the PMC_IN_KERNEL() macro definition.
Add missing macros to extract the return address (LR) from the trapframe.
Discussed with: andrew
Obtained from: Cambridge/L41
Sponsored by: DARPA, AFRL
MFC after: 2 weeks
This is pretty much a complete rewrite based on the existing i386 code. The
patches have been circulating for a couple years and have been looked at by
plenty of people, but I'm not putting anybody on the hook as having reviewed
this in any formal sense except myself.
After this has gotten wider testing from the user community, ARM_NEW_PMAP
will become the default and various dregs of the old pmap code will be
removed.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz>
the startup trampoline code. The old code allocated a kva page, mapped it
using using pmap_kenter_nocache(), then freed the kva without destroying
the mapping. This is the only use of pmap_kenter_nocache() in the system,
so redoing this one use of allows it to be garbage collected in the
near future.
the Raspberry Pi B we support most of the devices are already supported,
however the base address has changed.
A few items are not working, or missing. The main ones are:
* DMA doesn't work in the sdhci driver.
* Enabling vchiq halts the boot, may be interrupt related.
* There is no U-Boot port yet so the DTB is embedded in the kernel.
The last point will make it difficult to boot FreeBSD, however there is
support for the Raspberry Pi 2 in the U-Boot git repo. As I have not tested
this it is left as an open task to create a port to build.
X-MFC: When the above issues are fixed
Sponsored by: ABT Systems Ltd
we're not looking at it.
Fix this by increasing l2->l2_occupancy before we try to alloc (and decrease
it if the allocation failed, or if another thread did a similar allocation).
Submitted by: Kohji Okuno <okuno.kohji@jp.panasonic.com>
MFC after: 1 week
to get the default frequency of the sdhci device.
While here use a u_int to hold the frequency as it may be too large to fit
in a 32-bit signed integer. This is the case when we have a 250MHz clock.
Each plaform performs virtual memory split between kernel and user space
and assigns kernel certain amount of memory space. However, is is sometimes
reasonable to change the default values. Such situation may happen on
systems where the demand for kernel buffers is high, many devices occupying
memory etc. This of course comes with the cost of decreasing user space
memory range so shall be used with care. Most embedded systems will not
suffer from this limtation but rather take advantage of this potential
since default behavior is left unchanged.
Submitted by: Wojciech Macek <wma@semihalf.com>
Reviewed by: imp
Obtained from: Semihalf
These are left over from long ago when there was no way to load modules
on early armv6 platforms, and when there was a build problem with ahc
that has long since been fixed, and they just keep getting copy-pasted
into new configs.
ntpd is hard-coded to use /dev/ppsN, and typically when multiple PPS sources
are available the sysadmin will symlink the right one(s) to /dev/ppsN, so
it's best to not usurp those names directly.
Also ensure dsb precedes isb in all icache maintenance routines (first
do a data sync, then stall the instruction stream until it finishes).
Submitted by: Michal Meloun
been done by U-Boot. This allows the USB to work when we load the kernel
directly.
No dma sync is performed after these operations as the data we read/write
is not used by the cpu after the calls to the maimbox driver.
Differential Revision: https://reviews.freebsd.org/D1940
Reviewed by: imp, Michal Meloun (meloun AT miracle.cz)
MFC after: 1 Week
Sponsored by: ABT Systems Ltd
both the post-filter and post-thread callbacks.
Also eliminate a completely unecessary write to INTC_ISR_CLEAR register,
which clears a software-generated interrupt that can only happen with a
write to INTC_ISR_SET (which nothing does).
uart implementations, and export them using the new linker-set mechanism.
Differential Revision: https://reviews.freebsd.org/D1993
Submitted by: Michal Meloun
PL (programmable logic) uses FCLK0..FCLK3 as a clock sources.
