When I did the use_numa support, I missed the fact that there is
a separate hash function for send tag nic selection. So when
use_numa is enabled, ktls offload does not work properly, as it
does not reliably allocate a send tag on the proper egress nic
since different egress nics are selected for send-tag allocation
and packet transmit. To fix this, this change:
- refectors lacp_select_tx_port_by_hash() and
lacp_select_tx_port() to make lacp_select_tx_port_by_hash()
always called by lacp_select_tx_port()
- pre-shifts flowids to convert them to hashes when calling lacp_select_tx_port_by_hash()
- adds a numa_domain field to if_snd_tag_alloc_params
- plumbs the numa domain into places where we allocate send tags
In testing with NIC TLS setup on a NUMA machine, I see thousands
of output errors before the change when enabling
kern.ipc.tls.ifnet.permitted=1. After the change, I see no
errors, and I see the NIC sysctl counters showing active TLS
offload sessions.
Reviewed by: rrs, hselasky, jhb
Sponsored by: Netflix
This silences an "unused label" warning as well as fixes the attach fail
path that wasn't releasing resources.
Submitted by: Nicholas O'Brien <nickisobrien_gmail.com>
Sponsored by: Axiado
Differential Revision: https://reviews.freebsd.org/D24004
Before skipping the current cpu when trying to find the ones that
have the same opp, record that this one have this opp.
Reported by: mmel
MFC after: 2 weeks
X-MFC-With: r358555
We were reusing a structure for multiple operations, but failing to
reinitialize one member. The result is that a server that cares about FUSE
file handle IDs would see one correct FUSE_FSYNC operation, and one with the
FHID unset.
PR: 244431
Reported by: Agata <chogata@gmail.com>
MFC after: 2 weeks
The vectx API, computes the hash for verifying a file as it is read.
This avoids the overhead of reading files twice - once to verify, then
again to load.
For doing an install via loader, avoiding the need to rewind
large files is critical.
This API is only used for modules, kernel and mdimage as these are the
biggest files read by the loader.
The reduction in boot time depends on how expensive the I/O is
on any given platform. On a fast VM we see 6% improvement.
For install via loader the first file to be verified is likely to be the
kernel, so some of the prep work (finding manifest etc) done by
verify_file() needs to be factored so it can be reused for
vectx_open().
For missing or unrecognized fingerprint entries, we fail
in vectx_open() unless verifying is disabled.
Otherwise fingerprint check happens in vectx_close() and
since this API is only used for files which must be verified
(VE_MUST) we panic if we get an incorrect hash.
Reviewed by: imp,tsoome
MFC after: 1 week
Sponsored by: Juniper Networks
Differential Revision: https://reviews.freebsd.org//D23827
The vectx API, computes the hash for verifying a file as it is read.
This avoids the overhead of reading files twice - once to verify, then
again to load.
For doing an install via loader, avoiding the need to rewind
large files is critical.
This API is only used for modules, kernel and mdimage as these are the
biggest files read by the loader.
The reduction in boot time depends on how expensive the I/O is
on any given platform. On a fast VM we see 6% improvement.
For install via loader the first file to be verified is likely to be the
kernel, so some of the prep work (finding manifest etc) done by
verify_file() needs to be factored so it can be reused for
vectx_open().
For missing or unrecognized fingerprint entries, we fail
in vectx_open() unless verifying is disabled.
Otherwise fingerprint check happens in vectx_close() and
since this API is only used for files which must be verified
(VE_MUST) we panic if we get an incorrect hash.
Reviewed by: imp,tsoome
MFC after: 1 week
Sponsored by: Juniper Networks
Differential Revision: https://reviews.freebsd.org//D23827
because it clobbers the super speed link status when a device is in super
speed mode. Currently the power bit is not needed for anything in the USB
hub driver.
This fixes USB warm reset for super speed devices.
Tested by: Shichun.Ma@dell.com
MFC after: 3 days
Sponsored by: Mellanox Technologies
This has a side effect of eliminating filedesc slock/sunlock during path
lookup, which in turn removes contention vs concurrent modifications to the fd
table.
Reviewed by: markj, kib
Differential Revision: https://reviews.freebsd.org/D23889
This makes it easier to write libkvm programs that access UMA data
structures.
- Remove a couple of unused slab functions and make others local to
uma_core.c. Similarly move SLAB_BITSETS, which affects the layout of
slab structures, to uma_core.c.
- Stop defining the slab structures under _KERNEL. There's no real
reason they can't be visible to userspace like the rest of UMA's
structures are.
- Group KEG_ASSERT_COLD with other keg macros.
- Convert an assertion about MAXMEMDOM to use _Static_assert.
No functional change intended.
Discussed with: jeff
Reviewed by: rlibby
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23980
Summary:
Allow src.conf to override the inferred COMPAT_ARCH and COMPAT_CPUTYPE
variables, such that a different CPU target can be specified explicitly
for the general target vs the compat target.
Reviewed by: imp
Differential Revision: https://reviews.freebsd.org/D23992
The intent is to provide a header that can be included by other headers
without introducing too much pollution. smr.h depends on various
headers and will likely grow over time, but is less likely to be
required by system headers.
Rename SMR_TYPE_DECLARE() to SMR_POINTER():
- One might use SMR to protect more than just pointers; it
could be used for resizeable arrays, for example, so TYPE seems too
generic.
