https://bugzilla.kernel.org/show_bug.cgi?id=43292

--- Comment #16 from Jan Kara <jack@suse.cz>  2012-06-01 16:26:59 ---

Grumble, it's the same, just for a different allocation group:
253,2    0     6514    25.375884394   744  Q   R 707526656 + 8 [nfsd]
  9,0    0     5451    25.375887257   744  A   R 749576576 + 8 <- (253,2)
707526
656
  9,0    0     5452    25.375888584   744  Q   R 749576576 + 8 [nfsd]
  8,16   0    17311    25.375891308   744  A   R 749576639 + 8 <- (8,17)
7495765
76
  8,16   0    17312    25.375892077   744  Q   R 749576639 + 8 [nfsd]
  8,16   0    17313    25.375894381   744  G   R 749576639 + 8 [nfsd]
  8,16   0    17314    25.375895918   744  P   N [nfsd]
  8,16   0    17315    25.375898572   744  I   R 749576639 + 8 [nfsd]
  8,16   0    17316    25.375900038   744  U   N [nfsd] 1
  8,16   0    17317    25.375901715   744  D   R 749576639 + 8 [nfsd]

Let's try the following:
for (( i = 0; i < 25293; i++ )); do
  dd if=/dev/dm-2 of=/dev/null bs=4k count=1 skip=$(($i*32768))
done

Again run it before and after the hang. If it finishes even after the hang,
check whether nfsd is still hung.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.