It's not necessarily caching the content of pages on closed tabs. I'm not exactly sure how the bfcache works. However, there are several other caches, such as "the image cache (keeps images in memory to help load them faster), font cache, textrun cache (short lived, but used to cache computed glyph indicies and metrics and such), etc." As far as I can tell, most plugins also cache themselves so they don't need to start from scratch every time you go to a page that uses the plugin.
Now I'm still not sure what "memory and performance issues" you're referring to. I'm not able to reproduce any, even though I'm trying hard. How would I see one of these memory and performance issues you speak of? If I can reproduce one, no matter how trivial, I will be sure to file a proper bug report so the problem can be fixed. Can you get Firefox to use significantly more CPU or memory than another browser? If so, please explain in detail how.
What do you mean, fix the caching? Is there a problem with the way Firefox is caching? If so, please explain what it is so the problem can be fixed.
The fragmentation is currently being looked into. I'm not sure how much of an impact that will have. If you're seeing truly serious memory problems, I doubt reducing fragmentation will help significantly. It's a way of reducing memory usage, but only typical usage. If there's a serious bug you're experiencing, reducing fragmentation will not compensate for that much.
Firefox is infamous for its memory bugs and performance degradation. A lot of people consider this Firefox's main problem, and even its only important problem.
And yet, no one can seem to explain how to see the problem. It sounds like a rumor to me. Could you point me to one shred of real evidence, hard data, that Firefox has any kind of serious memory bug or performance problem? How would I see these infamous bugs?
People complain that when they close tabs that memory is not released. For years, I've said that's due to caching and fragmentation. That is what Stuart Parmenter is also saying. It is a problem with every browser, not just Firefox. Even with this memory fragmentation, I never, ever have any memory problems. It is only a problem for people who check Firefox's memory usage every few minutes and think they're seeing a memory leak when in fact they're merely seeing some fragmentation. That is the "trying to figure out why after loading a lot of pages much of your memory seems to disappear" he is referring to.
If you're really having any problem at all with the amount of memory Firefox is using, such as needing to restart Firefox due to your computer running out of memory, or Firefox using more memory than other browsers when performing the same operations, please do give a set of steps to reproduce the problem. Instead of posting it here, you'd be better off posting it in a Bugzilla bug report or on the Firefox Bugs forum at MozillaZine. As far as I can tell, no one has given the details of any problem like this in over a year. It is not a "well known problem" but a problem few people can see, and no one can seem to describe in any detail.
They didn't drop Thunderbird support at all. They set up a new corporation to develop Thunderbird and other email/communication technologies and set them up with US $3 million. All this talk about dropping Thunderbird is just the same sensationalism we see in many Mozilla stories.
I use Firefox every day. I have Firefox 2.0.0.9 at work and the latest nightly version of Firefox 3 at home. I never need to close it due to freeing up memory, even after a week of use every day. I hang out in the Firefox Bugs forum every day, and none of the regulars seems to have any problems with memory, either. Now and then, someone comes by and mentions a problem, but we can almost never reproduce any problem whatsoever. Usually it turns out that creating a new profile or other simple procedure fixes the problem immediately.
Perhaps Firefox eats up all the memory on your computer. In that case, either you have a problem on your computer to fix or you have a bug to report. Please do one of those things instead on continuing to claim that there's some sort of awful and obvious memory problem. If it's so obvious, why can nearly no one reproduce it?
Yep, that's pretty much what the MozillaZine Knowledge Base article on Firefox CPU usage says. If anyone can demonstrate a CPU use bug in Firefox, please do. Make sure you give steps we can follow so we can reproduce the problem. If we cannot, there is little hope of troubleshooting and fixing the problem.
We've said for years (just check Bugzilla and the Firefox Bugs forum) that memory fragmentation is one reason memory use does not go down as much as users might expect when they close tabs. As far as I can tell, all browsers suffer from memory fragmentation. If it were a big problem in Firefox (compared to other browsers), we'd still be able to see the Firefox process using much more memory than other browsers when performing the same actions. I cannot reproduce such a problem, so it looks to me like Firefox suffers from memory fragmentation about the same as the other browsers. Still, there doesn't seem to be any "memory gobbling bug". If you can demonstrate such a bug, please do. Just do it in Bugzilla or the Firefox Bugs forum, which are the proper places to discuss bugs. Discussing them on Slashdot does little good.
