the ones which run the message ring handler.
Some bits of the interrupt mask are part of the status register which is
saved with the process context, and these bits are initialized from the
cpu on which the process is created. This means that all the processes
should have the same value for these interrupt mask bits, so that the
interrupt mask remains the same regardless of what thread is scheduled
on the cpu.
Submitted by: Sriram Gorti (srgorti at netlogicmicro dot com)
o) Clear/acknowledge receive interrupt at end of thread. This gives the
management interfaces performance on the order of 100Mbps rather than
the previous level of 10Mbps on my MR-730.
o) Add 'octm', a trivial driver for the 10/100 management ports found on some
Octeon systems.
o) Make the Simple Executive's management port helper routines compile on
FreeBSD (namely by not doing math on void pointers.)
o) Add a cvmx_mgmt_port_sendm routine to the Simple Executive to send an mbuf
so there is only one copy in the transmit path, rather than having to first
copy the mbuf to an intermediate buffer and then copy that to the Simple
Executive's transmit ring.
o) Properly work out MII addresses of management ports on the Lanner MR-730.
XXX The MR-730 also needs some patches to the MII read/write routines, but
this is sufficient for now. Media detection will be fixed in the future
when I can spend more time reading the vendor-supplied patches.
quite awful, because e.g. 4 packets will come in and get processed on 4
different cores at the same time, really battling with the TCP stack quite
painfully. For now, just run one task at a time.
This gets performance up in most cases to where it was before the correctness
fixes that got interrupts to run on all cores (except in high-load TCP transmit
cases where all we're handling receive for is ACKs) and in some cases it's
better now. What would be ideal would be to use a more advanced interrupt
mitigation strategy and possibly to use different workqueue groups per port for
multi-port systems, and so on, but this is a fine stopgap.
and pointers don't always have the same size, e.g. the __mips_n32 ABI
(ILP32) has 64 bit registers but 32 bit pointers.
On mips introduce PRIptr to fix the format specifier for (u)intptr_t.
Prefix PRI64 and PRIptr with underscores because macro names starting with
PRI[a-zX] are reserved for future use.
Approved by: kib (mentor)
architecture macros (__mips_n64, __powerpc64__) when 64 bit types (and
corresponding macros) are different from 32 bit. [1]
Correct the type of INT64_MIN, INT64_MAX and UINT64_MAX.
Define (U)INTMAX_C as an alias for (U)INT64_C matching the type definition
for (u)intmax_t. Do this on all architectures for consistency.
Suggested by: bde [1]
Approved by: kib (mentor)
of (unsigned) int __attribute__((__mode__(__DI__))). This aligns better
with macros such as (U)INT64_C, (U)INT64_MAX, etc. which assume (u)int64_t
has type (unsigned) long long.
The mode attribute was used because long long wasn't standardised until
C99. Nowadays compilers should support long long and use of the mode
attribute is discouraged according to GCC Internals documentation.
The type definition has to be marked with __extension__ to support
compilation with "-std=c89 -pedantic".
Discussed with: bde
Approved by: kib (mentor)
On some architectures UCHAR_MAX and USHRT_MAX had type unsigned int.
However, lacking integer suffixes for types smaller than int, their type
should correspond to that of an object of type unsigned char (or short)
when used in an expression with objects of type int. In that case unsigned
char (short) are promoted to int (i.e. signed) so the type of UCHAR_MAX and
USHRT_MAX should also be int.
Where MIN/MAX constants implicitly have the correct type the suffix has
been removed.
While here, correct some comments.
Reviewed by: bde
Approved by: kib (mentor)
It was used mainly to discover and fix some 64-bit portability problems
before 64-bit arches were widely available.
Discussed with: bde
Approved by: kib (mentor)
The macros here for generating coprocessor 0 accessors are named like:
MIPS_RDRW32_COP0
That macro would produce mips_rd_<register>() and mips_wr_<register>()
inlines to access the specified register by name from C. The problem is that
the R and the W were swapped in the macros originally; it was meant to be named
RDWR because it generated mips_rd_* and mips_wr_* functions, but was instead
spelled RDRW, which nobody should be expected to get right by anything other
than copy and paste.
It's too many consonants in a row to keep straight anyway, so just prefer e.g.:
MIPS_RW32_COP0
While here, add a missing #undef.
running an o32 kernel safely, and would have to add interrupt disabling and
reenabling to a bunch of macros in the Simple Executive sources to support it.
The only reason one would run an o32 kernel on Octeon would be to run o32 world,
which is better worked towards by adding o32 binary compatibility to n64 kernels
along with, eventually, supporting multilib systems so o32 binaries can run
alongside n32 and n64 ones.
Discussed with: imp
o) Make the octeon_wdog driver work on multi-CPU systems and to also print more
information on NMI that may aid debugging. Simplify and clean up internal
API and structure.
o) There's no good reason to make the low bits of the ebase the core
number. While per-CPU exception bases would be nice, for now we just
need to make ebase the same on all cores.
not just that it is greater than the minimal kernel virtual address, but also
that it is less than the maximal kernel virtual address. On n64 kernels, the
pcpup comes out of a direct-mapped address that, with an unsigned compare, is
rather greater than the minimal kernel virtual address.
o) Turn the panic if interrupts are disabled in cpu_idle into a KASSERT since on
other architectures it's behind INVARIANTS anyway.
o) Add a check that not all interrupts are masked, too.
o) Add cpu_idleclock() and cpu_activeclock() use to cpu_idle as is done on other
architectures.
