Get rid of the original puffs_req(3) framework and use puffs_framebuf(3)
instead for file system requests. It has the advantage of being
suitable for transporting a distributed message passing protocol
and therefore us being able to run the file system server on any
host.
Ok, puffs is not quite here yet: libpuffs needs to grow request
routing support and the message contents need to be munged into a
host independent format. Saying which format would be telling,
but it might begin with an X, end in an L and have the 13th character
in the middle. Keep an eye out for the sequels: Parts 3+m/n.
Ok, ok, a few more words about it: stop holding puffs_cc as a holy
value and passing it around to almost every possible place (popquiz:
which kernel variable does this remind you of?). Instead, pass
the natural choice, puffs_usermount, and fetch puffs_cc via
puffs_cc_getcc() only in routines which actually need it. This
not only simplifies code, but (thanks to the introduction of
puffs_cc_getcc()) enables constructs which weren't previously sanely
possible, say layering as a curious example.
There's still a little to do on this front, but this was the major
fs interface blast.
separately
* provide puffs_cc_getcc()
This is in preparation for the removal of you-should-guess-what as
an argument to routines here and there and everywhere.
is a forwards jump instead of a backwards jump that a P4 will mispredect (and
probably immediatly get a cache-line miss due to the way the code gets
aligned).
extra argument to ensure that 'off_t' parameters are even numbered
parameters for stack alignment reasons on 32 bit systems) to use a normal
'syscall' wrapper for __foo instead of using __syscall(SYS_foo, ... ).
This saves all the faffing needed to get a 32bit return value on sparc.
Auto-generate the extra syscall wrappers.
Move swapon.c into the correct list (swapon() uses swapctl()).
This probably means there are very few users of syscall() and __syscall() left.
also synchronizes with puffs_mount() and does not return (exit) in the
parent process until the file system has been mounted. This makes
it possible to reliably run e.g. mount_foo jippi /kai ; cd /kai/ee
- Play scrooge again and chop more cycles off acquire/release.
- Spin while the lock holder is running on another CPU (adaptive mutexes).
- Do non-atomic release.
Threadreg:
- Add the necessary hooks to use a thread register.
- Add the code for i386, using %gs.
- Leave i386 code disabled until xen and COMPAT_NETBSD32 have the changes.
- Override __libc_thr_init() instead of using our own constructor.
- Add pthread__getenv() and use instead of getenv(). This is used before
we are up and running and unfortunatley getenv() takes locks.
Other changes:
- Cache the spinlock vectors in pthread__st. Internal spinlock operations
now take 1 function call instead of 3 (i386).
- Use pthread__self() internally, not pthread_self().
- Use __attribute__ ((visibility("hidden"))) in some places.
- Kill PTHREAD_MAIN_DEBUG.
for misc support routines and put it there.
- Add a libc constructor. Use this to initialize threading and the
stack protector stuff. libpthread cannot be initialized safely using
its own constructor because libc and libpthread are deeply intertwined.
PR bin/37347
This adds reachover Makefiles to build the libatf library and enables it in
the parent Makefile.
Things to review in this change:
* Add proper version numbers in the shlib_version files.
* Is libatf properly listed in lib/Makefile? It theoretically needs
libstdc++, but the resulting binary library is not linked against it.
This function allows the caller finer control of how the writes happen
and doesn't force stdio as interface. Optimise clear_gap a bit to not
fill the gap byte-wise. Bump minor version of libkvm.
we try to set both in one ioctl, to the same value, which fails if not
both record and play direction support that many channels. (Having
more play than record channels is common these days.)
This does likely address part of the problem described in PR pkg/35444
by Mihai Chelaru, and it helped Matthew R. Green about half a year
ago -- see the thread in tech-misc in April.
"round to zero" variants of the softfloat conversion functions.
Add a variant to convert long double to unsigned long - the "to long"
variant checked for overflows that do not apply to unsigned results.
This fixes the regress/lib/libc/convfp tests for sparc64.
each line that we copy.
We also need to touch the destination window here.
Fixes display bug in atc(6). Bug reported by and fix tested by
David Holland.
