| Age | Commit message (Collapse) | Author |
|---|
|
ok jsg@
|
|
|
|
ok jsg@
|
|
|
|
|
|
saa_luns instead of adapter_buswidth and luns in the prototype link.
ok dlg@, miod@
|
|
32-bit on 32-bit architectures; we need off_t to support >4GB vnd(4)
images.
Discovered by, tested, and ok bluhm@
|
|
ok deraadt
|
|
|
|
device signature so i left it blank, and the identify is the cached copy
in the port structure.
i filled the SAT vendor stuff in as "OpenBSD", "atascsi", osrelease.
|
|
to do WRITE SAME 16 for now (which is all we do).
|
|
|
|
use less magic numbers when check the block limits for trim. fill in the
block limits vpd page with some conservative numbers about how much
unmap we can do at a time (~32MB over 64 descriptors).
|
|
structs and calling vn_lock+VOP_READ/WRITE+VOP_UNLOCK.
|
|
xxstrategy() methods, and punt in validating it in vndread() and
vndwrite() (also like we do in other xx{read,write}() methods...).
|
|
use it instead.
|
|
the freebsd firmwares are compressed and get the kernel to uncompress them
to use them. we uncompress in build.c so the kernel can just read off disk
onto the card without any extra work. this lets us copy future fbsd fw
updates directly without any extra work.
myri have also rescinded one of the clauses on their license.
ok claudio@ deraadt@
|
|
added in POSIX Issue 7: openat(), mknodat(), mkfifoat(), linkat(),
symlinkat(), unlinkat(), faccessat(), fstatat(), readlinkat(),
fchmodat(), fchownat(), utimensat(), renameat(), and mkdirat().
This diff mostly just refactors the existing sys_foo() logic into a
common dofooat() function that can then be called by both sys_foo()
and sys_fooat(). Some of the new system calls support new flags to
control their behavior, and proper support for these will be added in
subsequent diffs.
Incorporating suggestions from thib@, guenther@, and tedu@.
ok tedu@, thib@, deraadt@, guenther@
|
|
with this nothing in the tree fiddles if ifqueue internals any more, of
course except if.c and if.h (and some altq)
|
|
|
|
it is the only place in the tree (minus if.{c,h} which are allowed to) that
reads ifq_maxlen without setting it, ryan ok
|
|
horrid and must be wrong. now in this case it is actually harmless - but
setting them to 0 just after having malloc'd it with M_ZERO is pointless
to begin with. ok claudio
|
|
hold the kernel lock, but still need call one function that needs it.
Instead of grabbing the lock all over the place, move the locks into
the affected functions: trapsignal, scdebug*, ktrsyscall, ktrsysret,
systrace_redirect and ADDUPROF. In the cases we already hold the biglock
we'll just recurse.
kettenis@, beck@ ok
|
|
is a partial block. Also correctly handle the case where there is no
partial block. Whilst here remove a pointless variable.
Issues spotted by and based on diffs from Piotr Durlej - thanks!
ok marco@
|
|
avoid duplication. Implement a sr_shutdownhook() function that simply
calls sr_shutdown() and use this for the shutdown hook.
ok marco@
|
|
|
|
use consistent network types as requested and ok deraadt.
|
|
ok deraadt matthew
|
|
|
|
having one per volume.
ok marco@
|
|
anymore.
|
|
ok krw@
|
|
however since this code exists it still should work correctly.
|
|
it again.
|
|
to indicate if we should force a metadata write.
ok marco@
|
|
and POOLBYTES for write.
looks right to deraadt
|
|
bounds_check_with_label().
|
|
the logic slightly and makes vnd(4) more like any other disk driver.
To avoid races, this means vndopen() can only set VNF_HAVELABEL if
dk_openmask == 0. Otherwise, it's possible for userspace to open
rvnd0c, call VNDIOCSET, open vnd0a, then while vndreaddisklabel() (via
vndstrategy) is waiting for VOP_READ() to finish, you could issue a
read or write on the still open rvnd0c and have VNF_HAVELABEL set but
the disklabel might be in a weird state.
Note that this makes VNF_HAVELABEL nicely analogous to sd(4)/cd(4)'s
SDEV_MEDIA_LOADED flag, which is handled similarly in
{sd,cd}{open,close,strategy}.
ok dlg@, krw@, deraadt@
|
|
that they're implemented consistently in bounds_check_with_label().
Also, per krw's request, change bounds_check_with_label() to return 0
if the checks succeed, and change the drivers to test == -1 instead of
<= 0. (Man page update to follow; intentionally omitting
arch/vax/mba/hp.c from this commit because it doesn't even build
currently and miod@ promises to kill it soon.)
ok krw@
|
|
|
|
ironlake interrupt handler is changed. some nasty bug lurks in here,
and any binary change randomly exposes this and screws up the chip.
back this out for now; until it is found.
ok oga
|
|
something. Doesn't matter right now because scsibus_attach_args only
has one field and it's mandatory, but I'm planning to move some more
fields from scsi_link to scsibus_attach_args+scsibus_softc.
ok dlg@
|
|
link->bus->sc_dev.dv_unit.
|
|
Instead of allocating a crypto op and the optional dma buffer on each
and every io, preallocate a list of softraid crypto wus that contain a
buffer of the max size we will use (MAXPHYS). since we know the number
of openings we have in advance this means that on each io we just pick
one, shorten the list of crypto descs, init any values then do the io.
ok jsing (who provided many useful comments. he also provided a smarter way of
handling the cryptop lists which is not in this diff but will be implemented
soonish), marco@.
dlg@ pointed out that this should probably use iopools but letting
disciplines allocate their own iopool involves more softraid rejigging
that will have to be done first. For now this is sufficient.
|
|
the MSI enabled flag there such that the driver actually pays attention to it.
Found out the hard way by Chris Smith on an 82540EM, which defenitely does
not like MSI.
ok deraadt@
|
|
become empty. In that case, the hme driver could not recover as
the ring was only filled after receiving data. Check and potentially
fill an empty receive ring every second in hme_tick().
ok kettenis@
|
|
into DSM/TRIM commands.
found by tedu
|
|
Some gm45 and 965 are having issues and this fixes it for halex at least.
Art's machine (945gm) is also playing up but he didn't update for a year
(and the problem doesn't look like what i would expect).
|
|
SR_DEBUG enabled. Broken since r1.227.
|
|
scsi_link::scsibus hack to determine which channel the link is
associated with.
"looks sane" dlg@, but haven't found any testers yet; committing so
further SCSI refactorings can go in. dlg@ or I will back out or fix
if anything breaks.
|
