's not a memory leak, so in practice the memory usage is a problem only when you're displaying pages with lots of large images, but it can cause hundreds of megabytes of memory usage on certain pages.
Good news... There are several reports that Firefox 2 uses less memory than IE 7. Only a small percentage of users ever had problems with memory usage to begin with.
The product with the fewest advisories is not necessarily the safest to use. It could be that black hats have discovered a serious vulnerability in Opera and are exploiting it right now. And besides, you can't necessarily predict the future from the past, otherwise making predictions would be trivial.
The results (Mem Usage/Peak Mem Usage/VM Size):
IE 7: 153/162/130
Fx 2: 96/113/85
It looks like browser users who don't like memory leaks should start complaining about IE 7. Actually, IE 7 didn't even finish the test; the Flash plug-in crashed before it was done. Otherwise, memory use probably would have climbed even higher.