o Fall back to default m_ext free mech, using function pointer in
m_ext_free, and remove sf_ext_free() called directly from mbuf code.
Testing on modern CPUs showed no regression.
o Provide internally used flag EXT_FLAG_SYNC, to mark that I/O uses
SF_SYNC flag. Lack of the flag allows us not to dereference
ext_arg2, saving from a cache line miss.
o Create function sendfile_free_page() that later will be used, for
multi-page mbufs. For now compiler will inline it into
sendfile_free_mext().
In collaboration with: gallatin
Differential Revision: https://reviews.freebsd.org/D12615
in mb_free_ext() always use fields from the original refcount holding
mbuf (see. r296242) mbuf. Cuts another cache miss from mb_free_ext().
However, treat EXT_EXTREF mbufs differently, since they are different -
they don't have a refcount holding mbuf.
Provide longer comments in m_ext declaration to explain this change
and change from r296242.
In collaboration with: gallatin
Differential Revision: https://reviews.freebsd.org/D12615
All of these arguments are stored in m_ext, so there is no reason
to pass them in the argument list. Not all functions need the second
argument, some don't even need the first one. The second argument
lives in next cache line, so not dereferencing it is a performance
gain. This was discovered in sendfile(2), which will be covered by
next commits.
The second goal of this commit is to bring even more flexibility
to m_ext mbufs, allowing to create more fields in m_ext, opaque to
the generic mbuf code, and potentially set and dereferenced by
subsystems.
Reviewed by: gallatin, kbowling
Differential Revision: https://reviews.freebsd.org/D12615
"iflib is a library to eliminate the need for frequently duplicated device
independent logic propagated (poorly) across many network drivers."
Participation is purely optional. The IFLIB kernel config option is
provided for drivers that want to transition between legacy and iflib
modes of operation. ixl and ixgbe driver conversions will be committed
shortly. We hope to see participation from the Broadcom and maybe
Chelsio drivers in the near future.
Submitted by: mmacy@nextbsd.org
Reviewed by: gallatin
Differential Revision: D5211
m_unshare passes on the source mbuf's flags as-is to m_getcl and this
results in a leak if the flags include M_NOFREE. The fix is to clear
the bits not listed in M_COPYALL before calling m_getcl. M_RDONLY
should probably be filtered out too but that's outside the scope of this
fix.
Add assertions in the zone_mbuf and zone_pack ctors to catch similar
bugs.
Update netmap_get_mbuf to not pass M_NOFREE to m_getcl. It's not clear
what the original code was trying to do but it's likely incorrect.
Updated code is no different functionally but it avoids the newly added
assertions.
Reviewed by: gnn@
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D5698
The mbuf provider is made up of a set of Statically Defined Tracepoints
which help us look into mbufs as they are allocated and freed. This can be
used to inspect the buffers or for a simplified mbuf leak detector.
New tracepoints are:
mbuf:::m-init
mbuf:::m-gethdr
mbuf:::m-get
mbuf:::m-getcl
mbuf:::m-clget
mbuf:::m-cljget
mbuf:::m-cljset
mbuf:::m-free
mbuf:::m-freem
There is also a translator for mbufs which gives some visibility into the structure,
see mbuf.d for more details.
Reviewed by: bz, markj
MFC after: 2 weeks
Sponsored by: Rubicon Communications (Netgate)
Differential Revision: https://reviews.freebsd.org/D5682
EXT_MOD_TYPE, EXT_DISPOSABLE types we should first execute the free
callback, then free the mbuf, otherwise we will derefernce memory that
was just freed.
Reported and tested: jhibbits
The m_ext.ext_cnt pointer becomes a union. It can now hold the refcount
value itself. To tell that m_ext.ext_flags flag EXT_FLAG_EMBREF is used.
The first mbuf to attach a cluster stores the refcount. The further mbufs
to reference the cluster point at refcount in the first mbuf. The first
mbuf is freed only when the last reference is freed.
The benefit over refcounts stored in separate slabs is that now refcounts
of different, unrelated mbufs do not share a cache line.
