pmc_process_interrupt takes 5 arguments when only 3 are needed.
cpu is always available in curcpu and inuserspace can always be
derived from the passed trapframe.
While facially a reasonable cleanup this change was motivated
by the need to workaround a compiler bug.
core2_intr(cpu, tf) ->
pmc_process_interrupt(cpu, ring, pmc, tf, inuserspace) ->
pmc_add_sample(cpu, ring, pm, tf, inuserspace)
In the process of optimizing the tail call the tf pointer was getting
clobbered:
(kgdb) up
at /storage/mmacy/devel/freebsd/sys/dev/hwpmc/hwpmc_mod.c:4709
4709 pmc_save_kernel_callchain(ps->ps_pc,
(kgdb) up
1205 error = pmc_process_interrupt(cpu, PMC_HR, pm, tf,
resulting in a crash in pmc_save_kernel_callchain.
memset fills the target buffer from a byte-sized value passed in as the
second argument.
The fully-sized (8 bytes) register containing it is named %rsi. Lower 4 bytes
can be referred to as %esi and finally the lowest byte is %sil.
Vast majority of all the callers just zero the target buffer and set it up by
doing xor %esi,%esi which has a side-effect of zeroing the upper parts of
the register as well. Some others do a word-sized move to %esi which has the
same result.
However, there are callers which only fill %sil. This does *not* clear up
the rest of the register.
The value of %rsi is multiplied by $0x0101010101010101 to create a 8-byte sized
pattern for 8-byte stores.
Prior to the patch, the func just blindly took %rsi assuming the unwanted bytes
are zeroed out. Since this is not the case for the callers which only play with
%sil (the rest of the register can have absolutely anything), the resulting
pattern can be garbage.
This has potential for funny bugs. One side effect (which was not amusing)
after enabling it instead of bzero was that the kernel was hanging on boot
as a xen domU.
Reported by: Trond Endrestøl <Trond.Endrestol fagskolen.gjovik.no>
Pointy hat: me
pagetables.
physmap[] can be inconsistent with the physical memory limit due to
buggy bios, or to the hw.physmem tunable. Since bootstrap pagetables
are initialized by accesses through the DMAP, we must ensure that DMAP
really cover the selected pages. This is only relevant when machine
has less than 4G RAM and buggy BIOS, which is the combination on Acer
Chromebook 720.
The call to mp_bootaddress() is moved later to have Maxmem initialized.
An alternative could be to always cover 4G for DMAP, but this change
seems to be simpler.
Reported and tested by: grembo
Reviewed by: royger
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D15675
- increase pmc cpuid field from 8 to 12 bits
- add cpuid version string to initialize entry in the log
so that filter can identify which counter index an
event name maps to
- GC unused config flags
- make fixed counter assignment more robust as well as the
changes needed to be properly identified for filter
Currently all the primitives are waiting for a rewrite, tidy them up in the
meantime.
Vast majority of cases pass sizes which are multiple of 8. Which means the
following rep stosb/movb has nothing to do. Turns out testing first if there
is anything to do is a big win across the board (cpus with and without ERMS,
Intel and AMD) while not pessimizing the case where there is work to do.
Sample results for zeroing 64 bytes (ops/second):
Ryzen Threadripper 1950X 91433212 -> 147265741
Intel(R) Xeon(R) CPU X5675 @ 3.07GHz 90714044 -> 121992888
bzero and bcopy are on their way out and were not modified. Nothing in the
tree uses them.
file in /sys/conf, so was unavailable in configurations that don't use
modules, and was not testable or notable in NOTES. Its normal
configuration (not using a module) is still silently deprecated in
aout(4) by not mentioning it there.
Update i386 NOTES for COMPAT_AOUT. It is not i386-only, or even very MD.
Sort its entry better.
Finish gzip configuration (but not support) for amd64. gzip is really
gzipped aout. It is currently broken even for i386 (a call to vm fails).
amd64 has always attempted to configure and test it, but it depends on
COMPAT_AOUT (as noted). The bug that it depends on unconfigured files
was not detected since it is configured as a device. All other optional
image activators are configured properly using an option.
time, especially for SMP. If configured, it turns itself on at boot
time for calibration, so is fragile even if never otherwise used.
Both types of kernel profiling were supposed to use a global spinlock
in the SMP case. If hi-res profiling is configured (but not necessarily
used), this was supposed to be optimized by only using it when
necessary, and slightly more efficiently, in asm. But it was not done
at all for mcount entry where it is necessary. This caused crashes
in the SMP case when either type of profiling was enabled. For mcount
exit, it only caused wrong times. The times were wrongest with an
i8254 timer since using that requires exclusive access to the hardware.
The i8254 timer was too slow to use here 20 years ago and is much less
usable now, but it is the default for the SMP case since TSCs weren't
invariant when SMP was new. Do the locking in all hi-res SMP cases for
simplicity.
Calibration uses special asms, and the clobber lists in these were sort
of inverted. They contained the arg and return registers which are not
clobbered, but on amd64 they didn't contain the residue of the call-used
registers which may be clobbered (%r10 and %r11). This usually caused
hangs at boot time. This usually affected even the UP case.
kernel profiling remains broken).
memmove() was broken using ALTENTRY(). ALTENTRY() is only different from
ENTRY() in the profiling case, and its use in that case was sort of
backwards. The backwardness magically turned memmove() into memcpy()
instead of completely breaking it. Only the high resolution parts of
profiling itself were broken. Use ordinary ENTRY() for memmove().
Turn bcopy() into a tail call to memmove() to reduce complications.
This gives slightly different pessimizations and profiling lossage.
The pessimizations are minimized by not using a frame pointer() for
bcopy().
Calls to profiling functions from exception trampolines were not
relocated. This caused crashes on the first exception. Fix this using
function pointers.
Addresses of exception handlers in trampolines were not relocated. This
caused unknown offsets in the profiling data. Relocate by abusing
setidt_disp as for pmc although this is slower than necessary and
requires namespace pollution. pmc seems to be missing some relocations.
Stack traces and lots of other things in debuggers need similar relocations.
Most user addresses were misclassified as unknown kernel addresses and
then ignored. Treat all unknown addresses as user. Now only user
addresses in the kernel text range are significantly misclassified (as
known kernel addresses).
The ibrs functions didn't preserve enough registers. This is the only
recent breakage on amd64. Although these functions are written in
asm, in the profiling case they call profiling functions which are
mostly for the C ABI, so they only have to save call-used registers.
They also have to save arg and return registers in some cases and
actually save them in all cases to reduce complications. They end up
saving all registers except %ecx on i386 and %r10 and %r11 on amd64.
Saving these is only needed for 1 caller on each of amd64 and i386.
Save them there. This is slightly simpler.
Remove saving %ecx in handle_ibrs_exit on i386. Both handle_ibrs_entry
and handle_ibrs_exit use %ecx, but only the latter needed to or did
save it. But saving it there doesn't work for the profiling case.
amd64 has more automatic saving of the most common scratch registers
%rax, %rcx and %rdx (its complications for %r10 are from unusual use
of %r10 by SYSCALL). Thus profiling of handle_ibrs_exit_rs() was not
broken, and I didn't simplify the saving by moving the saving of these
registers from it to the caller.
Intel now provides comprehensive tables for all performance counters
and the various valid configuration permutations as text .json files.
Libpmc has been converted to use these and hwpmc_core has been greatly
simplified by moving to passthrough of the table values.
The one gotcha is that said tables don't support pentium pro and and pentium
IV. There's very few users of hwpmc on _amd64_ kernels on new hardware. It is
unlikely that anyone is doing low level optimization on 15 year old Intel
hardware. Nonetheless, if someone feels strongly enough to populate the
corresponding tables for p4 and ppro I will reinstate the files in to the
build.
Code for the K8 counters and !x86 architectures remains unchanged.
This is a follow-up to r321483, which disabled -Wmacro-redefined for
some lib/msun tests.
If an application included both fenv.h and ieeefp.h, several macros such
as __fldcw(), __fldenv() were defined in both headers, with slightly
different arguments, leading to conflicts.
Fix this by putting all the common macros in the machine-specific
versions of ieeefp.h. Where needed, update the arguments in places
where the macros are invoked.
