sys/dev/xen/blkback/blkback.c:
o Implement front-end request coalescing. This greatly improves the
performance of front-end clients that are unaware of the dynamic
request-size/number of requests negotiation available in the
FreeBSD backend driver. This required a large restructuring
in how this driver records in-flight transactions and how those
transactions are mapped into kernel KVA. For example, the driver
now includes a mini "KVA manager" that allocates ranges of
contiguous KVA to patches of requests that are physically
contiguous in the backing store so that a single bio or UIO
segment can be used to represent the I/O.
o Refuse to open any backend files or devices if the system
has yet to mount root. This avoids a panic.
o Properly handle "onlined" devices. An "onlined" backend
device stays attached to its backing store across front-end
disconnections. This feature is intended to reduce latency
when a front-end does a hand-off to another driver (e.g.
PV aware bootloader to OS kernel) or during a VM reboot.
o Harden the driver against a pathological/buggy front-end
by carefully vetting front-end XenStore data such as the
front-end state.
o Add sysctls that report the negotiated number of
segments per-request and the number of requests that
can be concurrently in flight.
Submitted by: kdm
Reviewed by: gibbs
Sponsored by: Spectra Logic Corporation
MFC after: 1 week
of the devices we manage. These changes can be due to writes
we make ourselves or due to changes made by the control domain.
The goal of these changes is to insure that all state transitions
can be detected regardless of their source and to allow common
device policies (e.g. "onlined" backend devices) to be centralized
in the XenBus bus code.
sys/xen/xenbus/xenbusvar.h:
sys/xen/xenbus/xenbus.c:
sys/xen/xenbus/xenbus_if.m:
Add a new method for XenBus drivers "localend_changed".
This method is invoked whenever a write is detected to
a device's XenBus tree. The default implementation of
this method is a no-op.
sys/xen/xenbus/xenbus_if.m:
sys/dev/xen/netfront/netfront.c:
sys/dev/xen/blkfront/blkfront.c:
sys/dev/xen/blkback/blkback.c:
Change the signature of the "otherend_changed" method.
This notification cannot fail, so it should return void.
sys/xen/xenbus/xenbusb_back.c:
Add "online" device handling to the XenBus Back Bus
support code. An online backend device remains active
after a front-end detaches as a reconnect is expected
to occur in the near future.
sys/xen/interface/io/xenbus.h:
Add comment block further explaining the meaning and
driver responsibilities associated with the XenBus
Closed state.
sys/xen/xenbus/xenbusb.c:
sys/xen/xenbus/xenbusb.h:
sys/xen/xenbus/xenbusb_back.c:
sys/xen/xenbus/xenbusb_front.c:
sys/xen/xenbus/xenbusb_if.m:
o Register a XenStore watch against the local XenBus tree
for all devices.
o Cache the string length of the path to our local tree.
o Allow the xenbus front and back drivers to hook/filter both
local and otherend watch processing.
o Update the device ivar version of "state" when we detect
a XenStore update of that node.
sys/dev/xen/control/control.c:
sys/xen/xenbus/xenbus.c:
sys/xen/xenbus/xenbusb.c:
sys/xen/xenbus/xenbusb.h:
sys/xen/xenbus/xenbusvar.h:
sys/xen/xenstore/xenstorevar.h:
Allow clients of the XenStore watch mechanism to attach
a single uintptr_t worth of client data to the watch.
This removes the need to carefully place client watch
data within enclosing objects so that a cast or offsetof
calculation can be used to convert from watch to enclosing
object.
Sponsored by: Spectra Logic Corporation
MFC after: 1 week
cpuset_t objects.
That is going to offer the underlying support for a simple bump of
MAXCPU and then support for number of cpus > 32 (as it is today).
Right now, cpumask_t is an int, 32 bits on all our supported architecture.
cpumask_t on the other side is implemented as an array of longs, and
easilly extendible by definition.
The architectures touched by this commit are the following:
- amd64
- i386
- pc98
- arm
- ia64
- XEN
while the others are still missing.
Userland is believed to be fully converted with the changes contained
here.