2MB of memory in the bootmem allocator for the SDK to use internally at a later
point. It'd be nice if there were some functions we could call before
allocating memory to let various facilities reserve some memory, but for now
this seems sufficient. Previously some unfortunate systems could give up all
(or at least most) of their memory to the kernel from bootmem, and then
allocating command queues for packet output and the like would fail later in
the boot process (which in turn would lead to crashes even later.)
Reported by: kan
- Remove the -shared flag for the trampoline binary, generate an
ELF executable instead of a shared object.
- No need to generate tmphack.S, move the code to sys/mips/mips/inckern.S
- No need generate opt_kernname.h, KERNNAME can be passed with -D
Reviewed by: gonzo, imp
- ds1374u : use multi-byte write.
- at24co2n, max6657: remove mutex, iicbus has the necessary locking.
Submitted by: Sreekanth M. S. (kanthms at netlogicmicro com)
1) 32-bit assignment are expected to always be atomic.
2) Release/acquire memory barrier semantics doesn't seem to be needed here.
So a simple assignment can be used.
Remove unused port_set_counter() while here, it also used to mis-use
atomic_set_int().
Reported by: jhb
Pointyhat to: avg
MFC after: 3 weeks
- Major update to xlr_i2c.c: do multi-byte ops correctly, remove unnecessary
code, add mutex to protect bus operations, style(9) fixes.
- Drivers for I2C devices on XLR/XLS engineering boards, ds1374u RTC, max6657
temparature sensor and at24co2n EEPROM.
Submitted by: Sreekanth M. S. (kanthms at netlogicmicro com)
- remove unused code in mips/rmi/xlr_pci.c
- remove unused variable in mips/rmi/dev/nlge/if_nlge.c
- fix reference to old function in mips/mips/pmap.c
Reported by: Prabhath Raman (prabhath at netlogicmicro com)
Implement uma_small_alloc() and uma_small_free() for mips that allocates
pages from direct mapped memory. Uses the same mechanism as the page table
page allocator, so that we allocate from KSEG0 in 32 bit, and from XKPHYS
on 64 bit.
Reviewed by: alc, jmallett
Passing a count of zero on i386 and amd64 for [I386|AMD64]_BUS_SPACE_MEM
causes a crash/hang since the 'loop' instruction decrements the counter
before checking if it's zero.
PR: kern/80980
Discussed with: jhb
link state polling or media-specific ones, while avoidiing changing link state
on interfaces that use miibus; this substantially speeds up link time on
interface (re)initialization.
list on exit from the transmit path. The scatter-gather list itself can be
asynchronously DMAed to the transmit hardware, and we could actually lock up
the transmitter if any of a number of races around this were lost.
Instead, let the PKO free the scatter-gather list when it is done with it, and
use the "i" bit in each segment of the packet to avoid having them go into the
FPA.
This fixes an unrecoverable transmit stall under transmit load.
MFC after: 3 days
facilities as well as support for the Octeon 2 family of SoCs.
XXX Note that with our antediluvian assembler, we can't support some Octeon 2
instructions and fall back to using the old ones instead.
o) Remove some options that are configurable on Linux but not FreeBSD.
o) Centralize open/poll/stop routines for XAUI and SGMII and use the common
uninit routine directly rather than providing a wrapper for it. The init
functions for these interfaces are now identical and the common init routine
could merge in setting those function pointers except that some hardware
seems to use no open/poll/stop method?
The current implementation of vm_page_alloc_freelist() does not handle
order > 0 correctly. Remove order parameter to the function and use it
only for order 0 pages.
Submitted by: alc
DPCPU_DEFINE and VNET_DEFINE macros, as these cause problems for various
people working on the affected files. A better long-term solution is
still being considered. This reversal may give some modules empty
set_pcpu or set_vnet sections, but these are harmless.
Changes reverted:
------------------------------------------------------------------------
r215318 | dim | 2010-11-14 21:40:55 +0100 (Sun, 14 Nov 2010) | 4 lines
Instead of unconditionally emitting .globl's for the __start_set_xxx and
__stop_set_xxx symbols, only emit them when the set_vnet or set_pcpu
sections are actually defined.
------------------------------------------------------------------------
r215317 | dim | 2010-11-14 21:38:11 +0100 (Sun, 14 Nov 2010) | 3 lines
Apply the STATIC_VNET_DEFINE and STATIC_DPCPU_DEFINE macros throughout
the tree.
------------------------------------------------------------------------
r215316 | dim | 2010-11-14 21:23:02 +0100 (Sun, 14 Nov 2010) | 2 lines
Add macros to define static instances of VNET_DEFINE and DPCPU_DEFINE.
mipsel' or 'machine mips mipseb' into the config file (with a few 64's
tossed in for good measure). This will let us build the proper
kernels with different worlds as part of make universe.
contents of the ones that were not empty were stale and unused.
- Now that <machine/mutex.h> no longer exists, there is no need to allow it
to override various helper macros in <sys/mutex.h>.
- Rename various helper macros for low-level operations on mutexes to live
in the _mtx_* or __mtx_* namespaces. While here, change the names to more
closely match the real API functions they are backing.
