- Remove 4 extra bytes from the ethernet payload.
- The maximum RX buffer was incorrectly set. Increase it to 64K for
now, until the exact limit is understood.
- Enable hardware checksumming again.
- Make hardware data structure packed.
MFC after: 3 days
The Tx interrupt is now kept disabled in the common case, only
enabled when the number of free descriptors in the queue falls
below a threshold. Transmitted frames are cleared from the VQ
before subsequent transmit, or in the watchdog timer.
This was a very big performance improvement for an experimental
Netmap bhyve backend.
MFC after: 1 month
- add missing rcvif in mbuf
- add missing ipacket stat
- remove uncessary mbuf copy on output path
- fix deadlock of the TX engine in case of error
Obtained from: NETASQ
MFC after: 2 weeks
This includes:
o All directories named *ia64*
o All files named *ia64*
o All ia64-specific code guarded by __ia64__
o All ia64-specific makefile logic
o Mention of ia64 in comments and documentation
This excludes:
o Everything under contrib/
o Everything under crypto/
o sys/xen/interface
o sys/sys/elf_common.h
Discussed at: BSDcan
Previously ISID was changed every time, that made impossible correct
persistent reservation, because reconnected session was identified as
completely new one.
Reviewed by: trasz
MFC after: 1 week
Update some comments on code, specifying the correct vlans used on switch
setup.
Advertise the proper switch operation mode (the rtl8366rb only support
dot1q vlans).
This fixes the breakage that i introduced on r249752 and make the rtl8366rb
switch works again with etherswitchcfg(8).
Tested on TP-Link 1043ND.
Tested by: me, Harm Weites (harm at weites.com)
- Don't discard frames if the dropped or error flag is set.
- Don't remove the last 4-bytes of every packet.
- Add extra range check for data position offset when receiving data.
MFC after: 1 day
PR: 191432
The array index for the callchain is getting double-incremented -- both in the
loop and the storing. It should only be incremented in one location.
Also, constrain the stack pointer range check.
MFC after: 2 weeks
1. oce_multiq_start(): make sure the buffer is consumed even on ENXIO
2. oce_multiq_transmit(): there is an extra call to drbr_enqueue()
causing the mbuf to be enqueued twice when the NIC's queue is full,
and potential panics
3. oce_multiq_transmit(): same problem fixed recently in ixgbe (r267187)
and other drivers: if the mbuf is enqueued, the proper return value is 0
Submitted by: Stefano Garzarella
MFC after: 3 days
rules prevent the USB serial module to be unloaded before any client
modules. This patch ensures that the "ucom_mtx" mutex is destroyed
last when doing a system uninit in a monotolith build aswell.
MFC after: 3 days
The ixgbe(4) hardware is capable of RSS hashing RX packets and doing RSS
queue selection for up to 8 queues.
However, even if multi-queue is enabled for ixgbe(4), the RX path doesn't use
the RSS flowid from the received descriptor. It just uses the MSIX queue id.
This patch does a handful of things if RSS is enabled:
* Instead of using a random key at boot, fetch the RSS key from the RSS code
and program that in to the RSS redirection table.
That whole chunk of code should be double checked for endian correctness.
* Use the RSS queue mapping to CPU ID to figure out where to thread pin
the RX swi thread and the taskqueue threads for each queue.
* The software queue is now really an "RSS bucket".
* When programming the RSS indirection table, use the RSS code to
figure out which RSS bucket each slot in the indirection table maps
to.
* When transmitting, use the flowid RSS mapping if the mbuf has
an RSS aware hash. The existing method wasn't guaranteed to align
correctly with the destination RSS bucket (and thus CPU ID.)
This code warns if the number of RSS buckets isn't the same as the
automatically configured number of hardware queues. The administrator
will have to tweak one of them for better performance.
There's currently no way to re-balance the RSS indirection table after
startup. I'll worry about that later.
Additionally, it may be worthwhile to always use the full 32 bit flowid if
multi-queue is enabled. It'll make things like lagg(4) behave better with
respect to traffic distribution.
The igb(4) hardware is capable of RSS hashing RX packets and doing RSS
queue selection for up to 8 queues. (I believe some hardware is limited
to 4 queues, but I haven't tested on that.)
