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
This fixes fallout from r256425.
Reported by: Pavel Timofeev <timp87@gmail com>
Tested by: Pavel Timofeev <timp87@gmail com>
Reviewed by: Roger Pau Monnц╘
MFC after: 3 days
do not map the b_pages pages into buffer_map KVA. The use of the
unmapped buffers eliminate the need to perform TLB shootdown for
mapping on the buffer creation and reuse, greatly reducing the amount
of IPIs for shootdown on big-SMP machines and eliminating up to 25-30%
of the system time on i/o intensive workloads.
The unmapped buffer should be explicitely requested by the GB_UNMAPPED
flag by the consumer. For unmapped buffer, no KVA reservation is
performed at all. The consumer might request unmapped buffer which
does have a KVA reserve, to manually map it without recursing into
buffer cache and blocking, with the GB_KVAALLOC flag.
When the mapped buffer is requested and unmapped buffer already
exists, the cache performs an upgrade, possibly reusing the KVA
reservation.
Unmapped buffer is translated into unmapped bio in g_vfs_strategy().
Unmapped bio carry a pointer to the vm_page_t array, offset and length
instead of the data pointer. The provider which processes the bio
should explicitely specify a readiness to accept unmapped bio,
otherwise g_down geom thread performs the transient upgrade of the bio
request by mapping the pages into the new bio_transient_map KVA
submap.
The bio_transient_map submap claims up to 10% of the buffer map, and
the total buffer_map + bio_transient_map KVA usage stays the
same. Still, it could be manually tuned by kern.bio_transient_maxcnt
tunable, in the units of the transient mappings. Eventually, the
bio_transient_map could be removed after all geom classes and drivers
can accept unmapped i/o requests.
Unmapped support can be turned off by the vfs.unmapped_buf_allowed
tunable, disabling which makes the buffer (or cluster) creation
requests to ignore GB_UNMAPPED and GB_KVAALLOC flags. Unmapped
buffers are only enabled by default on the architectures where
pmap_copy_page() was implemented and tested.
In the rework, filesystem metadata is not the subject to maxbufspace
limit anymore. Since the metadata buffers are always mapped, the
buffers still have to fit into the buffer map, which provides a
reasonable (but practically unreachable) upper bound on it. The
non-metadata buffer allocations, both mapped and unmapped, is
accounted against maxbufspace, as before. Effectively, this means that
the maxbufspace is forced on mapped and unmapped buffers separately.
The pre-patch bufspace limiting code did not worked, because
buffer_map fragmentation does not allow the limit to be reached.
By Jeff Roberson request, the getnewbuf() function was split into
smaller single-purpose functions.
Sponsored by: The FreeBSD Foundation
Discussed with: jeff (previous version)
Tested by: pho, scottl (previous version), jhb, bf
MFC after: 2 weeks
kern_timeout_callwheel_alloc() where it is actually used.
This is a mechanical move and no tuning parameters are changed.
The pre-allocated callout array is only used for legacy timeout(9)
calls and is only allocated and active on cpu0. Eventually all
remaining users of timeout(9) should switch to the callout_* API.
Reviewed by: davide
precise time event generation. This greatly improves granularity of
callouts which are not anymore constrained to wait next tick to be
scheduled.
- Extend the callout KPI introducing a set of callout_reset_sbt* functions,
which take a sbintime_t as timeout argument. The new KPI also offers a
way for consumers to specify precision tolerance they allow, so that
callout can coalesce events and reduce number of interrupts as well as
potentially avoid scheduling a SWI thread.
- Introduce support for dispatching callouts directly from hardware
interrupt context, specifying an additional flag. This feature should be
used carefully, as long as interrupt context has some limitations
(e.g. no sleeping locks can be held).
- Enhance mechanisms to gather informations about callwheel, introducing
a new sysctl to obtain stats.
This change breaks the KBI. struct callout fields has been changed, in
particular 'int ticks' (4 bytes) has been replaced with 'sbintime_t'
(8 bytes) and another 'sbintime_t' field was added for precision.
Together with: mav
Reviewed by: attilio, bde, luigi, phk
Sponsored by: Google Summer of Code 2012, iXsystems inc.
Tested by: flo (amd64, sparc64), marius (sparc64), ian (arm),
markj (amd64), mav, Fabian Keil
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().
When maxusers was unrestricted and maxfiles was allowed to autotune
much higher the result was that ncallout which was based on maxfiles
and maxproc grew much higher than was needed.
To fix this clip autotuning to the same number we would get with
the old maxusers algorithm which would stop scaling at 384
maxusers.
Growing ncalout higher is not likely to be needed since most consumers
of timeout(9) are gone and any higher value for ncallout causes the
callwheel hashes to be much larger than will even be needed for
most applications.
MFC after: 1 month
Reviewed by: mav
variable as they may overflow on i386/PAE and i386 with > 2GB RAM.
Use 64bit quad_t instead. It has broader kernel infrastructure support
with TUNABLE_QUAD_FETCH() and qmin/qmax() than other available types.
Pointed out by: alc, bde
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
A default install on large memory machines with multiple 10gigE interfaces
were not being given enough mbufs to do full bandwidth TCP or NFS traffic.
To keep the value somewhat reasonable, we scale back the number of
maxuers by 1/6 past the 384 point. This gives us enough mbufs for most
of our pretty basic 10gigE line-speed tests to complete.
Also, express this new maximum as a fraction of the kernel's address
space size rather than a constant so that increasing KVA_PAGES will
automatically increase this maximum. As a side-effect of this change,
kern.maxvnodes will automatically increase by a proportional amount.
