first and the native ia32 compat as middle (before other things).
o(ld)brandinfo as well as third party like linux, kfreebsd, etc.
stays on SI_ORDER_ANY coming last.
The reason for this is only to make sure that even in case we would
overflow the MAX_BRANDS sized array, the native FreeBSD brandinfo
would still be there and the system would be operational.
Reviewed by: kib
MFC after: 1 month
Remove the altkstacks, instead instantiate threads with kernel stack
allocated with the right size from the start. For the thread that has
kernel stack cached, verify that requested stack size is equial to the
actual, and reallocate the stack if sizes differ [1].
This fixes the bug introduced by r173361 that was committed several days
after r173004 and consisted of kthread_add(9) ignoring the non-default
kernel stack size.
Also, r173361 removed the caching of the kernel stacks for a non-first
thread in the process. Introduce separate kernel stack cache that keeps
some limited amount of preallocated kernel stacks to lower the latency
of thread allocation. Add vm_lowmem handler to prune the cache on
low memory condition. This way, system with reasonable amount of the
threads get lower latency of thread creation, while still not exhausting
significant portion of KVA for unused kstacks.
Submitted by: peter [1]
Discussed with: jhb, julian, peter
Reviewed by: jhb
Tested by: pho (and retested according to new test scenarious)
MFC after: 1 week
allocated with the right size from the start. For the thread that has
kernel stack cached, verify that requested stack size is equial to the
actual, and reallocate the stack if sizes differ [1].
This fixes the bug introduced by r173361 that was committed several days
after r173004 and consisted of kthread_add(9) ignoring the non-default
kernel stack size.
Also, r173361 removed the caching of the kernel stacks for a non-first
thread in the process. Introduce separate kernel stack cache that keeps
some limited amount of preallocated kernel stacks to lower the latency
of thread allocation. Add vm_lowmem handler to prune the cache on
low memory condition. This way, system with reasonable amount of the
threads get lower latency of thread creation, while still not exhausting
significant portion of KVA for unused kstacks.
Submitted by: peter [1]
Discussed with: jhb, julian, peter
Reviewed by: jhb
Tested by: pho
MFC after: 1 week
- The device is based on Marvell 88F6281 system on chip.
- More info about the platform at http://www.plugcomputer.org
- To build the FreeBSD kernel:
make buildkernel TARGET_ARCH=arm KERNCONF=SHEEVAPLUG
- Installation notes at: http://wiki.freebsd.org/FreeBSDMarvell
Submitted by: Michal Hajduk
Obtained from: Semihalf
Modules on Marvell SOC can be selectively PM-disabled, and we must not access
disabled devices' registers (attempt to initialize them) unconditionally, as
this leads to the system hang. This patch introduces graceful handling of the
PM state during devices init.
Submitted by: Michal Hajduk
Obtained from: Semihalf
will initialize the FIFO memory correctly on attach. Before
that this values was intialized in only in at91_usart_bus_attach
which is called after the uart(4) memory allocation happens.
Approved by: re (kib)
MFC after: 1 week
vnet.h, we now use jails (rather than vimages) as the abstraction
for virtualization management, and what remained was specific to
virtual network stacks. Minor cleanups are done in the process,
and comments updated to reflect these changes.
Reviewed by: bz
Approved by: re (vimage blanket)
a device pager (OBJT_DEVICE) object in that it uses fictitious pages to
provide aliases to other memory addresses. The primary difference is that
it uses an sglist(9) to determine the physical addresses for a given offset
into the object instead of invoking the d_mmap() method in a device driver.
Reviewed by: alc
Approved by: re (kensmith)
MFC after: 2 weeks
On some ARM variations CPU func dispatcher has the D-cache invalidate method
point to write-back invalidate, which is wrong, and can lead to a crash/panic
on affected platforms.
Spotted by: HPS
Reviewed by: cognet
Approved by: re (kib)
a) nocache-remap problem
When a page is remapped into a non-cacheable virtual memory region there
was no associated write-back invalidate operation performed. We remove
writeback of the original buffer size from bus_dmamem_alloc() and add
appropriate L1/L2 flush operation.
b) missing write-back invalidate operation
In pmap_kremove a page is removed so we must do a write-back
invalidate operation aligned to the page virtual address.
Submitted by: Michal Hajduk
Reviewed by: Mark Tinguely, rpaulo, stas
Approved by: re (kib)
Obtained from: Semihalf
dependent memory attributes:
Rename vm_cache_mode_t to vm_memattr_t. The new name reflects the
fact that there are machine-dependent memory attributes that have
nothing to do with controlling the cache's behavior.
Introduce vm_object_set_memattr() for setting the default memory
attributes that will be given to an object's pages.
