ngie [Sat, 3 Dec 2016 18:14:29 +0000 (18:14 +0000)]
MFC r301754,r301769:
r301754 (by pfg):
libc/rpc: Make use of some xdr_* macros.
xdr_rpcprog and xdr_rpcvers were broken in older versions of FreeBSD
but were fixed in r296394. Give them some use hoping they help make
the code somewhat more readable.
r301769 (by pfg):
libc/rpc: Make use of some xdr_* macros. (part 2)
xdr_rpcproc, xdr_rpcprog and xdr_rpcvers were broken in older
versions of FreeBSD but fixed in r296394. Give them some use
hoping they help make the code somewhat more readable.
convname and dst are guaranteed to be non-NULL by iconv_open(3).
src is an array. Remove these tests for NULL pointers.
While I'm here, eliminate a strlcpy with a correct but suspicious-looking
calculation for the third parameter (i.e. not a simple sizeof).
Compare the strings in-place instead of copying.
Found by: bdrewery
Found by: Coverity
CID: 1130050, 1130056
jhb [Sat, 3 Dec 2016 01:10:45 +0000 (01:10 +0000)]
MFC 303348:
cxgbe(4): Initialize the adapter queues (fwq and mgmtq) instead of
returning EAGAIN if they aren't available when the user tries to program
a filter. Do this after validating the filter so that the driver
doesn't bring up the queues if it doesn't have to.
jhb [Sat, 3 Dec 2016 00:18:38 +0000 (00:18 +0000)]
MFC 304854: cxgbe/iw_cxgbe: Various fixes to the iWARP driver.
- Return appropriate error code instead of ENOMEM when sosend() fails in
send_mpa_req.
- Fix for problematic race during destroy_qp.
- Abortive close in the failure of send_mpa_reject() instead of normal close.
- Remove the unnecessary doorbell flowcontrol logic.
303522:
Various fixes to the t4/5nex character device.
- Remove null open/close methods.
- Don't set d_flags to 0 explicitly.
- Remove t5_cdevsw as the .d_name member isn't really used and doesn't
warrant a separate cdevsw just for the name.
- Use ENOTTY as the error value for an unknown ioctl request.
- Use make_dev_s() to close race with setting si_drv1.
303647:
Store the offset of the KDOORBELL and GTS registers in the softc.
VF devices use a different register layout than PF devices. Storing
the offset in a value in the softc allows code to be shared between the
PF and VF drivers.
303860:
Reserve an adapter flag IS_VF to mark VF devices vs PF devices.
303880:
Track the base absolute ID of ingress and egress queues.
Use this to map an absolute queue ID to a logical queue ID in interrupt
handlers. For the regular cxgbe/cxl drivers this should be a no-op as
the base absolute ID should be zero. VF devices have a non-zero base
absolute ID and require this change. While here, export the absolute ID
of egress queues via a sysctl.
304168:
Make SGE parameter handling more VF-friendly.
Add fields to hold the SGE control register and free list buffer sizes to
the sge_params structure. Populate these new fields in
t4_init_sge_params() for PF devices and change t4_read_chip_settings() to
pull these values out of the params structure instead of reading
registers directly. This will permit t4_read_chip_settings() to be reused
for VF devices which cannot read SGE registers directly.
While here, move the call to t4_init_sge_params() to
get_params__post_init(). The VF driver will populate the SGE parameters
structure via a different method before calling t4_read_chip_settings().
304169:
Update mailbox writes to work with VF devices.
- Use alternate register locations for the data and control registers for
VFs.
- Do a dummy read to force the writes to the mailbox data registers to
post before the write to the control register on VFs.
- Do not check the PCI-e firmware register for errors on VFs.
304170:
Add support for register dumps on VF devices.
- Add handling of VF register sets to t4_get_regs_len() and t4_get_regs().
- While here, use t4_get_regs_len() in the ioctl handler for regdump
instead of inlining it.
304479:
Add structures for VF-specific adapter parameters.
While here, mark which parameters are PF-specific and which are
VF-specific.
