| Age | Commit message (Collapse) | Author |
|---|
|
|
|
existing one for earlier arm revisions.
|
|
From Patrick Wildt in Bitrig.
|
|
now that the disklabel code will reduce MAXDISKSIZE to the real size if a Sun
label is found.
|
|
ok miod@
|
|
usually invoked after initdisklabel() which proactively changes a zero disk
size to MAXDISKSIZE, causing this test to fail.
Allow for MAXDISKSIZE too in that test. This makes spoofed disklabels of SMD
disks have a proper `c' slice size.
luna88k disklabel_om_to_bsd() is modified accordingly, to keep diffability,
even though luna88k can't - to the best of my knowledge - sport SMD disk
controllers.
ok deraadt@ krw@
|
|
This commit adds generic infrastructure to do binary code patching on amd64.
The existing code patching for SMAP is converted to the new infrastruture.
More consumers and support for i386 will follow later.
This version of the diff has some simplifications in codepatch_fill_nop()
compared to a version that was:
OK @kettenis @mlarkin @jsg
|
|
because it teaches us the drive geometry, but from then on there is no reason
not to implement disklabel spoof support and a proper DIOCGPDINFO ioctl.
As a result, this makes disklabel -A now work on SMD disks.
The drawback of this, is that, since native labels do not have a pcylinders
value, the code will now always assume pcylinders == ncylinders + acylinders.
(this ought to be the case by default for labels produced by SunOS format(8).)
|
|
|
|
Only pmap.c was not including them explicitly. Fix that.
"The direction is good" deraadt@
|
|
This is set by many hypervisors, including kvm, vmware, hyper-v.
|
|
|
|
be sure to adjust the IOPB data pointer before redirecting the I/O for this
particular sector to the replacement location.
Otherwise, the data pointer still points to the first sector of the I/O, which
may not necessarily be the one which failed.
(This is yet another 19 years old bug, making your filesystems self-destruct
even faster than intended)
|
|
further pending requests, since they will be forcefully failed by the reset
routine.
|
|
member, rather than a global. No functional change.
|
|
Remove obsolete comments in x[dy]dump().
|
|
noone wrote an equivalent of SunOS format(8) and noone ever will.
|
|
asking for the device type upon labeling a freshly formatted disk.
|
|
spotted by Patrick Wildt
|
|
make install
from Simon Nicolussi
ok jsing tedu
|
|
|
|
|
|
- when entering bad144 mode, correctly compute the actual C/H/S values for the
logical block number.
- when leaving bad144 mode and resuming normal I/O operation, when computing
back the C/H/S values for the next logical block number, compute the
sector number correctly.
|
|
thanks jsg
|
|
floppy. A few drivers are missing, but the world has moved on (the drivers
included are always a work in progress)
Speeds up make release substantially, of course.
|
|
the installer. We used a full release cycle to learn that suspend/resume
is reliable enough for this default. Personal policy can disable this using
machdep.lidsuspend=0 in /etc/sysctl.conf
ok more people begging, and less people whining
|
|
|
|
pointed out by Helg (xx404 (at) msn.com)
ok deraadt@ miod@
|
|
This should catch all the same bad cases, but be a little less aggravating
in circumstances where a prototype isn't necessary
ok deraadt
|
|
than the maximum DMA transfer size the PROM reports.
Crank version to 2.9.
|
|
instead of, well, bcopy and bzero. This lets ie@vme transmit packets instead of
panicing due to memset() trying to use 32-bit stores on a device address.
|
|
a sun4 system - the prom i/o routines depend on it.
|
|
|
|
use these on armv7 as a backend for byteswapping things that endian.h
provide.
i dunno if its faster, but it makes smaller code. saves 30k on
GENERIC-OMAP.
ok jsing@ bmercer@ jsg@
|
|
|
|
|
|
always load all allocated pages instead of the size specified in the
bus_dmamap_load_raw(9) call. Also fixes the corner case where a specified
boundary is less than the page size, which would always create multiple
segments, even if the specified size was smaller than the boundary.
Fixes xhci(4) on sparc64.
|
|
backed out because of libkvm breakage, recommitting now with libkvm fix.
ok deraadt@
|
|
can check if we're entering code we think should only be used from
lower ipls.
modelled a bit on how sparc64 does things.
with help from and ok kettenis@
|
|
to replace x86_atomic_testset_{u32,u64}.
help from guenther@ kettenis@
ok kettenis@
|
|
|
|
|
|
This changes the logic to prevent a recursion when processing soft
interrupts. Previously a per-CPU flag was set before re-enabling
interrupts. Now the IPL level is raised to SOFTTTY which makes
splsoftassert() happy, greatly inspired by mips64.
As a side effect, the ppc_intr_{disable,enable}() dance is now done
only once instead of twice per splx(9).
While here, make use of dosoftint() instead of having 3 different
functions for dispatching soft interrupts.
Tested by deraadt@ on G4 smp and by myself G5 smp, G3, G4 and socppc.
No objection from the usual (and over busy) suspects.
|
|
|
|
|
|
with the base address of the window, which only works if its size is smaller
than the address it begins at.
Until these assumptions are carefully fixed, revert r1.9, and go back to using
only one direct map window.
|
|
|
|
OK miod@
|
|
CompuLab Utilite, SolidRun HummingBoard/CuBox-i and UDOO.
|
|
|
