Benjamin Herrenschmidt [Mon, 1 Jun 2020 22:32:17 +0000 (08:32 +1000)]
litedram: Test bench
The test bench test simple access forms for now, it's a starting point
but it already helped find/fix a bug.
Includes a litedram update to be able to operate the sim model without
inits.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Wed, 27 May 2020 01:06:51 +0000 (11:06 +1000)]
litedram: Add an L2 cache with store queue
This adds a cache between the wishbone and litedram with the following
features (at this point, it's still evolving)
- 128 bytes line width in order to have a reasonable amount of
litedram pipelining on the 128-bit wide data port.
- Configurable geometry otherwise
- Stores are acked immediately on wishbone whether hit or miss
(minus a 2 cycles delay if there's a previous load response in the
way) and sent to LiteDRAM via 8 entries (configurable) store queue
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Mon, 25 May 2020 10:20:59 +0000 (20:20 +1000)]
litedram: Add support for booting without BRAM
This adds an option to disable the main BRAM and instead copy a
payload stashed along with the init code in the secondary BRAM
into DRAM and boot from there
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 22 May 2020 08:43:50 +0000 (18:43 +1000)]
litedram: Add simulation support
This adds a simulated litedram model along with the necessary
Makefile gunk to verilate it and wrap it for use by ghdl.
The core_dram_tb test bench is a variant of core_tb with
LiteDRAM simulated. It's not built by default, an explicit
make core_dram_tb
is necessary as to not require verilator to be installed for
the normal build process (also it's slow'ish).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras [Thu, 4 Jun 2020 23:29:24 +0000 (09:29 +1000)]
Merge pull request #185 from ozbenh/misc
Small fixes to mw_debug, bin2hex and improve BRAM generation in dcache
Michael Neuling [Thu, 4 Jun 2020 22:38:18 +0000 (08:38 +1000)]
Merge pull request #188 from ozbenh/openocd-tweaks
Openocd tweaks
Benjamin Herrenschmidt [Thu, 4 Jun 2020 13:29:29 +0000 (23:29 +1000)]
flash-arty: Add support for specifying the file type
By default openocd tries to "guess" the file type and interpret
it accordingly. For example it will detect an ELF file based on
the presence of an ELF header and will try to load the relevant
segments into the flash.
This may not be what we want. For example, I want to load the raw
ELF file into the flash.
Additionally the ELF parser in most distro's OpenOCD version
only supports ELF32 and will error out.
This adds a "-t" argument to flash-arty to allow us to specify the
file format. For example "-t bin" will treat the file as raw binary.
Unfortunately I had to copy and modify jtagspi.cfg from OpenOCD
to achieve this.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 4 Jun 2020 13:13:45 +0000 (23:13 +1000)]
flash-arty: Support hex values for address
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras [Wed, 3 Jun 2020 04:20:02 +0000 (14:20 +1000)]
Merge pull request #168 from shenki/flash-arty
Scripts to write data to the Arty's SPI flash
Benjamin Herrenschmidt [Mon, 25 May 2020 06:48:47 +0000 (16:48 +1000)]
dcache: Rework RAM wrapper to synthetize better on Xilinx
The global wr_en signal is causing Vivado to generate two TDP (True Dual Port)
block RAMs instead of one SDP (Simple Dual Port) for each cache way. Remove
it and instead apply a AND to the individual byte write enables.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Mon, 1 Jun 2020 08:24:15 +0000 (18:24 +1000)]
bin2hex: Make sure to generate little endian files
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 28 May 2020 14:04:46 +0000 (00:04 +1000)]
mw_debug: Fix memory overflow with "sim" backend
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Anton Blanchard [Tue, 2 Jun 2020 01:54:00 +0000 (11:54 +1000)]
Merge pull request #178 from antonblanchard/intercon
Interconnect timing improvements from Ben
Anton Blanchard [Tue, 2 Jun 2020 00:36:06 +0000 (10:36 +1000)]
Merge pull request #184 from antonblanchard/verific
Delete old verific script
Michael Neuling [Tue, 26 May 2020 06:55:20 +0000 (16:55 +1000)]
Delete bit rotted verific script
We can use ghdl-synth and fusesoc now, so verific is a dead path which
has bit rotted.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Benjamin Herrenschmidt [Wed, 20 May 2020 11:00:27 +0000 (21:00 +1000)]
litedram: Split the init memory from the main wrapper
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Sun, 17 May 2020 05:04:23 +0000 (15:04 +1000)]
irq: Simplify xics->core irq input
Use a simple wire. common.vhdl types are better kept for things
local to the core. We can add more wires later if we need to for
HV irqs etc...