304482:
Adjust t4_port_init() to work with VF devices.
Specifically, the FW_PORT_CMD may or may not work for a VF (the PF
driver can choose whether or not to permit access to this command),
so don't attempt to fetch port information on a VF if permission is
denied by the PF.
304485:
Reorder sysctls so that nodes shared with the VF driver are added first.
This permits a single early return for VF devices in the routines that
add sysctl nodes.
305548:
Don't break out of the m_advance() loop if len drops to zero.
If a packet contains the Ethernet header (14 bytes) in the first mbuf
and the payload (IP + UDP + data) in the second mbuf, then the attempt
to fetch the l3hdr will return a NULL pointer. The first loop iteration
will drop len to zero and exit the loop without setting 'p'. However,
the desired data is at the start of the second mbuf, so the correct
behavior is to loop around and let the conditional set 'p' to m_data of
the next mbuf (and leave offset as 0).
305549:
Chelsio T4/T5 VF driver.
The cxgbev/cxlv driver supports Virtual Function devices for Chelsio
T4 and T4 adapters. The VF devices share most of their code with the
existing PF4 driver (cxgbe/cxl) and as such the VF device driver
currently depends on the PF4 driver.
Similar to the cxgbe/cxl drivers, the VF driver includes a t4vf/t5vf
PCI device driver that attaches to the VF device. It then creates
child cxgbev/cxlv devices representing ports assigned to the VF.
By default, the PF driver assigns a single port to each VF.
t4vf_hw.c contains VF-specific routines from the shared code used to
fetch VF-specific parameters from the firmware.
t4_vf.c contains the VF-specific PCI device driver and includes its
own attach routine.
VF devices are required to use a different firmware request when
transmitting packets (which in turn requires a different CPL message
to encapsulate messages). This alternate firmware request does not
permit chaining multiple packets in a single message, so each packet
results in a firmware request. In addition, the different CPL message
requires more detailed information when enabling hardware checksums,
so parse_pkt() on VF devices must examine L2 and L3 headers for all
packets (not just TSO packets) for VF devices. Finally, L2 checksums
on non-UDP/non-TCP packets do not work reliably (the firmware trashes
the IPv4 fragment field), so IPv4 checksums for such packets are
calculated in software.
Most of the other changes in the non-VF-specific code are to expose
various variables and functions private to the PF driver so that they
can be used by the VF driver.
Note that a limited subset of cxgbetool functions are supported on VF
devices including register dumps, scheduler classes, and clearing of
statistics. In addition, TOE is not supported on VF devices, only for
the PF interfaces.
jhb [Fri, 2 Dec 2016 22:27:54 +0000 (22:27 +0000)]
MFC 308564: Don't place threads on the run queue after waking up other CPUs.
The other CPU might resume and see a still-empty runq and go back to
sleep before sched_add() adds the thread to the runq. This results
in a lost wakeup and a potential hang if the system is otherwise
completely idle.
The race originated due to a micro-optimization (my fault) in 4BSD in
that it avoided putting a thread on the run queue if the scheduler was
going to preempt to the new thread. To avoid complexity while fixing
this race, just drop this optimization. 4BSD now always sets the
"owepreempt" flag when a preemption is warranted and defers the actual
preemption to the thread_unlock of the caller the same as ULE.
jhb [Fri, 2 Dec 2016 21:36:21 +0000 (21:36 +0000)]
MFC 303204: Install a handler for firmware work request error messages.
If a driver sends an malformed or disallowed work request, the firmware
responds with a work request error. Previously the driver treated this is
as an unexpected message and panicked. Now it decodes the error message
to aid in debugging.
jhb [Fri, 2 Dec 2016 21:35:14 +0000 (21:35 +0000)]
MFC 308005: Add powerd(8) support for several families of AMD CPUs.
Use the same logic to calculate the nominal CPU frequency from the P-state
MSRs on family 0x12, 0x15, and 0x16 CPUs as is used for family 0x10.