- Drop support for including <sys/mutex.h> in assembly source files.
Suggested by: bde (1, 2)
work properly with single-stepping in a kernel debugger. Specifically,
these routines have always disabled interrupts before increasing the nesting
count and restored the prior state of interrupts after decreasing the nesting
count to avoid problems with a nested interrupt not disabling interrupts
when acquiring a spin lock. However, trap interrupts for single-stepping
can still occur even when interrupts are disabled. Now the saved state of
interrupts is not saved in the thread until after interrupts have been
disabled and the nesting count has been increased. Similarly, the saved
state from the thread cannot be read once the nesting count has been
decreased to zero. To fix this, use temporary variables to store interrupt
state and shuffle it between the thread's MD area and the appropriate
registers.
In cooperation with: bde
MFC after: 1 month
the NIC drivers as well as the PHY drivers to take advantage of the
mii_attach() introduced in r213878 to get rid of certain hacks. For
the most part these were:
- Artificially limiting miibus_{read,write}reg methods to certain PHY
addresses; we now let mii_attach() only probe the PHY at the desired
address(es) instead.
- PHY drivers setting MIIF_* flags based on the NIC driver they hang
off from, partly even based on grabbing and using the softc of the
parent; we now pass these flags down from the NIC to the PHY drivers
via mii_attach(). This got us rid of all such hacks except those of
brgphy() in combination with bce(4) and bge(4), which is way beyond
what can be expressed with simple flags.
While at it, I took the opportunity to change the NIC drivers to pass
up the error returned by mii_attach() (previously by mii_phy_probe())
and unify the error message used in this case where and as appropriate
as mii_attach() actually can fail for a number of reasons, not just
because of no PHY(s) being present at the expected address(es).
Reviewed by: jhb, yongari
using miibus, since for some devices that use multiple addresses on the bus,
going through miibus may be unclear, and for devices that are not standard
MII PHYs, miibus may throw a fit, necessitating complicated interfaces to
fake the interface that it expects during probe/attach.
o) Make the mv88e61xx SMI interface in octe attach a PHY directly and fix some
mistakes in the code that resulted from trying too hard to present a nice
interface to miibus.
o) Add a PHY driver for the mv88e61xx. If attached (it is optional in kernel
compiles so the default behavior of having a dumb switch is preserved) it
will place the switch in a VLAN-tagging mode such that each physical port
has a VLAN associated with it and interfaces for the VLANs can be created to
address or bridge between them.
XXX It would be nice for this to be part of a single module including the
SMI interface, and for it to fit into a generic switch configuration
framework and for it to use DSA rather than VLANs, but this is a start
and gives some sense of the parameters of such frameworks that are not
currently present in FreeBSD. In lieu of a switch configuration
interface, per-port media status and VLAN settings are in a sysctl tree.
XXX There may be some minor nits remaining in the handling of broadcast,
multicast and unknown destination traffic. It would also be nice to go
through and replace the few remaining magic numbers with macros at some
point in the future.
XXX This has only been tested with the MV88E6161, but it should work with
minimal or no modification on related switches, so support for probing
them was included.
Thanks to Pat Saavedra of TELoIP and Rafal Jaworowski of Semihalf for their
assistance in understanding the switch chipset.
physical address. Adds a dma tag to the XLR/XLS pci bus with the
lowaddr if the CPU happens to be a XLR C rev.
Submitted by: Sreekanth M. S. (kanthms at netlogicmicro dot com))
- nlge_ioctl handles IFF_UP and IFF_PROMISC flags
- Translate table code, to enable flow based CPU assignment added
disabled by default (can be enabled by a tunable).
- Changed signature of nlge_port_disable to make it consistent with nlge_port_enable
- Removed TXCSUM and VLAN_HW_TAGGING from i/f capabilities.
Submitted by: Sriram Gorti (srgorti at netlogicmicro dot com)
sysctl and counters for message ring threads (intial version). Update
watermark values, and and decrease the maximum threads to 3 (this will leave
a few CPUs for other processes)
Minor comment fix in nlge.
- Wakeup multiple threads per core using message ring watermark interrupts.
- Update message ring handler registration, use the real device station id
for registering interrupts.
- rge/nlge: update for the new message ring registration code.
- rge/nlge: use 2 message ring stations for incoming packets, this will
allow more messages to be queued.
- nlge: comment fixes, remove unused variable
- style and whitespace fixes
the miibus attached to octe interfaces.
o) Add an SMI/MDIO interface to the MV88E61XX and use it for the switch PHY on
the Lanner MR-320. An actual driver for the switch PHY will come later.
Note that for now it intercepts and fakes MII_BMSR reads to prevent the
miibus from talking to anything but the switch itself.
bus interface does that's special here now is to use a 64-bit register size.
In theory, uart(4) ought to support a regsz as well as regshft and support
64-bit registers directly.
Also use the UART class's range rather than a hand-coded 1024 for the address
range.
o) Give a virtual address for I/O ports on n64.
o) On the Portwell CAM-0100, return the right IRQ for the on-board SATA.
o) Except on bridges, only set PORTEN and MEMEN on devices that have I/O or
memory BARs respectively.
o) Disable PORTEN and MEMEN while reprogramming BARs.
o) On the Lanner MR-955, set the Tx DMA power register for the on-board Promise
SATA controller.
