Date: Tue, 30 Oct 2001 11:21:39 -0800 (PST) From: Julian Elischer <julian@elischer.org> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Bruce Evans <bde@zeta.org.au>, Nate Williams <nate@yogotech.com>, Poul-Henning Kamp <phk@critter.freebsd.dk>, Peter Wemm <peter@wemm.org>, arch@FreeBSD.ORG Subject: Re: 64 bit times revisited.. Message-ID: <Pine.BSF.4.21.0110301120420.26174-100000@InterJet.elischer.org> In-Reply-To: <200110301834.f9UIYeR94440@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 30 Oct 2001, Matthew Dillon wrote: > :On Mon, 29 Oct 2001, Julian Elischer wrote: > : > :> and you really want us to believe that data file came form 1910? > : > :I simply refuted a false claim that timestamps can't be before 1970. > : > :> On Tue, 30 Oct 2001, Bruce Evans wrote: > :> > :> > On Mon, 29 Oct 2001, Julian Elischer wrote: > :> > > :> > > be used for multiplying the seconds scale by 4. (in UFS timestamps.. > :> > > you would never write a non-normalised time to disk) > :> > > also timestamps can't be before 1970 so making it unsigned > :> > > allows us to go to 2100+ and mutiplying it by for takes us to about 2600.. > :> > > :> > Script started on Tue Oct 30 06:55:55 2001 > :> > ttyp1:bde@gamplex:/tmp> touch -t 191001011230 foo > :> > ttyp1:bde@gamplex:/tmp> ls -l foo > :> > -rw-r--r-- 1 bde wheel 0 Jan 1 1910 foo > :> > ttyp1:bde@gamplex:/tmp> exit > :> > > :> > Script done on Tue Oct 30 06:56:10 2001 > : > :This mailing list is approaching the quality of the average newsgroup, > :complete with top posting by newbies. > : > :Bruce > > We can't make time_t unsigned anyway. A huge amount of code > (hundreds of instances) do delta comparisons and simply assume > that time_t is signed. With it unsigned many of those comparisons > would blow up (return true when they should return false). I'm not advocating making time_t unsigned.. just the timestamp fields in the UFS inodes.. > > I'm about ready to give up trying to convince people that it's worth > moving IA32 to 64 bit timestamps. The bigger issue, however, is > what to do about 64 bit platforms. I keep finding system calls > that need to be rolled. The list is now getting pretty big. I am > up to 22 system calls. > > This means that if we are NOT going to change IA32 but we ARE going > to change 64 bit architectures, then we should do it *without* > rolling system calls at all which would mean having to eat any > binary incompatibilities from older code. It might conceivably be > worth rolling the syscalls if we were to change IA32, but if we aren't > it just isn't worth it to roll the syscalls just to support > pre-time-change 64 bit platforms. > > -Matt > Matthew Dillon > <dillon@backplane.com> > > > 7 MCOMPAT2 BSD { int wait4(int pid, int *status, int options, \ > 83 MCOMPAT2 BSD { int setitimer(u_int which, struct o2itimerval *itv, \ > 86 MCOMPAT2 BSD { int getitimer(u_int which, struct o2itimerval *itv); } > 93 MCOMPAT2 BSD { int select(int nd, fd_set *in, fd_set *ou, \ > 116 MCOMPAT2 BSD { int gettimeofday(struct o2timeval *tp, \ > 117 MCOMPAT2 BSD { int getrusage(int who, struct o2rusage *rusage); } > 122 MCOMPAT2 BSD { int settimeofday(struct o2timeval *tv, \ > 138 COMPAT2 BSD { int utimes(char *path, struct o2timeval *tptr); } > 140 MCOMPAT2 BSD { int adjtime(struct o2timeval *delta, \ > 188 COMPAT2 POSIX { int stat(char *path, struct o2stat *ub); } > 189 MCOMPAT2 POSIX { int fstat(int fd, struct o2stat *sb); } > 190 COMPAT2 POSIX { int lstat(char *path, struct o2stat *ub); } > 206 COMPAT2 BSD { int futimes(int fd, struct o2timeval *tptr); } > 232 MCOMPAT2 POSIX { int clock_gettime(clockid_t clock_id, \ > 233 MCOMPAT2 POSIX { int clock_settime(clockid_t clock_id, \ > 234 MCOMPAT2 POSIX { int clock_getres(clockid_t clock_id, \ > 240 MCOMPAT2 POSIX { int nanosleep(const struct o2timespec *rqtp, \ > 276 COMPAT2 BSD { int lutimes(char *path, struct o2timeval *tptr); } > 299 COMPAT2 BSD { int fhstat(const struct fhandle *u_fhp, struct o2stat *sb); } > 315 COMPAT2 BSD { int aio_suspend(struct aiocb * const * aiocbp, int nent, const struct o2timespec *timeout); } > 334 MCOMPAT2 POSIX { int sched_rr_get_interval (pid_t pid, struct o2timespec *interval); } > 363 MCOMPAT2 BSD { int kevent(int fd, \ > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-arch" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0110301120420.26174-100000>