re: CVS commit: src/external/gpl3
"David A. Holland" writes: > Module Name: src > Committed By: dholland > Date: Sun Jul 27 04:16:09 UTC 2014 > > Modified Files: > src/external/gpl3/gcc.old/usr.bin: Makefile.inc > src/external/gpl3/gcc/usr.bin: Makefile.inc > > Log Message: > Update the bug report URL as per PR 48712. I assume in order for this > to propagate to where it'll be seen we need to rerun mknative, which > I'm not in a position to do right now; but this way it'll update the > next time someone else does that. hmmm, we should try to get the URL working again. it's baked into old binaries we will support for a while yet... .mrg.
Re: CVS commit: src/sys/external/bsd/dwc2/dist
(07/26/14 18:18), Nick Hudson wrote: Log Message: Recover from channel halt errors by using the 3 strikes xacterr rule. Adapted from the RaspberryPI linux driver. This allows the rum(4) I was sent to be somewhat usable. Need to investigate further what is causing the problem in the first place. Might help PR/49019 (RPI: interrupt storm when url0 is up) I tried this but still get intr storm when if is up. -- t-hash
Re: CVS commit: src/usr.sbin/sysinst
dholland@ wrote: > Module Name: src > Committed By: dholland > Date: Sat Jul 26 19:56:30 UTC 2014 > > Modified Files: > src/usr.sbin/sysinst: Makefile > Added Files: > src/usr.sbin/sysinst/arch/dummy: Makefile md.c md.h msg.md.de msg.md.en > msg.md.es msg.md.fr msg.md.pl > > Log Message: > Add a dummy sysinst architecture for ports that don't have their own yet. You don't have to put empty translation files. (I guess empty "LANGUAGES=" line like sysinst/arch/hp300/Makefile is enough) Anyway, this also reminds PR install/45976, i.e. unmaintained translations which don't have proper native maintainers should be removed before the next release, as paranoiac people did on www. There is no mechanism/rule to see "which messages are translated or not." --- Izumi Tsutsui
Re: CVS commit: src/sys/external/bsd/drm2/include/linux
On Sat, 26 Jul 2014, Taylor R Campbell wrote: Make Linux usecs_to_jiffies work with hz up to 2000. usecs_to_jiffies(unsigned int usec) { - return mstohz((usec + (1000 / hz) - 1) / (1000 / hz)); + if (hz <= 100) + return mstohz(roundup(usec, (1000 / hz))); + + /* +* Avoid integer overflow on 32-bit platforms. The cutoff is +* kinda arbitrary; for hz <= 2000, 0x20 is safe, but both +* values could wiggle around a little. +*/ + KASSERT(hz <= 2000); + if (usec <= 0x20) + return ((usec * hz) / 100); + else + return ((usec / 100) * hz); } You could just populate a struct timeval, and call tvtohz(9). What is the desired behaviour for small inputs? The code quoted above appears to round up when hz <= 100 (and to return 1 for very small but non-zero inputs) but to round down (and to return 0 for very small inputs) when hz > 100. Once the desired behaviour is clarified, you could implement it by making adjustments to the value before or after calling tvtohz. --apb (Alan Barrett)