Page 5 of 12 FirstFirst 12345678910 ... LastLast
Results 81 to 100 of 221
  1. Stihl's Avatar
    Posts
    741 Posts
    Global Posts
    747 Global Posts
    #81  
    mama say mama say mamakusa!
  2. #82  
    how exactly did u do that through webosquickinstall could u explaine how exactly
  3. jhp
    jhp is offline
    jhp's Avatar
    Posts
    30 Posts
    #83  
    I just umounted /media/ram and played youtube, some local .mp3 and .mp4 files and streamed some shoutcast stations without any problems, and nothing showed up in /media/ram (which would now just be part of /media). I'm going to leave things like this until I see problems; if I don't see anything after a few days I'll comment it out of /etc/fstab.
  4. Stihl's Avatar
    Posts
    741 Posts
    Global Posts
    747 Global Posts
    #84  
    One of the best features of these two patches is the alleviation of the sort of 'web page limit' that the pre used to have. One of the reasons I've wanted a facebook application, is because it was difficult to have more than two or three web pages running serious content open at once.

    Open one page. Good. Open another. Read and article. Swipe to page 1. Good. Open that third (like Facebook). Check someone's page. Swipe to webpage 1. It is now a blank page that is reloading.

    This used to infuriate me as it sort of obviated the point of allowing multiple instances of the web browser if they would just autorefresh (since the content could not be stored in the ram).

    Now that does not seem to be an issue. While I have no doubt that one day I will actually have the desire to run a panoply of individual applications, my first order of business is going to be running the five or six web pages (simultaneously) that I usually check out while I browse on my desktop. This is huge.

    Thanks.
    Last edited by Stihl; 10/15/2009 at 09:43 PM.
  5. #85  
    Hello folks.

    As someone who has some experience with Linux and tmpfs, I'm trying to figure out HOW this could possibly work.

    Let me explain.

    tmpfs is a great RAMdisk implementation. The main reason is that tmpfs will only allocate pages of system memory when you actually STORE DATA in the filesystem.

    This means that you should be able to say 16M, 32M, even 256M and it shouldn't make any difference. As long as you don't actually put that much data there, the tmpfs filesystem SHOULD only use however much memory as the data in it occupies.

    I'm writing SHOULD here because there are so many people in this thread claiming that their Pre is much faster after this change. So either I'm wrong, or there is some massive placebo effect going on and affecting all of us.

    I'm going to do more research, comparing my Pre's performance with this patch and without. But call me skeptical; my experience tweaking tmpfs on desktop systems tells me that this patch should make NO difference whatsoever.
  6. #86  
    Quote Originally Posted by pinodb View Post
    how exactly did u do that through webosquickinstall could u explaine how exactly
    Here's quick simple tutorial of using .patch files in WebOSQuick Install:

    Step 1:


    Step 2:


    Step 3:
  7. diomark's Avatar
    Posts
    752 Posts
    Global Posts
    770 Global Posts
       #87  
    Quote Originally Posted by zorinlynx View Post
    Hello folks.

    As someone who has some experience with Linux and tmpfs, I'm trying to figure out HOW this could possibly work.

    Let me explain.

    tmpfs is a great RAMdisk implementation. The main reason is that tmpfs will only allocate pages of system memory when you actually STORE DATA in the filesystem.

    This means that you should be able to say 16M, 32M, even 256M and it shouldn't make any difference. As long as you don't actually put that much data there, the tmpfs filesystem SHOULD only use however much memory as the data in it occupies.

    I'm writing SHOULD here because there are so many people in this thread claiming that their Pre is much faster after this change. So either I'm wrong, or there is some massive placebo effect going on and affecting all of us.

    I'm going to do more research, comparing my Pre's performance with this patch and without. But call me skeptical; my experience tweaking tmpfs on desktop systems tells me that this patch should make NO difference whatsoever.
    I realized the same thing when I started playing with different /media/ram settings. I'm not sure why it'd make a difference, but it does. Memory for me was lower after a day or two of runtime when it was set to the defaults then it is now.