I'm looking at Mem Usage, Peak Mem Usage, and VM Size in the Windows Task Manager. I usually report the VM Size, as on Windows it seems to show the actual memory used. When pages are swapped to disk, Mem Usage can go way down but VM Size stays high. All the numbers I reported for VM Size here are very similar to the Mem Usage and Peak Mem Usage, because I'm not getting anywhere near the point where Firefox needs to swap memory out to disk, even with only 1 GB of RAM and opening and closing tabs like crazy.
After opening the browser and going around Google maps for several minutes, I got memory use up to 55 MB. After opening 20 Yahoo articles in 20 tabs, memory use went up to 94 MB. After closing all tabs but this one, memory use went down to 66 MB. After going to Slashdot and opening the pages for comments for all articles in current stories, memory usage went up to 148 MB. After closing all tabs but this one, memory use went down to 89 MB. After opening about 30 articles from the NY Times in different tabs, memory use went up to 118 MB. After closing all tabs but this one, memory use went back down to 85 MB. I'm using the latest nightly build of Firefox 3 on Windows.
Sorry, I don't think the problem is my attitude. I simply cannot reproduce the problem of Firefox using hundreds of megabytes for no reason, no matter how hard I try. All I see is memory use hanging around 100 MB, just like I see in all browsers. That's due to memory fragmentation and caching, which I see in all browsers. Believe me, if I could see a problem, I'd be right there reporting it. Why would I not want a serious problem fixed? If you think the problem is my attitude, feel free to discuss the problem on the Firefox Bugs forum at MozillaZine. Certainly there can't be such a widespread conspiracy that we're all going to deny a serious, obvious problem?
Yes, I hang out at the Firefox Bug forum on MozillaZine every day, just waiting for someone to come by and point out a bug. People rarely do. The problems discussed there rarely turn out to be bugs in Firefox. They are typically fixed by uninstalling a buggy extension, creating a new profile, deleting Windows prefetch files, updating hardware drivers, or some other simple operation. As I've said many times before, if you're still seeing problems, file a bug report or discuss it in the Firefox Bugs forum so it can be fixed.
I got up to 272 MB after opening up 20 pages in 20 tabs, and after closing all tabs but one memory usage went back down to 131 MB. I opened up 30 different pages in 30 tabs and memory use went up to 205 MB, and after closing all tabs but one memory use went back down to 136 MB. It seems like memory use is going down after closing tabs to me. You may want to discuss the matter further in the Firefox Bugs forum on MozillaZine to see if anyone else can see the problem. Like I said, I have not been able to reproduce any memory problem described to me in over a year. I should note that I'm using the latest nightly build of Firefox 3 on Windows.
Close firefox and re-open (with the same session), and FireFox will drop to between 100-200MB.
Yes, of course, you're removing all the cached content and all the memory fragmentation. It's like completely clearing every single cache and completely defragging memory. Of course memory dropped. That's not necessarily a memory leak. It's memory use caused by fragmentation and caching. If it were a memory leak, memory use would keep rising without limit, or a memory leak detector would report a leak.
I've never seen it, and nearly everyone I see on the Firefox Bugs forum has never seen it. It sounds pretty rare to me. There may be some problem, but few people seem to ever see it, and no one can describe it in enough detail to file a proper bug report. Without a bug report, it won't be fixed. Sorry you see the problem, but no amount of continuing to state that you see the problem will be productive. We need details of the problem, such as a set of steps to reproduce it, to track it down and fix it. If no one ever reports this information, it cannot be fixed.
No, I'm not kidding. Not one person has stopped by the Firefox Bugs forum and pointed out any problem like you are describing. Why not do it? You might find that after you give a few details of your problem, someone can submit a bug report and the problem can be fixed. If no one does, however, do not be surprised if the problem is not fixed.
I can see people complaining on Slashdot and a few other places, yes. I hang out on the Firefox Bugs forum, and nearly no one there sees any problem at all. No one comes by and points out any problem. Come to think of it, no one here is giving any details of any problem. So what problem are you complaining about, specifically? Could you give a set of steps with which I could reproduce the problem? Could you give any details at all, whatsoever?
No, developers have not pretended that memory leaks aren't there. There are plenty that are easily reproducible, and about 100 memory leaks have been fixed in the past year. In fact, they can find so few leaks now that they've needed to look at other ways of reducing memory use. Please stop fabricating lies about memory problems in Firefox. If you want to point out a specific "memory gobbling" or CPU hogging bug, fine, let's have it so a bug report can be filed and the problem fixed.