Family 0x14 was included in the original patch in the PR but I left that
out as the BIOS writer's guide for family 0x14 CPUs show a different layout
for the relevant MSR and include a different formulate for calculating the
frequency.
While here, simplify a few expressions and print out the family of
unsupported CPUs in hex rather than decimal.
jhb [Fri, 2 Dec 2016 21:29:52 +0000 (21:29 +0000)]
MFC 302339:
cxgbe(4): Changes to the CPL-handler registration mechanism and code
related to "shared" CPLs.
a) Combine t4_set_tcb_field and t4_set_tcb_field_rpl into a single
function. Allow callers to direct the response to any iq. Tidy up
set_ulp_mode_iscsi while there to use names from t4_tcb.h instead of
magic constants.
b) Remove all CPL handler tables from struct adapter. This reduces its
size by around 2KB. All handlers are now registered at MOD_LOAD instead
of attach or some kind of initialization/activation. The registration
functions do not need an adapter parameter any more.
c) Add per-iq handlers to deal with CPLs whose destination cannot be
determined solely from the opcode. There are 2 such CPLs in use right
now: SET_TCB_RPL and L2T_WRITE_RPL. The base driver continues to send
filter and L2T_WRITEs over the mgmtq and solicits the reply on fwq.
t4_tom (including the DDP code) now uses the port's ctrlq to send
L2T_WRITEs and SET_TCB_FIELDs and solicits the reply on an ofld_rxq.
fwq and ofld_rxq have different handlers that know what kind of tid to
expect in the reply. Update t4_write_l2e and callers to to support any
wrq/iq combination.
jhb [Fri, 2 Dec 2016 20:16:52 +0000 (20:16 +0000)]
MFC 292736:
cxgbe(4): Updates to the base NIC driver and t4_tom to support the iSCSI
offload driver. These changes come from projects/cxl_iscsi.
Note that these changes make use of the mbufq API from 11.0, but that
API is not present in 10.x in the same form. Borrow an implementation
from the CAM CTL ha code that uses m_nextpkt to implement mbufq for use
in 10.
julian [Fri, 2 Dec 2016 08:21:25 +0000 (08:21 +0000)]
MFH: r309295
bhyve: stability and performance improvement for dbgport
The TCP server implementation in dbgport does not track clients, so it
may try to write to a disconected socket resulting in SIGPIPE.
Avoid that by setting SO_NOSIGPIPE socket option.
Because dbgport emulates an I/O port to guest, the communication is done
byte by byte. Reduce latency of the TCP/IP transfers by using
TCP_NODELAY option. In my tests that change improves performance of
kgdb commands with lots of output (e.g. info threads) by two orders of
magnitude.
A general note. Since we have a uart emulation in bhyve, that can be
used for the console and gdb access to guests. So, bvmconsole and bvmdebug
could be de-orbited now. But there are many existing deployments that
still dependend on those.
julian [Fri, 2 Dec 2016 05:42:04 +0000 (05:42 +0000)]
MFH: r303611
slite style changes. There is an incoming patch that rewrites a
lot of this module and I want to get the style and whitespace changes in
a separate commit (or maybe more).
Pull in most of the Chelsio and iWARP related changes from stable/11 into
stable/10. A few changes from 278886 (OFED 1.2) were also included though
the full merge is not:
- The find_gid_port() function in infiband/core/cma.c.
- Addition of the 'ord' and 'ird' fields to 'struct iw_cm_event'.
273806:
Userspace library for Chelsio's Terminator 5 based iWARP RNICs (pretty
much every T5 card that does _not_ have "-SO" in its name is RDMA
capable).
This plugs into the OFED verbs framework and allows userspace RDMA
applications to work over T5 RNICs. Tested with rping.
289103:
iw_cxgbe: fix for page fault in cm_close_handler().
This is roughly the iw_cxgbe equivalent of
https://github.com/torvalds/linux/commit/be13b2dff8c4e41846477b22cc5c164ea5a6ac2e
-----------------
RDMA/cxgb4: Connect_request_upcall fixes
When processing an MPA Start Request, if the listening endpoint is
DEAD, then abort the connection.