Normally they're configured by first stage boot loader (FSBL)
and normal user never has to touch them. These sysctls may come
useful for hardware developers
hw.fpga.fclk.N.source: clock source (IO, DDR, ARM)
hw.fpga.fclk.N.freq: requested frequency in Hz
hw.fpga.fclk.N.actual_freq: actual frequency in Hz (R/O)
hw.fgpa.level_shifters: 0/1 to enable/disable PS-PL level shifters,
normally they're enabled either by FSBL or after programming
FPGA through devcfg(4)
emulate the instructions used in function entry and exit.
For function entry ARM will use a push instruction to push up to 16
registers to the stack. While we don't expect all 16 to be used we need to
handle any combination the compiler may generate, even if it doesn't make
sense (e.g. pushing the program counter).
On function return we will either have a pop or branch instruction. The
former is similar to the push instruction, but with care to make sure we
update the stack pointer and program counter correctly in the cases they
are either in the list of registers or not. For branch we need to take the
24-bit offset, sign-extend it, and add that number of 4-byte words to the
program counter. Care needs to be taken as, due to historical reasons, the
address the branch is relative to is not the current instruction, but 8
bytes later.
This allows us to use the following probes on ARM boards:
dtrace -n 'fbt::malloc:entry { stack() }'
and
dtrace -n 'fbt:🆓return { stack() }'
Differential Revision: https://reviews.freebsd.org/D2007
Reviewed by: gnn, rpaulo
Sponsored by: ABT Systems Ltd
the translation table (this may be left over from armv5 days). It's
especially bad to do so using a cache operation that isn't coherent on
SMP systems.
Submitted by: Michal Meloun
used by other places that expect to unwind the stack, e.g. dtrace and
stack(9).
As I have written most of this code I'm changing the license to the
standard FreeBSD license. I have received approval from the other
developers who have changed any of the affected code.
Approved by: ian, imp, rpaulo, eadler (all license change)
Switch the cache line size during invalidations/flushes
to be read from CP15 cache type register.
Submitted by: Wojciech Macek <wma@semihalf.com>
Reviewed by: ian, imp
Obtained from: Semihalf
With this commit any of the GPIO pins can now be programmed to act as an
interrupt source for GPIO devices (i.e. limited to devices directly
attached to gpiobus - at least for now).
Differential Revision: https://reviews.freebsd.org/D1000
The DISP DPLL clock is slower and was making the PRU programs slower
on FreeBSD than on Linux.
Submitted by: Manuel Stuehn <freebsdnewbie at freenet.de>
MFC after: 1 week
const. On x86, even after the machine context is supposedly read into
the struct ucontext, lazy FPU state save code might only mark the FPU
data as hardware-owned. Later, set_fpcontext() needs to fetch the
state from hardware, modifying the *mcp.
The set_mcontext(9) is called from sigreturn(2) and setcontext(2)
implementations and old create_thread(2) interface, which throw the
*mcp out after the set_mcontext() call.
Reported by: dim
Discussed with: jhb
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Allocate all four, we will use them soon.
Simplificate the allocation of memory and interrupt resources with a single
bus_alloc_resources() call instead of doing them separately.
Destroy the mutex in case of errors.
With this change, it is now possible to verify the pin function and level
of reserved pins (but not set them).
The use of reserved pins on Raspberry pi can lead to short circuits and
real damage to the SoC.
While here, remove duplicated code, make use of OF_getencprop_alloc()
instead of using fixed sized variables and reduce the dmesg spam by
printing reserved pin ranges (when possible) instead of printing each pin
in the range.
string.
With this change it is not necessary redeclare the driver relations with
gpiobus and gpioc.
This also prevents redundant declarations when gpiobusvar.h is included
(which is going to happen soon).
Minimum LCDC is 2 so clock freq shouild be 2*max_pixel_clock. Maximum pixel
clock for HDMI is 148500 (1920x1080). But AM335x can not run in this mode
due to bandwidth and clock limitations
lock around the mapping and uiomove(). Before r277643, it was
partially protected by Giant (but potential sleeping in fault from
uiomove() would still allow other thread to reuse the mapping).