This also slightly reduces the differences between the amd64 and i386
versions of ieeefp.h.
Reviewed by: kib
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D15633
The copied data is accessed in part soon after and it results with additional
cache misses during a -j 1 buildkernel WITHOUT_CTF=yes KERNFAST=1, as measured
with pmc stat.
before:
256165411 cache-references # 0.003 refs/inst
15105408 cache-misses # 5.897%
20.70 real # 99.67% cpu
13.24 user # 63.94% cpu
7.40 sys # 35.73% cpu
after:
256764469 cache-references # 0.003 refs/inst
11913551 cache-misses # 4.640%
20.70 real # 99.67% cpu
13.19 user # 63.73% cpu
7.44 sys # 35.95% cpu
Note the real time did not change, but traffic to RAM was reduced (multiple
measurements performed with switching the implementation at runtime).
Since nobody else is using non-temporal for this and there is no apparent
benefit at least these days, don't use them either.
Side note is that pagecopy arguments should probably get reversed to not
have to flip them around in the primitive.
Discussed with: jeff
The TSC-s are checked and synchronized only if they were good
originally. That is, invariant, synchronized, etc.
This is necessary on an AMD-based system where after a wakeup from STR I
see that BSP clock differs from AP clocks by a count that roughly
corresponds to one second. The APs are in sync with each other. Not
sure if this is a hardware quirk or a firmware bug.
This is what I see after a resume with this change:
SMP: passed TSC synchronization test after adjustment
acpi_timer0: restoring timecounter, ACPI-fast -> TSC-low
Reviewed by: kib
MFC after: 3 weeks
Differential Revision: https://reviews.freebsd.org/D15551
Instead, construct an auxargs array and copy it out all at once.
Use an array of Elf_Auxinfo rather than pairs of Elf_Addr * to represent
the array. This is the correct type where pairs of words just happend
to work. To reduce the size of the diff, AUXARGS_ENTRY is altered to act
on this array rather than introducing a new macro.
Return errors on copyout() and suword() failures and handle them in the
caller.
Incidentally fixes AT_RANDOM and AT_EXECFN in 32-bit linux on amd64
which incorrectly used AUXARG_ENTRY instead of AUXARGS_ENTRY_32
(now removed due to the use of proper types).
Reviewed by: kib
Comments from: emaste, jhb
Obtained from: CheriBSD
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D15485
We certainly should clear PSL_T when calling the SIGTRAP signal
handler, which is already done by all x86 sendsig(9) ABI code. On the
other hand, there is no obvious reason why PSL_T needs to be cleared
when returning from the signal handler. For instance, Linux allows
userspace to set PSL_T and keep tracing enabled for the desired
period. There are userspace programs which would use PSL_T if we make
it possible, for instance sbcl.
Remember if PSL_T was set by PT_STEP or PT_SETSTEP by mean of TDB_STEP
flag, and only clear it when the flag is set.
Discussed with: Ali Mashtizadeh
Reviewed by: jhb (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D15054
- Add constants for fields in DR6 and the reserved fields in DR7. Use
these constants instead of magic numbers in most places that use DR6
and DR7.
- Refer to T_TRCTRAP as "debug exception" rather than a "trace trap"
as it is not just for trace exceptions.
- Always read DR6 for debug exceptions and only clear TF in the flags
register for user exceptions where DR6.BS is set.
- Clear DR6 before returning from a debug exception handler as
recommended by the SDM dating all the way back to the 386. This
allows debuggers to determine the cause of each exception. For
kernel traps, clear DR6 in the T_TRCTRAP case and pass DR6 by value
to other parts of the handler (namely, user_dbreg_trap()). For user
traps, wait until after trapsignal to clear DR6 so that userland
debuggers can read DR6 via PT_GETDBREGS while the thread is stopped
in trapsignal().
Reviewed by: kib, rgrimes
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D15189
Speculative Store Bypass (SSB) is a speculative execution side channel
vulnerability identified by Jann Horn of Google Project Zero (GPZ) and
Ken Johnson of the Microsoft Security Response Center (MSRC)
https://bugs.chromium.org/p/project-zero/issues/detail?id=1528.
Updated Intel microcode introduces a MSR bit to disable SSB as a
mitigation for the vulnerability.
Introduce a sysctl hw.spec_store_bypass_disable to provide global
control over the SSBD bit, akin to the existing sysctl that controls
IBRS. The sysctl can be set to one of three values:
0: off
1: on
2: auto
Future work will enable applications to control SSBD on a per-process
basis (when it is not enabled globally).
SSBD bit detection and control was verified with prerelease microcode.
Security: CVE-2018-3639
Tested by: emaste (previous version, without updated microcode)
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
When we issue shootdown IPIs, we first assign zero to pm_gens to
indicate the need to flush on the next context switch in case our IPI
misses the context, next we read pm_active. On context switch we set
our bit in pm_active, then we read pm_gen. It is crucial that both
threads see the memory in the program order, otherwise invalidation
thread might read pm_active bit as zero and the context switching
thread might read pm_gen as zero.
IA32 allows CPU for both reads to see zero. We must use the barriers
between write and read. The pm_active bit set is already locked, so
only the invalidation functions need it.
I never saw it in real life, or at least I do not have a good
reproduction case. I found this during code inspection when hunting
for the Xen TLB issue reported by cperciva.
Reviewed by: alc, markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D15506
This turns on support for kernel dump encryption and compression, and
netdump. arm and mips platforms are omitted for now, since they are more
constrained and don't benefit as much from these features.
Reviewed by: cem, manu, rgrimes
Tested by: manu (arm64)
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D15465
Currently, when using dd(1) to take a VM memory image, the capture never ends,
reading zeroes when it's beyond VM system memory max address.
Return EFAULT when trying to read beyond VM system memory max address.
Reviewed by: imp, grehan, anish
Approved by: grehan
Differential Revision: https://reviews.freebsd.org/D15156
Kernel debuggers depend on symbol names to find stack frames with a
trapframe rather than a normal stack frame. The labels used for the
shared interrupt entry point for the PTI and non-PTI cases did not
match the existing patterns confusing debuggers. Add the '.L' prefix
to mark these symbols as local so they are not visible in the symbol
table.
Reviewed by: kib
MFC after: 1 week
Sponsored by: Chelsio Communications
From now on, linking amd64 kernel requires either lld or newer ld.bfd.
Reviewed by: jhb (as part of the large patch)
Discussed with: emaste
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D13838
Adapt assembly generated by clang for memcmp and use it for <= 64 sized
compares (which are the vast majority).
Sample result of doing stats on Broadwell (% of samples):
before: 4.0 kernel bcmp cache_lookup
after : 0.7 kernel bcmp cache_lookup
The routine is most definitely still not optimal. Anyone interested in
spending time improving it is welcome to take over.
Reviewed by: kib
Evaluate cpu_stdext_feature early to have moved link_elf_ireloc() see
correct flags, most important is SMAP.
Tested by: mjg
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D15367
Supposedly, they PG_U bits there were set to easier making some kernel
page accessible to userspace in-place. Since it was not used for the
whole existence of the amd64 pmap.c and current design of the shared
pages prefers double-mapping over the in-place access, remove PG_U
both from the direct map and KVA slots.
Reviewed by: alc, markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
This PML4 page is never used for the userspace process, so there is no
security implications. But the configuration trips SMAP check, which
should be corrected.
Reviewed by: alc, markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Since pop %ss/mov %ss instructions defer all interrupts and exceptions
for the next instruction, it is possible that the userspace watchpoint
trap executes on the first instruction of the kernel entry for
syscall/bpt.
In this case, DB# should be treated similarly to NMI: on amd64 we must
always load GSBASE even if the trap comes from kernel mode, and load
the kernel page table root into %cr3. Moreover, the trap must
use the dedicated stack, because we are still on the user stack when
trapped on syscall entry.
For i386, we must reload %cr3. The syscall instruction is not configured,
so there is no issue with executing on user stack when trapping.
Due to some CPU erratas it is not always possible to detect that the
userspace watchpoint triggered by inspecting %dr6. In trap(), compare the
trap %rip with the known unsafe entry points and if matched pretend that
the watchpoint did not fire at all.