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Mon, 11 May 2020 11:22:07 +0000 (21:22 +1000)]
soc: Rework interconnect
This changes the SoC interconnect such that the main 64-bit wishbone out
of the processor is first split between only 3 slaves (BRAM, DRAM and a
general "IO" bus) instead of all the slaves in the SoC.
The IO bus leg is then latched and down-converted to 32 bits data width,
before going through a second address decoder for the various IO devices.
This significantly reduces routing and timing pressure on the main bus,
allowing to get rid of frequent timing violations when synthetizing on
small'ish FPGAs such as the Artix-7 35T found on the original Arty board.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 15 May 2020 07:43:51 +0000 (17:43 +1000)]
sw: Add full memory map to .h and use it for litedram .lds
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Anton Blanchard [Sat, 23 May 2020 06:50:12 +0000 (16:50 +1000)]
Merge pull request #181 from antonblanchard/Makefile-rework-2
Pass clock frequency to UART sim wrapper
Anton Blanchard [Sat, 23 May 2020 03:46:43 +0000 (13:46 +1000)]
Pass clock frequency to UART sim wrapper
The UART sim wrapper is currently hard wired to 50 MHz.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Thu, 21 May 2020 02:29:55 +0000 (12:29 +1000)]
Merge pull request #180 from antonblanchard/Makefile-rework
Makefile rework
Anton Blanchard [Wed, 20 May 2020 06:27:06 +0000 (16:27 +1000)]
A little less shouting in the Makefile
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Wed, 20 May 2020 06:18:58 +0000 (16:18 +1000)]
Fix the simulated DMI
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Wed, 20 May 2020 06:07:13 +0000 (16:07 +1000)]
Exit cleanly from testbench on success
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Wed, 20 May 2020 05:37:49 +0000 (15:37 +1000)]
Merge Makefile and Makefile.synth
We still need to a way to our FPGA target on the command line, but this
at least gets us down to a common Makefile.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Wed, 20 May 2020 05:01:42 +0000 (15:01 +1000)]
Add Makefile command line variables to enable docker and podman
Instead of having to edit the Makefile, we can now do:
make DOCKER=1
make PODMAN=1
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Wed, 20 May 2020 04:29:50 +0000 (14:29 +1000)]
Rework Makefile
Instead of building each file one by one (and having to track all
the dependencies manually), use the ghdl -c command that does
analysis and elaboration in one go.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Tue, 19 May 2020 05:53:54 +0000 (15:53 +1000)]
Merge pull request #179 from antonblanchard/yosys-verilator
Add yosys/verilator support
Anton Blanchard [Tue, 19 May 2020 05:32:04 +0000 (15:32 +1000)]
Improve make clean
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Tue, 19 May 2020 05:18:42 +0000 (15:18 +1000)]
Add yosys/verilator support
Add a microwatt-verilator target that simulates the
ghdl -> yosys -> verilog -> verilator path. A good test of
ghdl/yosys synthesis.
