* To display conversion candidate in KouhoWindow, we need to change skipping
index bytes. In CannaInterface::GenerateKouhoString(), we have skipped
2 bytes as index before conversion candidate, but index is UTF8 FULL
WIDTH NUMBERS, so we need to skip 3 bytes.
* To show highlightRect correctly, change Invalidate(highlightRect) to
Invalidate() in KouhoView::HighlightLine().
* Change BeOS's Haru font to VL PGothic (CannaLooper::ReadSettings()
and KouhoWindow::KouhoWindow()).
* hrev43729 switched the interrupt handler to a threaded handler, thus requiring
to implement HAIKU_CHECK_DISABLE_INTERRUPTS()
* should help fixing #8327
* Prepend x86_ to non-static x86 code
* Add x86_init_fpu function to kernel header
* Don't init fpu multiple times on smp systems
* Verified fpu is still started on smp and non-smp
* SSE code still generates general protection faults
on smp systems though
* Add new function to check for protocol support
* If IPv6 support is present assign ::1 to loopback
* ping6 ::1 functions 100%
* Still some route strangeness with link local addresses
* Check to ensure the address doesn't already exist on card
* In the future we will want to check interface address flags
for any local scoped addresses on a card.
* Remove old ioctl code, cleanup AutoLooper.
* Move link local code into NetServer
AutoLooper should only be used for things that
count as "auto-configuration" such as DHCP, router
advertisements, and DHCPv6
* Properly form IPv6 link local address from MAC address
* I think some IPv6 routes are needed still for proper
local link connectivity.
* Duplicate Address Detection is still a TODO
* Style cleanup
* yuck, glibc uses yet another version of mbstate_t (__c_mbstate_t),
adjusting this to match the other glibc-internal version (__mbstate_t)
fixes another crash triggered by fwide_test.
* instead of unification, we now keep both our and glibc's fields
separate in order to allow for both our code and glibc's to use
such a struct at the same time (independently)
* _IO_USER_BUF was being used to indicate a user-owned buffer without
taking into account that there are two of those: one for the normal
and another one for the wide version of a stream. Backport
_IO_FLAGS2_USER_WBUF from current glibc version to fix that.
* wcsftime invokes strftime and then converts the resulting string
internally, so it relies on the correct ctype locale being active
* minor style cleanup
