search for: cache_flusharray

...46e84 e386cc00 bbfb9f0b fffffff5 f8e7cf20 f8e6ac04 00000000 f8e7eae4 Feb 28 14:40:53 tempmailstore1a kernel: 00000000 f7dd9b50 f7dd9a00 e386cc00 f6acbf55 f7dd9a00 f8e6923c c19f31f4 Feb 28 14:40:53 tempmailstore1a kernel: Call Trace: Feb 28 14:40:53 tempmailstore1a kernel: [<c0146a64>] cache_flusharray+0x1a/0x9c Feb 28 14:40:53 tempmailstore1a kernel: [<c0146e84>] kfree+0x43/0x51 Feb 28 14:40:53 tempmailstore1a kernel: [<f8e6ac04>] cache_check+0x11c/0x126 [sunrpc] Feb 28 14:40:53 tempmailstore1a kernel: [<f8e6923c>] svcauth_unix_set_client+0x68/0xb5 [sunrpc] Feb 28 14:40:53 t...

..._pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kern...

..._pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kern...

...tack: 41e0d010 39f19000 09fa4540 41e0d010 0000003c 02140c5e 41e0d000 41f6f280 kernel: 41e0d000 09fa4540 41e0d010 00000202 0214102d 09fa4548 00000000 3f1bb808 kernel: 2674d1e0 42d02886 2674d1e0 41f47e00 3d5b2320 3d11ce8c 42d0291e 1a7b8880 kernel: Call Trace: kernel: [<02140c5e>] cache_flusharray+0x6e/0x9c kernel: [<0214102d>] kfree+0x43/0x51 kernel: [<42d02886>] free_rb_tree_fname+0x31/0x6c [ext3] kernel: [<42d0291e>] ext3_htree_free_dir_info+0x8/0x10 [ext3] kernel: [<42d02cd3>] ext3_release_dir+0xf/0x14 [ext3] kernel: [<021549ca>] __fput+0x55/0x100 kerne...

Hi Waiman, As promised; here is the paravirt stuff I did during the trip to BOS last week. All the !paravirt patches are more or less the same as before (the only real change is the copyright lines in the first patch). The paravirt stuff is 'simple' and KVM only -- the Xen code was a little more convoluted and I've no real way to test that but it should be stright fwd to make work.

Hi Waiman, As promised; here is the paravirt stuff I did during the trip to BOS last week. All the !paravirt patches are more or less the same as before (the only real change is the copyright lines in the first patch). The paravirt stuff is 'simple' and KVM only -- the Xen code was a little more convoluted and I've no real way to test that but it should be stright fwd to make work.

...-- free_pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kernel.kal...

...-- free_pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kernel.kal...

...-- free_pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kernel.kal...

...-- free_pcppages_bulk |--1.88%-- do_wp_page |--1.77%-- handle_pte_fault |--1.58%-- do_anonymous_page |--1.56%-- rmqueue_bulk.clone.0 |--1.35%-- copy_pte_range |--1.25%-- zap_pte_range |--1.13%-- cache_flusharray |--0.88%-- __pmd_alloc |--0.70%-- wake_up_new_task |--0.66%-- __pud_alloc |--0.59%-- ext4_discard_preallocations --6.53%-- [...] With the qspinlock patch, the perf profile at 1000 users was: 3.25% reaim [kernel.kal...