Bug 896196

+++ This bug was initially created as a clone of Bug #814410 +++

Description of problem:
After upgrade to KMail2, kmail consume almost all CPU, typically 60-95% of CPU. Example from top:

PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
2637 vjuranek  20   0  547m 373m  22m S 82.8  9.5  31:36.93 kmail


Version-Release number of selected component (if applicable):
KMail 4.8.1
kdepim-4.8.1-4.fc16.i686

How reproducible:
almost always


Steps to Reproduce:
1. run KMail
  
Actual results:
KMail consumes more 60% of CPU all the time.

Expected results:
KMail consumes less ~ 10% of CPU once started


Additional info:

--- Additional comment from Vojtech Juranek on 2012-04-19 16:25:18 EDT ---

Here is sample output from strace (when taken, according to progress window, kmail didn't fetch new mails or didn't do any other activity):

clock_gettime(CLOCK_MONOTONIC, {5996, 63683081}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
ioctl(10, FIONREAD, [20340])            = 0
read(10, "* 105682 FETCH (UID 105682 REV 0"..., 20340) = 20340
clock_gettime(CLOCK_MONOTONIC, {5996, 66585545}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=8, events=POLLIN}, {fd=7, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=5, events=POLLIN}, {fd=12, events=POLLIN}], 7, 44) = 1 ([{fd=10, revents=POLLIN}])
clock_gettime(CLOCK_MONOTONIC, {5996, 66733399}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
ioctl(10, FIONREAD, [22507])            = 0
read(10, "* 105598 FETCH (UID 105598 REV 0"..., 22507) = 22507
clock_gettime(CLOCK_MONOTONIC, {5996, 69964046}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=8, events=POLLIN}, {fd=7, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=5, events=POLLIN}, {fd=12, events=POLLIN}], 7, 41) = 1 ([{fd=10, revents=POLLIN}])
clock_gettime(CLOCK_MONOTONIC, {5996, 70108898}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
ioctl(10, FIONREAD, [22988])            = 0
read(10, "* 105566 FETCH (UID 105566 REV 0"..., 22988) = 22988
clock_gettime(CLOCK_MONOTONIC, {5996, 75723927}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=8, events=POLLIN}, {fd=7, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=5, events=POLLIN}, {fd=12, events=POLLIN}], 7, 35) = 1 ([{fd=10, revents=POLLIN}])
clock_gettime(CLOCK_MONOTONIC, {5996, 75860746}) = 0
read(7, 0x85525c8, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
ioctl(10, FIONREAD, [22574])            = 0
read(10, "* 105512 FETCH (UID 105512 REV 0"..., 22574) = 22574

--- Additional comment from Vojtech Juranek on 2012-04-23 04:53:34 EDT ---

My observation is that some folders (IMAP as well as local) show wheel icon instead of folder icon (means loading something?). If I click on given folder, the content almost immediately appears and wheel icon is replaced by folder icon. If I do it for all folders with wheel icon, CPU goes back to normal values.