If the IWCM returns an error, then we must abort the connection and
release resources. Also abort_connection() should not post a CLOSE
event, so clean that up too.
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
-----------------
289201:
iw_cxgbe: MPA v2 is always available.
289338:
iw_cxgbe: use correct RFC number.
289578:
Merge LinuxKPI changes from DragonflyBSD:
- Define the kref structure identical to the one found in Linux.
- Update clients referring inside the kref structure.
- Implement kref_sub() for FreeBSD.
293185:
iw_cxgbe: Shut down the socket but do not close the fd in case of error.
The fd is closed later in this case. This fixes a "SS_NOFDREF on enter"
panic.
294474:
iw_cxgbe: fix a couple of problems int the RDMA_TERMINATE handler.
a) Look for the CPL in the payload buffer instead of the descriptor.
b) Retrieve the socket associated with the tid with the inpcb lock held.
294610:
Fix for iWARP servers that listen on INADDR_ANY.
The iWARP Connection Manager (CM) on FreeBSD creates a TCP socket to
represent an iWARP endpoint when the connection is over TCP. For
servers the current approach is to invoke create_listen callback for
each iWARP RNIC registered with the CM. This doesn't work too well for
INADDR_ANY because a listen on any TCP socket already notifies all
hardware TOEs/RNICs of the new listener. This patch fixes the server
side of things for FreeBSD. We've tried to keep all these modifications
in the iWARP/TCP specific parts of the OFED infrastructure as much as
possible.
297124:
iw_cxgbe/libcxgb4: Pull in many applicable fixes from the upstream Linux
iWARP driver and userspace library to the FreeBSD iw_cxgbe and libcxgb4.
297368:
cxgbe/iw_cxgbe: Fix for stray "start_ep_timer timer already started!"
messages.
297406:
Remove unnecessary dequeue_mutex (added in r294610) from the iWARP
connection manager. Examining so_comp without synchronization with
iw_so_event_handler is a harmless race.
300875:
iw_cxgbe: Use vmem(9) to manage PBL and RQT allocations.
300888:
iw_cxgbe: Plug a lock leak in process_mpa_request().
If the parent is DEAD or connect_request_upcall() fails, the parent
mutex is left locked. This leads to a hang when process_mpa_request()
is called again for another child of the listening endpoint.
301158:
iw_cxgbe: Fix panic that occurs when c4iw_ev_handler tries to acquire
comp_handler_lock but c4iw_destroy_cq has already freed the CQ memory
(which is where the lock resides).
301896:
Fix bug in iwcm that caused a panic in iw_cm_wq when krping is run
repeatedly in a tight loop.
301897:
iw_cxgbe: Make sure that send_abort results in a TCP RST and not a FIN.
Release the hold on ep->com immediately after sending the RST. This
fixes a bug that sometimes leaves userspace iWARP tools hung when the
user presses ^C.
304838:
Do not free an uninitialized pointer on soaccept failure in the iWARP
connection manager.
jhb [Thu, 1 Dec 2016 20:36:48 +0000 (20:36 +0000)]
MFC 308456: Pass the correct flag to find_symdef() from _rtld_bind().
When symbol versioning was added to rtld, the boolean 'in_plt' argument
to find_symdef() was converted to a bitmask of flags. The first flag
added was 'SYMLOOK_IN_PLT' which replaced the 'in_plt' bool. This
happened to still work by accident as SYMLOOK_IN_PLT had the value of 1
which is the same as 'true', so there should be no functional change.
vangyzen [Wed, 30 Nov 2016 22:20:23 +0000 (22:20 +0000)]
MFC r306577 r306652 306830
Add GARP retransmit capability
A single gratuitous ARP (GARP) is always transmitted when an IPv4
address is added to an interface, and that is usually sufficient.
However, in some circumstances, such as when a shared address is
passed between cluster nodes, this single GARP may occasionally be
dropped or lost. This can lead to neighbors on the network link
working with a stale ARP cache and sending packets destined for
that address to the node that previously owned the address, which
may not respond.