I didn't say there's no memory gobbling bug. I said there's no sign of one that I can see. Would you like to point out the bug to me so I can see it? How would I reproduce the problem? If you do, I would be happy to file a bug report. As it is, no one has been able to point one out to me for over a year. It must be quite rare and elusive if no one can demonstrate it out of over 120 million users.
Sorry, I have never seen such a bug, and the vast majority of people I talk to has never seen it, either. I have used Firefox for up to a week, and memory use goes up to only 200 MB. It doesn't go up when nothing is going on. Anyway, the set of steps to reproduce needs to be, guess what, reproducible. I can't reproduce the random sites you visit. Feel free to discuss the bug in the Firefox Bugs forum at MozillaZine and see if anyone else is seeing a problem like that caused by a bug in Firefox.
Things aren't fine just for me. I see very few people complaining about Firefox's memory usage. When I ask people who say they are experiencing memory problems how I can reproduce them, they cannot seem to find any demonstration of a memory problem at all. So, just what is this horrible memory problem people keep vaguely referring to? It seems to be quite rare and elusive. I wish developers would fix the more noticeable bugs, ones that can easily be demonstrated and are experienced by many.
Sorry, I don't see anyone else complaining of the same problem. It must be quite rare. If you're one of the very few people that experience it, you'll need to give lots details so the problem can be found and fixed. You can start by finding the regression window if it's recently become worse. Include that information in your bug report.
I'm not hiding my head in the sand. I'm asking which important memory leak bugs you think need fixing. Why don't you simply say which ones specifically you think need fixing rather than try to start an argument? How would a Mac user reproduce the memory problems you see? If you can explain, someone can file a bug report and the problem can be fixed. On the other hand, if you keep trying to fight about it nothing productive will result.
The Firefox CPU hogging and memory gobbling bug would take some serious troubleshooting to find, and no one wants to do the work, apparently.
First, the Firefox CPU bug you've been complaining about (Firefox consumers lots of CPU after the computer wakes up from standby or hibernate) was fixed in Firefox 2.0.0.8. If you're still having any problems with the latest release of Firefox, let developers know by filing a proper bug report, including steps to reproduce the problem.
Second, there is no sign of any "memory gobbling bug" that I can see, just a few little leaks here and there and some memory fragmentation. If you're still having any problems with the latest release of Firefox, let developers know by filing a proper bug report, including steps to reproduce the problem.
Slashdot Mirror
It's not necessarily caching the content of pages on closed tabs. I'm not exactly sure how the bfcache works. However, there are several other caches, such as "the image cache (keeps images in memory to help load them faster), font cache, textrun cache (short lived, but used to cache computed glyph indicies and metrics and such), etc." As far as I can tell, most plugins also cache themselves so they don't need to start from scratch every time you go to a page that uses the plugin.
Now I'm still not sure what "memory and performance issues" you're referring to. I'm not able to reproduce any, even though I'm trying hard. How would I see one of these memory and performance issues you speak of? If I can reproduce one, no matter how trivial, I will be sure to file a proper bug report so the problem can be fixed. Can you get Firefox to use significantly more CPU or memory than another browser? If so, please explain in detail how.
What do you mean, fix the caching? Is there a problem with the way Firefox is caching? If so, please explain what it is so the problem can be fixed.
The fragmentation is currently being looked into. I'm not sure how much of an impact that will have. If you're seeing truly serious memory problems, I doubt reducing fragmentation will help significantly. It's a way of reducing memory usage, but only typical usage. If there's a serious bug you're experiencing, reducing fragmentation will not compensate for that much.
People complain that when they close tabs that memory is not released. For years, I've said that's due to caching and fragmentation. That is what Stuart Parmenter is also saying. It is a problem with every browser, not just Firefox. Even with this memory fragmentation, I never, ever have any memory problems. It is only a problem for people who check Firefox's memory usage every few minutes and think they're seeing a memory leak when in fact they're merely seeing some fragmentation. That is the "trying to figure out why after loading a lot of pages much of your memory seems to disappear" he is referring to.
If you're really having any problem at all with the amount of memory Firefox is using, such as needing to restart Firefox due to your computer running out of memory, or Firefox using more memory than other browsers when performing the same operations, please do give a set of steps to reproduce the problem. Instead of posting it here, you'd be better off posting it in a Bugzilla bug report or on the Firefox Bugs forum at MozillaZine. As far as I can tell, no one has given the details of any problem like this in over a year. It is not a "well known problem" but a problem few people can see, and no one can seem to describe in any detail.
