svn commit: r308691 - in head/sys: cddl/compat/opensolaris/sys cddl/contrib/opensolaris/uts/common/fs/zfs fs/tmpfs kern vm

[Ruslan Bukin]

On Wed, Nov 16, 2016 at 11:53:55AM -0600, Alan Cox wrote:
> On 11/16/2016 10:59, Ruslan Bukin wrote:
> > On Wed, Nov 16, 2016 at 06:53:43PM +0200, Konstantin Belousov wrote:
> >> On Wed, Nov 16, 2016 at 01:37:18PM +0000, Ruslan Bukin wrote:
> >>> I have a panic with this on RISC-V. Any ideas ?
> >> How did you checked that the revision you replied to, makes the problem ?
> >> Note that the backtrace below is not reasonable.
> > I reverted this commit like that and rebuilt kernel:
> > git show 2fa36073055134deb2df39c7ca46264cfc313d77 | patch -p1 -R
> >
> > So the problem is reproducible on dual-core with 32mb mdroot.
> 
> This change amounted to dead code removal, so I'm not sure how it could
> have an effect.  There were only a couple places where the changes were
> other than mechanical in nature.  Also, that the number of cores matters
> is no less puzzling.
> 
> Can you send Kostik and me the output of "sysctl vm.stats.vm" from
> shortly after boot on the kernel with the patch reverted?
> 

Here is result with patch reverted:
# sysctl vm.stats.vm
vm.stats.vm.v_vm_faults: 1578
vm.stats.vm.v_io_faults: 213
vm.stats.vm.v_cow_faults: 135
vm.stats.vm.v_cow_optim: 0
vm.stats.vm.v_zfod: 360
vm.stats.vm.v_ozfod: 0
vm.stats.vm.v_swapin: 0
vm.stats.vm.v_swapout: 0
vm.stats.vm.v_swappgsin: 0
vm.stats.vm.v_swappgsout: 0
vm.stats.vm.v_vnodein: 0
vm.stats.vm.v_vnodeout: 0
vm.stats.vm.v_vnodepgsin: 0
vm.stats.vm.v_vnodepgsout: 0
vm.stats.vm.v_intrans: 0
vm.stats.vm.v_reactivated: 0
vm.stats.vm.v_pdwakeups: 0
vm.stats.vm.v_pdpages: 2
vm.stats.vm.v_pdshortfalls: 0
vm.stats.vm.v_tcached: 0
vm.stats.vm.v_dfree: 0
vm.stats.vm.v_pfree: 142
vm.stats.vm.v_tfree: 340
vm.stats.vm.v_page_size: 4096
vm.stats.vm.v_page_count: 235637
vm.stats.vm.v_free_reserved: 356
vm.stats.vm.v_free_target: 5064
vm.stats.vm.v_free_min: 1533
vm.stats.vm.v_free_count: 231577
vm.stats.vm.v_wire_count: 3779
vm.stats.vm.v_active_count: 251
vm.stats.vm.v_inactive_target: 7596
vm.stats.vm.v_inactive_count: 29
vm.stats.vm.v_laundry_count: 0
vm.stats.vm.v_cache_count: 0
vm.stats.vm.v_pageout_free_min: 34
vm.stats.vm.v_interrupt_free_min: 2
vm.stats.vm.v_forks: 4
vm.stats.vm.v_vforks: 0
vm.stats.vm.v_rforks: 0
vm.stats.vm.v_kthreads: 20
vm.stats.vm.v_forkpages: 132
vm.stats.vm.v_vforkpages: 0
vm.stats.vm.v_rforkpages: 0
vm.stats.vm.v_kthreadpages: 0
# 

And here is patch not reverted, but 800m of physical memory:
# sysctl sysctl vm.stats.vm
vm.stats.vm.v_vm_faults: 1580
vm.stats.vm.v_io_faults: 213
vm.stats.vm.v_cow_faults: 135
vm.stats.vm.v_cow_optim: 0
vm.stats.vm.v_zfod: 362
vm.stats.vm.v_ozfod: 0
vm.stats.vm.v_swapin: 0
vm.stats.vm.v_swapout: 0
vm.stats.vm.v_swappgsin: 0
vm.stats.vm.v_swappgsout: 0
vm.stats.vm.v_vnodein: 0
vm.stats.vm.v_vnodeout: 0
vm.stats.vm.v_vnodepgsin: 0
vm.stats.vm.v_vnodepgsout: 0
vm.stats.vm.v_intrans: 0
vm.stats.vm.v_reactivated: 0
vm.stats.vm.v_pdwakeups: 0
vm.stats.vm.v_pdpages: 4
vm.stats.vm.v_pdshortfalls: 0
vm.stats.vm.v_tcached: 0
vm.stats.vm.v_dfree: 0
vm.stats.vm.v_pfree: 142
vm.stats.vm.v_tfree: 340
vm.stats.vm.v_page_size: 4096
vm.stats.vm.v_page_count: 179753
vm.stats.vm.v_free_reserved: 284
vm.stats.vm.v_free_target: 3872
vm.stats.vm.v_free_min: 1181
vm.stats.vm.v_free_count: 176074
vm.stats.vm.v_wire_count: 3396
vm.stats.vm.v_active_count: 253
vm.stats.vm.v_inactive_target: 5808
vm.stats.vm.v_inactive_count: 29
vm.stats.vm.v_laundry_count: 0
vm.stats.vm.v_cache_count: 0
vm.stats.vm.v_pageout_free_min: 34
vm.stats.vm.v_interrupt_free_min: 2
vm.stats.vm.v_forks: 4
vm.stats.vm.v_vforks: 0
vm.stats.vm.v_rforks: 0
vm.stats.vm.v_kthreads: 20
vm.stats.vm.v_forkpages: 132
vm.stats.vm.v_vforkpages: 0
vm.stats.vm.v_rforkpages: 0
vm.stats.vm.v_kthreadpages: 0
# 

Ruslan