Thank you to the MSRC Incident Response Team, and in particular Greg
Lenti and Nate Warfield, for coordinating the response to this issue
across multiple vendors.
Thanks to Computer Recycling at The Working Center of Kitchener for
making hardware available to allow us to test the patch on additional
CPU families.
Reviewed by: jhb
Discussed with: Matthew Dillon
Tested by: emaste
Sponsored by: The FreeBSD Foundation
Security: CVE-2018-8897
Security: FreeBSD-SA-18:06.debugreg
The parameter is effectively controllable by userspace. It does not matter
what it is set to as it is being passed to copyin - worst case the operation
will just fail.
While here stop computing it unless it is going to be used.
Noted by: dillon@backplane.com
There was a missing trick expanding the passed pattern to a full word
by multiplication. As a side effect non-zero patterns would be
incorrectly laid down.
This stems from the use of rep stosq which is word-sized, while the passed
argument is byte-sized.
I initially repurposed memcpy into memset without taking this into account.
All but non-bzero testing was performed with a variant utilizing ERMS, i.e.
using only stosb which happens to not into the problem whatsoever. So my bad
twice.
Thanks to Oliver Pinter for noting the problem and providing a testcase.
memmove is repurposed bcopy (arguments swapped, return value added)
The libkern variant is a wrapper around bcopy, so this is a big
improvement.
memset is repurposed memcpy. The librkern variant is doing fishy stuff,
including branching on 0 and calling bzero.
Both functions are rather crude and subject to partial depessimization.
This is a soft prerequisite to adding variants utilizing the
'Enhanced REP MOVSB/STOSB' bit and let the kernel patch at runtime.
The code was unnecessarily conditionally copying either 5 or 6 args.
It can blindly copy 6, which also means the size is known at compilation
time and the operation can be depessimized.
Note the entire syscall handling code is rather slow.
Tested on Skylake, sample result for getppid (calls/s):
without pti: 7310106 -> 10653569
with pti: 3304843 -> 4148306
Some syscalls (like read) did not note any difference, other have typically
very modest wins.
Required MD bits are only provided for x86.
Reviewed by: jhb (previous version, as part of the larger patch)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D13838
A large number of devices don't support PCIe FLR, in particular
graphics adapters. Use PCI power management to perform the
reset if FLR fails or isn't available, by cycling the device
through the D3 state.
This has been tested by a number of users with Nvidia and AMD GPUs.
Submitted and tested by: Matt Macy
Reviewed by: jhb, imp, rgrimes
MFC after: 3 weeks
Differential Revision: https://reviews.freebsd.org/D15268
Dumpers may wish to print messages from an initialization hook; this
change ensures that such messages aren't mixed with output from the
generic dump code.
MFC after: 1 week
GCC warns about the potentially confusing use of the binary AND ('&')
operator with a left operand containing an addition expression. (The
confusion would be around the operator precedence between the + and & infix
operators.) The warning is converted into an error with -Werror.
No functional change.
This construct was actually introduced in r328083, but r333059 (re)moved the
closing parentheses.
For reference, see http://en.cppreference.com/w/c/language/operator_precedence .
- Microsemi SCSI driver for PQI controllers.
- Found on newer model HP servers.
- Restrict to AMD64 only as per developer request.
The driver provides support for the new generation of PQI controllers
from Microsemi. This driver is the first SCSI driver to implement the PQI
queuing model and it will replace the aacraid driver for Adaptec Series 9
controllers. HARDWARE Controllers supported by the driver include:
HPE Gen10 Smart Array Controller Family
OEM Controllers based on the Microsemi Chipset.
Submitted by: deepak.ukey@microsemi.com
Relnotes: yes
Sponsored by: Microsemi
Differential Revision: https://reviews.freebsd.org/D14514
hardware will ensure the stack pointer is aligned to a 16-byte
boundary before saving the fault state on the stack.
In the PTI case, handle this potential alignment adjustment by copying
both frames independently while unwinding the stack in between.
Reviewed by: kib
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D15183
Currently both the page lock and a page queue lock must be held in
order to enqueue, dequeue or requeue a page in a given page queue.
The queue locks are a scalability bottleneck in many workloads. This
change reduces page queue lock contention by batching queue operations.
To detangle the page and page queue locks, per-CPU batch queues are
used to reference pages with pending queue operations. The requested
operation is encoded in the page's aflags field with the page lock
held, after which the page is enqueued for a deferred batch operation.
Page queue scans are similarly optimized to minimize the amount of
work performed with a page queue lock held.
Reviewed by: kib, jeff (previous versions)
Tested by: pho
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14893
Remove auxarg_size as it was only used once right after a confusing
assignment in each of the variants of exec_copyout_strings().
Reviewed by: emaste
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D15123
This will allow to hook a ddb script to "kdb.enter.trap" event.
Previously there was no specific name for this event, so it could only
be handled by either "kdb.enter.unknown" or "kdb.enter.default" hooks.
Both are very unspecific.
Having a specific event is useful because the fatal trap condition is
very similar to panic but it has an additional property that the current
stack frame is the frame where the trap occurred. So, both a register
dump and a stack bottom dump have additional information that can help
analyze the problem.
I have added the event only on architectures that have trap_fatal()
function defined. I haven't looked at other architectures. Their
maintainers can add support for the event later.
Sample script:
kdb.enter.trap=bt; show reg; x/aS $rsp,20; x/agx $rsp,20
Reviewed by: kib, jhb, markj
MFC after: 11 days
Sponsored by: Panzura
Differential Revision: https://reviews.freebsd.org/D15093
Half of implementations always failed (returned (-1)) and they were
previously used in only one place.
Reviewed by: kib, andrew
Obtained from: CheriBSD
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D15102
Trampoline mappings are better treated as global since they are valid
in all address spaces, even for PTI. pmap_invalidate_range() must work
on global mappings for pti since kernel_pmap invalidations are really
same as for non-PTI.
Reviewed by: alc, markj
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D15052
The miscellaneous x86 sysent->sv_setregs() implementations tried to
migrate PSL_T from the previous program to the new executed one, but
they evaluated regs->tf_eflags after the whole regs structure was
bzeroed. Make this functional by saving PSL_T value before zeroing.
Note that if the debugger is not attached, executing the first
instruction in the new program with PSL_T set results in SIGTRAP, and
since all intercepted signals are reset to default dispostion on
exec(2), this means that non-debugged process gets killed immediately
if PSL_T is inherited. In particular, since suid images drop
P_TRACED, attempt to set PSL_T for execution of such program would
kill the process.
Another issue with userspace PSL_T handling is that it is reset by
trap(). It is reasonable to clear PSL_T when entering SIGTRAP
handler, to allow the signal to be handled without recursion or
delivery of blocked fault. But it is not reasonable to return back to
the normal flow with PSL_T cleared. This is too late to change, I
think.
Discussed with: bde, Ali Mashtizadeh
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
Differential revision: https://reviews.freebsd.org/D14995
In pti-enabled pmap, the PCID allocation scheme assigns temporal id
for the kernel page table, and user page table twin PCID is
calculating by setting high bit in the kernel PCID. So the kernel AS
is mapped with per-vmspace PCID, and we must completely shut down all
mappings in KVA when switching contexts, so that newly switched thread
would see all changes in KVA occured while it was not executing.
After all, KVA is same between all threads.
Currently the pti context switch for the user part of the page table
gets its TLB entries flushed too. It is excessive. The same PCID
flushing algorithm that is used for non-pti pmap, correctly works for
the UVA mappings. The only shared TLB entries are the pages from KVA
accessed by the kernel entry trampoline. All of them are static
except per-thread TSS and LDT. For TSS and LDT, the lifetime of newly
allocated entries is the whole thread life, so it is fine as well. If
not fine, then explicit shutdowns for current pmap of the newly
allocated LDT and TSS pages would be enough.
Also restore the constant value for the pm_pcid for the kernel_pmap.
Before, for PTI pmap, pm_pcid was erronously rolled same as user
pmap's pm_pcid, but it was not used.
Reviewed by: markj (previous version)
Discussed with: alc
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D14961
Previously linuxulator had three identical copies of
linux_exec_imgact_try. Deduplicate before adding another arch to
linuxulator.