- XLS B0 and later revision chips have PCIe link 2 & 3 mapped to different
PIC interrupts. Update pic.h, board.h and xlr_pci.c to reflect this.
- remove debug prints in xlr_pci.c
- add more processor IDs to board.h, add function xlr_is_xls_b0()
- some style(9) and whitespace fixes
these could be made dependent on either of the octusb or octe options, but
making them standard fixes a number of option combinations that were previously
broken.
driver to try to switch interrupt handlers at setup. It's not a very
good implementation of bus_teardown_intr, though.
o) Set cache line size and latency timers for PCI devices per Linux.
o) Reset and configure the bus from scratch rather than expecting U-Boot to
do it for us. Values and configuration from Linux, U-Boot and comments
in the Cavium Simple Executive sources.
o) Do a resource assignment and bus numbering pass in the absence of a PCI
BIOS or firmware that will do it for us.
XXX This has to be the third or fourth instance of this in FreeBSD and
it would be nice to have it become part of the PCI bus driver itself,
like it is on Linux.
o) Fix interrupt mapping for and adjust bus configuration for the Lanner
MR-955, based on information provided by Lanner.
- 64 bit fixes for ifnlge.c
- Use m_nextpkt to save mbuf vaddr on 64 bit, we cannot store the
64 bit vaddr in the 40bit freeback field.
- remove unused code and unnecessary variables.
- use xlr_io_mmio macro instead of adding io base address
- rewrite GPIO related code to fixup nlge using xlr_write_reg and DELAY
- support for engg boards major num 11 and 12
- add xlr_paddr_lw() to load 32bit value from physical address, fix
inline assembly
- style fixes
- Process some tx done messages in the transmit path, to ensure that
the XLR NA tx done FIFO does not overflow.
- Add a message ring handler API to process atmost a given number of
messages from a specified bucket mask. This will be used to process
the tx done messages
- Add a callout to restart transmit in the case transmit gets blocked.
- Update enable_msgring_int() and disable_msgring_int(), remove unused
args and make static.
Obtained from: Sriram Gorti (srgorti at netlogicmicro dot com)
- Compile fixes for 9.0, the previous version of this driver was
for FreeBSD 6.
- Add virtual address field in OperationDescriptor_t, we cannot use
MIPS_PHYS_TO_KSEG0 on physical address.
- Fixes for new message ring API
- Remove unused sys/mips/rmi/dev/sec/stats.h
- Whitespace fixes
- Move RMI MIPS extension to atomic increment word (LDADDWU) to common
header file sys/mips/rmi/rmi_mips_exts.h
- Fix xlr_ldaddwu() for 64 bit, it is a 32 bit operation, use
unsigned int* instead of unsigned long* argument
- Provide dummy xlr_enable_kx/xlr_restore_kx for n32 and n64.
- Provide xlr_paddr_ld() instead of xlr_paddr_lw(), so that the
descriptor formats are same for 32 and 64 bit
- update nlge and rge for the changes
These changes are also needed by the security driver which will be
added later.
Do not explicitly enable interrupts in smp_init_secondary() because it
renders any spinlock protected code after that point to run with
interrupts enabled. This is because the processor is executing in the
context of idlethread whose 'md_spinlock_count' is already set to 1.
Instead just let sched_throw() re-enable interrupts when it releases
the spinlock.
The original powerpc commit log for r212559 is available here:
http://svn.freebsd.org/viewvc/base?view=revision&revision=212559
- Remove sync from msgrng_send, sync needs to be called just once before
sending.
- Fix retry logic - don't reload registers when retrying in message_send,
also fix check for send pending fail.
- remove unused message_send_block_fast()
- merge message_receive_fast() to message_receive
- style(9) fixes, and comments
- rge and nlge updated for the sys/mips/rmi/msgring.h changes
The main goal of this is to generate timer interrupts only when there is
some work to do. When CPU is busy interrupts are generating at full rate
of hz + stathz to fullfill scheduler and timekeeping requirements. But
when CPU is idle, only minimum set of interrupts (down to 8 interrupts per
second per CPU now), needed to handle scheduled callouts is executed.
This allows significantly increase idle CPU sleep time, increasing effect
of static power-saving technologies. Also it should reduce host CPU load
on virtualized systems, when guest system is idle.
There is set of tunables, also available as writable sysctls, allowing to
control wanted event timer subsystem behavior:
kern.eventtimer.timer - allows to choose event timer hardware to use.
On x86 there is up to 4 different kinds of timers. Depending on whether
chosen timer is per-CPU, behavior of other options slightly differs.
kern.eventtimer.periodic - allows to choose periodic and one-shot
operation mode. In periodic mode, current timer hardware taken as the only
source of time for time events. This mode is quite alike to previous kernel
behavior. One-shot mode instead uses currently selected time counter
hardware to schedule all needed events one by one and program timer to
generate interrupt exactly in specified time. Default value depends of
chosen timer capabilities, but one-shot mode is preferred, until other is
forced by user or hardware.
kern.eventtimer.singlemul - in periodic mode specifies how much times
higher timer frequency should be, to not strictly alias hardclock() and
statclock() events. Default values are 2 and 4, but could be reduced to 1
if extra interrupts are unwanted.
kern.eventtimer.idletick - makes each CPU to receive every timer interrupt
independently of whether they busy or not. By default this options is
disabled. If chosen timer is per-CPU and runs in periodic mode, this option
has no effect - all interrupts are generating.