For EXT_EXTREF mbufs the zone_ext_refcnt is no longer needed, and m_extadd()
becomes void, making widely used M_EXTADD macro safe.
For EXT_SFBUF mbufs the sf_ext_ref() is removed, which was an optimization
exactly against the cache aliasing problem with regular refcounting.
Discussed with: rrs, rwatson, gnn, hiren, sbruno, np
Reviewed by: rrs
Differential Revision: https://reviews.freebsd.org/D5396
Sponsored by: Netflix
mbuf(9) manipulation functions into uipc_mbuf.c. This looks like
the initial intent, but had diffused in the last decade.
o Gather all declarations in mbuf.h in one place and sort them.
o Uninline m_clget() and m_cljget().
There are no functional changes in this patch.
The patch comes from a larger version, where all mbuf(9) allocation was
uninlined, which allowed to make mbuf(9) UMA zones private to kern_mbuf.c.
The performance impact of the total uninlining is still unclear, so we
are holding on now with larger version.
Together with: melifaro, olivier
exhausted.
It is possible for a bug in the code (or, theoretically, even unusual
network conditions) to exhaust all possible mbufs or mbuf clusters.
When this occurs, things can grind to a halt fairly quickly. However,
we currently do not call mb_reclaim() unless the entire system is
experiencing a low-memory condition.
While it is best to try to prevent exhaustion of one of the mbuf zones,
it would also be useful to have a mechanism to attempt to recover from
these situations by freeing "expendable" mbufs.
This patch makes two changes:
a) The patch adds a generic API to the UMA zone allocator to set a
function that should be called when an allocation fails because the
zone limit has been reached. Because of the way this function can be
called, it really should do minimal work.
b) The patch uses this API to try to free mbufs when an allocation
fails from one of the mbuf zones because the zone limit has been
reached. The function schedules a callout to run mb_reclaim().
Differential Revision: https://reviews.freebsd.org/D3864
Reviewed by: gnn
Comments by: rrs, glebius
MFC after: 2 weeks
Sponsored by: Juniper Networks
A couple of internal functions used by malloc(9) and uma truncated
a size_t down to an int. This could cause any number of issues
(e.g. indefinite sleeps, memory corruption) if any kernel
subsystem tried to allocate 2GB or more through malloc. zfs would
attempt such an allocation when run on a system with 2TB or more
of RAM.
Note to self: When this is MFCed, sparc64 needs the same fix.
Differential revision: https://reviews.freebsd.org/D2106
Reviewed by: kib
Reported by: Michael Fuckner <michael@fuckner.net>
Tested by: Michael Fuckner <michael@fuckner.net>
MFC after: 2 weeks
o Do not use UMA refcount zone. The problem with this zone is that
several refcounting words (16 on amd64) share the same cache line,
and issueing atomic(9) updates on them creates cache line contention.
Also, allocating and freeing them is extra CPU cycles.
Instead, refcount the page directly via vm_page_wire() and the sfbuf
via sf_buf_alloc(sf_buf_page(sf)) [1].
o Call refcounting/freeing function for EXT_SFBUF via direct function
call, instead of function pointer. This removes barrier for CPU
branch predictor.
o Do not cleanup the mbuf to be freed in mb_free_ext(), merely to
satisfy assertion in mb_dtor_mbuf(). Remove the assertion from
mb_dtor_mbuf(). Use bcopy() instead of manual assignments to
copy m_ext in mb_dupcl().
[1] This has some problems for now. Using sf_buf_alloc() merely to
increase refcount is expensive, and is broken on sparc64. To be
fixed.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
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
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
2 predictable branches nowadays. However as a pre-condition the
caller had to ensure that the mbuf pkthdr did not have any mtags
attached to it, costing some potential branches again.
