No, plugin writers are not being blamed for all the memory issues. At any point in time, the total memory used by Firefox is a combination of necessary memory use, memory used for caching, memory fragmentation, memory leaks in Firefox, plugin memory use, extension memory use. Pointing to any one of them and saying it's the sole problem, or even the main problem, is simplistic. Certainly memory leaks in extensions are the reason for a significant percentage of memory complaints.
I don't see that memory usage remains a problem for most users. It's just the vocal few who are having memory problems. The main problem is that these users assume this is part of the "normal" experience of using Firefox, so they complain that every user must also be seeing the same thing. They take no steps to fix or report their problems, as they consider the problem to be "well-known" and think developers must be idiots for not being able to see it.
If you're still having serious problems with Firefox, try creating a new profile and installing the Firefox 3 Beta. If you still have problems, discuss them on the MozillaZine Builds forum. If the problems do not get resolved, just switch to another browser. It's not normal to experience serious problems when browsing, so I don't see why anyone accepts it as part of the "normal" experience.
I agree that the damage to Firefox's reputation is already done. I've found that no matter how many reports come out that Firefox doesn't have a severe and obvious memory problem, the few reports that show a problem are the ones that become popular. If any of them just included instructions to reproduce the problem on other computers, those reports would be productive. Somehow, they always seem to leave that part out.
I'm sure Firefox developers don't want to hear you rant. However, they do want you to clearly and patiently explain what the problem is. Remember, developers are people, too, and will likely ignore you or get argumentative if you start complaining too vigorously. Why not file a polite bug report?
A lot of problems with keeping Firefox open longer than 24 hours can be attributed to heap fragmentation.
No, fragmentation is not causing people to have to restart Firefox. Fragmentation looks like a memory leak to people who expect memory use to go down to a previous level after closing a tab. Fragmentation also looks like a memory leak to people who point out that Firefox uses less memory after they close Firefox and restore their session. It's not the problem with Firefox gobbling up hundreds of megabytes for no reason that some are reporting. Those problems sound like severe memory leaks that are going unreported, because users think that it's a "well-known problem" that everyone is experiencing.
Those of us who leave Firefox running for days at a time have problems. Firefox consumes GIGABYTES of memory in short order for me, and yes, I see this as a major programming fault.
If you're seeing a problem like that, please file a bug report. When I run Firefox for a week, it uses only about 200 MB. Please reports problems so they can be fixed.
If Firefox is taking more memory than it should, it's a problem and you should file a bug report. The problem is knowing how much it "should" take. One guideline is looking at the memory usage of other browsers. If you can make Firefox use much more memory than other browsers, you can consider it a problem and file a bug report. Please include the steps you used to make Firefox use more memory than it should.
No, there's no flamewar. No one has ever called the memory leak a feature. David Baron explained that some increased memory usage was due to a new cache introduced in Firefox 1.5. Many people have mistaken caching for a memory leak. Many people have also mistaken fragmentation for a memory leak. There are those who violently protest that these are merely excuses and that developers are denying that Firefox leaks memory. I suppose you could say it's half of a flamewar. Yes, hopefully that is FINALLY over now.
Open "about:blank" repeatedly and watch the memory footprint rise and rise.
It doesn't for me. Don't assume that others can see the same problems you can; you need to ask if others can reproduce the problem. Perhaps you could try creating a new profile and see if that problem goes away. If not, head on over to the MozillaZine forums and discuss the problem.
It depends on the "problem" you are referring to. If you mean the problem of Firefox using hundreds of megabytes for no apparent reason, no, fragmentation is not the majority of the problem. Fragmentation will cause perhaps tens of megabytes to be wasted, not hundreds. In the example above, 14 MB ("a whopping 13,881,800 bytes") is wasted due to fragmentation.
The fragmentation is the problem people point out when they say that the memory leak is very easy to see. They say to look at memory usage, open a link in a new tab, close the tab, and observe that the memory does not go all the way back down to the previous level. Yes, the majority of that problem is fragmentation.
Anyway, it doesn't matter whether the problem is a true leak or fragmentation. If you can demonstrate a sequence of steps that causes Firefox to use far more memory than other browsers, you have demonstrated a memory problem.
Could you give an example of a file that Firefox has a problem with that Opera and IE seem to handle just fine? If we can reproduce the problem, we can write a bug report and the problem can be fixed.
