On Sep 18, 2026, at 20:51, Warner Losh <[email protected]> wrote (off list)
> On Fri, Sep 18, 2026 at 9:43 PM Mark Millard <[email protected]> wrote: > > . . . > > Yea. I thought I'd pushed this, but I hadn't. > > The stack that ends at https://reviews.freebsd.org/D59835 has what I > thought I'd pushed. > > Warner So I built, installed, and booted a separate personal kernel based in part on using the copies: # ls -C1 /usr/main-src/D5982*.diff /usr/main-src/D59827.diff /usr/main-src/D59828.diff (The 2 smaller numbered diffs in the stack.) It looked like those 2 were all that are involved for the formatting test so I limited my nvme additions to the 2. The sequence was (format_timeout_period is new): # sysctl -a | grep 'nvme.*timeout' dev.nvme.5.format_timeout_period: 600 dev.nvme.5.timeout_period: 30 dev.nvme.5.admin_timeout_period: 60 . . . # nvmecontrol devlist nvme0: INTEL SSDPE21K015TA nvme0ns1 (1430799MB) . . . # sysctl dev.nvme.0.format_timeout_period=1800 dev.nvme.0.format_timeout_period: 600 -> 1800 # time -l \ > nvmecontrol format -f 3 -m 0 nvme0 1377.27 real 0.00 user 0.00 sys 0 maximum resident set size 0 average shared memory size 0 average unshared data size 0 average unshared stack size 148 page reclaims 0 page faults 0 swaps 0 block input operations 0 block output operations 0 messages sent 0 messages received 0 signals received 4 voluntary context switches 0 involuntary context switches So, a little under 23 minutes. No console messages. Thanks much. -- === Mark Millard marklmi at yahoo.com