As soon as this patch modifies cpu_idle() on some platforms, I have also
refactored one on x86. Now it makes use of MONITOR/MWAIT instrunctions
(if supported) under high sleep/wakeup rate, as fast alternative to other
methods. It allows SMP scheduler to wake up sleeping CPUs much faster
without using IPI, significantly increasing performance on some highly
task-switching loads.
Tested by: many (on i386, amd64, sparc64 and powerc)
H/W donated by: Gheorghe Ardelean
Sponsored by: iXsystems, Inc.
This reflects actual type used to store and compare child device orders.
Change is mostly done via a Coccinelle (soon to be devel/coccinelle)
semantic patch.
Verified by LINT+modules kernel builds.
Followup to: r212213
MFC after: 10 days
- Provide 64 bit implementations for some macros. On n64 and n32,
don't split 64 bit values.
- No need for 32 bit ops for control registers.
- Fix few bugs (write control reg, write_c0_register64).
- Re-write EIRR/EIMR/CPUID operations using read_c0_registerXX, no
need of inline assembly.
- rename control reg access functions to avoid phnx, update callers.
- stlye/whitespace fixes.
- Updates for the message ring clean up in r212321.
- Instead of dropping Tx packet on credit fail, retry send until it
succeeds.
- Fix freeing mbufs in case of P2P descriptors:
We cannot free the mbuf when the P2P descriptor freeback is received. The
mbuf may be still in use by the GMAC, since the P2P freeback indicates that
it read the P2D descriptors in the P2P message.
Now we free just the P2P descriptor when the P2P freeback message is
received. Another freeback P2D message has been added to the end of
the packet descriptors, the mbuf will be freed only when we received
this.
The P2P descriptor issue was reported by srgorti at netlogicmicro dot com.
Fix message ring send path:
- define msgrng_access_enable() which disables local interrupts
and enables message ring access. Also define msgrng_restore() which
restores interrupts
- remove all other msgrng enable/disable macros, no need of critical_enter
and other locking here.
- message_send() fixup: re-read status until pending bit clears
- message_send_retry() fixup: retry only few times with interrupts disabled
- Fix up message_send/message_send_retry callers - call
msgrng_access_enable() and msgrng_restore() correctly so that interrupts
are not disabled for long.
- removed unused and obsolete code from sys/mips/rmi/msgring.h
- some style fixes - more later
rge.c (XLR GMAC driver):
- updated for the message ring changes
- remove unused message_send_block()
- retry on credit failure, this is not a permanent failure when credits
are configured correctly. Add panic if credits are not available to
send for a long time.
Fix interrupt routing so that the irq returned is correct for XLR and
XLS. This also updates the MSI hack we had earlier - we still don't
really support MSI, but we support some drivers that use MSI, by providing
support for allocating one MSI per pci link - this MSI is directly
mapped to the link IRQ.
- set cache_coherent_dma flag in cpuinfo for XLR, this will make sure that
BUS_DMA_COHERENT flag is handled correctly in busdma_machdep.c
- iodi.c, call device_get_name() just once
- clear RMI specific EIRR while intializing CPUs
- remove debug print in intr_machdep.c
unused files.
- remove clock.c and clock.h, these are not used after the new timer
code was added.
- remove duplicated include files, fix header file ordering, remove
some unneeded includes.
- rename mips/rmi/shared_structs.h which contains the RMI boot loader
interface to mips/rmi/rmi_boot_info.h. Remove unused files
mips/rmi/shared_structs_func.h and sys/mips/rmi/shared_structs_offsets.h
- merge mips/rmi/xlrconfig.h and mips/rmi/rmi_mips_exts.h, and remove
duplicated functions.
- nlge - minor change to remove unneeded argument.
- Add FreeBSD svn keyword for headers
PMAP_DIAGNOSTIC was eliminated from amd64/i386, and, in fact, the
non-MIPS parts of the kernel, several years ago. Any of the interesting
checks were turned into KASSERT()s. Basically, the motivation was that
lots of people run with INVARIANTS but no one runs with DIAGNOSTIC.
panic strings needn't and shouldn't have a terminating newline.
Finally, there is one functional change. The sched_pin() in
pmap_remove_pages() is an artifact of the way we temporarily map page
table pages on i386. (The mappings are processor private. We don't do
a system-wide shootdown.) It isn't needed by MIPS.
Tested by: jchandra
Submitted by: alc
and XAUI 10G interfaces in addition RGMII/SGMII 1G interfaces. This driver
is work in progress.
board.c and board.h expanded to include more info.
Only one of rge and nlge can be enabled at a time, rge will be deprecated
when nlge stabilizes.
Submitted by: Sriram Gorti <srgorti at netlogicmicro com>
SMP.
We used to route all PIC based interrupts to cpu 0, and used the per-CPU
interrupt mask to enable/disable interrupts. But the interrupt threads can
run on any cpu on SMP, and the interrupt thread will re-enable the interrupts
on the CPU it runs on when it is done, and not on cpu0 where the PIC will
still send interrupts to.
The fix is move the disable/enable for PIC based interrupts to PIC, we will
ack on PIC only when the interrupt thread is done, and we do not use the
per-CPU interrupt mask.