To avoid this situation, GARP retransmissions can be enabled by setting
the net.link.ether.inet.garp_rexmit_count sysctl to a value greater
than zero. The setting represents the maximum number of retransmissions.
The interval between retransmissions is calculated using an exponential
backoff algorithm, doubling each time, so the retransmission intervals
are: {1, 2, 4, 8, 16, ...} (seconds).
Due to the exponential backoff algorithm used for the interval
between GARP retransmissions, the maximum number of retransmissions
is limited to 16 for sanity. This limit corresponds to a maximum
interval between retransmissions of 2^16 seconds ~= 18 hours.
Increasing this limit is possible, but sending out GARPs spaced
days apart would be of little use.
Update arp(4) to document the net.link.ether.inet.garp_rexmit_count sysctl.
Submitted by: dab
Relnotes: yes
Sponsored by: Dell EMC
vangyzen [Wed, 30 Nov 2016 20:48:44 +0000 (20:48 +0000)]
MFC r308904
Fix error reporting from wcstof()
When wcstof() skipped initial space and then parsing failed, it set
endptr to the first non-space character. Fix it to correctly report
failure by setting endptr to the beginning of the input string.
The fix is from theraven@, who fixed this bug in wcstod() and
wcstold() in r227753.
While I'm here:
Move assignments out of declarations in wcstod() and wcstold().
This is against my personal preference, but it is our agreed style(9).
Set endptr correctly on malloc() failure in all three functions.
Remove an incorrect comment: This is pointer arithmetic,
so the code was not actually making that assumption.
wcstold() advanced the wcp pointer beyond leading whitespace
and then reset it back to the beginning of the string.
Do not reset it. This seems to have no functional effect,
since strtold_l() also skips leading whitespace. I'm making
the change to keep this function consistent with wcstof() and
wcstod(), and because the C11 spec prescribes the use of iswspace()
to skip leading space.
Reported by: libc++ unit test for std::stof(std::wstring)
Sponsored by: Dell EMC
vangyzen [Wed, 30 Nov 2016 18:11:35 +0000 (18:11 +0000)]
MFC r308824
locale: fix display of "grouping" and "mon_grouping" values
The "grouping" and "mon_grouping" values are arrays of one-byte
integers, not arrays of ASCII characters. Display them in a format
similar to GNU and MacOS.
dexuan [Wed, 30 Nov 2016 07:51:46 +0000 (07:51 +0000)]
MFC r308797-308799,309082
Approved by: sephe (mentor)
r308797
update the hv_vmbus(4) manual by adding a dependency on pci
We enhanced the vmbus driver to support PCIe pass-through recently.
Reviewed by: sephe
Approved by: sephe (mentor)
Sponsored by: Microsoft
r308798
remove the hv_ata_pci_disengage(4) manual
A few months ago, we removed the driver, which was not necessary any longer.
Reviewed by: sephe
Approved by: sephe (mentor)
Sponsored by: Microsoft
r308799
fix share/man/man4/Makefile for hv_ata_pci_disengage.4
We need to remove the line since we removed the related manual just now.
Reviewed by: sephe
Approved by: sephe (mentor)
Sponsored by: Microsoft
r309082
share/man/man4/Makefile: Only install Hyper-V man pages on amd64 and i386
We shouldn't install them on the architectures not supported by Hyper-V.
And, hv_ata_pci_disengage.4.gz should be removed from all architectures:
1) It should have only applied to Hyper-V;
2) For Hyper-V platforms (amd64 and i386), the related driver was removed by
r306426 | sephe | 2016-09-29 09:41:52 +0800 (Thu, 29 Sep 2016),
because now we have a better mechanism to disble the ata driver for hard
disks when the VM runs on Hyper-V.
dexuan [Wed, 30 Nov 2016 06:26:37 +0000 (06:26 +0000)]
MFC: 308723-308725,308793-308795,309127
Approved by: sephe (mentor)
r308723
hyperv/vmbus: add a new method to get vcpu_id
vcpu_id is host's representation of guest CPU.