Introduce and use pmap_page_{get,set}_memattr() for getting and
setting a page's machine-dependent memory attributes. Add full
support for these functions on amd64 and i386 and stubs for them on
the other architectures. The function pmap_page_set_memattr() is also
responsible for any other machine-dependent aspects of changing a
page's memory attributes, such as flushing the cache or updating the
direct map. The uses include kmem_alloc_contig(), vm_page_alloc(),
and the device pager:
kmem_alloc_contig() can now be used to allocate kernel memory with
non-default memory attributes on amd64 and i386.
vm_page_alloc() and the device pager will set the memory attributes
for the real or fictitious page according to the object's default
memory attributes.
Update the various pmap functions on amd64 and i386 that map pages to
incorporate each page's memory attributes in the mapping.
Notes: (1) Inherent to this design are safety features that prevent
the specification of inconsistent memory attributes by different
mappings on amd64 and i386. In addition, the device pager provides a
warning when a device driver creates a fictitious page with memory
attributes that are inconsistent with the real page that the
fictitious page is an alias for. (2) Storing the machine-dependent
memory attributes for amd64 and i386 as a dedicated "int" in "struct
md_page" represents a compromise between space efficiency and the ease
of MFCing these changes to RELENG_7.
In collaboration with: jhb
Approved by: re (kib)
o add to platforms where it was missing (arm, i386, powerpc, sparc64, sun4v)
o define as "1" on amd64 and i386 where there is no restriction
o make the type returned consistent with ALIGN
o remove _ALIGNED_POINTER
o make associated comments consistent
Reviewed by: bde, imp, marcel
Approved by: re (kensmith)
DPCPU area was not properly mapped into kernel VA space, which caused page
fault on the first DPCPU access. This patch fixes the problem by mapping DPCPU
area into kernel VA space.
Submitted by: Michal Hajduk, Piotr Ziecik
Reviewed by: cognet, stas
Approved by: re (kib)
Obtained from: Semihalf
IF_ADDR_UNLOCK() across network device drivers when accessing the
per-interface multicast address list, if_multiaddrs. This will
allow us to change the locking strategy without affecting our driver
programming interface or binary interface.
For two wireless drivers, remove unnecessary locking, since they
don't actually access the multicast address list.
Approved by: re (kib)
MFC after: 6 weeks
required by video card drivers. Specifically, this change introduces
vm_cache_mode_t with an appropriate VM_CACHE_DEFAULT definition on all
architectures. In addition, this changes adds a vm_cache_mode_t parameter
to kmem_alloc_contig() and vm_phys_alloc_contig(). These will be the
interfaces for allocating mapped kernel memory and physical memory,
respectively, with non-default cache modes.
In collaboration with: jhb
- Modules and kernel code alike may use DPCPU_DEFINE(),
DPCPU_GET(), DPCPU_SET(), etc. akin to the statically defined
PCPU_*. Requires only one extra instruction more than PCPU_* and is
virtually the same as __thread for builtin and much faster for shared
objects. DPCPU variables can be initialized when defined.
- Modules are supported by relocating the module's per-cpu linker set
over space reserved in the kernel. Modules may fail to load if there
is insufficient space available.
- Track space available for modules with a one-off extent allocator.
Free may block for memory to allocate space for an extent.
Reviewed by: jhb, rwatson, kan, sam, grehan, marius, marcel, stas
driver's i/o ops must be locked to avoid chaos. Extend the cambria
bus tag to support ata and add a spin lock. The ata driver is
hacked to use that instead of it's builtin hack for ixp425. Once
the ata driver is fixed to not be confused about byte order we can
generalize the cambria bus tag code and make it generally useful.
While here take advantage of our being ixp435-specific to remove
delays when switching between byte+word accesses and to eliminate
the 2us delay for the uarts (the spin lock overhead looks to do
this for us).
used for the optional GPS+RS485 uarts on the Gateworks Cambria boards
which otherwise are unreliable
o setup the hack bus space tag for the GPS+RS485 uarts
o program the gpio interrupts for the uarts to be edge-rising
o force timing on the expansion bus for the uarts to be "slow"
Thanks to Chris Lang of Gateworks for these tips.
structure. When the page is shared, the kernel mapping becomes a special
type of managed page to force the cache off the page mappings. This is
needed to avoid stale entries on all ARM VIVT caches, and VIPT caches
with cache color issue.
Submitted by: Mark Tinguely
Reviewed by: alc
Tested by: Grzegorz Bernacki, thompsa
causes both to become inoperative; this apparently was done by the original
IAL code as a workaround for IMEM parity errors which we've not seen so
just disable the reset.
Note this problem does not occur on IXP425 boards. The linux driver does
fuse-resets on each NPE but in the order NPE-A < NPE-B < NPE-C (when probing
for which NPE's are present/operational); we may want to switch to a similar
scheme but for now disable the resets until we see an issue.
normally taken from the hints file so this should have no effect
o set the port address "just in case"
o add NPE-A support to the tx done qmgr callback
so that it isn't exposured unless needed. In particular this means
that it's easier to tune the memory layout based on board details.