Noted by: ian
Reviewed by: alc, ian
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
1 second is not enugh for TDA19988 HDMI framer (e.g. on Beaglebone Black)
- Add per-device i2c_timout sysctl (dev.iichb.X.i2c_timeout) to control
I2C bus timeout manually
- Pass softc instead of device_t to all sysctl handlers
o Digital Audio Multiplexer (AUDMUX)
o Smart Direct Memory Access Controller (SDMA)
o Synchronous Serial Interface (SSI)
Disable by default as it depends on SDMA firmware.
Sponsored by: Machdep, Inc.
for i386, and from the code inspection, nothing in the
arm/mips/sparc64 implementations depends on it.
Discussed with: imp, nwhitehorn
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
the data the inline functions access together at the start of the bus_space
struct. The start-of part isn't so important, it's the grouping-together
that's the point: now all the most-accessed data should be in one cache line.
Suggested by: cognet
Fill in some formerly NULL members where the implementation function
exists. Add a dummy function that panics and use it as a placeholder
for thigns that are still unimplemented. Remove a few unused includes.
every operation to retrieve the bs_cookie value almost nothing actually uses.
The bus_space struct contains a private data pointer (poorly named bs_cookie,
now renamed to bs_privdata) which is used only by a few old armv4 xscale
implementations. The bus_space functions were all defined to take this
value as the first parameter instead of the bus_space_tag_t, requiring all
the inline macro and function expansions to dereference the tag to pass it
to another function, which never uses it. Now all the functions take the tag
as the first parameter and retrieve the privdata if they need it.
Also fix a couple bus_space_unmap() implementations that were calling
kva_free() instead of pmap_unmapdev().
Discussed with: cognet
driver on Rockchip boards. It currently supports PIO mode
and dma mode needs external dma controller to be used.
Submitted by: jmcneill
Approved by: stas (mentor)
This doesn't actually change any behavior, because it just allows a 16-bit
read of the command register to return the correct value, and nothing
actually does a 16-bit read of that register.
sdhci controllers, such as the one on a Raspberry Pi, mishandle the signal
timing in high speed signaling mode, but run just fine in standard mode
with the bus running at frequencies between 25-50MHz (which shouldn't work).
This is the solution adopted by U-Boot and other OSes (linux and *BSD)
for the timeouts on Raspberry Pi boards with certain SD cards. Some
research shows that this quirk is also used on a few other boards, so the
fix is a generic quirk instead of being in the RPi-specific driver code.
This change is based on information discovered by Michal Meloun.
the stack for secondary cores, the other two values are only used for zeroing
bss on the primary core. No need to store the size of the stack at the
top of the stack (seems to be a leftover instruction from some cut-n-paste).
While in theory this should have been a transparent change (and was for all
other drivers), cpsw(4) never used the proper accessor macros in a few
places but spelt the indirect m_hdr.mh_* out itself. Convert those to
use m_len and m_data and unbreak the driver build.
To cut off the power we need to start the shutdown sequence by writing
the OFF bit on PMIC.
Once the PMIC is programmed the SoC needs to toggle the PMIC_PWR_ENABLE
pin when it is ready for the PMIC to cut off the power. This is done by
triggering the ALARM2 interrupt on SoC RTC.
The RTC driver only works in power management mode which means it won't
provide any kind of time keeping functionality. It only implements a way
to trigger the ALARM2 interrupt when requested.
Differential Revision: https://reviews.freebsd.org/D1489
Reviewed by: rpaulo
MFC after: 2 weeks
The driver inherently does dma in 512 byte chunks, but it's possible that
such a buffer can span two physically discontiguous pages (such as when
a userland program does IO on the raw /dev/mmcsdN devices). Now the driver
can handle a buffer that's split across two pages.
It could in theory handle any number of segments now, but as long as IO is
being done in 512 byte blocks it will never need more than two.
requires that each 512 byte IO be in a single contiguous buffer, but if a
buffer crosses a page boundary and the physical pages aren't contiguous
you can get an EFBIG failure (too many segments).