However, even if multi-queue is enabled for igb(4), the RX path doesn't use
the RSS flowid from the received descriptor. It just uses the MSIX queue id.
This patch does a handful of things if RSS is enabled:
* Instead of using a random key at boot, fetch the RSS key from the RSS code
and program that in to the RSS redirection table.
That whole chunk of code should be double checked for endian correctness.
* Use the RSS queue mapping to CPU ID to figure out where to thread pin
the RX swi thread and the taskqueue threads for each queue.
* The software queue is now really an "RSS bucket".
* When programming the RSS indirection table, use the RSS code to
figure out which RSS bucket each slot in the indirection table maps
to.
* When transmitting, use the flowid RSS mapping if the mbuf has
an RSS aware hash. The existing method wasn't guaranteed to align
correctly with the destination RSS bucket (and thus CPU ID.)
This code warns if the number of RSS buckets isn't the same as the
automatically configured number of hardware queues. The administrator
will have to tweak one of them for better performance.
There's currently no way to re-balance the RSS indirection table after
startup. I'll worry about that later.
Additionally, it may be worthwhile to always use the full 32 bit flowid if
multi-queue is enabled. It'll make things like lagg(4) behave better with
respect to traffic distribution.
reset device task request from the driver. If the drive fails to respond
with a signature FIS, the driver would previously get into an endless retry
loop, stalling all I/O to the drive and keeping user processes stranded.
Instead, fail the i/o and invalidate the device if the task management
command times out. This is controllable with the sysctl and tunable
hw.isci.fail_on_task_timeout
dev.isci.0.fail_on_task_timeout
The default for these is 1.
Reviewed by: jimharris
Obtained from: Netflix, Inc.
MFC after: 2 days
If a controller is set to JBOD, it has no RAID functions turned on.
Populate even more of the firmware specification headers, copied from
cciss_vol_status.
Reviewed by: Benesh, Scott <scott.benesh@hp.com>
MFC after: 2 weeks
If cam_periph_find() doesn't locate the path we requested, bail to error
condition.
Acquire ciss->mtx for this operation.
Reviewed by: "Benesh, Scott" <scott.benesh@hp.com>
MFC after: 2 weeks
These changes prevent sysctl(8) from returning proper output,
such as:
1) no output from sysctl(8)
2) erroneously returning ENOMEM with tools like truss(1)
or uname(1)
truss: can not get etype: Cannot allocate memory
implement options TERMINAL_{KERN,NORM}_ATTR. These are aliased to
SC_{KERNEL_CONS,NORM}_ATTR and like these latter, allow to change the
default colors of normal and kernel text respectively.
Note on the naming: Although affecting the output of vt(4), technically
kern/subr_terminal.c is primarily concerned with changing default colors
so it would be inconsistent to term these options VT_{KERN,NORM}_ATTR.
Actually, if the architecture and abstraction of terminal+teken+vt would
be perfect, dev/vt/* wouldn't be touched by this commit at all.
Reviewed by: emaste
MFC after: 3 days
Sponsored by: Bally Wulff Games & Entertainment GmbH
With this change and previous work from ray@ it will be possible to put
both in GENERIC, and have one enabled by default, but allow the other to
be selected via the loader.
(The previous implementation had separate kern.vt.disable and
hw.syscons.disable tunables, and would panic if both drivers were
compiled in and neither was explicitly disabled.)
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.
Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.
MFC after: 2 weeks
Sponsored by: Mellanox Technologies
configs. Switch the BERI_NETFPGA_MDROOT to 64bit by default.
Give we have working interrupts also cleanup the extra polling CFLAGS from
the module Makefile.
MFC after: 2 weeks
time by setting NF10BMAC_64BIT and using a REGWTYPE #define to set correct
variable and return value widths.
Adjust comments to indicate the 32 or 64bit register widths.
MFC after: 2 weeks
PCI root bridges except for the one known-valid case on x86 where bridges
claim the I/O port registers used for PCI config space access.
Tested by: Hilko Meyer <hilko.meyer@gmx.de>
MFC after: 1 week
Prevent the Xen and VirtIO balloon drivers from marking pages as
wired. This prevents them from increasing the system wired page count,
which can lead to mlock failing because of hitting the limit in
vm.max_wired.
