On Mon, Aug 22, 2016 at 06:55:58PM +1000, Bruce Evans wrote: > On Sun, 21 Aug 2016, Slawa Olhovchenkov wrote: > > > On Mon, Aug 22, 2016 at 02:49:07AM +1000, Bruce Evans wrote: > > > >> On Sun, 21 Aug 2016, Slawa Olhovchenkov wrote: > >> > >>> On Sun, Aug 21, 2016 at 11:39:02PM +1000, Bruce Evans wrote: > >>> > >>>> On Sun, 21 Aug 2016, Slawa Olhovchenkov wrote: > >>>>> I am remeber about platforms with missaligment trap when > >>>>> accessing int16 by odd address. Now platforms like this do not exist > >>>>> anymore? > >>>> > >>>> i386 still exists, and it supports trapping on misalignement for at least > >>>> CPL 3 (not kernel CPL 0). IIRC, amd64 drops support for this. > >>> > >>> Someone enable and support this? I am don't see. > >>> May be PPC trap on this? > >>> Alpha trap on this, but support of Alpha is droped. > >> > >> It is a 1-line change in asm (or a little more in C with #includes) to > >> enable the trap: > > > > OK, we can turn amd64 in this mode. > > And cat do request to kernel with unalligned access, this cause trap > > and panic, yes? > > No. PSL_AC is ignored in kernel mode.
OK. I.e. i386 and amd64 is not target. cloudabi work in kernel mode, yes? > >> It is a trillion-line change to fix the compilers and applications to not > >> do misaligned accesses :-). I only tried to use this ~25 years ago. Then > >> the most obvious compiler bug was generating 32-bit acccesses to assign > >> large but misaligned structs. If the compiler just generated calls to > >> memcpy(), that might work, but in practice libraries also assume alignment. > > > > This issuse can be trigerred and by two-bytes assigmen, yes? > > Not quite that short. i386 has the 1-byte cli instruction for conveniently > setting the interrupt enable flag, but setting PSL_AC seems to take at > least 3 instructions and 6-7 bytes (pushf; orb $N,$M(%[re][bs]p); popf). I am miss something. Why you talk about bytes per instruction? I think this is about returning value to applications unaligning buffer? 8 bytes in this commit or 2 bytes in my example. (for this commit, as I see, td_retval always aligned) > >>>>>> There are also endianness problems. The old version was even more > >>>>>> broken > >>>>>> on big endian systems. The current version needs some magic to reverse > >>>>>> the memcpy() of the bits. We already depend on this for some 64-bit > >>>>>> syscalls like lseek(). > >>>>> > >>>>> Can you explain some more? > >>>>> This is not transfer over network and don't read from external media. > >>>>> Where is problem? > >>>> > >>>> It is similar to a network transfer. It needs a protocol to pass values > >>>> to applications. Type puns are fragile even within a single compilation > >>>> unit. > >>> > >>> Application ad kernel run with same byte order, not? > >> > >> The application can do anything it wants, but has to translate if it uses > >> the kernel or a library written in another language. > > > > You talk about different byte order in differenr languages? > > Could be, or the same language with a different ABI. ABI enforced by `server`, ex: kernel, or cloudabi in this case. If language need communicate -- language need adopted, not kernel or cloadabi. No simple way from kernel/cloudabi/etc detect caller abi. Or I am missunderstund you. _______________________________________________ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