Sponsored by: The FreeBSD Foundation
features. The changes in particular are:
o Remove rarely used "header" pointer and replace it with a 64bit protocol/
layer specific union PH_loc for local use. Protocols can flexibly overlay
their own 8 to 64 bit fields to store information while the packet is
worked on.
o Mechanically convert IP reassembly, IGMP/MLD and ATM to use pkthdr.PH_loc
instead of pkthdr.header.
o Extend csum_flags to 64bits to allow for additional future offload
information to be carried (e.g. iSCSI, IPsec offload, and others).
o Move the RSS hash type enumerator from abusing m_flags to its own 8bit
rsstype field. Adjust accessor macros.
o Add cosqos field to store Class of Service / Quality of Service information
with the packet. It is not yet supported in any drivers but allows us to
get on par with Cisco/Juniper in routing applications (plus MPLS QoS) with
a modernized ALTQ.
o Add four 8 bit fields l[2-5]hlen to store the relative header offsets
from the start of the packet. This is important for various offload
capabilities and to relieve the drivers from having to parse the packet
and protocol headers to find out location of checksums and other
information. Header parsing in drivers is a lot of copy-paste and
unhandled corner cases which we want to avoid.
o Add another flexible 64bit union to map various additional persistent
packet information, like ether_vtag, tso_segsz and csum fields.
Depending on the csum_flags settings some fields may have different usage
making it very flexible and adaptable to future capabilities.
o Restructure the CSUM flags to better signify their outbound (down the
stack) and inbound (up the stack) use. The CSUM flags used to be a bit
chaotic and rather poorly documented leading to incorrect use in many
places. Bring clarity into their use through better naming.
Compatibility mappings are provided to preserve the API. The drivers
can be corrected one by one and MFC'd without issue.
o The size of pkthdr stays the same at 48/56bytes (32/64bit architectures).
Sponsored by: The FreeBSD Foundation
to 8 bits. ext_type is an enumerator and the number of types we
have is a mere dozen.
A couple of ext_types are renumbered to fit within 8 bits.
EXT_VENDOR[1-4] and EXT_EXP[1-4] types for vendor-internal and
experimental local mapping.
The ext_flags field is currently unused but has a couple of flags
already defined for future use. Again vendor and experimental
flags are provided for local mapping.
EXT_FLAG_BITS is provided for the printf(9) %b identifier.
Initialize and copy ext_flags in the relevant mbuf functions.
Improve alignment and packing of struct m_ext on 32 and 64 archs
by carefully sorting the fields.
for a very long time, if ever.
Should such a functionality ever be needed again the appropriate and
much better way to do it is through a custom EXT_SOMETHING external mbuf
type together with a dedicated *ext_free function.
Discussed with: trociny, glebius
transparent layering and better fragmentation.
- Normalize functions that allocate memory to use kmem_*
- Those that allocate address space are named kva_*
- Those that operate on maps are named kmap_*
- Implement recursive allocation handling for kmem_arena in vmem.
Reviewed by: alc
Tested by: pho
Sponsored by: EMC / Isilon Storage Division
Now that r253351 moved sendfile() stats to a separate struct, the
last field used in mbstat is m_mcfail, which is updated, but never
read or obtained from userland.
be a GOOD IDEA (TM).
Apparently MOST users set this (e.g. tcp and friends) but there are a few
users that just assume that it is a sensible value but then go on to read it.
These include SCTP, pf and the FLOWTABLE option (and maybe others).
instead of kernel_map size to prevent kernel memory exhaustion
by mbufs and a subsequent panic on physical page allocation
failure.
On architectures without a direct map all mbuf memory (except
for jumbo mbufs larger than PAGE_SIZE) comes from kmem_map.
It is the limiting factor hence.
For architectures with a direct map using the size of kmem_map
is a good proxy of available kernel memory as well. If it is
much smaller the mbuf limit may be sub-optimal but remains
reasonable, while avoiding panics under exhaustion.
The overall mbuf memory limit calculation may be reconsidered
again later, however due to the many different mbuf sizes and
different backing KVM maps it is a tricky subject.
Found by: pho's new network stress test
Pointed out by: alc (kmem_map instead of kernel_map)
Tested by: pho
tunable_mbinit() where it is next to where it is used later.