In the Xen case make sure pages are zeroed before giving them back to
the hypervisor, or else we might be leaking data. Also remove the
balloon_{append/retrieve} and link pages directly into the
ballooned_pages queue using the plinks.q field in the page struct.
Sponsored by: Citrix Systems R&D
Reviewed by: kib, bryanv
Approved by: gibbs
dev/virtio/balloon/virtio_balloon.c:
- Don't allocate pages with VM_ALLOC_WIRED.
dev/xen/balloon/balloon.c:
- Don't allocate pages with VM_ALLOC_WIRED.
- Make sure pages are zeroed before giving them back to the
hypervisor.
- Remove the balloon_entry struct and the balloon_{append/retrieve}
functions and use the page plinks.q entry to link the pages
directly into the ballooned_pages queue.
peripheral devices. When transmitting (rx) from slave to master,
sometimes nAKC delays. As a result, some slaves fails their
transmission.
Submitted by: Masanori OZAWA <ozawa@ongs.co.jp>
Reviewed by: brix
MFC after: 1 week
The sound drivers that use own buffer management can use sndbuf_setup
and not do any busdma allocation, so the driver will end up with the
managed buffer but no valid dma map and tag for it. Avoid calling
bus_dmamem_free in such cases.
Reported by: ache
Missed in review by: kan
separate argument structure with added level_type field for
CPUID_CPUID_COUNT request.
Reviewed by: attilio (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Assume the number of description used is reasonable value to
increment this otherwise opaque field by.
While here, reduce a minor difference between the legacy and
multiqueue transmit paths.
MFC after: 1 week
This requires the VMware vmxnet3 device to flip the start of packet
descriptor's generation before the rest of the packet's descriptors
have been loaded into the Rx ring. I've never observed this behavior,
and it seems to make the most sense not to do it this way. But it is
not a lot of work for the driver to handle this situation just in case.
MFC after: 1 week
The sbp_cam_detach_target can be called from sbp_post_explore function
on the first target that is not really attached and it was written with
the corresponding safety check in place to tolerate that. Unfortunately
the recent locking cleanup did add a locking assertion that tries to
dereference the target->sbp pointer unconditionally, which causes less
than desirable outcome. Since the assertion is useful, just initialize
the target sbp pointer once when sbp device is being initialized instead
of when the target is being attached. This makes assertion work in all
cases and fixes the crash on boot.
performing cpuid calls.
Add also a new way to specify the level type to cpucontrol(8) as
reported in the manpage.
Sponsored by: EMC / Isilon storage division
Reviewed by: bdrewery, gcooper
Testerd by: bdrewery
instead of trying to cache it.
Previously, we only trusted the state if we did not have a cached state.
However, once a state was cached, the _STA method was always ignored.
Specifically, once a power resource had been turned on once (e.g.
during resume), the driver assumed it was always on even if _STA said it
was off and never turned it back on. This prevented the power resource
from being turned back on if a laptop was resumed twice, for example.
To fix, just remove the cached state entirely and always use the results
of _STA. The loops already skip any resources where _STA fails.
Submitted by: trasz (initial patch to invoke _ON)
MFC after: 1 week
- Don't compare the DMA map to NULL to determine if bus_dmamap_unload()
should be called when releasing a static allocation. Instead, compare
the bus address against 0.
- Don't assume that the DMA map for static allocations is NULL. Instead,
save the value set by bus_dmamem_alloc() so it can later be passed to
bus_dmamem_free(). Also, add missing calls to bus_dmamap_unload() in
these cases before freeing the buffer.
- Use the bus address from the bus_dma callback instead of calling
vtophys() on the address allocated by bus_dmamem_alloc().
Reviewed by: kan
- Add missing calls to bus_dmamap_unload() in et(4).
- Check the bus address against 0 to decide when to call
bus_dmamap_unload() instead of comparing the bus_dma map against NULL.
- Check the virtual address against NULL to decide when to call
bus_dmamem_free() instead of comparing the bus_dma map against NULL.
- Don't clear bus_dma map pointers to NULL for static allocations.
Instead, treat the value as completely opaque.