- It is useful to be able to define anonymous SMR-protected pointer
types and the _DECLARE suffix makes that look wrong.
Reviewed by: jeff, mjg, rlibby
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23988
Replace hardcoded sizes by nitems and sizeof
Replace CTLFLAG_NEEDGIANT with CTLFLAG_MPSAFE, I run this driver since a few
years with CTLFLAG_MPSAFE w/o issues.
Replace hardcoded sizes by nitems and sizeof
Replace CTLFLAG_NEEDGIANT with CTLFLAG_MPSAFE, I run this driver since a few
years with CTLFLAG_MPSAFE w/o issues.
Add a HACK to handle a special case for a sensor location.
Since switch to the lockless grab, shared busy for ahead/behind pages
allows other threads to validate and map the pages readonly.
Reviewed by: avg, jeff, markj
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D23986
32-bit Book-E doesn't set UMA_MD_SMALL_ALLOC, and 32-bit OEA platforms
have a 32-bit vm_paddr_t. Moreover, this code was wrong in that it
leaked the page if the check failed.
Reviewed by: jhibbits
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23991
The aim is to reduce the boilerplate needed today to define and
initialize global counters. Also add SI_SUB_COUNTER to the sysinit
ordering.
Reviewed by: kib
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23977
file systems to safely access their disk devices, and adapt FFS to use it.
Also add a new BO_NOBUFS flag to allow enforcing that file systems using
mntfs vnodes do not accidentally use the original devfs vnode to create buffers.
Reviewed by: kib, mckusick
Approved by: imp (mentor)
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D23787
Reland r374450 with Richard Smith's comments and test fixed.
The behavior from the original patch has changed, since we're no
longer allowing LLVM to just ignore the alignment. Instead, we're
just assuming the maximum possible alignment.
Differential Revision: https://reviews.llvm.org/D68824
llvm-svn: 374562
This fixes 'Assertion failed: (Alignment != 0 && "Invalid Alignment"),
function CreateAlignmentAssumption', when building recent versions of
v8, which invoke __builtin_assume_aligned() with its alignment argument
set to 4GiB or more.
Clang will now report a warning, and show the maximum possible alignment
instead, e.g.:
huge-align.cpp:1:27: warning: requested alignment must be 536870912 bytes or smaller; maximum alignment assumed [-Wbuiltin-assume-aligned-alignment]
void *f(void *g) { return __builtin_assume_aligned(g, 4294967296); }
^ ~~~~~~~~~~
Upstream PR: https://bugs.llvm.org/show_bug.cgi?id=43839
Reported by: cem
MFC after: 3 days
- Add more registers needed by bhyve [1]
- Move EL2 registers from armreg.h to hypervisor.h
- Add the register name to hypervisor.h
Obtained from: https://github.com/FreeBSD-UPB/freebsd [1]
This fixes some errors on PPC64, during attach and when trying to assign an IP
to an interface. With this change, basic operation of X710 NICs is now
possible.
This also fixes builds with IXL_DEBUG enabled
Reviewed by: erj
Sponsored by: Eldorado Research Institute (eldorado.org.br)
Differential Revision: https://reviews.freebsd.org/D23975
As written, the condition of (cdb[0] != 0x28 || cdb[0] != 0x2A) will always
be true, since if it's one, it's obviously not the other. Reading the code,
the intent appears to be that it should only perform the operation if it's
neither, otherwise the conditional can be elided.
Found by clang 10.
Summary:
POWER architecture CPUs (Book-S) require natural alignment for
cache-inhibited storage accesses. Since we can't know the caching model
for a page ahead of time, always enforce natural alignment in memcpy.
This fixes a SIGBUS in X with acceleration enabled on POWER9.
As part of this, revert r358672, it's no longer necessary with this fix.
Regression tested by alfredo.
Reviewed by: alfredo
Differential Revision: https://reviews.freebsd.org/D23969
r358556 added alignas() use to the functional tests, which isn't defined
until C11. Raise the -std to C11 to fix the build under freebsd-gcc{6,9}.
Reported by: mhorne, Jenkins/CI
Previously _libelf_cvt_NOTE_tom (to host) returned false if a note's
namesz + descsz exceeded the buffer size, while _libelf_cvt_NOTE_tof
(to file) silently truncated. Return false in the latter case too.
Sponsored by: The FreeBSD Foundation
Port aacraid driver to big-endian (BE) hosts.
The immediate goal of this change is to make it possible to use the
aacraid driver on PowerPC64 machines that have Adaptec Series 8 SAS
controllers.
Adapters supported by this driver expect FIB contents in little-endian
(LE) byte order. All FIBs have a fixed header part as well as a data
part that depends on the command being issued to the controller.
In this way, on BE hosts, the FIB header and all FIB data structures
used in aacraid.c and aacraid_cam.c need to be converted to LE before
being sent to the adapter and converted to BE when coming from it.
The functions to convert each struct are on aacraid_endian.c.
For little-endian (LE) targets, they are macros that expand
to nothing.
In some cases, when only a few fields of a large structure are used,
the fields are converted inline, by the code using them.
PR: 237463
Reviewed by: jhibbits
Sponsored by: Eldorado Research Institute (eldorado.org.br)
Differential Revision: https://reviews.freebsd.org/D23887
