| Age | Commit message (Collapse) | Author |
|---|
|
remaining instances of them with appropriate copy(9) usage.
ok art@, tested on all arches unless my memory is non-ECC
|
|
|
|
set from <sys.mk>
|
|
Instead of allocating a static amount of memory for the data
structures via valloc() in allocsys(), allocate things dynamically
using pool(9) when possible and malloc(9) when not. The various
members of struct seminfo and struct shminfo are in kern.seminfo
and kern.shminfo respectively (not all members of kern.seminfo are
changable).
The data structures used still leave something to be desired but
things are not made worse in that respect by this commit.
|
|
|
|
produce typo fixes, but the typo pool is infinite.
|
|
altq is merged into pf and handled through pf.
|
|
attack protection scheme, into gcc.
This protection is enabled by default. It can be turned off by using the
-fno-stack-protector flag.
Code by Hiroaki Etoh (etoh at jp dot ibm dot com); work on openbsd-specific
integration by fgsch@, deraadt@ and myself; tests by fgsch@, naddy@ and
myself; beer drinking by myself.
Please note that system upgrades with this new code will require a new
libc and ld.so to be build and installed before the propolice-enabled
compiler can be installed.
|
|
The macro PG_PALCODE in pte.h is supposed to be used when comparing
two ptes to see if we need a tlb flush. For that it uses the macro
ALPHA_PTE_PALCODE which in turn is defined to ~ALPHA_PTE_SOFTWARE.
Unfortunately ALPHA_PTE_SOFTWARE is small enough to fit in an int,
so ALPHA_PTE_PALCODE becomes an int to and the masking in PG_PALCODE
masks off the pfn in the pte and pmap_enter will then fail to flush
certain TLB entries when it needs to (this situation shouldn't
happen too often).
This might or might not be the solution for the memory corruption
bug I've been hunting for the last three months (the machine still
borks up, but in different ways now).
|
|
In sys_sigreturn - don't uvm_useracc the context address and then copyin
it afterwards. The copyin deals better with finding if the user address is
correct and it's faster.
|
|
only one.
X11 is configured here to start on ttyC5, like on i386 again. This requires
alpha-specific changes to the X servers, that went in the tree before 3.2
was released.
On tga-compatible boards, there are no virtual consoles available, and these
changes will have no effect.
|
|
(which I will leave for Dale since it needs special handling).
From NetBSD (and same as sparc64). espie@ OK
|
|
Diff generated by Chris Kuethe.
|
|
|
|
pages into the queue already containing allocated pages.
breaks i386:setup_buffers() because of this.
|
|
|
|
|
|
overall.
|
|
resolved. sorry -- i've been warning people for some time that i would
start to take this stance.
|
|
and return a VM_PAGE. This is to allow sparc64 to cheaply record the
VAC color for those pages.
|
|
instead of the pa. Most callers already had it handy and those who didn't
only called it for managed pages and were outside time-critical code.
This will allow us to make those functions clean and fast on sparc and
sparc64 letting us to avoid unnecessary cache flushes.
deraadt@ miod@ drahn@ ok.
|
|
to allow access to VGA card's memory by mmapping /dev/xf86.
The macdep.allowaperture does also control /dev/pci access.
Ok miod@, deraadt@
|
|
|
|
This makes sure it will be regenerated if you run config(8) again.
|
|
The only OSes I've seen that use SIZE_T_MAX are 4.4BSD-derived whereas
SYSV things seem to use SIZE_MAX. It is also consistent with SSIZE_MAX
(which we already have). deraadt@ OK
|
|
Instead of using FOE for just emulating references, we also keep track
of a pages executability and don't remove the FOE bit if the page
is not executable.
This is implmented with horrible hacks. Maybe when I have time, I'll
reimplment the whole pmap to allow this without ugly hacks (read: probably
not this decade).
The stack on alpha is now non-exec.
|
|
|
|
|
|
an uvm aobj, copy out the signal trampoline into it and share that page
among all processes for the same emulation.
This also requires us to actually be able to tell signal code where the
trampoline is located, so introduce a new field in struct proc - p_sigcode
that is a pointer to sigcode. This allows us to remove all the ugly
calculations of the signal trampoline address done in every sendsig
function in the tree (that's why so many files are changed).
Tested by various people. ok deraadt@
|
|
|
|
|
|
|
|
It doesn't seem to cause problems anymore and it should simplify some things
for X.
|
|
Found the hard way by pvalchev@
|
|
|
|
And fix a comment while there.
|
|
|
|
Added to floppy31 still fits.
|
|
|
|
|
|
|
|
|
|
We kept too much of the old code in trap handling.
This should fix floppies and bsd.rd on some machines.
|
|
squashed already.
|
|
usage of cdev_decl() in the driver code is more tricky since i have
kqueue mods here and there and will fix that on that way through.
|
|
|
|
|
|
|
|
wrong than using cdev_decl_random for the same purpose
|
|
|