- Pass the correct virtual address to bus_dmamem_free() in wpi(4) instead
of trying to free a pointer to the virtual address.
Reviewed by: yongari
the queue where to enqueue pages that are going to be unwired.
- Add stronger checks to the enqueue/dequeue for the pagequeues when
adding and removing pages to them.
Of course, for unmanaged pages the queue parameter of vm_page_unwire() will
be ignored, just as the active parameter today.
This makes adding new pagequeues quicker.
This change effectively modifies the KPI. __FreeBSD_version will be,
however, bumped just when the full cache of free pages will be
evicted.
Sponsored by: EMC / Isilon storage division
Reviewed by: alc
Tested by: pho
Switch the initialization of gnttab to use an unused physical memory
range for both PVHVM and PVH.
In the past PVHVM was using the xenpci BAR, but there's no reason to
do that, and in fact FreeBSD was probably doing it because it was the
way it was done in Windows, were drivers cannot probably request for
unused physical memory ranges, but it was never enforced in the
hypervisor.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
xen/gnttab.c:
- Allocate contiguous physical memory for grant table frames for both
PVHVM and PVH.
- Since gnttab is not a device, use the xenpv device in order to
request for this allocation.
dev/xen/xenpci/xenpcivar.h:
dev/xen/xenpci/xenpci.c:
- Remove the now unused xenpci_alloc_space and xenpci_alloc_space_int
functions.
xen/gnttab.h:
- Change the prototype of gnttab_init and gnttab_resume, that now
takes a device_t parameter.
dev/xen/control/control.c:
x86/xen/xenpv.c:
- Changes to accomodate the new prototype of gnttab_init and
gnttab_resume.
Add the PV shutdown hook to PVH.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
dev/xen/control/control.c:
- Make xen_pv_shutdown_final available on XENHVM builds.
- Register the Xen PV shutdown hook for PVH guests.
Create a dummy bus so top level Xen devices can attach to it (instead
of attaching directly to the nexus). This allows to have all the Xen
related devices grouped under a single bus.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
x86/xen/xenpv.c:
- Attach the xenpv bus when running as a Xen guest.
- Attach the ISA bus if needed, in order to attach syscons.
conf/files.amd6:
conf/files.i386:
- Include the xenpv.c file in the build of i386/amd64 kernels using
XENHVM.
dev/xen/console/console.c:
dev/xen/timer/timer.c:
xen/xenstore/xenstore.c:
- Attach to the xenpv bus instead of the Nexus.
dev/xen/xenpci/xenpci.c:
- Xen specific devices on PVHVM guests are no longer attached to the
xenpci device, they are instead attached to the xenpv bus, remove
the now unused methods.
The header structure consists of two 1-byte elements, but it must always
be describable by a single SG entry. Note for consistency, specify the
alignment everywhere, even if the structure has the appropriate natural
alignment since it contains a uint16_t.
Obtained from: DragonFlyBSD
MFC after: 1 week
These defines are applicable to userland too, but virtqueue.h contains
the kernel virtqueue interface, and is therefore not usable in userland.
Note that Linux places these defines in virtio_ring.h, but I don't want
the drivers including this header file to keep the VirtIO ring opaque to
everything but the virtqueue.
MFC after: 1 week
The eventual goal is to share this file with userland, so
remove the macro that is only specific for virtio_pci(4).
Instead, add the VIRTIO_PCI_CONFIG_OFF macro from Linux to
get the config size whether MSIX is enabled or not.
MFC after: 1 week
NULL to determine if bus_dmamap_unload() or bus_dmamem_free() should be
called. Instead, check the associated bus and virtual addresses.
- Don't clear static DMA maps to NULL.
Reviewed by: jfv
freeing them instead of after.
- Check the bus address of a static DMA buffer to decide if the associated
map should be unloaded.
- Don't try to destroy bus dma maps for static DMA buffers.
Reviewed by: davidcs
- Don't call xpt_free_path() in os_query_remove_device() and
always return TRUE.
- Update os_buildsgl() to support build logical SG table which
will be used by lower RAID module.
- Return CAM_SEL_TIMEOUTstatus for SCSIcommand failed as target
missing.