Some technical notes:
- This commit may be considered an ABI nop for all the architectures
different from amd64 and ia64 (and sparc64 in the future)
- per-cpu members, which are now converted to cpuset_t, needs to be
accessed avoiding migration, because the size of cpuset_t should be
considered unknown
- size of cpuset_t objects is different from kernel and userland (this is
primirally done in order to leave some more space in userland to cope
with KBI extensions). If you need to access kernel cpuset_t from the
userland please refer to example in this patch on how to do that
correctly (kgdb may be a good source, for example).
- Support for other architectures is going to be added soon
- Only MAXCPU for amd64 is bumped now
The patch has been tested by sbruno and Nicholas Esborn on opteron
4 x 12 pack CPUs. More testing on big SMP is expected to came soon.
pluknet tested the patch with his 8-ways on both amd64 and i386.
Tested by: pluknet, sbruno, gianni, Nicholas Esborn
Reviewed by: jeff, jhb, sbruno
adding appropriate #ifdefs. For module builds the framework needs
adjustments for at least carp.
Reviewed by: gnn
Sponsored by: The FreeBSD Foundation
Sponsored by: iXsystems
MFC after: 4 days
receive processing.
Remove unnecessary restrictions on the mbuf chain length built during an
LRO receive. This restriction was copied from the Linux netfront driver
where the LRO implementation cannot handle more than 18 discontinuities.
The FreeBSD implementation has no such restriction.
MFC after: 1 week
mechanical change. This opens the door for using PV device drivers
under Xen HVM on i386, as well as more general harmonisation of i386
and amd64 Xen support in FreeBSD.
Reviewed by: cperciva
MFC after: 3 weeks
max_request_segments * PAGE_SIZE if the I/O is page-aligned; the
largest I/O we can guarantee will work is PAGE_SIZE less than that.
This unbreaks 'diskinfo -t'.
In xbb_detach() only perform cleanup of our taskqueue and
device statistics structures if they have been initialized.
This avoids a panic when xbb_detach() is called on a partially
initialized device instance, due to an early failure in
attach.
Sponsored by: Spectra Logic Corporation
o Add support for backend devices (e.g. blkback)
o Implement extensions to the Xen para-virtualized block API to allow
for larger and more outstanding I/Os.
o Import a completely rewritten block back driver with support for fronting
I/O to both raw devices and files.
o General cleanup and documentation of the XenBus and XenStore support code.
o Robustness and performance updates for the block front driver.
o Fixes to the netfront driver.
Sponsored by: Spectra Logic Corporation
sys/xen/xenbus/init.txt:
Deleted: This file explains the Linux method for XenBus device
enumeration and thus does not apply to FreeBSD's NewBus approach.
sys/xen/xenbus/xenbus_probe_backend.c:
Deleted: Linux version of backend XenBus service routines. It
was never ported to FreeBSD. See xenbusb.c, xenbusb_if.m,
xenbusb_front.c xenbusb_back.c for details of FreeBSD's XenBus
support.
sys/xen/xenbus/xenbusvar.h:
sys/xen/xenbus/xenbus_xs.c:
sys/xen/xenbus/xenbus_comms.c:
sys/xen/xenbus/xenbus_comms.h:
sys/xen/xenstore/xenstorevar.h:
sys/xen/xenstore/xenstore.c:
Split XenStore into its own tree. XenBus is a software layer built
on top of XenStore. The old arrangement and the naming of some
structures and functions blurred these lines making it difficult to
discern what services are provided by which layer and at what times
these services are available (e.g. during system startup and shutdown).
sys/xen/xenbus/xenbus_client.c:
sys/xen/xenbus/xenbus.c:
sys/xen/xenbus/xenbus_probe.c:
sys/xen/xenbus/xenbusb.c:
sys/xen/xenbus/xenbusb.h:
Split up XenBus code into methods available for use by client
drivers (xenbus.c) and code used by the XenBus "bus code" to
enumerate, attach, detach, and service bus drivers.
sys/xen/reboot.c:
sys/dev/xen/control/control.c:
Add a XenBus front driver for handling shutdown, reboot, suspend, and
resume events published in the XenStore. Move all PV suspend/reboot
support from reboot.c into this driver.