Sponsored by: Turing Robotic Industries Inc
Differential Revision: https://reviews.freebsd.org/D14856
from userland without the need to use sysctls, it allows the old
sysctls to continue to function, but deprecates them at
FreeBSD_version 1200060 (Relnotes for deprecate).
The command line of bhyve is maintained in a backwards compatible way.
The API of libvmmapi is maintained in a backwards compatible way.
The sysctl's are maintained in a backwards compatible way.
Added command option looks like:
bhyve -c [[cpus=]n][,sockets=n][,cores=n][,threads=n][,maxcpus=n]
The optional parts can be specified in any order, but only a single
integer invokes the backwards compatible parse. [,maxcpus=n] is
hidden by #ifdef until kernel support is added, though the api
is put in place.
bhyvectl --get-cpu-topology option added.
Reviewed by: grehan (maintainer, earlier version),
Reviewed by: bcr (manpages)
Approved by: bde (mentor), phk (mentor)
Tested by: Oleg Ginzburg <olevole@olevole.ru> (cbsd)
MFC after: 1 week
Relnotes: Y
Differential Revision: https://reviews.freebsd.org/D9930
SKZ63 Processor May Hang When Executing Code In an HLE Transaction
Region
Problem: Under certain conditions, if the processor acquires an HLE
(Hardware Lock Elision) lock via the XACQUIRE instruction in the Host
Physical Address range between 40000000H and 403FFFFFH, it may hang
with an internal timeout error (MCACOD 0400H) logged into
IA32_MCi_STATUS.
Move the pages from the range into the blacklist. Add a tunable to
not waste 4M if local DoS is not the issue.
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D15001
This is used as part of implementing run control in bhyve's debug
server. The hypervisor now maintains a set of "debugged" CPUs.
Attempting to run a debugged CPU will fail to execute any guest
instructions and will instead report a VM_EXITCODE_DEBUG exit to
the userland hypervisor. Virtual CPUs are placed into the debugged
state via vm_suspend_cpu() (implemented via a new VM_SUSPEND_CPU ioctl).
Virtual CPUs can be resumed via vm_resume_cpu() (VM_RESUME_CPU ioctl).
The debug server suspends virtual CPUs when it wishes them to stop
executing in the guest (for example, when a debugger attaches to the
server). The debug server can choose to resume only a subset of CPUs
(for example, when single stepping) or it can choose to resume all
CPUs. The debug server must explicitly mark a CPU as resumed via
vm_resume_cpu() before the virtual CPU will successfully execute any
guest instructions.
Reviewed by: avg, grehan
Tested on: Intel (jhb), AMD (avg)
Differential Revision: https://reviews.freebsd.org/D14466
opt_compat.h is mentioned in nearly 180 files. In-progress network
driver compabibility improvements may add over 100 more so this is
closer to "just about everywhere" than "only some files" per the
guidance in sys/conf/options.
Keep COMPAT_LINUX32 in opt_compat.h as it is confined to a subset of
sys/compat/linux/*.c. A fake _COMPAT_LINUX option ensure opt_compat.h
is created on all architectures.
Move COMPAT_LINUXKPI to opt_dontuse.h as it is only used to control the
set of compiled files.
Reviewed by: kib, cem, jhb, jtl
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14941
we patted the watchdog approximately once per 4KB page of memory. After
this change, we pat the watchdog approximately once per 128MB of memory.
On a sample machine, this translated to patting the watchdog approximately
every 5.4 seconds, which "seems reasonable". We can choose a different
value in the future, if warranted.
This has extensive field experience. It is a performance improvement, and
has not caused any known problems.
Reviewed by: imp, kib
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D14988
Add the missing breaks in the for loops, in order to exit the loop
when a suitable entry is found.
Also switch amd64 native_start_all_aps to use PHYS_TO_DMAP in order to
find the virtual address of the boot_trampoline and the initial page
tables.
Reported and tested by: pho
Sponsored by: Citrix Systems R&D
So that it doesn't rely on physmap[1] containing an address below
1MiB. Instead scan the full physmap and search for a suitable address
to place the trampoline code (below 1MiB) and the initial memory pages
(below 4GiB).
Sponsored by: Citrix Systems R&D
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D14878
The lcall trampoline enters kernel by int $0x80, which sets up invalid
length of the instruction for %rip rewind.
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Having the IDT entry specify ring 0 DPL caused delivery of #GP instead
of #OF.
The instruction is not valid in 64bit mode, which probably explains
why the IDT entry for #OF was initially set this way. It is
interesting to note that the BOUND instruction works with the IDT #BR
entry DPL 0, most likely CPU considers #BR from BOUND as generated by
a machine, not user.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
x86/cpu_machdep.c now needs to include elan_mmcr.h when CPU_ELAN is set.
While here, also remove the now unneeded inclusion of isareg.h in i386
and amd64 vm_machdep.c.
Reported by: lwhsu
MFC after: 14 days
X-MFC with: r331878
Because I didn't see any reason not too.
I've been making some changes to the code and couldn't help but notice
that the i386 and am64 code was nearly identical.
MFC after: 17 days
If cpu_reset() is called on an AP and if it somehow fails to wake the
BSP, then it's better to attempt the reset on the AP than just sit there
spinning on an unusable and undebuggable system.
MFC after: 16 days
The processor is "parked" in a spin-loop already and that's sufficient
for the reset. There is nothing that stop_cpus() would add here, only
extra complexity and fragility.
The original processor does not need to enable interrupts now, in fact,
it must not do that.
MFC after: 2 weeks
The ocs_fc(4) driver supports the following hardware:
Emulex 16/8G FC GEN 5 HBAS
LPe15004 FC Host Bus Adapters
LPe160XX FC Host Bus Adapters
Emulex 32/16G FC GEN 6 HBAS
LPe3100X FC Host Bus Adapters
LPe3200X FC Host Bus Adapters
The driver supports target and initiator mode, and also supports FC-Tape.
Note that the driver only currently works on little endian platforms. It
is only included in the module build for amd64 and i386, and in GENERIC
on amd64 only.
Submitted by: Ram Kishore Vegesna <ram.vegesna@broadcom.com>
Reviewed by: mav
MFC after: 5 days
Relnotes: yes
Sponsored by: Broadcom
Differential Revision: https://reviews.freebsd.org/D11423
platforms. Original commit message as follows:
Only use CPUs in the domain the device is attached to for default
assignment. Device drivers are able to override the default assignment
if they bind directly. There are severe performance penalties for
handling interrupts on remote CPUs and this should only be done in
very controlled circumstances.
Reviewed by: jhb, kib
Tested by: pho
Sponsored by: Netflix, Dell/EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14838
These have been supplanted by the MI signal information codes in
<sys/signal.h> since 7.0. The FPE_*_TRAP ones were deprecated even
earlier in 1999.
PR: 226579 (exp-run)
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D14637
assignment. Device drivers are able to override the default assignment
if they bind directly. There are severe performance penalties for
handling interrupts on remote CPUs and this should only be done in
very controlled circumstances.
Reviewed by: jhb, kib
Tested by: pho (earlier version)
Sponsored by: Netflix, Dell/EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14838
Current code, which copies the potential syscall arguments into the
current frame, puts an arbitrary limit on the number of syscall
arguments. Apparently, mmap(2) and lseek(2) (?) require larger
number. But there is an issue that stack is only need to be mapped to
contain the number of arguments required by the syscall, so copying
arbitrary large number of words from the stack is not completely safe.
Use different approach to convert lcall frame into int $0x80 frame in
place, by doing the retl in kernel. This also allows to stop proceed
vfork case specially, and stop making assumptions about %cs at the
syscall time.
Also, improve comments with the formulations provided by bde.
Reviewed and tested by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
controlled by the TCP_BLACKBOX option.
Enable this as part of amd64 GENERIC. For now, leave it disabled on
other platforms.
Sponsored by: Netflix, Inc.
Bring #includes closer to style(9) and reduce differences between the
(three) MD versions of linux_machdep.c and linux_sysvec.c.
Sponsored by: Turing Robotic Industries Inc.
ptrace_xstate_info).
struct ptrace_xstate_info has 64bit member but ends up with 32bit
one. As result, on amd64 there is a 32bit padding at the end, but not
on i386.