The changes also introduce a way for subsystems to add a function that
will be called to clear the interrupt on the subsystem. Currently This is
used by the PCI/PCIe for doing additional work during the interrupt
handling.
- Use timer 7 in XLR PIC as a 32 counter
- provide pic_init_timer(), pic_set_timer(), pic_timer_count32() and
pic_timer_count() PIC timer operations.
- register this timer as platform_timecounter on rmi platform.
- style(9) fixes to mips/rmi platform files
- update pic.h to add pic_setup_intr() and use pic_setup_intr() for setting
up interrupts which are routed thru PIC.
- remove rmi_spin_mutex_safe and haslock, and make sure that the functions
are called only after mutexes are available.
- move PIC code to xlr_machdep.c
- move fast message ring code completely to on_chip.c
- move memory initialization to a new function xlr_mem_init()
- style fixes
* Add a function to write to the relevant PLL register
* Break out the PLL configuration for the AR71XX into the CPU ops,
lifted from if_arge.c.
* Add the AR91XX PLL configuration ops, using the AR91XX register
definitions.
This is untested but should at least allow an AR724X to boot.
The current code is lacking the detail needed to expose the PCIe bus.
It is also lacking any NIC, PLL or flush/WB code.
This works well enough to bring a system up to single-user mode
using an MDROOT.
Known Issues:
* The EHCI USB doesn't currently work and will panic the kernel during
attach.
* The onboard ethernet won't work until the PLL routines have been
fleshed out and shoe-horned into if_arge.
* The WMAC device glue (and quite likely the if_ath support)
hasn't yet been implemented.
* Implement a SoC probe function, from Linux, which determines the
SoC family, type and revision. This only probes the AR71xx series
SoC and (currently) panics on others.
* Migrate some of the AR71XX specific hardware init (USB device, determining
system frequencies) into using the cpuops introduced in an earlier commit.
Other SoC specific hardware stuff (per-device flush/WB, GPIO pin wiring,
Ethernet PLL setup, other things I've likely missed) will be introduced in
subsequent commits.
Reviewed by: imp@
Obtained from: (partially) Linux
1. On n64, use XKPHYS to map page table pages instead of KSEG0. Maintain
just one freepages list on n64.
The changes are mainly to introduce MIPS_PHYS_TO_DIRECT(pa),
MIPS_DIRECT_TO_PHYS(), which will use KSEG0 in 32 bit compilation
and XKPHYS in 64 bit compilation.
2. Change macro based PMAP_LMEM_MAP1(), PMAP_LMEM_MAP2(), PMAP_LMEM_UNMAP()
to inline functions.
3. Introduce MIPS_DIRECT_MAPPABLE(pa), which will further reduce the cases
in which we will need to have a special case for 64 bit compilation.
4. Update CP0 hazard definitions for CPU_RMI - the cpu does not need any
nops
Reviewed by: neel
Each of these SoCs have different devices, different hardware initialisation
methods and, quite likely, different quirks. These functions will abstract
out the SoC differences and keep these differences out of the drivers (eg
USB init, if_arge, etc.)
In particular, provide pagesize and pagesizes array, the canary value
for SSP use, number of host CPUs and osreldate.
Tested by: marius (sparc64)
MFC after: 1 month
- Enable KX and UX bits on CPU startup for non-boot CPUs
- Keep the KX bit when in userspace - XTLB handler needs it to access
PCPU data
- revert r210638 partly - we don't need to enable KX on kernel entry
now
Reviewed by: jmallett, imp
1. Move dirty bit emulation code that is duplicted for kernel and user
in trap.c to a function pmap_emulate_modified() in pmap.c.
2. While doing dirty bit emulation, it is not necessary to update the
TLB entry on all CPUs using smp_rendezvous(), we can just update the
TLB entry on the current CPU, and let the other CPUs update their TLB
entry lazily if they get an exception.
Reviewed by: alc, neel
per-cpu variants are also available to be called. The per-cpu variants
are needed for some later optimizations.
Also remove unnecessary casts, do some style fixes.
Reviewed by: alc, neel
r211130 in favor of this more general fix.
This fixes a compilation error for mips 64-bit little endian build.
libexec/rtld-elf/mips/reloc.c:196: warning: right shift count >= width of type
Suggested by: stefanf, jchandra, bde
IPI to a specific CPU by its cpuid. Replace calls to ipi_selected() that
constructed a mask for a single CPU with calls to ipi_cpu() instead. This
will matter more in the future when we transition from cpumask_t to
cpuset_t for CPU masks in which case building a CPU mask is more expensive.
Submitted by: peter, sbruno
Reviewed by: rookie
Obtained from: Yahoo! (x86)
MFC after: 1 month
MIPS doesn't really need to use atomic_cmpset_int() in situations like
this because the software dirty bit emulation in trap.c acquires
the pmap lock. Atomics like this appear to be a carryover from i386
where the hardware-managed TLB might concurrently set the modified bit.
Reviewed by: alc
pmap_page_wired_mappings() counts the number of pv entries for the
specified page that have the pv entry wired flag set to TRUE.
pmap_enter() correctly initializes this flag. However,
pmap_change_wiring() doesn't update the corresponding pv entry flag,
only the PTE. So, the count returned by pmap_page_wired_mappings()
will sometimes be wrong.