While I'm here ensure that this change doesn't result in an unintended
increase in maxpipekva on i386. Calculate maxpipekva based upon the
size of the kernel address space and the amount of physical memory
instead of the size of the kmem map. The memory backing pipes is not
allocated from the kmem map. It is allocated from its own submap of
the kernel map. In short, it has no real connection to the kmem map.
(In fact, the commit messages for the maxpipekva auto-sizing talk
about using the kernel map size, cf. r117325 and r117391, even though
the implementation actually used the kmem map size.) Although the
calculation is now done differently, the resulting value for
maxpipekva should remain almost the same on i386. However, on amd64,
the value will be reduced by 2/3. This is intentional. The recent
change to VM_KMEM_SIZE_SCALE on amd64 for the benefit of ZFS also had
the unnecessary side-effect of increasing maxpipekva. This change is
effectively restoring maxpipekva on amd64 to its prior value.
Eliminate init_param3() since it is no longer used.
the virtualization detection successfully disabling the clflush instruction.
This fixes insta-panics for XEN hvm users when the hw.clflush_disable
tunable is -1 or 0 (-1 by default).
Discussed with: jhb
that the virtual machine monitor has enabled machine check exceptions.
Unfortunately, on AMD Family 10h processors the machine check hardware
has a bug (Erratum 383) that can result in a false machine check exception
when a superpage promotion occurs. Thus, I am disabling superpage
promotion when the FreeBSD kernel is running as a guest operating system
on an AMD Family 10h processor.
Reviewed by: jhb, kib
MFC after: 3 days
too high due to several overflows. The actual limit is somewhere in the
neighborhood of INT_MAX/4 on 64-bit machines, but most systems could not
support such a limit due to a lack of memory and the cost of duplicate
credentials.
Reported by: bde
kern.ngroups+1. kern.ngroups can range from NGROUPS_MAX=1023 to
INT_MAX-1. Given that the Windows group limit is 1024, this range
should be sufficient for most applications.
MFC after: 1 month
under VM environments, it's too slow for FreeBSD to work
properly. For example, ping at 10hz pings about every 600ms
instead of about every second.
Approved by: re (kib)
address space sizes to be longs instead of ints. Specifically, the follow
values are now longs: runningbufspace, bufspace, maxbufspace,
bufmallocspace, maxbufmallocspace, lobufspace, hibufspace, lorunningspace,
hirunningspace, maxswzone, maxbcache, and maxpipekva. Previously, a
relatively small number (~ 44000) of buffers set in kern.nbuf would result
in integer overflows resulting either in hangs or bogus values of
hidirtybuffers and lodirtybuffers. Now one has to overflow a long to see
such problems. There was a check for a nbuf setting that would cause
overflows in the auto-tuning of nbuf. I've changed it to always check and
cap nbuf but warn if a user-supplied tunable would cause overflow.
Note that this changes the ABI of several sysctls that are used by things
like top(1), etc., so any MFC would probably require a some gross shims
to allow for that.
MFC after: 1 month
arrays under #ifndef XEN to make XEN config compile again.
In case of Xen vm_guest is hard coded.
Move the list for the vm_guest sysctl out of the restictive
bounds as the sysctl is there in either case.
it running under a virtual environment. This also introduces a globally
accessible variable vm_guest that can be used where appropriate in the
kernel to inspect this environment.
To make it easier for the long run, an enum VM_GUEST is also introduced,
which could possibly be factored out in a header somewhere (but the
question is where - vm/vm_param.h? sys/param.h?) so it eventually becomes
a part of the standard KPI. In any case, it's a start.
The purpose of all this isn't to absolutely detect that the OS is running
under a virtual environment (cf. "redpill") but to allow the parts of the
kernel and the userland that care about this particular aspect and can do
something useful depending on it to have a standardised interface. Reducing
kern.hz is one example but there are other things that could be done like
avoiding context switches, not using CPU instructions that are known to be
slow in emulation, possibly different strategies in VM (memory) allocation,
CPU scheduling, etc.
It isn't clear if the JAILS/VIMAGE functionality should also be exposed
by this particular mechanism (probably not since they're not "full"
virtual hardware environments). Sometime in the future another sysctl and
a variable could be introduced to reflect if the kernel supports any kind
of virtual hosting (e.g. VMWare VMI, Xen dom0).
Reviewed by: silence from src-commiters@, virtualization@, kmacy@
Approved by: gnn (mentor)
Security: Obscurity doesn't help.
Due to the nature of the beast it causes lot of unproductive overhead. This
is especially bad when running SMP kernel on VMWare with several virtual
processors - idle FreeBSD guest with SMP kernel takes 150% host CPU time on my
dual-core MacBook Pro when I am enabling two virtual CPUs, making even host
not very usable. Detect when we are running in the sandbox and reduce HZ
to 10 (can be adjusted via VM_HZ in the kernel config) in such cases. This
brings host CPU usage of idle FreeBSD/SMP on two virtual processors down
to 10%.
Detect most popular VM platforms out there - VMWare, Parallels, VirtualBox
and VirtualPC.
MFC after: 2 weeks
correspond to the commit log. It changed the maxswzone and maxbcache
parameters from int to long, without changing the extern definitions
in <sys/buf.h>.
In fact it's a good thing it did not, because other parts of the system
are not yet ready for this, and on large-memory sparc machines it causes
severe filesystem damage if you try.
The worst effect of the change was that the tunables controlling the
above variables stopped working. These were necessary to allow such
large sparc64 machines (with >12GB RAM) to boot, since sparc64 did not
set a hard-coded upper limit on these parameters and they ended
up overflowing an int, causing an infinite loop at boot in bufinit().
Reviewed by: mlaier