We must clear the padding before doing the copyout. For compat32 case,
we must copyout the structure which does not have the padding at the
end. The later fixes 32bit gdb display of the YMM registers when
running on amd64 kernel.
Reported by: Vlad Tsyrklevich
Reviewed by: brooks (previous version)
Sponsored by: The FreeBSD Foundation
admbugs: 765
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D14794
On amd64, efi_enter calls fpu_kern_enter(). This may not be called until
fpuinitstate has been invoked, resulting in a kernel panic with
efirt_load="YES" in loader.conf(5).
Move fpuinitstate a little earlier in SI_SUB_DRIVERS so that we can squeeze
efirt between it and efirtc at SI_SUB_DRIVERS, SI_ORDER_ANY. efidev must be
after efirt and doesn't really need to be at SI_SUB_DEVFS, so drop it at
SI_SUB_DRIVER, SI_ORDER_ANY.
The not immediately obvious dependency of fpuinitstate by efirt has been
noted in both places.
Discussed with: kib, andrew
Reported by: Jakob Alvermark <jakob@alvermark.net>
X-MFC-With: r330868
I accidentally swapped 'linux_fixup_elf' to 'linux_elf_fixup' in amd64's
declaration (only), while bringing this change over from git and
encountering a conflict.
assym is only to be included by other .s files, and should never
actually be assembled by itself.
Reviewed by: imp, bdrewery (earlier)
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D14180
context switch code.
Some BIOSes give control to the OS with CR0.WP already set, making the
kernel text read-only before cpu_startup().
Reported by: Peter Lei <peter.lei@ieee.org>
Reviewed by: jtl
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D14768
This is a pure syntax patch to create an interface to enable and later
restore write access to the kernel text and other read-only mapped
regions. It is in line with e.g. vm_fault_disable_pagefaults() by
allowing the nesting.
Discussed with: Peter Lei <peter.lei@ieee.org>
Reviewed by: jtl
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D14768
It's preferable to have a consistent prefix. This also reduces
differences between the three linux*_sysvec.c files.
Sponsored by: Turing Robotic Industries Inc.
There's a fair amount of duplication between MD linuxulator files.
Make indentation and comments consistent between the three versions of
linux_sysvec.c to reduce diffs when comparing them.
Sponsored by: Turing Robotic Industries Inc.
Three copies of the linuxulator linux_sysvec.c contained identical
BSD to Linux errno translation tables, and future work to support other
architectures will also use the same table. Move the table to a common
file to be used by all. Make it 'const int' to place it in .rodata.
(Some existing Linux architectures use MD errno values, but x86 and Arm
share the generic set.)
This change should introduce no functional change; a followup will add
missing errno values.
MFC after: 3 weeks
Sponsored by: Turing Robotic Industries Inc.
Differential Revision: https://reviews.freebsd.org/D14665
This fixes a problem encountered on the Lenovo Thinkpad X220/Yoga 11e where
runtime services would try to inexplicably jump to other parts of memory
where it shouldn't be when attempting to enumerate EFI vars, causing a
panic.
The virtual mapping is enabled by default and can be disabled by setting
efi_disable_vmap in loader.conf(5).
Reviewed by: kib (earlier version)
MFC after: 3 weeks
Differential Revision: https://reviews.freebsd.org/D14677
Migrate to modern types before creating MD Linuxolator bits for new
architectures.
Reviewed by: cem
Sponsored by: Turing Robotic Industries Inc.
Differential Revision: https://reviews.freebsd.org/D14676
correctly for the data contained on each memory page.
There are several components to this change:
* Add a variable to indicate the start of the R/W portion of the
initial memory.
* Stop detecting NX bit support for each AP. Instead, use the value
from the BSP and, if supported, activate the feature on the other
APs just before loading the correct page table. (Functionally, we
already assume that the BSP and all APs had the same support or
lack of support for the NX bit.)
* Set the RW and NX bits correctly for the kernel text, data, and
BSS (subject to some caveats below).
* Ensure DDB can write to memory when necessary (such as to set a
breakpoint).
* Ensure GDB can write to memory when necessary (such as to set a
breakpoint). For this purpose, add new MD functions gdb_begin_write()
and gdb_end_write() which the GDB support code can call before and
after writing to memory.
This change is not comprehensive:
* It doesn't do anything to protect modules.
* It doesn't do anything for kernel memory allocated after the kernel
starts running.
* In order to avoid excessive memory inefficiency, it may let multiple
types of data share a 2M page, and assigns the most permissions
needed for data on that page.
Reviewed by: jhb, kib
Discussed with: emaste
MFC after: 2 weeks
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D14282
therefore, it should be safe to set the NX bit on the PML4E for the
recursive page table mappings. According to the Intel docs, the effect
of the NX bit should propogate to any page reached through a PML4E which
has the NX bit set.
Reviewed by: kib, markj
MFC after: 2 weeks
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D14333
first available virtual address to a 2MB boundary. After r329071,
create_pagetables() rounds firstaddr up to a 2MB boundary. This ensures
the kernel is mapped in super-pages, which is the point of the logic
in pmap_kmem_choose(). Therefore, it is no longer necessary for
pmap_bootstrap() to round up to the 2MB boundary again.
As pmap_bootstrap() was the only user of pmap_kmem_choose(), we can
delete pmap_kmem_choose().
Reviewed by: kib
MFC after: 2 weeks
X-MFC-with: r329071
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D14355
Rename ACPI_IVRS_HARDWARE_NEW to ACPI_IVRS_HARDWARE_EFRSUP, since new definitions add Extended Feature Register support. Use IvrsType to distinguish three types of IVHD - 0x10(legacy), 0x11 and 0x40(with EFR). IVHD 0x40 is also called mixed type since it supports HID device entries.
Fix 2 coverity bugs reported by cem.
Reported by:jkim, cem
Approved by:grehan
Differential Revision://reviews.freebsd.org/D14501
Since that change the system call stack traces look like this:
...
sys___sysctl() at sys___sysctl+0x5f/frame 0xfffffe0028e13ac0
amd64_syscall() at amd64_syscall+0x79b/frame 0xfffffe0028e13bf0
fast_syscall_common() at fast_syscall_common+0x101/frame 0xfffffe0028e13bf0
So, db_nextframe() stopped recognizing the system call frame.
This commit should fix that.
Reviewed by: kib
MFC after: 4 days
imcsmb(4) provides smbus(4) support for the SMBus controller functionality
in the integrated Memory Controllers (iMCs) embedded in Intel Sandybridge-
Xeon, Ivybridge-Xeon, Haswell-Xeon, and Broadwell-Xeon CPUs. Each CPU
implements one or more iMCs, depending on the number of cores; each iMC
implements two SMBus controllers (iMC-SMBs).
*** IMPORTANT NOTE ***
Because motherboard firmware or the BMC might try to use the iMC-SMBs for
monitoring DIMM temperatures and/or managing an NVDIMM, the driver might
need to temporarily disable those functions, or take a hardware interlock,
before using the iMC-SMBs. Details on how to do this may vary from board to
board, and the procedure may be proprietary. It is strongly suggested that
anyone wishing to use this driver contact their motherboard vendor, and
modify the driver as described in the manual page and in the driver itself.
(For what it's worth, the driver as-is has been tested on various SuperMicro
motherboards.)
Reviewed by: avg, jhb
MFC after: 1 week
Relnotes: yes
Sponsored by: Panasas
Differential Revision: https://reviews.freebsd.org/D14447
Discussed with: avg, ian, jhb
Tested by: allanjude (previous version), Panasas
Many licenses on Linuxolator files contained small variations from the
standard FreeBSD license text. To avoid license proliferation switch to
the standard 2-clause FreeBSD license for those files where I have
permission from each of the listed copyright holders. Additional files
still waiting on permission from others are listed in review D14210.
Approved by: dchagin, rdivacky, sos
MFC after: 1 week
MFC with: r329370
Sponsored by: The FreeBSD Foundation
Unlike the existing GLA2GPA ioctl, GLA2GPA_NOFAULT does not modify
the guest. In particular, it does not inject any faults or modify
PTEs in the guest when performing an address space translation.