We get the mapping between vcpu_id and FreeBSD kernel's cpu id when VMBus
driver is loaded. Later, when a driver, like the coming pcib driver, talks
to the host and needs to refer to a guest CPU, the driver must use the
vcpu_id.
The feature enables us to pass through physical PCIe devices to FreeBSD VM
running on Hyper-V (Windows Server 2016) to get near-native performance with
low CPU utilization.
The patch implements a PCI bridge driver to support the feature:
1) The pcib driver talks to the host to discover device(s) and presents
the device(s) to FreeBSD's pci driver via PCI configuration space (note:
to access the configuration space, we don't use the standard I/O port
0xCF8/CFC method; instead, we use an MMIO-based method supplied by Hyper-V,
which is very similar to the 0xCF8/CFC method).
2) The pcib driver allocates resources for the device(s) and initialize
the related BARs, when the device driver's attach method is invoked;
3) The pcib driver talks to the host to create MSI/MSI-X interrupt
remapping between the guest and the host;
4) The pcib driver supports device hot add/remove.
brooks [Wed, 30 Nov 2016 01:22:12 +0000 (01:22 +0000)]
MFC r309027:
Allocate a struct ifreq rather than using a (wrong) computed size for
the BIOCSETIF ioctl.
The kernel always copies an entire struct ifreq and IPv4 addresses will
always fit in an ifreq.
On systems with pointers larger than 64-bits, the computed size will be
less than the size of struct ifreq, potentially resulting in the kernel
attempting to copyin memory from outside the allocation.
jhb [Mon, 28 Nov 2016 18:36:37 +0000 (18:36 +0000)]
MFC 307756: Define max_align_t for C11.
libc++'s stddef.h includes an existing definition of max_align_t for
C++11, but it is only defined for C++, not for C. In addition, GCC and
clang both define an alternate version of max_align_t that uses a
union of multiple types rather than a plain long double as in libc++.
This adds a __max_align_t to <sys/_types.h> that matches the GCC and
clang definition that is mapped to max_align_t in <stddef.h>.
hselasky [Mon, 28 Nov 2016 17:24:24 +0000 (17:24 +0000)]
MFC r308730:
Make sure MAC address is reprogrammed when if_init() callback is
invoked. Else promiscious mode must be used to pass traffic. While at
it fix a debug print macro.
kib [Sun, 27 Nov 2016 09:14:52 +0000 (09:14 +0000)]
MFC r308618:
Provide simple mutual exclusion between mount point update and unmount.
In the update path in ffs_mount(), drop vfs_busy() reference around namei().
jhb [Fri, 25 Nov 2016 22:12:03 +0000 (22:12 +0000)]
MFC 307333: Reprogram I/O APIC interrupt pins when registering an I/O APIC.
All I/O APIC pins are masked when an I/O APIC is first probed. The
APIC enumerator (MP Table or MADT) then parses its associated tables to
configure individual pins to set custom delivery modes or alternate
routing (e.g. routing IRQ 0 to intpin 2). Pins for regular interrupt
pins are left masked until the first interrupt is assigned. However,
pins with unusual settings (e.g. NMI or SMI) are never assigned an
interrupt and thus never re-programmed. The I/O APIC code used to
reprogram all interrupt pins during registration but this was lost in
r151979.
In theory, this is mostly a no-op as the ACPI APIC table does not
include a way to enumerate NMI or SMI pins for the I/O APIC, so only
systems using an MP Table would be affected.
jch [Thu, 24 Nov 2016 14:48:46 +0000 (14:48 +0000)]
MFC r286227, r286443:
r286227:
Decompose TCP INP_INFO lock to increase short-lived TCP connections scalability:
- The existing TCP INP_INFO lock continues to protect the global inpcb list
stability during full list traversal (e.g. tcp_pcblist()).
- A new INP_LIST lock protects inpcb list actual modifications (inp allocation
and free) and inpcb global counters.