The driver really should handle multiple segment IO, but before adding that
I wanted to make sure that it's handling failure properly while the failure
is easily recreatable.
that some #ifdef SMP code is also conditional on __ARM_ARCH >= 7; we don't
support SMP on armv6, but some drivers and modules are compiled with it
forced on via the compiler command line.
For this to work the driver needs to know the bank it has attached to since
the registers for the first 12 pins are at a different location.
Move the lock initialization to simplify the code.
isn't supposed to manage all the GPIO pins in the system from a single
instance, instead it will attach to each one of the four available GPIO
controllers and only deal with one bank at time (32 pins per bank).
Rework part of the driver to take advantage of that, this simplify the
code a lot.
Also fix a bug in rk30_gpio_get_function() which wouldn't return the
correct values.
While here fix a typo in register name.
code in sys/kern/kern_dump.c. Most dumpsys() implementations are nearly
identical and simply redefine a number of constants and helper subroutines;
a generic implementation will make it easier to implement features around
kernel core dumps. This change does not alter any minidump code and should
have no functional impact.
PR: 193873
Differential Revision: https://reviews.freebsd.org/D904
Submitted by: Conrad Meyer <conrad.meyer@isilon.com>
Reviewed by: jhibbits (earlier version)
Sponsored by: EMC / Isilon Storage Division
mostly paves the way for the new pmap code, and shouldn't result in any
noticible behavior differences.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz
The ancient gas we've been using interprets .align 0 as align to the
minimum required alignment for the current section. Clang's integrated
assembler interprets it as align to a byte boundary. Fortunately both
assemblers interpret a non-zero value as align to 2^N so just make sure
we have appropriate non-zero values everywhere.
The elftoolchain project includes these additional defines for various
userland programs. Given that arch-specific defines are still interesting
in the context of userland programs reading or writing ELF metadata, they
should be included in top-level ELF headers.
Remove duplicate defines from ARM and MIPS elf headers.
Submitted by: will (initial version)
Reviewed by: imp, will
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D844
raft of new warnings that appear to be on by default in clang 3.5.0.
Fix RPI-B build issues with new clang not liking the ability to pass
arbitrary flags to as, since some flags are more arbitrary (and thus
verboten) than others.
These warnings should be actually fixed in the code, but this is a
band-aide to get things (almost) building again.
end of the actual instruction sequence for the function but before some
misc data in the text segment. This eliminates a strange "size must be
constant" error from the integrated assembler. Also, the build_pagetables
function was missing an END(), but really the problem is that it shouldn't
have an ASENTRY() because it's not a function that needs to be a global
symbol with unwind info and all, it's just a little private subroutine
used in very early kernel init.
the cache before clean/invalidate ensured that no new lines can come into
the cache or migrate between levels during the operation, but may not be
safe on some chips. Instead, if the cache was enabled on entry, do the
wbinv while it's still enabled, and then disable it and do a separate
invalidate pass. After the intitial writeback we know there are no
dirty lines left and no new dirty lines can be created as long as we
carefully avoid touching memory before turning the cache off. Add a
comment about that so no new code gets inserted between those points.
The kernel build machinery really wants the entry point to be in a file
named locore.S so doing this avoids a bunch of changes to the build system
for relatively little benefit.
armv6/7 systems. We need to use some new armv6/7 features at startup
and splitting the implemenations to separate files will be more maintainable
than adding even more #ifdef sections to locore.S.
Because of the standardized interfaces to cache and MMU operations in armv7,
we can tolerate the kernel being entered with caches enabled. This allows
running u-boot and loader(8) with caches enabled, and the performance
improvement can be dramatic (boot times can be cut from over a minute
to under 30 seconds). The new implementation also has more robust cache
and mmu sequences for launching AP cores, and it paves the way for
upcoming changes to the pmap code which will use the TEX remap feature.