Because the everything is run through synthesis, the instruction
image is baked into the build via the RAM_INIT_FILE generic.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Tue, 19 May 2020 04:27:42 +0000 (14:27 +1000)]
Merge pull request #171 from shenki/mw-debug-features
mw debug features
Anton Blanchard [Tue, 19 May 2020 04:04:19 +0000 (14:04 +1000)]
Merge pull request #173 from Jbalkind/core-vcs-syntax
Changing use of others in core files to satisfy VCS
Joel Stanley [Mon, 11 May 2020 04:22:41 +0000 (13:52 +0930)]
mw_debug: Add README
This describes how to build the tool on Fedora, and on Debian which lacks a packaged
liburjtag as of mid 2020.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Joel Stanley [Sun, 10 May 2020 04:00:26 +0000 (13:30 +0930)]
mw_debug: Add usage text
Signed-off-by: Joel Stanley <joel@jms.id.au>
Joel Stanley [Mon, 11 May 2020 03:52:00 +0000 (13:22 +0930)]
mw_debug: Add CFLAGS and fix warnings
CFLAGS was defined but not used anywhere. This adds them to the compile
line, and fixes the warnings (and errors!) that result.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Anton Blanchard [Tue, 19 May 2020 03:32:33 +0000 (13:32 +1000)]
Merge pull request #177 from antonblanchard/litedram
LiteDRAM fixes from Ben
Anton Blanchard [Tue, 19 May 2020 02:28:02 +0000 (12:28 +1000)]
Merge branch 'master' into litedram
Anton Blanchard [Tue, 19 May 2020 01:53:34 +0000 (11:53 +1000)]
Merge pull request #176 from antonblanchard/console-improv
Console improvements from Ben
Anton Blanchard [Tue, 19 May 2020 01:38:16 +0000 (11:38 +1000)]
Merge pull request #175 from antonblanchard/yosys-fixes-2
Fix yosys build after MMU merge
Jonathan Balkind [Thu, 14 May 2020 21:17:15 +0000 (17:17 -0400)]
Changes for compilation with VCS:
- Changing use of others in core files to satisfy VCS
- Adding workaround for VCS subtype constraint inconsistencies in common.vhdl
Signed-off-by: Jonathan Balkind <jbalkind@princeton.edu>
Anton Blanchard [Tue, 19 May 2020 01:22:29 +0000 (11:22 +1000)]
Fix yosys build after MMU merge
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Tue, 19 May 2020 00:44:04 +0000 (10:44 +1000)]
Merge pull request #174 from antonblanchard/yosys-fixes
Some yosys fixes
Anton Blanchard [Mon, 18 May 2020 00:15:03 +0000 (10:15 +1000)]
Some yosys fixes
This gets the yosys build further along, but I'm now chasing what looks
like a yosys bug.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Anton Blanchard [Mon, 18 May 2020 23:34:41 +0000 (09:34 +1000)]
Merge pull request #169 from paulusmack/mmu
Add radix MMU with dTLB and iTLB
Benjamin Herrenschmidt [Fri, 15 May 2020 03:31:11 +0000 (13:31 +1000)]
litedram: Regenerate
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 15 May 2020 03:30:01 +0000 (13:30 +1000)]
soc/core: Add reset latches
This adds one-cycle latches to the various resets out of the soc and
into the various core modules. It *seems* to help vivado P&R a bit
and has shown to avoid timing violations under some circumstances.
Interestingly those resets never seem to appear in the bad timing
path. It looks like those long resets simply impose placement
constraints that Vivado satisfies at the expense of timing elsewhere.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 15 May 2020 03:20:27 +0000 (13:20 +1000)]
arty/nexys: Rework reset with litedram
When using litedram, request a much longer PLL reset. This seems to
help get rid of all the grabled output after config.
Also use the clean system_rst out of litedram as our source of reset
for the rest of the SoC (it is synchronized with system_clk and takes
pll_locked into account already)
Benjamin Herrenschmidt [Fri, 15 May 2020 03:15:48 +0000 (13:15 +1000)]
soc_reset: Use counters, add synchronizers
In some cases we need to keep the reset held for much longer,
so use counters rather than shift registers.