sys/xen/blkif.h:
New file from Xen vendor with macros and structures used by
a block back driver to service requests from a VM running a
different ABI (e.g. amd64 back with i386 front).
sys/conf/files:
Adjust kernel build spec for new XenBus/XenStore layout and added
Xen functionality.
sys/dev/xen/balloon/balloon.c:
sys/dev/xen/netfront/netfront.c:
sys/dev/xen/blkfront/blkfront.c:
sys/xen/xenbus/...
sys/xen/xenstore/...
o Rename XenStore APIs and structures from xenbus_* to xs_*.
o Adjust to use of M_XENBUS and M_XENSTORE malloc types for allocation
of objects returned by these APIs.
o Adjust for changes in the bus interface for Xen drivers.
sys/xen/xenbus/...
sys/xen/xenstore/...
Add Doxygen comments for these interfaces and the code that
implements them.
sys/dev/xen/blkback/blkback.c:
o Rewrite the Block Back driver to attach properly via newbus,
operate correctly in both PV and HVM mode regardless of domain
(e.g. can be in a DOM other than 0), and to deal with the latest
metadata available in XenStore for block devices.
o Allow users to specify a file as a backend to blkback, in addition
to character devices. Use the namei lookup of the backend path
to automatically configure, based on file type, the appropriate
backend method.
The current implementation is limited to a single outstanding I/O
at a time to file backed storage.
sys/dev/xen/blkback/blkback.c:
sys/xen/interface/io/blkif.h:
sys/xen/blkif.h:
sys/dev/xen/blkfront/blkfront.c:
sys/dev/xen/blkfront/block.h:
Extend the Xen blkif API: Negotiable request size and number of
requests.
This change extends the information recorded in the XenStore
allowing block front/back devices to negotiate for optimal I/O
parameters. This has been achieved without sacrificing backward
compatibility with drivers that are unaware of these protocol
enhancements. The extensions center around the connection protocol
which now includes these additions:
o The back-end device publishes its maximum supported values for,
request I/O size, the number of page segments that can be
associated with a request, the maximum number of requests that
can be concurrently active, and the maximum number of pages that
can be in the shared request ring. These values are published
before the back-end enters the XenbusStateInitWait state.
o The front-end waits for the back-end to enter either the InitWait
or Initialize state. At this point, the front end limits it's
own capabilities to the lesser of the values it finds published
by the backend, it's own maximums, or, should any back-end data
be missing in the store, the values supported by the original
protocol. It then initializes it's internal data structures
including allocation of the shared ring, publishes its maximum
capabilities to the XenStore and transitions to the Initialized
state.
o The back-end waits for the front-end to enter the Initalized
state. At this point, the back end limits it's own capabilities
to the lesser of the values it finds published by the frontend,
it's own maximums, or, should any front-end data be missing in
the store, the values supported by the original protocol. It
then initializes it's internal data structures, attaches to the
shared ring and transitions to the Connected state.
o The front-end waits for the back-end to enter the Connnected
state, transitions itself to the connected state, and can
commence I/O.
Although an updated front-end driver must be aware of the back-end's
InitWait state, the back-end has been coded such that it can
tolerate a front-end that skips this step and transitions directly
to the Initialized state without waiting for the back-end.
sys/xen/interface/io/blkif.h:
o Increase BLKIF_MAX_SEGMENTS_PER_REQUEST to 255. This is
the maximum number possible without changing the blkif
request header structure (nr_segs is a uint8_t).
o Add two new constants:
BLKIF_MAX_SEGMENTS_PER_HEADER_BLOCK, and
BLKIF_MAX_SEGMENTS_PER_SEGMENT_BLOCK. These respectively
indicate the number of segments that can fit in the first
ring-buffer entry of a request, and for each subsequent
(sg element only) ring-buffer entry associated with the
"header" ring-buffer entry of the request.
o Add the blkif_request_segment_t typedef for segment
elements.
o Add the BLKRING_GET_SG_REQUEST() macro which wraps the
RING_GET_REQUEST() macro and returns a properly cast
pointer to an array of blkif_request_segment_ts.
o Add the BLKIF_SEGS_TO_BLOCKS() macro which calculates the
number of ring entries that will be consumed by a blkif
request with the given number of segments.