Many thanks to HighPoint for providing this driver update.
Submitted by: Steve Chang
Reviewed by: mav
MFC after: 3 days
In particular, don't check the value of the bus_dma map against NULL
to determine if either bus_dmamem_alloc() or bus_dmamap_load() succeeded.
Instead, assume that bus_dmamap_load() succeeeded (and thus that
bus_dmamap_unload() should be called) if the bus address for a resource
is non-zero, and assume that bus_dmamem_alloc() succeeded (and thus
that bus_dmamem_free() should be called) if the virtual address for a
resource is not NULL.
In many cases these bugs could result in leaks when a driver was detached.
Reviewed by: yongari
MFC after: 2 weeks
don't create a map before calling bus_dmamem_alloc() (such maps were
leaked). It is believed that the extra destroy of the map was generally
harmless since bus_dmamem_alloc() often uses special maps for which
bus_dmamap_destroy() is a no-op (e.g. on x86).
Reviewed by: scottl
shutdown by putting the former under !rebooting and turning the latter into
debug messages.
Reviewed by: hps
MFC after: 1 week
Sponsored by: Bally Wulff Games & Entertainment GmbH
- Use the existing vbus locks instead of Giant for the CAM sim lock.
- Use callout(9) instead of timeout(9).
- Mark the interrupt handler as MPSAFE.
- Don't attempt to pass data in the softc from probe() to attach().
Reviewed by: Steve Chang <ychang@highpoint-tech.com>
Assisted by: delphij
* The way rings are updated changed with the last API bump.
Also sync ->head when moving slots in netmap_sw_to_nic().
* Remove a crashing selrecord() call.
* Unclog the logic surrounding netmap_rxsync_from_host().
* Add timestamping to RX host ring.
* Remove a couple of obsolete comments.
Submitted by: Franco Fichtner
MFC after: 3 days
Sponsored by: Packetwerk
Apparently for VMware Fusion (and presumably VMware Workstation/Player
since the PR states TSO is broken there too, but I cannot test), the
TCP header pseudo checksum calculated should only include the protocol
(IPPROTO_TCP) value, not also the lengths as the stack does instead.
VMware ESXi seems to ignore whatever value is in the TCP header checksum,
and it is a bit surprising there is a different behavior between the
VMware products. And it is unfortunate that on ESXi we are forced to do
this extra bit of work.
PR: kern/185849
MFC after: 3 days
on USB HUBs by moving the code into the USB explore threads. The
deadlock happens because child devices of the USB HUB don't have the
expected reference count when called from outside the explore
thread. Only the HUB device itself, which the IOCTL interface locks,
gets the correct reference count.
MFC after: 3 days
- Revert r265427. It appears we are halting the DWC OTG host
controller schedule if we process events only at every SOF. When doing
split transactions we rely on that events are processed quickly and
waiting too long might cause data loss.
- We are not always able to meet the timing requirements of interrupt
endpoint split transactions. Switch from INTERRUPT to CONTROL endpoint
type for interrupt endpoint events until further, hence CONTROL
endpoint events are more relaxed, reducing the chance of data
loss. See comment in code for more in-depth explanation.
- Simplify TT scheduling.
MFC after: 3 days
- Remove double buffering interrupt and isochronous traffic via the
transaction translator. It can be avoided because the DWC OTG will
always delay the start split transactions for interrupt and
isochronous traffic, but will not delay the complete split
transactions, if we set the odd frame bit correctly.
- Need to check the transfer cache field in the device done function
to be sure all allocated channels are freed and not the transfer first
one. This seems to resolve the control endpoint transfer type quirk
which is now removed.
- Make sure any received data upon TX is dumped else RX path will
stop.
- Transmit isochronous data before receiving isochronous data as a
means to optimise the TT schedule.
- Implement a simple TT bandwidth scheduler.
- Cleanup use of old "td->error" variable.
- On interrupt IN traffic via the transaction translator we simply
ignore missed transfer opportunities and silently retry the
transaction upon next available time slot.
MFC after: 3 days
"Terminus BSD Console" is a derivative of Terminus that is provided
by Mr. Dimitar Zhekov under the 2-clause BSD license for use by the FreeBSD vt(4) console.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
