Update:

I logged into my system using another laptop and ran straces on xorg.

I saw tons of loops of these calls during the freezes:

10:19:35.422974 writev(54, 
[{"\1\0\333\r\6\0\0\0\312\7\3\0\360\0\32\0\34\10\3\0h\10\35"..., 56}], 1) = 56
10:19:35.424835 clock_gettime(CLOCK_MONOTONIC, {44940, 658485608}) = 0
10:19:35.424961 setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 0}}, 
NULL) = 0
10:19:35.425092 select(256, [1 3 5 12 16 17 18 19 30 35 46 47 48 49 50 51 52 53 
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 74 76 77 78], NULL, NULL, 
{50, 772000}) = 1 (in [54], left {50, 772000})
10:19:35.425335 setitimer(ITIMER_REAL, {it_interval={0, 20000}, it_value={0, 
20000}}, NULL) = 0
10:19:35.425464 clock_gettime(CLOCK_MONOTONIC, {44940, 659106144}) = 0
10:19:35.425581 read(54, "\232\n\2\0\204\343,\1", 4096) = 8
10:19:35.425713 read(54, 0xb5e7ca8, 4096) = -1 EAGAIN (Resource temporarily 
unavailable)


% time     seconds  usecs/call     calls    errors syscall
------ ----------- ----------- --------- --------- ----------------
 74.81    0.019764           1     17968        47 select
  5.83    0.001541           5       294           ioctl
  5.53    0.001461           0     11397      1355 read
  5.39    0.001423         109        13           munmap
  3.75    0.000990           0      4206           gettimeofday
  1.74    0.000459           0      3140           setitimer
  1.08    0.000285           0      6555           clock_gettime
  0.71    0.000187           4        47           poll
  0.42    0.000110           8        13           mmap2
  0.40    0.000107           0     10728           write
  0.27    0.000072           0      1787           writev
  0.08    0.000021           0      5834        62 sigreturn
  0.00    0.000000           0      3210           rt_sigprocmask
------ ----------- ----------- --------- --------- ----------------
100.00    0.026420                 65192      1464 total

So it appeared to me that something was unavailable to the system.
Debugging X.org gave me a loop inside the nvidia_drv module.

I was already at the current version by Ubuntu, 177.80.  So checked
nVidia's site for their beta driver 180.06.  Figured trying it wouldn't
hurt.  To my surprise,  180.06 fixed my problems.  I now can load that
website in fraction of a second.  Tab switching works as expected also.
CPU Usage is back below 10% now also.

But be aware, this fix it for me and is going to be unsupported by both
Ubuntu and nVidia (beta driver).

-- 
xorg 10-20% cpu usage after intrepid upgrade, general system sluggishness
https://bugs.launchpad.net/bugs/294972
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to