sys/xen/blkif.h:
o Update for changes in interface/io/blkif.h macros.
o Update the BLKIF_MAX_RING_REQUESTS() macro to take the
ring size as an argument to allow this calculation on
multi-page rings.
o Add a companion macro to BLKIF_MAX_RING_REQUESTS(),
BLKIF_RING_PAGES(). This macro determines the number of
ring pages required in order to support a ring with the
supplied number of request blocks.
sys/dev/xen/blkback/blkback.c:
sys/dev/xen/blkfront/blkfront.c:
sys/dev/xen/blkfront/block.h:
o Negotiate with the other-end with the following limits:
Reqeust Size: MAXPHYS
Max Segments: (MAXPHYS/PAGE_SIZE) + 1
Max Requests: 256
Max Ring Pages: Sufficient to support Max Requests with
Max Segments.
o Dynamically allocate request pools and segemnts-per-request.
o Update ring allocation/attachment code to support a
multi-page shared ring.
o Update routines that access the shared ring to handle
multi-block requests.
sys/dev/xen/blkfront/blkfront.c:
o Track blkfront allocations in a blkfront driver specific
malloc pool.
o Strip out XenStore transaction retry logic in the
connection code. Transactions only need to be used when
the update to multiple XenStore nodes must be atomic.
That is not the case here.
o Fully disable blkif_resume() until it can be fixed
properly (it didn't work before this change).
o Destroy bus-dma objects during device instance tear-down.
o Properly handle backend devices with powef-of-2 sector
sizes larger than 512b.
sys/dev/xen/blkback/blkback.c:
Advertise support for and implement the BLKIF_OP_WRITE_BARRIER
and BLKIF_OP_FLUSH_DISKCACHE blkif opcodes using BIO_FLUSH and
the BIO_ORDERED attribute of bios.
sys/dev/xen/blkfront/blkfront.c:
sys/dev/xen/blkfront/block.h:
Fix various bugs in blkfront.
o gnttab_alloc_grant_references() returns 0 for success and
non-zero for failure. The check for < 0 is a leftover
Linuxism.
o When we negotiate with blkback and have to reduce some of our
capabilities, print out the original and reduced capability before
changing the local capability. So the user now gets the correct
information.
o Fix blkif_restart_queue_callback() formatting. Make sure we hold
the mutex in that function before calling xb_startio().
o Fix a couple of KASSERT()s.
o Fix a check in the xb_remove_* macro to be a little more specific.
sys/xen/gnttab.h:
sys/xen/gnttab.c:
Define GNTTAB_LIST_END publicly as GRANT_REF_INVALID.
sys/dev/xen/netfront/netfront.c:
Use GRANT_REF_INVALID instead of driver private definitions of the
same constant.
sys/xen/gnttab.h:
sys/xen/gnttab.c:
Add the gnttab_end_foreign_access_references() API.
This API allows a client to batch the release of an array of grant
references, instead of coding a private for loop. The implementation
takes advantage of this batching to reduce lock overhead to one
acquisition and release per-batch instead of per-freed grant reference.
While here, reduce the duration the gnttab_list_lock is held during
gnttab_free_grant_references() operations. The search to find the
tail of the incoming free list does not rely on global state and so
can be performed without holding the lock.
sys/dev/xen/xenpci/evtchn.c:
sys/dev/xen/evtchn/evtchn.c:
sys/xen/xen_intr.h:
o Implement the bind_interdomain_evtchn_to_irqhandler API for HVM mode.
This allows an HVM domain to serve back end devices to other domains.
This API is already implemented for PV mode.
o Synchronize the API between HVM and PV.
sys/dev/xen/xenpci/xenpci.c:
o Scan the full region of CPUID space in which the Xen VMM interface
may be implemented. On systems using SuSE as a Dom0 where the
Viridian API is also exported, the VMM interface is above the region
we used to search.
o Pass through bus_alloc_resource() calls so that XenBus drivers
attaching on an HVM system can allocate unused physical address
space from the nexus. The block back driver makes use of this
facility.
sys/i386/xen/xen_machdep.c:
Use the correct type for accessing the statically mapped xenstore
metadata.
sys/xen/interface/hvm/params.h:
sys/xen/xenstore/xenstore.c:
Move hvm_get_parameter() to the correct global header file instead
of as a private method to the XenStore.
sys/xen/interface/io/protocols.h:
Sync with vendor.
sys/xeninterface/io/ring.h:
Add macro for calculating the number of ring pages needed for an N
deep ring.