Yes, I agree with you 100%. A browser taking up 2 GB of memory and making the computer slow to a crawl is obviously a very serious problem. No one could possibly disagree with you. Please report it as a bug, along with how to see that outrageous memory use. There is no flamewar. There is no way that memory use is caused by normal fragmentation or caching, which together normally account for only several tens of megabytes of memory usage.
Yes, many users were mistaking fragmentation (and caching) for memory leaks. We've been making this exact statement for years in the MozillaZine forums. But there are actual memory leaks, also. You can't point the finger at any one cause. At any point in time, Firefox memory use is some combination of memory needed to display the open pages, various caches (not just the two people talk about all the time), fragmentation, and possibly memory leaks. The only news is that the developers are getting diminishing returns on fixing leaks, and are now turning to reducing fragmentation to reduce normal memory use.
Yes, it still has memory leaks. I'm sure all popular browsers have memory leaks. Memory leaks are like crash bugs. You can make crashes less and less common, but it would take nearly a miracle to produce a browser that never, ever crashes no matter what. Similarly, Mozilla developers have worked for years to reduce the leaking of memory, but you can't expect a release with absolutely no memory leaks whatsoever. Mozilla developers report "Our extensive testing shows an occasional leak here and there and we are working to fix those, but in general we aren't seeing many leaks anymore."
Did you not read my replies? I tried to reproduce the problems, but I cannot. This is exactly the way things have been for over a year. I cannot reproduce any of these memory problems people keep complaining about. I also see that others in the Firefox Bugs forum cannot reproduce the problems. Without being able to reproduce any problem, we can't file a bug report and the problem cannot be fixed.
Please follow my advice, which I will give one last time:
IF
YOU
SEE
A
SERIOUS
PROBLEM
IN
FIREFOX
FILE
A
BUG
REPORT
OR
DISCUSS
IT
IN
THE
FIREFOX
BUGS
FORUM
ON
MOZILLAZINE.
That way, the bug will get the attention it deserves. Assuming that anyone else can reproduce it.
Firefox caches up to eight pages by default. That comes out to an average of about 32 MB used for the page cache (aka bfcache). I don't think 250 MB is used for the cache. Perhaps that's the total amount of memory used for pages you're currently viewing, the many different caches, memory fragmentation, plugins, and memory leaks all together after an extended period of time. Without a memory leak report or some other hard evidence, we can't really point to any one culprit and say that it's eating up too much memory in your case.
Firefox seems to free memory and return it to the OS relatively well. It doesn't return all of it due to memory fragmentation, but does it do worse than "other pieces of software"? If so, someone should be able to easily demonstrate Firefox using a larger VM Size (on Windows) than other browsers when performing the same operations. Do you have a specific set of operations that will demonstrate Firefox not returning memory to the OS as well as other browsers? If so, post them and we can have some solid information on the problem.
The highest memory usage due to fragmentation he mentions is "The rest, a whopping 13,881,800 bytes, is in free blocks!" That's a grand total of 14 MB of RAM consumed by memory fragmentation. Will the people still complaining about severe memory leaks be satisfied with a mere 14 MB reduction in memory usage? If not, then they should not look to reducing memory fragmentation as a fix for their issues.
Of course he finds that "Windows Low Fragmentation Heap" can reduce memory fragmentation, but that "certainly means that early on you're going to be using a lot more memory. However, over time you may end up using less memory as you will be less fragmented and thus new allocations as you load new pages should be able to fill in to the free space more easily." So you may end up using less memory after quite a while, but early on you'll be using lots of memory. Hmmm, doesn't sound like a fix for a severe memory problem, does it?
Anyone who interprets Stuart Parmenter's blog comments as insight into "high memory usage" is quite simply off their rocker.
I agree. It can be very hard to find a set of steps to reproduce problems sometimes. On the other hand, if the problem were as serious and widespread as many posters here are saying, I'm sure someone could produce a set of steps to reproduce a problem. And if someone has done such a thing, then perhaps people still saying that this problem is going ignored should find said set of steps and make sure someone files a proper bug report. If not, I suspect the problem will continue to be ignored, as no one has reported it.
