While having RAID-0/1 feature is nice on paper, the USB controller would be a bottleneck, as it is not suitable for this task, specially on a consumer router.
In five years watching the forums, never seen anyone actually using it, and mdadm functions are not exposed on GUI as this is meant for advanced users.
Also these advanced users wouldn't do RAID on a consumer router, for numerous reasonsm, so I believe we'd better sunset this feature.
Everything else for now, remains on the tree. If someone wants mdadm, just compile with RAID=y for AIO target (z).
While having RAID-0/1 feature is nice on paper, the USB controller would be a bottleneck, as it is not suitable for this task, specially on a consumer router.
In five years watching the forums, never seen anyone actually using it, and mdadm functions are not exposed on GUI as this is meant for advanced users.
Also these advanced users wouldn't do RAID on a consumer router, for numerous reasonsm, so I believe we'd better sunset this feature.
Everything else for now, remains on the tree. If someone wants mdadm, just compile with RAID=y for AIO target (z).
There seems to be some odd quirk/bug relating to some of the
automake/autoconf files used in flac, resulting in builds that sometimes
fail citing aclocal-1.15 missing (when automake 1.15 isn't anywhere on
the system). Run autoreconf in advance of configure to address this.
Details: http://www.linksysinfo.org/index.php?threads/getting-kille72-to-build-bugs-galore.74105/
Use of DEBUG_STIME was forcing calc(), thus save(), to write to disk
every 90 seconds rather than per NVRAM variable cstats_stime*60*60.
This bug was introduced in commit caa9070.
Details: http://www.linksysinfo.org/index.php?threads/fork-tomato-arm-by-kille72.73397/page-18#post-296002
Also comment out DEBUG_CSTATS and DEBUG_NOISY to shrink binary size.
This brings things in line with older MIPS releases and AsusWRT-Merlin
as well.
Use of DEBUG_STIME was forcing calc(), thus save(), to write to disk
every 90 seconds rather than per NVRAM variable cstats_stime*60*60.
This bug was introduced in commit caa9070.
Details: http://www.linksysinfo.org/index.php?threads/fork-tomato-arm-by-kille72.73397/page-18#post-296002
Also comment out DEBUG_CSTATS and DEBUG_NOISY to shrink binary size.
This brings things in line with older MIPS releases and AsusWRT-Merlin
as well.
kille72
pedro
AndreDVJ
Jeremy Chadwick