To avoid duplication within the macros, create and use the new
__RING_HEADER_SIZE() macro. This macro calculates the size of the
ring book keeping struct (producer/consumer indexes, etc.) that
resides at the head of the ring.
Add the __RING_PAGES() macro which calculates the number of shared
ring pages required to support a ring with the given number of
requests.
These APIs are used to support the multi-page ring version of the
Xen block API.
sys/xeninterface/io/xenbus.h:
Add Comments.
sys/xen/xenbus/...
o Refactor the FreeBSD XenBus support code to allow for both front and
backend device attachments.
o Make use of new config_intr_hook capabilities to allow front and back
devices to be probed/attached in parallel.
o Fix bugs in probe/attach state machine that could cause the system to
hang when confronted with a failure either in the local domain or in
a remote domain to which one of our driver instances is attaching.
o Publish all required state to the XenStore on device detach and
failure. The majority of the missing functionality was for serving
as a back end since the typical "hot-plug" scripts in Dom0 don't
handle the case of cleaning up for a "service domain" that is not
itself.
o Add dynamic sysctl nodes exposing the generic ivars of
XenBus devices.
o Add doxygen style comments to the majority of the code.
o Cleanup types, formatting, etc.
sys/xen/xenbus/xenbusb.c:
Common code used by both front and back XenBus busses.
sys/xen/xenbus/xenbusb_if.m:
Method definitions for a XenBus bus.
sys/xen/xenbus/xenbusb_front.c:
sys/xen/xenbus/xenbusb_back.c:
XenBus bus specialization for front and back devices.
MFC after: 1 month
- Re-enable TSO. This was broken previously due to CSUM_TSO clearing the
CSUM_TCP flag, so our checksum flags were incorrectly set going to the
netback driver. That was fixed in r206844 in tcp_output.c, so we can
turn TSO back on here.
- Fix the way transmit slots are calculated, so that we can't overfill
the ring.
- Avoid sending packets with more fragments/segments than netback can
handle. The Linux netback code can only handle packets of
MAX_SKB_FRAGS, which turns out to be 18 on machines with 4K pages. We
can easily generate packets with 32 or so fragments with TSO turned on.
Right now the solution is just to drop the packets (since netback
doesn't seem to handle it gracefully), but we should come up with a way
to allow a driver to tell the TCP stack the maximum number of fragments
it can handle in a single packet.
- Fix the way the consumer is tracked in the receive path. It could get
out of sync fairly easily.
- Use standard Xen ring macros to make it clearer how netfront is using
the rings.
- Get rid of Linux-ish negative errno return values.
- Added more documentation to the driver.
- Refactored code to make it easier to read.
- Some other minor fixes.
Reviewed by: gibbs
Reviewed by: gibbs
Sponsored by: Spectra Logic
MFC after: 7 days
have the delayed function take an argument as to the offset
to the SCTP header. This allows it to work for V4 and V6.
This of course means changing all callers of the function
to either pass the header len, if they have it, or create
it (ip_hl << 2 or sizeof(ip6_hdr)).
PR: 144529
MFC after: 2 weeks
blkif interface in Xen requires all I/O to be 512 byte aligned with each
segment bounded by a 4k page.
Note: This submission only documents the proper contraints for blkif I/O.
The alignment code in busdma does not yet handle alignment constraints
correctly in all cases.
- Use CONSOLE_DRIVER() instead of the deprecated CONS_DRIVER() declaration.
- This means we cannot use cn_checkc anymore, which is supposed to do
the same as cn_getc nowadays. Remove the cn_getc implementation (that
was never being called) and rename cn_checkc to cn_getc.
- Don't run-time patch cn_putc, but add the logic to xc_cnputc().
This means I could do some cleanups to our console code...