This is used by bhyve's debug server to read and write memory for
the remote debugger.
Reviewed by: grehan
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D14075
The conditional compilation support is now centralized in
tcp_fastopen.h and tcp_var.h. This doesn't provide the minimum
theoretical code/data footprint when TCP_RFC7413 is disabled, but
nearly all the TFO code should wind up being removed by the optimizer,
the additional footprint in the syncache entries is a single pointer,
and the additional overhead in the tcpcb is at the end of the
structure.
This enables the TCP_RFC7413 kernel option by default in amd64 and
arm64 GENERIC.
Reviewed by: hiren
MFC after: 1 month
Sponsored by: Limelight Networks
Differential Revision: https://reviews.freebsd.org/D14048
When I wrote the patch, I wanted to remove SYSINIT() usage from amd64 code.
There is no reason to keep the divergence any more because iwasaki merged
most amd64 suspend/resume code to i386 with r235622. Note this also fixed
an enge case reported by royger. [1]
Suggested by: jhb
Reviewed by: royger
Tested by: royger [1]
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D14400 [1]
No implementation of fpu_kern_enter() can fail, and it was causing needless
error checking boilerplate and confusion. Change the return code to void to
match reality.
(This trivial change took nine days to land because of the commit hook on
sys/dev/random. Please consider removing the hook or otherwise lowering the
bar -- secteam never seems to have free time to review patches.)
Reported by: Lachlan McIlroy <Lachlan.McIlroy AT isilon.com>
Reviewed by: delphij
Approved by: secteam (delphij)
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14380
These are a convenience for bhyve's debug server to use a single
ioctl for 'g' and 'G' rather than a loop of individual get/set
ioctl requests.
Reviewed by: grehan
MFC after: 2 months
Differential Revision: https://reviews.freebsd.org/D14074
Make vm_wait() take the vm_object argument which specifies the domain
set to wait for the min condition pass. If there is no object
associated with the wait, use curthread' policy domainset. The
mechanics of the wait in vm_wait() and vm_wait_domain() is supplied by
the new helper vm_wait_doms(), which directly takes the bitmask of the
domains to wait for passing min condition.
Eliminate pagedaemon_wait(). vm_domain_clear() handles the same
operations.
Eliminate VM_WAIT and VM_WAITPFAULT macros, the direct functions calls
are enough.
Eliminate several control state variables from vm_domain, unneeded
after the vm_wait() conversion.
Scetched and reviewed by: jeff
Tested by: pho
Sponsored by: The FreeBSD Foundation, Mellanox Technologies
Differential revision: https://reviews.freebsd.org/D14384
Many licenses on Linuxolator files contained small variations from the
standard FreeBSD license text. To avoid license proliferation switch to
the standard 2-clause FreeBSD license for those files where I have
permission from each of the listed copyright holders. Additional files
waiting on permission from others are listed in review D14210.
Approved by: kan, marcel, sos, rdivacky
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Small global symbols confuse ddb which matches them against small
unrelated displacements and makes the disassembly ugly.
Reported by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
IVRS can have entry of type legacy and non-legacy present at same time for same AMD-Vi device. ivhd driver will ignore legacy if new IVHD type is present as specified in AMD-Vi specification. Earlier both of IVHD entries used and two ivhd devices were created.
Add support for new IVHD type 0x11 and 0x40 in ACPI. Create new struct of type acpi_ivrs_hardware_new for these new type of IVHDs. Legacy type 0x10 will continue to use acpi_ivrs_hardware.
Reviewed by: avg
Approved by: grehan
Differential Revision:https://reviews.freebsd.org/D13160
The idea is, the pmap_qenter() API is now defined to not produce executable
mappings. If you need executable mappings, use another API.
Add pg_nx flag in pmap_qenter on x86 to make kernel pages non-executable.
Other architectures that support execute-specific permissons on page table
entries should subsequently be updated to match.
Submitted by: Darrick Lew <darrick.freebsd AT gmail.com>
Reviewed by: markj
Discussed with: alc, jhb, kib
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14062
compilation under FreeBSD. The mthca driver was temporarily removed as
part of the Linux 4.9 RoCE/infinband upgrade.
Top commit in Linux source tree:
69973b830859bc6529a7a0468ba0d80ee5117826
Sponsored by: Mellanox Technologies
significant source of cache line contention from vm_page_alloc(). Use
accessors and vm_page_unwire_noq() so that the mechanism can be easily
changed in the future.
Reviewed by: markj
Discussed with: kib, glebius
Tested by: pho (earlier version)
Sponsored by: Netflix, Dell/EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14273
makes them consistent with the way other page-table pages are allocated.
It also provides the rest of the VM system a good clue that these pages
are used.
Reviewed by: alc, kib, markj
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D14269
mappings for the pages used for the kernel and some initial allocations
used for the page table. It maps the kernel and the blocks used for
these initial allocations using 2MB pages.
However, if the kernel does not end on a 2MB boundary, it still maps the
last portion using a 2MB page, but reports that the unused 4K blocks
within this 2MB allocation are free physical blocks. This means that
these same physical blocks could also be mapped elsewhere - for example,
into a user process. Given the proximity to the kernel text and data
area, it seems wise to avoid allowing someone to write data to physical
blocks also mapped into these virtual addresses.
(Note that this isn't a security vulnerability: the direct map makes
most/all memory on the system mapped into kernel space. And, nothing
in the kernel should be trying to access these pages, as the virtual
addresses are unused. It simply seems wise to avoid reusing these
physical blocks while they are mapped to virtual addresses so close
to the kernel text and data area.)
Consequently, let's reserve the physical blocks covered by the
page-table mappings for these initial allocations.
Reviewed by: kib, markj
MFC after: 2 weeks
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D14268
global to per-domain state. Protect reservations with the free lock
from the domain that they belong to. Refactor to make vm domains more
of a first class object.
Reviewed by: markj, kib, gallatin
Tested by: pho
Sponsored by: Netflix, Dell/EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14000
A version of each of the MD files by necessity exists for each CPU
architecture supported by the Linuxolator. Clean these up so that new
architectures do not inherit whitespace issues.
Clean up shared Linuxolator files while here.
Sponsored by: Turing Robotic Industries Inc.
Branch Predictors) mitigation.
DOcument 336996-001 promises that CPUs which implement IBRS but not
STIBP silently ignore setting of the bit instead of trapping.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
It is coded according to the Intel document 336996-001, reading of the
patches posted on lkml, and some additional consultations with Intel.
For existing processors, you need a microcode update which adds IBRS
CPU features, and to manually enable it by setting the tunable/sysctl
hw.ibrs_disable to 0. Current status can be checked in sysctl
hw.ibrs_active. The mitigation might be inactive if the CPU feature
is not patched in, or if CPU reports that IBRS use is not required, by
IA32_ARCH_CAP_IBRS_ALL bit.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D14029
The virtual interrupt method uses V_IRQ, V_INTR_PRIO, and V_INTR_VECTOR
fields of VMCB to inject a virtual interrupt into a guest VM. This
method has many advantages over the direct event injection as it
offloads all decisions of whether and when the interrupt can be
delivered to the guest. But with a purely software emulated vAPIC the
advantage is also a problem. The problem is that the hypervisor does
not have any precise control over when the interrupt is actually
delivered to the guest (or a notification about that). Because of that
the hypervisor cannot update the interrupt vector in IRR and ISR in the
same way as real hardware would. The hypervisor becomes aware that the
interrupt is being serviced only upon the first VMEXIT after the
interrupt is delivered. This creates a window between the actual
interrupt delivery and the update of IRR and ISR. That means that IRR
and ISR might not be correctly set up to the point of the
end-of-interrupt signal.
The described deviation has been observed to cause an interrupt loss in
the following scenario. vCPU0 posts an inter-processor interrupt to
vCPU1. The interrupt is injected as a virtual interrupt by the
hypervisor. The interrupt is delivered to a guest and an interrupt
handler is invoked. The handler performs a requested action and
acknowledges the request by modifying a global variable. So far, there
is no VMEXIT and the hypervisor is unaware of the events. Then, vCPU0
notices the acknowledgment and sends another IPI with the same vector.