In the short term, the best fix would be to eliminate the pv entry
flag and use only the PTE. That flag is wasting non-trivial memory.
Remove pv_wired flag, and use PTE flag to count the wired mappings.
Reviewed by: alc
'counter_upper' and 'counter_lower_last'. The race exists because
interrupts are enabled even though tick_ticker() executes in a
critical section.
Fix a bug in clock_intr() in how it updates the cached values of
'counter_upper' and 'counter_lower_last'. They are updated only
when the COUNT register rolls over. More interestingly it will *never*
update the cached values if 'counter_lower_last' happens to be zero.
Get rid of superfluous critical section in clock_intr(). There is no
reason to do this because clock_intr() executes in hard interrupt
context.
Switch back to using 'tick_ticker()' as the cpu ticker for Sibyte.
Reviewed by: jmallett, mav
- 32 bit compilation will still use old 2 level page tables
- re-arrange pmap code so that adding another level is easier
- pmap code for 3 level page tables for n64
- update TLB handler to traverse 3 levels in n64
Reviewed by: jmallett
The emulation of 'ld' and 'sd' instructions only works for ABIs that support
64-bit registers and the instructions 'ldl' and 'ldr' that operate on those
registers.
Reviewed by: jmallett
that with a 32-bit ABI on a system with 64-bit registers can attempt to
access an invalid (well, kernel) memory address rather than the intended
user address for stack-relative loads and stores. Lowering the stack
pointer works around this. [1]
o) Make TRAP_DEBUG code conditional on the trap_debug variable. Make
trap_debug default to 0 instead of 1 now but make it possible to change it
at runtime using sysctl.
o) Kill programs that attempt an unaligned access of a kernel address. Note
that with some ABIs, calling useracc() is not sufficient since the register
may be 64-bit but vm_offset_t is 32-bit so a kernel address could be
truncated to what looks like a valid user address, allowing the user to
crash the kernel.
o) Clean up unaligned access emulation to support unaligned 16-bit and 64-bit
accesses. (For 16-bit accesses it was checking for user access to too much
memory (4 bytes) and there was no 64-bit support.) This still lacks support
for unaligned load-linked and store-conditional.
Reviewed by: [1] gonzo
now it uses a very dumb first-touch allocation policy. This will change in
the future.
- Each architecture indicates the maximum number of supported memory domains
via a new VM_NDOMAIN parameter in <machine/vmparam.h>.
- Each cpu now has a PCPU_GET(domain) member to indicate the memory domain
a CPU belongs to. Domain values are dense and numbered from 0.
- When a platform supports multiple domains, the default freelist
(VM_FREELIST_DEFAULT) is split up into N freelists, one for each domain.
The MD code is required to populate an array of mem_affinity structures.
Each entry in the array defines a range of memory (start and end) and a
domain for the range. Multiple entries may be present for a single
domain. The list is terminated by an entry where all fields are zero.
This array of structures is used to split up phys_avail[] regions that
fall in VM_FREELIST_DEFAULT into per-domain freelists.
- Each memory domain has a separate lookup-array of freelists that is
used when fulfulling a physical memory allocation. Right now the
per-domain freelists are listed in a round-robin order for each domain.
In the future a table such as the ACPI SLIT table may be used to order
the per-domain lookup lists based on the penalty for each memory domain
relative to a specific domain. The lookup lists may be examined via a
new vm.phys.lookup_lists sysctl.
- The first-touch policy is implemented by using PCPU_GET(domain) to
pick a lookup list when allocating memory.
Reviewed by: alc
booting again.
The code is a copy of the mips/mips/tick.c with minor modifications for
XLR interrupt handling. Disable mips/rmi/clock.c for now, the PIC based
timer code will be added later.
alc@.
The UMA zone based allocation is replaced by a scheme that creates
a new free page list for the KSEG0 region, and a new function
in sys/vm that allocates pages from a specific free page list.
This also fixes a race condition introduced by the UMA based page table
page allocation code. Dropping the page queue and pmap locks before
the call to uma_zfree, and re-acquiring them afterwards will introduce
a race condtion(noted by alc@).
The changes are :
- Revert the earlier changes in MIPS pmap.c that added UMA zone for
page table pages.
- Add a new freelist VM_FREELIST_HIGHMEM to MIPS vmparam.h for memory that
is not directly mapped (in 32bit kernel). Normal page allocations will first
try the HIGHMEM freelist and then the default(direct mapped) freelist.
- Add a new function 'vm_page_t vm_page_alloc_freelist(int flind, int
order, int req)' to vm/vm_page.c to allocate a page from a specified
freelist. The MIPS page table pages will be allocated using this function
from the freelist containing direct mapped pages.
- Move the page initialization code from vm_phys_alloc_contig() to a
new function vm_page_alloc_init(), and use this function to initialize
pages in vm_page_alloc_freelist() too.
- Split the function vm_phys_alloc_pages(int pool, int order) to create
vm_phys_alloc_freelist_pages(int flind, int pool, int order), and use
this function from both vm_page_alloc_freelist() and vm_phys_alloc_pages().