The default setting will store up to eight cache pages in RAM. That will use an average of about 32 MB. If you're seeing "800-1000MB of physical memory usage after a week of running" it sounds like you're suffering from a severe memory leak. When I run Firefox that long, it uses only about 200 MB of memory. Why not grab one of the many memory leak detectors available and help find the problem?
Slashdot Mirror
No, plugin writers are not being blamed for all the memory issues. At any point in time, the total memory used by Firefox is a combination of necessary memory use, memory used for caching, memory fragmentation, memory leaks in Firefox, plugin memory use, extension memory use. Pointing to any one of them and saying it's the sole problem, or even the main problem, is simplistic. Certainly memory leaks in extensions are the reason for a significant percentage of memory complaints.
I don't see that memory usage remains a problem for most users. It's just the vocal few who are having memory problems. The main problem is that these users assume this is part of the "normal" experience of using Firefox, so they complain that every user must also be seeing the same thing. They take no steps to fix or report their problems, as they consider the problem to be "well-known" and think developers must be idiots for not being able to see it.
If you're still having serious problems with Firefox, try creating a new profile and installing the Firefox 3 Beta. If you still have problems, discuss them on the MozillaZine Builds forum. If the problems do not get resolved, just switch to another browser. It's not normal to experience serious problems when browsing, so I don't see why anyone accepts it as part of the "normal" experience.
I agree that the damage to Firefox's reputation is already done. I've found that no matter how many reports come out that Firefox doesn't have a severe and obvious memory problem, the few reports that show a problem are the ones that become popular. If any of them just included instructions to reproduce the problem on other computers, those reports would be productive. Somehow, they always seem to leave that part out.
You can find the results of a more extensive memory test here.
Because users complained that Firefox was too slow after un-minimizing.
I'm sure Firefox developers don't want to hear you rant. However, they do want you to clearly and patiently explain what the problem is. Remember, developers are people, too, and will likely ignore you or get argumentative if you start complaining too vigorously. Why not file a polite bug report?
I cannot see memory footprint "rise and rise" no matter what I do. Firefox developers have been working on memory leak bugs for years. They have fixed about 100 in the past year alone. Memory leak bug reports are not at the bottom of the priority list. In fact, they've fixed so many memory leaks that "Our extensive testing shows an occasional leak here and there and we are working to fix those, but in general we aren't seeing many leaks anymore."
If Firefox is taking more memory than it should, it's a problem and you should file a bug report. The problem is knowing how much it "should" take. One guideline is looking at the memory usage of other browsers. If you can make Firefox use much more memory than other browsers, you can consider it a problem and file a bug report. Please include the steps you used to make Firefox use more memory than it should.
No, there's no flamewar. No one has ever called the memory leak a feature. David Baron explained that some increased memory usage was due to a new cache introduced in Firefox 1.5. Many people have mistaken caching for a memory leak. Many people have also mistaken fragmentation for a memory leak. There are those who violently protest that these are merely excuses and that developers are denying that Firefox leaks memory. I suppose you could say it's half of a flamewar. Yes, hopefully that is FINALLY over now.
It depends on the "problem" you are referring to. If you mean the problem of Firefox using hundreds of megabytes for no apparent reason, no, fragmentation is not the majority of the problem. Fragmentation will cause perhaps tens of megabytes to be wasted, not hundreds. In the example above, 14 MB ("a whopping 13,881,800 bytes") is wasted due to fragmentation.
The fragmentation is the problem people point out when they say that the memory leak is very easy to see. They say to look at memory usage, open a link in a new tab, close the tab, and observe that the memory does not go all the way back down to the previous level. Yes, the majority of that problem is fragmentation.
Anyway, it doesn't matter whether the problem is a true leak or fragmentation. If you can demonstrate a sequence of steps that causes Firefox to use far more memory than other browsers, you have demonstrated a memory problem.
Could you give an example of a file that Firefox has a problem with that Opera and IE seem to handle just fine? If we can reproduce the problem, we can write a bug report and the problem can be fixed.
Yes, I agree with you 100%. A browser taking up 2 GB of memory and making the computer slow to a crawl is obviously a very serious problem. No one could possibly disagree with you. Please report it as a bug, along with how to see that outrageous memory use. There is no flamewar. There is no way that memory use is caused by normal fragmentation or caching, which together normally account for only several tens of megabytes of memory usage.