The IPI gets collapsed into the previous IPI in the IRR of vCPU1. Only
after that a VMEXIT of vCPU1 occurs. At that time the vector is cleared
in the IRR and is set in the ISR. vCPU1 has vAPIC state as if the
second IPI has never been sent.
The scenario is impossible on the real hardware because IRR and ISR are
updated just before the interrupt handler gets started.
I saw several possibilities of fixing the problem. One is to intercept
the virtual interrupt delivery to update IRR and ISR at the right
moment. The other is to deliver the LAPIC interrupts using the event
injection, same as legacy interrupts. I opted to use the latter
approach for several reasons. It's equivalent to what VMM/Intel does
(in !VMX case). It appears to be what VirtualBox and KVM do. The code
is already there (to support legacy interrupts).
Another possibility was to use a special intermediate state for a vector
after it is injected using a virtual interrupt and before it is known
whether it was accepted or is still pending.
That approach was implemented in https://reviews.freebsd.org/D13828
That method is more complex and does not have any clear advantage.
Please see sections 15.20 and 15.21.4 of "AMD64 Architecture
Programmer's Manual Volume 2: System Programming" (publication 24593,
revision 3.29) for comparison between event injection and virtual
interrupt injection.
PR: 215972
Reported by: ajschot@hotmail.com, grehan
Tested by: anish, grehan, Nils Beyer <nbe@renzel.net>
Reviewed by: anish, grehan
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D13780
This avoids a nested page fault when obtaining a stack trace in DDB if
the address from the first frame does not resolve to a known symbol.
MFC after: 1 week
Sponsored by: Chelsio Communications
Use PCID to avoid complete TLB shootdown when switching between user
and kernel mode with PTI enabled.
I use the model close to what I read about KAISER, user-mode PCID has
1:1 correspondence to the kernel-mode PCID, by setting bit 11 in PCID.
Full kernel-mode TLB shootdown is performed on context switches, since
KVA TLB invalidation only works in the current pmap. User-mode part of
TLB is flushed on the pmap activations as well.
Similarly, IPI TLB shootdowns must handle both kernel and user address
spaces for each address. Note that machines which implement PCID but
do not have INVPCID instructions, cause the usual complications in the
IPI handlers, due to the need to switch to the target PCID temporary.
This is racy, but because for PCID/no-INVPCID we disable the
interrupts in pmap_activate_sw(), IPI handler cannot see inconsistent
state of CPU PCID vs PCPU pmap/kcr3/ucr3 pointers.
On the other hand, on kernel/user switches, CR3_PCID_SAVE bit is set
and we do not clear TLB.
I can imagine alternative use of PCID, where there is only one PCID
allocated for the kernel pmap. Then, there is no need to shootdown
kernel TLB entries on context switch. But copyout(3) would need to
either use method similar to proc_rwmem() to access the userspace
data, or (in reverse) provide a temporal mapping for the kernel buffer
into user mode PCID and use trampoline for copy.
Reviewed by: markj (previous version)
Tested by: pho
Discussed with: alc (some aspects)
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
Differential revision: https://reviews.freebsd.org/D13985
These files previously had a 3-clause license and 'THE REGENTS' text.
Switch to standard 2-clause text with kib's approval, and add the SPDX
tag.
Approved by: kib
Uses of mallocarray(9).
The use of mallocarray(9) has rocketed the required swap to build FreeBSD.
This is likely caused by the allocation size attributes which put extra pressure
on the compiler.
Given that most of these checks are superfluous we have to choose better
where to use mallocarray(9). We still have more uses of mallocarray(9) but
hopefully this is enough to bring swap usage to a reasonable level.
Reported by: wosch
PR: 225197
still active.
Map userspace portion of VA in the PTI kernel-mode page table as
non-executable. This way, if we ever miss reloading ucr3 into %cr3 on
the return to usermode, the process traps instead of executing in
potentially vulnerable setup. Catch the condition of such trap and
verify user-mode %cr3, which is saved by page fault handler.
I peek this trick in some article about Linux implementation.
Reviewed by: alc, markj (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 12 days
DIfferential revision: https://reviews.freebsd.org/D13956
kernel by PHYS_TO_DMAP() as previously present on amd64, arm64, riscv, and
powerpc64. This introduces a new MI macro (PMAP_HAS_DMAP) that can be
evaluated at runtime to determine if the architecture has a direct map;
if it does not (or does) unconditionally and PMAP_HAS_DMAP is either 0 or
1, the compiler can remove the conditional logic.
As part of this, implement PHYS_TO_DMAP() on sparc64 and mips64, which had
similar things but spelled differently. 32-bit MIPS has a partial direct-map
that maps poorly to this concept and is unchanged.
Reviewed by: kib
Suggestions from: marius, alc, kib
Runtime tested on: amd64, powerpc64, powerpc, mips64
Kernel Page Table Isolation (KPTI) was introduced in r328083 as a
mitigation for the 'Meltdown' vulnerability. AMD CPUs are not affected,
per https://www.amd.com/en/corporate/speculative-execution:
We believe AMD processors are not susceptible due to our use of
privilege level protections within paging architecture and no
mitigation is required.
Thus default KPTI to off for AMD CPUs, and to on for others. This may
be refined later as we obtain more specific information on the sets of
CPUs that are and are not affected.
Submitted by: Mitchell Horne
Reviewed by: cem
Relnotes: Yes
Security: CVE-2017-5754
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D13971
Similar to NMIs, machine check exceptions can fire at any time and are
not masked by IF. This means that machine checks can fire when the
kstack is too deep to hold a trap frame, or at critical sections in
trap handlers when a user %gs is used with a kernel %cs. Use the same
strategy used for NMIs of using a dedicated per-CPU stack configured
in IST 3. Store the CPU's pcpu pointer at the stop of the stack so
that the machine check handler can reliably find the proper value for
%gs (also borrowed from NMIs).
This should also fix a similar issue with PTI with a MC# occurring
while the CPU is executing on the trampoline stack.
While here, bypass trap() entirely and just call mca_intr(). This
avoids a bogus call to kdb_reenter() (there's no reason to try to
reenter kdb if a MC# is raised).
Reviewed by: kib
Tested by: avg (on AMD without PTI)
Differential Revision: https://reviews.freebsd.org/D13962
In the !PTI case the NMI handler jumped past the instructions that set
%rdi to point to the current PCB, but the target instructions assumed %rdi
were set.
Reviewed by: kib
Tested by: pho
Apparently machinde/cpu.h is supposed to contain MD implementations of
MI interfaces. Also, remove kernphys declaration from machdep.c,
since it is already provided by md_var.h.
Requested and reviewed by: bde
MFC after: 13 days
Currently most of the debug registers are not saved and restored
during VM transitions allowing guest and host debug register values to
leak into the opposite context. One result is that hardware
watchpoints do not work reliably within a guest under VT-x.
Due to differences in SVM and VT-x, slightly different approaches are
used.
For VT-x:
- Enable debug register save/restore for VM entry/exit in the VMCS for
DR7 and MSR_DEBUGCTL.
- Explicitly save DR0-3,6 of the guest.
- Explicitly save DR0-3,6-7, MSR_DEBUGCTL, and the trap flag from
%rflags for the host. Note that because DR6 is "software" managed
and not stored in the VMCS a kernel debugger which single steps
through VM entry could corrupt the guest DR6 (since a single step
trap taken after loading the guest DR6 could alter the DR6
register). To avoid this, explicitly disable single-stepping via
the trace flag before loading the guest DR6. A determined debugger
could still defeat this by setting a breakpoint after the guest DR6
was loaded and then single-stepping.
For SVM:
- Enable debug register caching in the VMCB for DR6/DR7.
- Explicitly save DR0-3 of the guest.
- Explicitly save DR0-3,6-7, and MSR_DEBUGCTL for the host. Since SVM
saves the guest DR6 in the VMCB, the race with single-stepping
described for VT-x does not exist.
For both platforms, expose all of the guest DRx values via --get-drX
and --set-drX flags to bhyvectl.
Discussed with: avg, grehan
Tested by: avg (SVM), myself (VT-x)
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D13229
The implementation of the Kernel Page Table Isolation (KPTI) for
amd64, first version. It provides a workaround for the 'meltdown'
vulnerability. PTI is turned off by default for now, enable with the
loader tunable vm.pmap.pti=1.