Reviewed by: alc
on-board USB controller. It is not currently enabled because there are
known problems with device communication and until those are fixed I am not
certain that it won't destabilize the system. [1]
o) Add the "cryptocteon" opencrypto device based on the OCF device written by
David McCullough. It is not currently enabled because until support for
saving/restoring coprocessor 2 state on context switch is available, it runs
with interrupts disabled, which tends to pessimize performance over using a
software crypto facility. Tests using this driver which are not negatively
affected by it running with interrupts disabled show it to be substantially
faster than software for large blocks.
Submitted by: hps [1]
library:
o) Increase inline unit / large function growth limits for MIPS to accommodate
the needs of the Simple Executive, which uses a shocking amount of inlining.
o) Remove TARGET_OCTEON and use CPU_CNMIPS to do things required by cnMIPS and
the Octeon SoC.
o) Add OCTEON_VENDOR_LANNER to use Lanner's allocation of vendor-specific
board numbers, specifically to support the MR320.
o) Add OCTEON_BOARD_CAPK_0100ND to hard-wire configuration for the CAPK-0100nd,
which improperly uses an evaluation board's board number and breaks board
detection at runtime. This board is sold by Portwell as the CAM-0100.
o) Add support for the RTC available on some Octeon boards.
o) Add support for the Octeon PCI bus. Note that rman_[sg]et_virtual for IO
ports can not work unless building for n64.
o) Clean up the CompactFlash driver to use Simple Executive macros and
structures where possible (it would be advisable to use the Simple Executive
API to set the PIO mode, too, but that is not done presently.) Also use
structures from FreeBSD's ATA layer rather than structures copied from
Linux.
o) Print available Octeon SoC features on boot.
o) Add support for the Octeon timecounter.
o) Use the Simple Executive's routines rather than local copies for doing reads
and writes to 64-bit addresses and use its macros for various device
addresses rather than using local copies.
o) Rename octeon_board_real to octeon_is_simulation to reduce differences with
Cavium-provided code originally written for Linux. Also make it use the
same simplified test that the Simple Executive and Linux both use rather
than our complex one.
o) Add support for the Octeon CIU, which is the main interrupt unit, as a bus
to use normal interrupt allocation and setup routines.
o) Use the Simple Executive's bootmem facility to allocate physical memory for
the kernel, rather than assuming we know which addresses we can steal.
NB: This may reduce the amount of RAM the kernel reports you as having if
you are leaving large temporary allocations made by U-Boot allocated
when starting FreeBSD.
o) Add a port of the Cavium-provided Ethernet driver for Linux. This changes
Ethernet interface naming from rgmxN to octeN. The new driver has vast
improvements over the old one, both in performance and functionality, but
does still have some features which have not been ported entirely and there
may be unimplemented code that can be hit in everyday use. I will make
every effort to correct those as they are reported.
o) Support loading the kernel on non-contiguous cores.
o) Add very conservative support for harvesting randomness from the Octeon
random number device.
o) Turn SMP on by default.
o) Clean up the style of the Octeon kernel configurations a little and make
them compile with -march=octeon.
o) Add support for the Lanner MR320 and the CAPK-0100nd to the Simple
Executive.
o) Modify the Simple Executive to build on FreeBSD and to build without
executive-config.h or cvmx-config.h. In the future we may want to
revert part of these changes and supply executive-config.h and
cvmx-config.h and access to the options contained in those files via
kernel configuration files.
o) Modify the Simple Executive USB routines to support getting and setting
of the USB PID.
Move inappropriate stuff in cpu.h elsewhere:
{s,g}et_intr_mask -> md_var.h
num_tlbentries -> tlb.h
Remove #define clockframe trapframe and fix clock, which was the only place
this was used.
All the rest of this stuff was unused.
# we're not quite minimal yet, since we duplicate a few status register things
# here...
Inspired by: bde@
calls mips_cpu_call via an obfuscated assembler call. Instead, delete
the current cpu_throw, and rename mips_cpu_throw to cpu_throw. This
is nicer to the cache on each context switch (since fixed jumps can be
prefected, while jumps through a register can't). Incidentally, it
also saves about 5 or 6 instructions.
Reviewed by: jmallet@
Use int32/intptr casts for exception vector names.
Define MIPS_SR_INT_MASK again
Change MIPS_XKPHYS_CCA_* to MIPS_CCA_* since we can use them in many contexts
Minor gratuitous whitespace churn
The problem with setting it there is that the last CPU to come up
wins, it seems. This also removes one more ifdef in locore.S, a noble
goal too. Since they are unused, and pollute cpu.h, remove them.
Submitted by: bde.h (cpu.h pollution)
Approved in theory by: jmallet@
Merge changes for initial n64 support in pmap.c. Use direct mapped (XKPHYS)
access for a lot of operations that earlier needed temporary mapping. Add
support for using XKSEG for kernel mappings.
Reviewed by: imp
Obtained from: jmallett (http://svn.freebsd.org/base/user/jmallett/octeon)
The existing code only checked the alignment of the first mbuf and
didn't enforce the size constraints.
This commit introduces a simple function to check the alignment and
size of all mbufs in the list. This fixes the initial issue in the
PR.
PR: kern/148307
Reviewed by: gonzo@
Updated PTE/PDE macros from http://svn.freebsd.org/base/user/jmallett/octeon
Introduce pmap_segshift() macro, use pmap_segmap() in place of pmap_pde, and
remove pmap_pde().
Approved by: rrs (mentor)
Obtained from: jmallett@