Yes, many users were mistaking fragmentation (and caching) for memory leaks. We've been making this exact statement for years in the MozillaZine forums. But there are actual memory leaks, also. You can't point the finger at any one cause. At any point in time, Firefox memory use is some combination of memory needed to display the open pages, various caches (not just the two people talk about all the time), fragmentation, and possibly memory leaks. The only news is that the developers are getting diminishing returns on fixing leaks, and are now turning to reducing fragmentation to reduce normal memory use.
Yes, it still has memory leaks. I'm sure all popular browsers have memory leaks. Memory leaks are like crash bugs. You can make crashes less and less common, but it would take nearly a miracle to produce a browser that never, ever crashes no matter what. Similarly, Mozilla developers have worked for years to reduce the leaking of memory, but you can't expect a release with absolutely no memory leaks whatsoever. Mozilla developers report "Our extensive testing shows an occasional leak here and there and we are working to fix those, but in general we aren't seeing many leaks anymore."
What flamewar? For nearly two years, Mozilla developers have asked users to file good memory leak bug reports and have even supplied tools for doing so. If you're still having problems, simply report them and they can be fixed. You can report any bugs in Firefox 3 beta directly to Bugzilla, or discuss them in the MozillaZine Firefox Builds forum first.
Did you not read my replies? I tried to reproduce the problems, but I cannot. This is exactly the way things have been for over a year. I cannot reproduce any of these memory problems people keep complaining about. I also see that others in the Firefox Bugs forum cannot reproduce the problems. Without being able to reproduce any problem, we can't file a bug report and the problem cannot be fixed.
Please follow my advice, which I will give one last time:
IF
YOU
SEE
A
SERIOUS
PROBLEM
IN
FIREFOX
FILE
A
BUG
REPORT
OR
DISCUSS
IT
IN
THE
FIREFOX
BUGS
FORUM
ON
MOZILLAZINE.
That way, the bug will get the attention it deserves. Assuming that anyone else can reproduce it.
Firefox caches up to eight pages by default. That comes out to an average of about 32 MB used for the page cache (aka bfcache). I don't think 250 MB is used for the cache. Perhaps that's the total amount of memory used for pages you're currently viewing, the many different caches, memory fragmentation, plugins, and memory leaks all together after an extended period of time. Without a memory leak report or some other hard evidence, we can't really point to any one culprit and say that it's eating up too much memory in your case.
Firefox seems to free memory and return it to the OS relatively well. It doesn't return all of it due to memory fragmentation, but does it do worse than "other pieces of software"? If so, someone should be able to easily demonstrate Firefox using a larger VM Size (on Windows) than other browsers when performing the same operations. Do you have a specific set of operations that will demonstrate Firefox not returning memory to the OS as well as other browsers? If so, post them and we can have some solid information on the problem.
The highest memory usage due to fragmentation he mentions is "The rest, a whopping 13,881,800 bytes, is in free blocks!" That's a grand total of 14 MB of RAM consumed by memory fragmentation. Will the people still complaining about severe memory leaks be satisfied with a mere 14 MB reduction in memory usage? If not, then they should not look to reducing memory fragmentation as a fix for their issues.
Of course he finds that "Windows Low Fragmentation Heap" can reduce memory fragmentation, but that "certainly means that early on you're going to be using a lot more memory. However, over time you may end up using less memory as you will be less fragmented and thus new allocations as you load new pages should be able to fill in to the free space more easily." So you may end up using less memory after quite a while, but early on you'll be using lots of memory. Hmmm, doesn't sound like a fix for a severe memory problem, does it?
Anyone who interprets Stuart Parmenter's blog comments as insight into "high memory usage" is quite simply off their rocker.
I agree. It can be very hard to find a set of steps to reproduce problems sometimes. On the other hand, if the problem were as serious and widespread as many posters here are saying, I'm sure someone could produce a set of steps to reproduce a problem. And if someone has done such a thing, then perhaps people still saying that this problem is going ignored should find said set of steps and make sure someone files a proper bug report. If not, I suspect the problem will continue to be ignored, as no one has reported it.
The default setting will store up to eight cache pages in RAM. That will use an average of about 32 MB. If you're seeing "800-1000MB of physical memory usage after a week of running" it sounds like you're suffering from a severe memory leak. When I run Firefox that long, it uses only about 200 MB of memory. Why not grab one of the many memory leak detectors available and help find the problem?