    I've been experimenting with this all night, and what I keep consistently seeing is that the swap gets full as you load more and more items, and doesn't necessarily go down. Sometimes, it gets to less then 64mb free (the threashold that we're re-setting down to 32mb in the first patch in the /etc/palm/luna.conf) - and stays there. Hince even though the system has memory, it doesn't allow new cards to be opened.

    I've been experimenting with some different parameters here in luna.conf - specifically, UseMemchute and EnableMemoryTracking - usememchute didn't make a difference, but I'm wondering if there might be a bug in the EnableMemoryTracking setting/program that takes advantage of it. As an experiment, I loaded my pre down (with every single app I could find plus a few web pages), then closed everything out. After having everything closed, the swap usage stayed high (less then 64mb free..) - and the free memory was ~10mb's or so, but there was 100mb free used in caching (so there was plenty of free memory..) - will write when I know more, but testing out w/ EnableMemoryTracking=false and FreeSwapThreshold=32 and getting fairly good results.

    The /etc/fstab might be a placebo effect after all (still testing this - to me, it still feels faster) - but setting FreeSwapThreshold=32 definately makes a difference.

    -mark
  8. #88  
    For the folks who have been waiting for this to show up in Preware, you should stop waiting for 2 very good reasons.

    1. Jason has posted such clear instructions on applying these patches through WOSQI that a badly trained monkey could follow the instructions. (Thank you Jason)

    2. I will not be submitting either of these to Preware at the present time. First of all, I only made the patch that edits /etc/fstab and don't have the right to be submitting the other patch to anybody. Second, the patch I did create requires a full reboot to function, and I'm not aware of a way to force this using a patch.

    The most important reason I won't be submitting either of them for Preware at the present time, is that there are still a lot of unanswered questions about how these patches may affect other aspects of the system, and I simply don't have a deep enough knowledge of webos to determine their long term usability.

    Right now, they appear to do no harm, while doing a lot of good. However, I just don't have the knowledge to put these somewhere people can install them without reading this thread. If another developer (i.e. a real developer instead of a hobbyist)feels confident enough in the function of these patches to submit my patch to preware, then by all means do so.
  9. jhp
    jhp is offline
    jhp's Avatar
    Posts
    30 Posts
    #89  
    Quote Originally Posted by diomark View Post
    I realized the same thing when I started playing with different /media/ram settings. I'm not sure why it'd make a difference, but it does. Memory for me was lower after a day or two of runtime when it was set to the defaults then it is now.

    I've been experimenting with this all night, and what I keep consistently seeing is that the swap gets full as you load more and more items, and doesn't necessarily go down. Sometimes, it gets to less then 64mb free (the threashold that we're re-setting down to 32mb in the first patch in the /etc/palm/luna.conf) - and stays there. Hince even though the system has memory, it doesn't allow new cards to be opened.

    I've been experimenting with some different parameters here in luna.conf - specifically, UseMemchute and EnableMemoryTracking - usememchute didn't make a difference, but I'm wondering if there might be a bug in the EnableMemoryTracking setting/program that takes advantage of it. As an experiment, I loaded my pre down (with every single app I could find plus a few web pages), then closed everything out. After having everything closed, the swap usage stayed high (less then 64mb free..) - and the free memory was ~10mb's or so, but there was 100mb free used in caching (so there was plenty of free memory..) - will write when I know more, but testing out w/ EnableMemoryTracking=false and FreeSwapThreshold=32 and getting fairly good results.

    The /etc/fstab might be a placebo effect after all (still testing this - to me, it still feels faster) - but setting FreeSwapThreshold=32 definately makes a difference.