It allows to use TCP INP_INFO_RLOCK lock in critical paths (e.g. tcp_input())
and INP_INFO_WLOCK only in occasional operations that walk all connections.
PR: 183659
Differential Revision: https://reviews.freebsd.org/D2599
Reviewed by: jhb, adrian
Tested by: adrian, nitroboost-gmail.com
Sponsored by: Verisign, Inc.
r286443:
Fix a kernel assertion issue introduced with r286227:
Avoid too strict INP_INFO_RLOCK_ASSERT checks due to
tcp_notify() being called from in6_pcbnotify().
Reported by: Larry Rosenman <ler@lerctr.org>
Submitted by: markj, jch
emaste [Thu, 24 Nov 2016 00:46:34 +0000 (00:46 +0000)]
MFC r308772: crunchide: report explicit error for combined string table
Some tools produce objects with a combined strtab and shstrtab.
These objects are not supported by crunchide since it rewrites the
symtab and strtab to "hide" symbols. This invalidates section header
offsets into a combined strtab/shstrtab.
In the future we could support these objects (by ensuring that we retain
unmodified section name strings in the output .strtab, and then rewriting
each section header's sh_name).
jhb [Wed, 23 Nov 2016 23:53:52 +0000 (23:53 +0000)]
MFC 308056: Fix formatting of tables.
Specifically, use .Ta instead of tabs to separate column entries. While
here fix a few other things:
- Use .Sy for all column headers (previously only the first column header
was bold)
- Use .Dv to markup constants used for MIB names.
- Use "1234" and "4321" for the byte order descriptions without
thousands separators.
- Mark up header files in the first table with .In.
jhb [Wed, 23 Nov 2016 23:45:42 +0000 (23:45 +0000)]
MFC 307975: Enable EFER_NXE properly on APs.
EFER_NXE is set in the EFER MSR by initializecpu() and must be set on all
CPUs in the system. When PG_NX support was added to PAE on i386, the
block to enable EFER_NXE was placed in a section of initializecpu() that
only runs if 'cpu == CPU_686'. During early boot, locore does an
initial pass to set cpu that sets it to CPU_686 on all CPUs later than
a Pentium. Later, printcpuinfo() adjusts the 'cpu' variable on
PII and later CPUs to one of CPU_PII, CPU_PIII, or CPU_P4. However,
printcpuinfo() is called after initializecpu() on the BSP, so the BSP
would enable EFER_NXE and pg_nx. The APs execute initializecpu() much
later after printcpuinfo() has run. The end result on a modern CPU was
that cpu was set to CPU_PIII when the APs invoked initializecpu(), so
they did not enable EFER_NXE. As a result, the APs would fault when
trying to access any pages marked with PG_NX set.
When booting a 2 CPU PAE kernel in bhyve this manifested as a hang before
single user mode. The attempt to execute /bin/init tried to copy out
the exec strings (argv, etc.) to a non-executable mapping while running
on the AP. The instruction kept faulting due to invalid bits in the PTE
in an infinite loop.
Fix this by moving the code to enable EFER_NXE out of the switch statement
on 'cpu' and always doing it if 'amd_feature' supports AMDID_NX.
kib [Wed, 23 Nov 2016 17:48:43 +0000 (17:48 +0000)]
MFC r308689:
Pass CPUID[1] %edx (cpu_feature), %ecx (cpu_feature2) and
CPUID[7].%ebx (cpu_stdext_feature), %ecx (cpu_stdext_feature2) to the
ifunc resolvers on x86.
MFC r308925:
Adjust r308689 to make rtld compilable with either in-tree or
(hopefully) stock gcc 4.2.1 on i386 and other arches.
asomers [Tue, 22 Nov 2016 21:02:35 +0000 (21:02 +0000)]
MFC r307584 (partially)
307584 added the __min_size macro and replaced some static array sizes with
__min_size. However, stable/10 doesn't yet have any static array sizes, so
all this MFC does is add the __min_size macro.