They didn't drop Thunderbird support at all. They set up a new corporation to develop Thunderbird and other email/communication technologies and set them up with US $3 million. All this talk about dropping Thunderbird is just the same sensationalism we see in many Mozilla stories.
I use Firefox every day. I have Firefox 2.0.0.9 at work and the latest nightly version of Firefox 3 at home. I never need to close it due to freeing up memory, even after a week of use every day. I hang out in the Firefox Bugs forum every day, and none of the regulars seems to have any problems with memory, either. Now and then, someone comes by and mentions a problem, but we can almost never reproduce any problem whatsoever. Usually it turns out that creating a new profile or other simple procedure fixes the problem immediately.
Perhaps Firefox eats up all the memory on your computer. In that case, either you have a problem on your computer to fix or you have a bug to report. Please do one of those things instead on continuing to claim that there's some sort of awful and obvious memory problem. If it's so obvious, why can nearly no one reproduce it?
Yep, that's pretty much what the MozillaZine Knowledge Base article on Firefox CPU usage says. If anyone can demonstrate a CPU use bug in Firefox, please do. Make sure you give steps we can follow so we can reproduce the problem. If we cannot, there is little hope of troubleshooting and fixing the problem.
We've said for years (just check Bugzilla and the Firefox Bugs forum) that memory fragmentation is one reason memory use does not go down as much as users might expect when they close tabs. As far as I can tell, all browsers suffer from memory fragmentation. If it were a big problem in Firefox (compared to other browsers), we'd still be able to see the Firefox process using much more memory than other browsers when performing the same actions. I cannot reproduce such a problem, so it looks to me like Firefox suffers from memory fragmentation about the same as the other browsers. Still, there doesn't seem to be any "memory gobbling bug". If you can demonstrate such a bug, please do. Just do it in Bugzilla or the Firefox Bugs forum, which are the proper places to discuss bugs. Discussing them on Slashdot does little good.
I'm looking at Mem Usage, Peak Mem Usage, and VM Size in the Windows Task Manager. I usually report the VM Size, as on Windows it seems to show the actual memory used. When pages are swapped to disk, Mem Usage can go way down but VM Size stays high. All the numbers I reported for VM Size here are very similar to the Mem Usage and Peak Mem Usage, because I'm not getting anywhere near the point where Firefox needs to swap memory out to disk, even with only 1 GB of RAM and opening and closing tabs like crazy.
After opening the browser and going around Google maps for several minutes, I got memory use up to 55 MB. After opening 20 Yahoo articles in 20 tabs, memory use went up to 94 MB. After closing all tabs but this one, memory use went down to 66 MB. After going to Slashdot and opening the pages for comments for all articles in current stories, memory usage went up to 148 MB. After closing all tabs but this one, memory use went down to 89 MB. After opening about 30 articles from the NY Times in different tabs, memory use went up to 118 MB. After closing all tabs but this one, memory use went back down to 85 MB. I'm using the latest nightly build of Firefox 3 on Windows.
Sorry, I don't think the problem is my attitude. I simply cannot reproduce the problem of Firefox using hundreds of megabytes for no reason, no matter how hard I try. All I see is memory use hanging around 100 MB, just like I see in all browsers. That's due to memory fragmentation and caching, which I see in all browsers. Believe me, if I could see a problem, I'd be right there reporting it. Why would I not want a serious problem fixed? If you think the problem is my attitude, feel free to discuss the problem on the Firefox Bugs forum at MozillaZine. Certainly there can't be such a widespread conspiracy that we're all going to deny a serious, obvious problem?
Yes, I hang out at the Firefox Bug forum on MozillaZine every day, just waiting for someone to come by and point out a bug. People rarely do. The problems discussed there rarely turn out to be bugs in Firefox. They are typically fixed by uninstalling a buggy extension, creating a new profile, deleting Windows prefetch files, updating hardware drivers, or some other simple operation. As I've said many times before, if you're still seeing problems, file a bug report or discuss it in the Firefox Bugs forum so it can be fixed.
I got up to 272 MB after opening up 20 pages in 20 tabs, and after closing all tabs but one memory usage went back down to 131 MB. I opened up 30 different pages in 30 tabs and memory use went up to 205 MB, and after closing all tabs but one memory use went back down to 136 MB. It seems like memory use is going down after closing tabs to me. You may want to discuss the matter further in the Firefox Bugs forum on MozillaZine to see if anyone else can see the problem. Like I said, I have not been able to reproduce any memory problem described to me in over a year. I should note that I'm using the latest nightly build of Firefox 3 on Windows.