The pmap page table is split into kernel-mode table and user-mode
table. Kernel-mode table is identical to the non-PTI table, while
usermode table is obtained from kernel table by leaving userspace
mappings intact, but only leaving the following parts of the kernel
mapped:
kernel text (but not modules text)
PCPU
GDT/IDT/user LDT/task structures
IST stacks for NMI and doublefault handlers.
Kernel switches to user page table before returning to usermode, and
restores full kernel page table on the entry. Initial kernel-mode
stack for PTI trampoline is allocated in PCPU, it is only 16
qwords. Kernel entry trampoline switches page tables. then the
hardware trap frame is copied to the normal kstack, and execution
continues.
IST stacks are kept mapped and no trampoline is needed for
NMI/doublefault, but of course page table switch is performed.
On return to usermode, the trampoline is used again, iret frame is
copied to the trampoline stack, page tables are switched and iretq is
executed. The case of iretq faulting due to the invalid usermode
context is tricky, since the frame for fault is appended to the
trampoline frame. Besides copying the fault frame and original
(corrupted) frame to kstack, the fault frame must be patched to make
it look as if the fault occured on the kstack, see the comment in
doret_iret detection code in trap().
Currently kernel pages which are mapped during trampoline operation
are identical for all pmaps. They are registered using
pmap_pti_add_kva(). Besides initial registrations done during boot,
LDT and non-common TSS segments are registered if user requested their
use. In principle, they can be installed into kernel page table per
pmap with some work. Similarly, PCPU can be hidden from userspace
mapping using trampoline PCPU page, but again I do not see much
benefits besides complexity.
PDPE pages for the kernel half of the user page tables are
pre-allocated during boot because we need to know pml4 entries which
are copied to the top-level paging structure page, in advance on a new
pmap creation. I enforce this to avoid iterating over the all
existing pmaps if a new PDPE page is needed for PTI kernel mappings.
The iteration is a known problematic operation on i386.
The need to flush hidden kernel translations on the switch to user
mode make global tables (PG_G) meaningless and even harming, so PG_G
use is disabled for PTI case. Our existing use of PCID is
incompatible with PTI and is automatically disabled if PTI is
enabled. PCID can be forced on only for developer's benefit.
MCE is known to be broken, it requires IST stack to operate completely
correctly even for non-PTI case, and absolutely needs dedicated IST
stack because MCE delivery while trampoline did not switched from PTI
stack is fatal. The fix is pending.
Reviewed by: markj (partially)
Tested by: pho (previous version)
Discussed with: jeff, jhb
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Focus on code where we are doing multiplications within malloc(9). None of
these ire likely to overflow, however the change is still useful as some
static checkers can benefit from the allocation attributes we use for
mallocarray.
This initial sweep only covers malloc(9) calls with M_NOWAIT. No good
reason but I started doing the changes before r327796 and at that time it
was convenient to make sure the sorrounding code could handle NULL values.
X-Differential revision: https://reviews.freebsd.org/D13837
domains can be done by the _domain() API variants. UMA also supports a
first-touch policy via the NUMA zone flag.
The slab layer is now segregated by VM domains and is precise. It handles
iteration for round-robin directly. The per-cpu cache layer remains
a mix of domains according to where memory is allocated and freed. Well
behaved clients can achieve perfect locality with no performance penalty.
The direct domain allocation functions have to visit the slab layer and
so require per-zone locks which come at some expense.
Reviewed by: Attilio (a slightly older version)
Tested by: pho
Sponsored by: Netflix, Dell/EMC Isilon
We already clear %RFLAGS.DF on the kernel entry due to the compiler's
ABI requirements.
Suggested by: jhb
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Hardware already did it for us due to the mask loaded into the
MSR_SF_MASK msr register.
Reviewed by: jhb
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D13838
The symbol is just an offset in the hardware TSS structure, it is not
limited to the common_tss instance.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
This is a followup to r307903.
struct svm_softc takes more than 200 kilobytes while what we really need
is 3 contiguous pages for I/O permission map and 2 contiguous pages for
MSR permission map. Other physically mapped structures have a size of
a single page, so a proper alignment is sufficient for their correct
mapping.
Thus, only the permission maps are allocated with contigmalloc now,
the softc is allocated with a regular malloc.
Additionally, this commit adds a check that malloc returns memory with the
expected page alignment and that contigmalloc does not fail.
Unfortunately, at present svm_vminit() is expected to always succeed and
there is no way to report an error.
So, a contigmalloc failure leads to a panic.
We should probably fix this.
MFC after: 2 weeks
Add cpuctl(4) ioctl CPUCTL_EVAL_CPU_FEATURES which forces re-read of
cpu_features, cpu_features2, cpu_stdext_features, and
std_stdext_features2.
The intent is to allow the kernel to see the changes in the CPU
features after micocode update. Of course, the update is not atomic
across variables and not synchronized with readers. See the man page
warning as well.
Reviewed by: imp (previous version), jilles
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D13770
It does not change anything in the behavior of trap_pfault(), while
eliminating obfuscation of jumping to the code which checks for the
condition reversed of the goto cause. Also avoid force initialize the
rv variable, since it is now only accessed after storing vm_fault()
return value.
Reviewed by: alc
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D13725
The entry must be logged "manually" using TSRAW rather than TSENTER
since PCPU data structures have not yet been initialized and thus
curthread cannot be accessed; &thread0 is what will become curthread
later in hammer_time.
Other MD initialization code should be similarly instrumented in order
to gain visibility into the time spent before entering mi_startup; this
will require some care and testing from people with access to such
hardware.
They provide relaxed-ordered atomic access semantic. Due to the
FreeBSD memory model, the operations are syntaxical wrappers around
the volatile accesses. The volatile qualifier is used to ensure that
the access not optimized out and in turn depends on the volatile
semantic as implemented by supported compilers.
The motivation for adding the operation is to help people coming from
other systems or knowing the C11/C++ standards where atomics have
special type and require use of the special access operations. It is
still the case that FreeBSD requires plain load and stores of aligned
integer types to be atomic.
Suggested by: jhb
Reviewed by: alc, jhb
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D13534
The DTrace fasttrap entry points expect a struct reg containing the
register values of the calling thread. Perform the conversion in
fasttrap rather than in the trap handler: this reduces the number of
ifdefs and avoids wasting stack space for traps that don't involve
DTrace.
MFC after: 2 weeks
This variable should be pure MI except possibly for reading it in MD
dump routines. Its initialization was pure MD in 4.4BSD, but FreeBSD
changed this in r36441 in 1998. There were many imperfections in
r36441. This commit fixes only a small one, to simplify fixing the
others 1 arch at a time. (r47678 added support for
special/early/multiple message buffer initialization which I want in
a more general form, but this was too fragile to use because hacking
on the msgbufp global corrupted it, and was only used for 5 hours in
-current...)
This is better than directly changing PCI configuration space of the
device because it makes the PCI bus aware of the configuration.
Also, the change allows to drop a bunch of code that duplicated
pci_enable_msi() functionality.
I wonder if it's possible to further simplify the code by using
pci_alloc_msi().
ivhd should attach after the root PCI bus and, thus, after the ACPI
Host-PCI bridge off which the bus hangs. This is because ivhd changes
PCI configuration of a PCI IOMMU device that is located on the root bus.
If the bus attaches after ivhd it clears the MSI portion of the
configuration. As a result IOMMU event interrupts would never be
delivered.
For regular ACPI devices the order is calculated as
ACPI_DEV_BASE_ORDER + level * 10
where level is a depth of the device in the ACPI namespace.
I expect the depth of the Host-PCI bridge to be two or three,
so ACPI_DEV_BASE_ORDER + 10 * 10 should be a sufficiently safe order
for ivhd.
This should fix the setup of the AMD-Vi event interrupt when vmm is
preloaded (as opposed to kldload-ed).
This ensures that we can receive further event interrupts.
See the description of the bits in the specification for
MMIO Offset 2020h IOMMU Status Register.
The bits are defined as set-by-hardware write-1-to-clear, same as all
the bits in the status register.
Discussed with: anish
