Those gaps in the trunk test for FBSD-32 (32-bit FreeBSD) seem to have cascaded from this:

Jul 31 23:01:34 vm-fbsd kernel: pid 33265 (managed_shared_memo), uid 1001: exited on signal 6 (core dumped)

Jul 31 23:03:30 vm-fbsd kernel: pid 33934 (managed_shared_memo), uid 1001: exited on signal 6 (core dumped)

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(16): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(8): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(16): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(12): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(16): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(12): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(6): failed

Jul 31 23:48:57 vm-fbsd kernel: swap_pager_getswapspace(16): failed

Jul 31 23:48:57 vm-fbsd kernel: pid 1674 (bjam), uid 1001, was killed: out of swap space

Jul 31 23:48:57 vm-fbsd kernel: pid 53297 (cc1plus), uid 1001, was killed: out of swap space

I think managed_shared_memo isn’t the cause, both from the timeline and from seeing this in older logs.

Suggestions welcome. I suspect we hit a bug in cc1plus and it wants infinite space, but I’m glad to tune my kernel if that’s the problem.

By the way, here’s “sysctl –a | grep –i swap”

vm.swap_enabled: 1

vm.nswapdev: 1

vm.swap_async_max: 4

vm.swap_reserved: 64126976

vm.swap_total: 2087518208

vm.swap_idle_threshold2: 10

vm.swap_idle_threshold1: 2

vm.stats.vm.v_swappgsout: 0

vm.stats.vm.v_swappgsin: 0

vm.stats.vm.v_swapout: 0

vm.stats.vm.v_swapin: 0

vm.disable_swapspace_pageouts: 0

vm.defer_swapspace_pageouts: 0

vm.swap_idle_enabled: 0

And g++ -v:

Using built-in specs.

Target: i386-undermydesk-freebsd

Configured with: FreeBSD/i386 system compiler

Thread model: posix

gcc version 4.2.1 20070719  [FreeBSD]