    -mark
    I thought about this a bit and I agree that tmpfs will only use as much memory as is needed to satisfy allocated files; limiting it's size will limit the amount of swap that will be used if the filesystem space is swapped out, so reducing it's size should have some benefits as far as the number of cards that can be opened is concerned, but any benefits that are non-swap related are probably more perceived than actual..
    Last edited by jhp; 10/15/2009 at 10:37 PM. Reason: fix typo
  10. #90  
    just thought I would add. I can now run 19 unique apps at once (ie only ine browser). Including all the pim apps, classic both gps programs doc viewer podcasters and others. The system was snappy up until 12 apps about.

    also as a trained monkey I have completed the steps to get both paches on. Thanks to all of your efforts


    I was wondering does enable/disable dev mode perform the complete reboot needed?
    Last edited by windzilla; 10/15/2009 at 10:47 PM.
  11. #91  
    Quote Originally Posted by windzilla View Post
    I was wondering does enable/disable dev mode perform the complete reboot needed?
    It does, but unless you need to enable/dissable it for some reason, there's an easier way. Just press sym+orange+r to reboot your pre.

    If you happen to already be at a root terminal for some reason, you can also just type reboot and press enter.
  12. Beefy's Avatar
    Posts
    91 Posts
    Global Posts
    96 Global Posts
    #92  
    What an awesome find. thanks for your work on this, Mark.

    I'm going to check this out when I next have some free time to mess around with WOSQI.
  13. xsavior's Avatar
    Posts
    80 Posts
    Global Posts
    89 Global Posts
    #93  
    From direct comparison before and after the patch, bringing up the launcher and moving between my pages, there was a stutter as it was loading the second and third screens of icons (this was with Email and Messaging open and the 4x4 version 3 patch), and immediately after a full reboot. After the patches were installed and another full reboot, pulling up the same programs, (which definitely *felt* like they loaded faster) I had much less stutter (a noticeable 4 quick pauses, down to 1) when moving between the launcher pages. Take that for what it's worth. It's as close as I can come to an empirical measurement of the before and after.
  14. #94  
    Any negative side effects so far?
    Palm Vx -> Treo 600 -> Treo 700p -> Centro -> Pre (Launch Phone 06/06/09) -> AT&T Pre Plus with Sprint EVDO swap -> Samsung Epic 4G w/ Froyo
  15. Beefy's Avatar
    Posts
    91 Posts
    Global Posts
    96 Global Posts
    #95  
    So, question. What's the danger if I lower the /tmp allocation even further? Say, to 16mb? I want the maximum amount of ram allocated to the GUI and not to tmp/media, since I rarely run a large amount of apps at the same time to preserve battery life through the workday.
  16. pmc2010's Avatar
    Posts
    25 Posts
    Global Posts
    27 Global Posts
    #96  
    Not sure this has been answered or if i didnt see it posted, but if there is an OTA update do we need to remove these and if so Jason will your Repair Utility work to fix this before. If not how do we manual uninstall this patch.

    Thanks for the help Precentral Community
  17. pmc2010's Avatar
    Posts
    25 Posts
    Global Posts
    27 Global Posts
    #97  
    Just had to add. Phone is noticeably faster with these patches. thanks for all the hard work
  18. SirWill's Avatar
    Posts
    439 Posts
    Global Posts
    492 Global Posts
    #98  
    Quote Originally Posted by pmc2010 View Post
    Not sure this has been answered or if i didnt see it posted, but if there is an OTA update do we need to remove these and if so Jason will your Repair Utility work to fix this before. If not how do we manual uninstall this patch.

    Thanks for the help Precentral Community
    To remove the patch, in WOQI, go to device management and select patches. From there you can remove it along with other patches.
  19. Beefy's Avatar
    Posts
    91 Posts
    Global Posts
    96 Global Posts
    #99  
    thanks so much for the patches. just applied them and while I don't notice an appreciable difference in the actual app launcher (or if there is one, it's very slight), apps themselves launch MUCH faster.
  20. #100  
    i just ran into two problems since i installed these. it seems that when i run classic 2.0 and try to install the free apps my phone restarted. then i opened classic and it froze again then restarted. then today i was moving icons around and it froze then restarted again. just fyi
Page 5 of 12 FirstFirst 12345678910 ... LastLast

Posting Permissions