Yes, of course, you're removing all the cached content and all the memory fragmentation. It's like completely clearing every single cache and completely defragging memory. Of course memory dropped. That's not necessarily a memory leak. It's memory use caused by fragmentation and caching. If it were a memory leak, memory use would keep rising without limit, or a memory leak detector would report a leak.I've never seen it, and nearly everyone I see on the Firefox Bugs forum has never seen it. It sounds pretty rare to me. There may be some problem, but few people seem to ever see it, and no one can describe it in enough detail to file a proper bug report. Without a bug report, it won't be fixed. Sorry you see the problem, but no amount of continuing to state that you see the problem will be productive. We need details of the problem, such as a set of steps to reproduce it, to track it down and fix it. If no one ever reports this information, it cannot be fixed.
No, I'm not kidding. Not one person has stopped by the Firefox Bugs forum and pointed out any problem like you are describing. Why not do it? You might find that after you give a few details of your problem, someone can submit a bug report and the problem can be fixed. If no one does, however, do not be surprised if the problem is not fixed.
I can see people complaining on Slashdot and a few other places, yes. I hang out on the Firefox Bugs forum, and nearly no one there sees any problem at all. No one comes by and points out any problem. Come to think of it, no one here is giving any details of any problem. So what problem are you complaining about, specifically? Could you give a set of steps with which I could reproduce the problem? Could you give any details at all, whatsoever?
No, developers have not pretended that memory leaks aren't there. There are plenty that are easily reproducible, and about 100 memory leaks have been fixed in the past year. In fact, they can find so few leaks now that they've needed to look at other ways of reducing memory use. Please stop fabricating lies about memory problems in Firefox. If you want to point out a specific "memory gobbling" or CPU hogging bug, fine, let's have it so a bug report can be filed and the problem fixed.
I didn't say there's no memory gobbling bug. I said there's no sign of one that I can see. Would you like to point out the bug to me so I can see it? How would I reproduce the problem? If you do, I would be happy to file a bug report. As it is, no one has been able to point one out to me for over a year. It must be quite rare and elusive if no one can demonstrate it out of over 120 million users.
Sorry, I have never seen such a bug, and the vast majority of people I talk to has never seen it, either. I have used Firefox for up to a week, and memory use goes up to only 200 MB. It doesn't go up when nothing is going on. Anyway, the set of steps to reproduce needs to be, guess what, reproducible. I can't reproduce the random sites you visit. Feel free to discuss the bug in the Firefox Bugs forum at MozillaZine and see if anyone else is seeing a problem like that caused by a bug in Firefox.
Things aren't fine just for me. I see very few people complaining about Firefox's memory usage. When I ask people who say they are experiencing memory problems how I can reproduce them, they cannot seem to find any demonstration of a memory problem at all. So, just what is this horrible memory problem people keep vaguely referring to? It seems to be quite rare and elusive. I wish developers would fix the more noticeable bugs, ones that can easily be demonstrated and are experienced by many.
Sorry, I don't see anyone else complaining of the same problem. It must be quite rare. If you're one of the very few people that experience it, you'll need to give lots details so the problem can be found and fixed. You can start by finding the regression window if it's recently become worse. Include that information in your bug report.
I'm not hiding my head in the sand. I'm asking which important memory leak bugs you think need fixing. Why don't you simply say which ones specifically you think need fixing rather than try to start an argument? How would a Mac user reproduce the memory problems you see? If you can explain, someone can file a bug report and the problem can be fixed. On the other hand, if you keep trying to fight about it nothing productive will result.
They're spending far, far more than $300,000 on their software developers. Do you really think they have just three developers?
BTW, what "important memory leak bugs" do you think need fixing? I'm seeing almost no signs of memory leaks. I'd rather they focus on bugs that I actually see, such as not remembering the scroll position or not importing IE favorites in the correct order.
Every dime they are spending on employees and infrastructure goes to OSS. Or do they produce or distribute any software that isn't open source?
First, the Firefox CPU bug you've been complaining about (Firefox consumers lots of CPU after the computer wakes up from standby or hibernate) was fixed in Firefox 2.0.0.8. If you're still having any problems with the latest release of Firefox, let developers know by filing a proper bug report, including steps to reproduce the problem.
Second, there is no sign of any "memory gobbling bug" that I can see, just a few little leaks here and there and some memory fragmentation. If you're still having any problems with the latest release of Firefox, let developers know by filing a proper bug report, including steps to reproduce the problem.
Search engines.
For placement of their search engines.
Read more here.