Tested by: nobody on hackers@
The newbus lock is responsible for protecting newbus internIal structures,
device states and devclass flags. It is necessary to hold it when all
such datas are accessed. For the other operations, softc locking should
ensure enough protection to avoid races.
Newbus lock is automatically held when virtual operations on the device
and bus are invoked when loading the driver or when the suspend/resume
take place. For other 'spourious' operations trying to access/modify
the newbus topology, newbus lock needs to be automatically acquired and
dropped.
For the moment Giant is also acquired in some key point (modules subsystem)
in order to avoid problems before the 8.0 release as module handlers could
make assumptions about it. This Giant locking should go just after
the release happens.
Please keep in mind that the public interface can be expanded in order
to provide more support, if there are really necessities at some point
and also some bugs could arise as long as the patch needs a bit of
further testing.
Bump __FreeBSD_version in order to reflect the newbus lock introduction.
Reviewed by: ed, hps, jhb, imp, mav, scottl
No answer by: ariff, thompsa, yongari
Tested by: pho,
G. Trematerra <giovanni dot trematerra at gmail dot com>,
Brandon Gooch <jamesbrandongooch at gmail dot com>
Sponsored by: Yahoo! Incorporated
Approved by: re (ksmith)
I don't want people to override the mutex when allocating a TTY. It has
to be there, to keep drivers like syscons happy. So I'm creating a
tty_alloc_mutex() which can be used in those cases. tty_alloc_mutex()
should eventually be removed.
The advantage of this approach, is that we can just remove a function,
without breaking the regular API in the future.
to dequeue a packet.
The tx path was trying to ensure that enough Xenbus TX ring slots existed but
it didn't check to see whether the mbuf TX ring slots were also available.
They get freed in xn_txeof() which occurs after transmission, rather than earlier
on in the process. (The same happens under Linux too.)
Due to whatever reason (CPU use, scheduling, memory constraints, whatever) the
mbuf TX ring may not have enough slots free and would allocate slot 0. This is
used as the freelist head pointer to represent "free" mbuf TX ring slots; setting
this to an actual mbuf value rather than an id crashes the code.
This commit introduces some basic code to track the TX mbuf ring use and then
(hopefully!) ensures that enough slots are free in said TX mbuf ring before it
enters the actual work loop.
A few notes:
* Similar logic needs to be introduced to check there are enough actual slots
available in the xenbuf TX ring. There's some logic which is invoked earlier
but it doesn't hard-check against the number of available ring slots.
Its trivial to do; I'll do it in a subsequent commit.
* As I've now commented in the source, it is likely possible to deadlock the
driver under certain conditions where the rings aren't receiving any changes
(which I should enumerate) and thus Xen doesn't send any further software
interrupts. I need to make sure that the timer(s) are running right and
the queues are periodically kicked.
PR: 134926
Slot 0 must always remain "free" and be a pointer to the first free entry in the
mbuf descriptor list. It is thus an error to have code allocate or push slot 0
back into the list.
we, like TCP and UDP, move the checksum calculation
into the IP routines when there is no hardware support
we call into the normal SCTP checksum routine.
The next round of SCTP updates will use
this functionality. Of course the IGB driver needs
a few updates to support the new intel controller set
that actually does SCTP csum offload too.
Reviewed by: gnn, rwatson, kmacy
Inside the kernel, the minor() function was responsible for obtaining
the device minor number of a character device. Because we made device
numbers dynamically allocated and independent of the unit number passed
to make_dev() a long time ago, it was actually a misnomer. If you really
want to obtain the device number, you should use dev2udev().
We already converted all the drivers to use dev2unit() to obtain the
device unit number, which is still used by a lot of drivers. I've
noticed not a single driver passes NULL to dev2unit(). Even if they
would, its behaviour would make little sense. This is why I've removed
the NULL check.
Ths commit removes minor(), minor2unit() and unit2minor() from the
kernel. Because there was a naming collision with uminor(), we can
rename umajor() and uminor() back to major() and minor(). This means
that the makedev(3) manual page also applies to kernel space code now.
I suspect umajor() and uminor() isn't used that often in external code,
but to make it easier for other parties to port their code, I've
increased __FreeBSD_version to 800062.