Original commit message:
Fix C++ includability of crypto headers with static array sizes
C99 allows array function parameters to use the static keyword for their
sizes. This tells the compiler that the parameter will have at least the
specified size, and calling code will fail to compile if that guarantee is
not met. However, this syntax is not legal in C++.
This commit reverts r300824, which worked around the problem for
sys/sys/md5.h only, and introduces a new macro: min_size(). min_size(x) can
be used in headers as a static array size, but will still compile in C++
mode.
mckusick [Tue, 22 Nov 2016 00:27:19 +0000 (00:27 +0000)]
MFC r307978:
Bug 180894 reports that rm -rf on a directory causes kernel panic and reboot.
Return EINVAL rather than panic for low directory link count.
jilles [Sun, 20 Nov 2016 13:44:27 +0000 (13:44 +0000)]
MFC r307755: swapoff: Remove only late devices with -aL.
Currently, '/etc/rc.d/swaplate stop' removes all swap devices. This can be
very slow and may not even be possible if there is a lot of swap space in
use. However, removing swap devices is only needed for late swap devices
that may depend on daemons that subsequent shutdown steps stop. Normal swap
devices such as hard disk partitions will remain available throughout the
shutdown process and need not be removed.
In swapoff, interpret -aL to remove late swap devices only, and use this in
etc/rc.d/swaplate. The meaning of -aL in swapon remains unchanged (add all
swap devices, both normal and late).
emaste [Fri, 18 Nov 2016 16:05:32 +0000 (16:05 +0000)]
MFC r306740, r307150: groff: use changelog date in man pages
r306740:
groff: use changelog date rather than file modification date in man pages
The source checkout date is not particularly relevant, and this makes
groff man pages build reproducibly.
r307150:
Avoid using 'head' in generating groff doc date
It may not be available in certain cross build cases.
Note that this is a slight change in functionality, in that now only the
first line of the source ChangeLog file is processed. This is acceptable
as groff will be retired and we won't encounter a possibly-different
ChangeLog format.
hselasky [Tue, 15 Nov 2016 09:00:01 +0000 (09:00 +0000)]
MFC r308416:
Add timer to watch the RQ when we are out of mbufs.
The firmware/hardware does not generate additional completion
events unless we post new buffers. Use a timer to try to post
more buffers in case we are temporarily out of mbufs. Else
the receive schedule completely stops.
mav [Sat, 12 Nov 2016 23:58:07 +0000 (23:58 +0000)]
MFC r308173:
Fix ZIL records ordering when ZVOL opened both with and without FSYNC.
Before this an earlier writes to a ZVOL opened without FSYNC could get to
ZIL after later writes to the same ZVOL opened with FSYNC. Fix this by
replicating functionality of ZPL (zv_sync_cnt equivalent to z_sync_cnt),
marking all log records sync if anybody opened the ZVOL with FSYNC.
mav [Sat, 12 Nov 2016 23:40:40 +0000 (23:40 +0000)]
MFC r308055: Add vdev_reopening support to vdev_geom.
It allows to avoid extra GEOM providers flapping without significant need.
Since GEOM got resize support, we don't need to reopen provider to get new
size. If provider was orphaned and no longer valid, ZFS should already
know that, and in such case reopen should be done in full as expected.
mav [Sat, 12 Nov 2016 23:38:04 +0000 (23:38 +0000)]
MFC r308051: Matching GUIDs, handle possible race on vdev detach.
In case of vdev detach, causing top level mirror vdev destruction, leaf
vdev changes its GUID to one of the destroyed mirror, that creates race
condition when GUID in vdev label may not match one in the pool config.
This change replicates logic nuance of vdev_validate() by adding special
exception, matching the vdev GUID against the top level vdev GUID.
Since this exception is not completely reliable (may give false positives
if we fail to erase label on detached vdev), use it only as last resort.
Quick way to reproduce this scenario now is detach vdev from a pool with
enabled autoextend. During vdev detach autoextend logic tries to reopen
remaining vdev, that always fails now since in-memory configuration is
already updated, while on-disk labels are not yet.