While here, remove inclusion of <machine/intr.h> from mvreg.h. This
also contains exposure to SoC specifics in MI drivers, because NIRQ
depends on the SoC.
the implementation can guarantee forward progress in the event of
a stuck interrupt or interrupt storm. This is especially critical
for fast interrupt handlers, as they can cause a hard hang in that
case. When first called, arm_get_next_irq() is passed -1.
Obtained from: Juniper Networks, Inc.
When pages are removed from virtual address space by calling pmap_remove_all()
CPU caches were not invalidated, which led to read corruption when another
page got mapped at this same virtual address at later time (the CPU was
retrieving stale contents).
Submitted by: Piotr Ziecik
Obtained from: Semihalf
CPU for too long period than necessary. Additively, interfaces are kept
polled (in the tick) even if no more packets are available.
In order to avoid such situations a new generic mechanism can be
implemented in proactive way, keeping track of the time spent on any
packet and fragmenting the time for any tick, stopping the processing
as soon as possible.
In order to implement such mechanism, the polling handler needs to
change, returning the number of packets processed.
While the intended logic is not part of this patch, the polling KPI is
broken by this commit, adding an int return value and the new flag
IFCAP_POLLING_NOCOUNT (which will signal that the return value is
meaningless for the installed handler and checking should be skipped).
Bump __FreeBSD_version in order to signal such situation.
Reviewed by: emaste
Sponsored by: Sandvine Incorporated
The system hostname is now stored in prison0, and the global variable
"hostname" has been removed, as has the hostname_mtx mutex. Jails may
have their own host information, or they may inherit it from the
parent/system. The proper way to read the hostname is via
getcredhostname(), which will copy either the hostname associated with
the passed cred, or the system hostname if you pass NULL. The system
hostname can still be accessed directly (and without locking) at
prison0.pr_host, but that should be avoided where possible.
The "similar information" referred to is domainname, hostid, and
hostuuid, which have also become prison parameters and had their
associated global variables removed.
Approved by: bz (mentor)
possible future I-cache coherency operation can succeed. On ARM
for example the L1 cache can be (is) virtually mapped, which
means that any I/O that uses temporary mappings will not see the
I-cache made coherent. On ia64 a similar behaviour has been
observed. By flushing the D-cache, execution of binaries backed
by md(4) and/or NFS work reliably.
For Book-E (powerpc), execution over NFS exhibits SIGILL once in
a while as well, though cpu_flush_dcache() hasn't been implemented
yet.
Doing an explicit D-cache flush as part of the non-DMA based I/O
read operation eliminates the need to do it as part of the
I-cache coherency operation itself and as such avoids pessimizing
the DMA-based I/O read operations for which D-cache are already
flushed/invalidated. It also allows future optimizations whereby
the bcopy() followed by the D-cache flush can be integrated in a
single operation, which could be implemented using on-chips DMA
engines, by-passing the D-cache altogether.
registers contents.
- Use memory barriers to preserve the order of buffer space operations.
This might be needed if we'll ever use this driver on architectures
where ordering is not guaranteed.
used in some cases):
- Ignore DMA tag boundaries when allocating bounce pages. The boundaries
don't determine whether or not parts of a DMA request bounce. Instead,
they are just used to carve up segments.
- Allow tags with sub-page alignment to share bounce pages since bounce
pages are always page aligned.
Reviewed by: scottl (amd64)
MFC after: 1 month
operates in the common memory mode and use polling mode to control
the status of operations as I don't have any board with interrupt
line routed yet. I'll add the GPIO interrupt driven mode as soon
as I get one.
respectivly. This will allow one to have a kernel with both devices
present and use it for multiple boards with different types of RTC
sitting on a bus.
Discussed with: imp
a fair number of static data structures, making this an unlikely
option to try to change without also changing source code. [1]
Change default cache line size on ia64, sparc64, and sun4v to 128
bytes, as this was what rtld-elf was already using on those
platforms. [2]
Suggested by: bde [1], jhb [2]
MFC after: 2 weeks
CACHE_LINE_SIZE constant. These constants are intended to
over-estimate the cache line size, and be used at compile-time
when a run-time tuning alternative isn't appropriate or
available.
Defaults for all architectures are 64 bytes, except powerpc
where it is 128 bytes (used on G5 systems).
MFC after: 2 weeks
Discussed on: arch@
a tag that has BUS_DMA_KEEP_PG_OFFSET set. Otherwise the page could be
reused with a non-zero offset by a tag that doesn't have
BUS_DMA_KEEP_PG_OFFSET leading to data corruption.
Sleuthing by: avg
Reviewed by: scottl