Change the sysinit level of tunable_mbinit() from SI_SUB_TUNABLES
to SI_SUB_KMEM after the VM is running. This allows to use better
methods to determine the effectively available physical and virtual
memory available to the kernel.
Update comments.
In a second step it can be merged into mbuf_init().
- unp_zone: kern.ipc.maxsockets limit reached
- socket_zone: kern.ipc.maxsockets limit reached
- zone_mbuf: kern.ipc.nmbufs limit reached
- zone_clust: kern.ipc.nmbclusters limit reached
- zone_jumbop: kern.ipc.nmbjumbop limit reached
- zone_jumbo9: kern.ipc.nmbjumbo9 limit reached
- zone_jumbo16: kern.ipc.nmbjumbo16 limit reached
Note that those warnings are printed not often than every five minutes and can
be globally turned off by setting sysctl/tunable vm.zone_warnings to 0.
Discussed on: arch
Obtained from: WHEEL Systems
MFC after: 2 weeks
kernel memory, whichever is lower. The overall mbuf related memory
limit must be set so that mbufs (and clusters of various sizes)
can't exhaust physical RAM or KVM.
The limit is set to half of the physical RAM or KVM (whichever is
lower) as the baseline. In any normal scenario we want to leave
at least half of the physmem/kvm for other kernel functions and
userspace to prevent it from swapping too easily. Via a tunable
kern.maxmbufmem the limit can be upped to at most 3/4 of physmem/kvm.
At the same time divorce maxfiles from maxusers and set maxfiles to
physpages / 8 with a floor based on maxusers. This way busy servers
can make use of the significantly increased mbuf limits with a much
larger number of open sockets.
Tidy up ordering in init_param2() and check up on some users of
those values calculated here.
Out of the overall mbuf memory limit 2K clusters and 4K (page size)
clusters to get 1/4 each because these are the most heavily used mbuf
sizes. 2K clusters are used for MTU 1500 ethernet inbound packets.
4K clusters are used whenever possible for sends on sockets and thus
outbound packets. The larger cluster sizes of 9K and 16K are limited
to 1/6 of the overall mbuf memory limit. When jumbo MTU's are used
these large clusters will end up only on the inbound path. They are
not used on outbound, there it's still 4K. Yes, that will stay that
way because otherwise we run into lots of complications in the
stack. And it really isn't a problem, so don't make a scene.
Normal mbufs (256B) weren't limited at all previously. This was
problematic as there are certain places in the kernel that on
allocation failure of clusters try to piece together their packet
from smaller mbufs.
The mbuf limit is the number of all other mbuf sizes together plus
some more to allow for standalone mbufs (ACK for example) and to
send off a copy of a cluster. Unfortunately there isn't a way to
set an overall limit for all mbuf memory together as UMA doesn't
support such a limiting.
NB: Every cluster also has an mbuf associated with it.
Two examples on the revised mbuf sizing limits:
1GB KVM:
512MB limit for mbufs
419,430 mbufs
65,536 2K mbuf clusters
32,768 4K mbuf clusters
9,709 9K mbuf clusters
5,461 16K mbuf clusters
16GB RAM:
8GB limit for mbufs
33,554,432 mbufs
1,048,576 2K mbuf clusters
524,288 4K mbuf clusters
155,344 9K mbuf clusters
87,381 16K mbuf clusters
These defaults should be sufficient for even the most demanding
network loads.
MFC after: 1 month
Some hooks are added to clamp down maxusers and nmbclusters for
small address space systems.
VM_MAX_AUTOTUNE_MAXUSERS - the max maxusers that will be autotuned based on
physical memory.
VM_MAX_AUTOTUNE_NMBCLUSTERS - max nmbclusters based on physical memory.
These are set to the old values on i386 to preserve the clamping that was
being done to all arches.
Another macro VM_AUTOTUNE_NMBCLUSTERS is provided to allow an override
for the calculation on a MD basis. Currently no arch defines this.
Reviewed by: peter
MFC after: 2 weeks