Changes in mp_machdep.c work with the new behavior in locore-v6 mp_entry,
and also reuse the original boot-time page tables to get transitioned
from physical to virtual addressing before installing the normal tables.
Submitted by Svatopluk Kraus and Michal Meloun with some changes by me.
the #ifdef stuff at multiple points the functions are called from. Also
rework the armv7 implementations so that the invalidate operations work
from outermost to innermost cache level, and the writeback works from
inner to outer levels.
initially set up the MMU. Some day they may also be useful as part of
suspend/resume handling, when we get better at power management.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz
are inline functions that handle all the routine maintenance operations
except the flush-all and invalidate-all routines which are required only
during early kernel init.
These inline functions should be very much faster than the old mechanism
that involved jumping through the big cpufuncs table, especially for
common operations such as invalidating a single TLB entry. Note that
nothing is calling these yet, this just is just required infrastructure
for upcoming changes to the pmap-v6 code.
mechanism defined for armv7 (and also present on some armv6 chips including
the arm1176 used on rpi). The information is parsed into a global cpuinfo
structure, which will be used by (upcoming) new cache and tlb maintenance
code to handle cpu-specific variations of the maintence sequences.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz
macro wasn't needed and was being used with swapped arguments which always
give the same result (0) defeating the overflow check.
On initialization, do not use bcm_mbox_intr() to read the pending messages,
with the new semaphore based implementation this will lead to semaphore
being incremented on the channels that contain pending data and will make
the first read for that channel return stale data.
This fixes the hang that happens on boot while initializing the cpufreq on
Raspberry Pi.
According to http://e2e.ti.com/support/arm/sitara_arm/f/791/t/210729 the
USB reset pulse has an undocumented duration of 200ns and during this
period the module must not be acessed.
We wait for 100us to take into account for some imprecision of the early
DELAY() loop.
This fixes the eventual 'External Non-Linefetch Abort (S)' that happens at
boot while resetting the musb subsystem.
While here, enable the USB subsystem clock before the first access.
Discussed with: ian, adrian
MFC after: 2 weeks
This enables the use of GPIO pins as interrupt sources for kernel devices
directly attached to gpiobus (userland notification will be added soon).
The use of gpio interrupts for other kernel devices will be possible when
intrng is complete.
All GPIO pins can be set to trigger on:
- active-low;
- active-high;
- rising edge;
- falling edge.
Tested on: Beaglebone-black
code, passing a 0/1 flag that indicates which type of abort it was. This
sets the stage for unifying the handling of page faults in a single routine.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz
If it seems like this is getting out of hand, I quite agree. I wonder if
it's safe, here in the 21st century, to lose the distinction between C and
ASM symbols?
around so that related things are more grouped together, rewrite comments.
No functional changes, this is all so that the functional changes in the
next commit will stand out.
'extra' entry points which are nested within or provide a synonym name
for another function. It's most likely not safe to be messing with the
IP and LR registers at anything other than the primary entry point to a
function. Anywhere beyond initial function entry, those registers may
be in use as scratch or variable registers.
semicolons between the code and comments instead of after the comments,
and line continuations in the arbitrary but now consistant column 76.
No functional changes.
- Eliminate unused irqframe
- Eliminate unused saframe
- Instead of splitting r4-sp storage between the stack and switchframe,
just put all the registers in switchframe and eliminate the un_32 struct.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>,
Michal Meloun <meloun@miracle.cz>
Fix the following issues:
- Removed revision from device softc, it isn't used anywhere else out of
device attach routine;
- Move the duplicated code for verification of valid banks (and pins) to
a single function;
- Use some macros to simplify the handling of some constants;
- Update some stale comments.
which means that the NFSCLIENT and NFSSERVER
kernel options will no longer work. This commit
only removes the kernel components. Removal of
unused code in the user utilities will be done
later. This commit does not include an addition
to UPDATING, but that will be committed in a
few minutes.
Discussed on: freebsd-fs