Additionally, some signals such as ext_rst and pll_locked
or signals going from the ext_clk domain to the pll_clk
domain need to be treated as async, and testing them without
synchronizers is asking for trouble.
Finally, make the external reset also reset the PLL.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 15 May 2020 03:11:58 +0000 (13:11 +1000)]
litedram: Forward system reset signal
The wrapper wouldn't forward it. Make it do so
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 15 May 2020 00:14:53 +0000 (10:14 +1000)]
litedram: Remove init delays
The clocks / resets are now stable
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Tue, 12 May 2020 10:27:15 +0000 (20:27 +1000)]
litedram: Update to new LiteX/LiteDRAM version
Things have changed a bit in upstream LiteX. LiteDRAM now exposes a
wishbone for the CSRs for example.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras [Fri, 15 May 2020 12:01:02 +0000 (22:01 +1000)]
dcache: Fix bug in store hit after dcbz case
This fixes a bug where a store that hits in the dcache immediately
following a dcbz has its write to the cache RAM suppressed (but not
its write to memory). If a load to the same location comes along
before the cache line gets replaced, the load will return incorrect
data.
Fixes: 4db1676ef8b3 ("dcache: Don't assert on dcbz cache hit")
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 04:26:14 +0000 (14:26 +1000)]
pp_soc_uart: Fix rx synchronizers and ensure stable tx init state
The rx synchronizers were ... non existent. Someone forgot to add
a if rising_edge(clk) to the process.
For tx, ensure that we have a default value so that TX stays high
from TPGA configuration to the reset being sampled on the first clock
cycle.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 02:30:11 +0000 (12:30 +1000)]
pp_fifo: Fix full fifo losing all data on simultaneous push & pop
The pp_fifo decides whether top = bottom means empty or full based
on whether the previous operation was a push or a pop.
If the fifo performs both in one cycle, it sets the previous op to
pop. That means that a full fifo being added a character and removed
one at the same time becomes empty.
Instead, just leave the previous op alone. If the fifo was empty, it
remains so, if it was full ditto.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 00:12:44 +0000 (10:12 +1000)]
Makefile: Improve clean a bit
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 02:03:39 +0000 (12:03 +1000)]
console: Remove putstr()
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 00:08:06 +0000 (10:08 +1000)]
console: Move console files
console.c goes to a new lib/ where we'll store other general utilities
and console.h goes to include/
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 02:03:01 +0000 (12:03 +1000)]
console: Replace putstr with puts
It makes things a bit more standard and a bit nicer to read
without all those strlen(). Also console.c takes care of adding
the carriage returns before the linefeeds.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 14 May 2020 00:09:36 +0000 (10:09 +1000)]
console: Improve putchar(), add puts()
Make putchar() match a standard prototype and add puts()
Also make puts() add carriage returns before linefeeds so the
users don't have to do it all over the place.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras [Thu, 14 May 2020 05:43:33 +0000 (15:43 +1000)]
soc: Work around compile error with ghdl 0.37-dev
The ghdl packaged in Fedora 31 doesn't like a port map of the form
"rst => rst or core_reset", so this works around the problem by
doing the OR in a separate statement.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 14 May 2020 05:41:51 +0000 (15:41 +1000)]
Merge branch 'mmu'
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Anton Blanchard [Thu, 14 May 2020 05:08:33 +0000 (15:08 +1000)]
Merge pull request #170 from antonblanchard/litedram
LiteDRAM integration
Joel Stanley [Wed, 6 May 2020 02:20:17 +0000 (11:50 +0930)]
Add script for writing to flash on arty
You must have openocd 0.10.0 installed.
$ ./openocd/flash-arty ~/microwatt-fusesoc/build/microwatt_0/arty_a7-35-vivado/microwatt_0.bit
Open On-Chip Debugger 0.10.0
Licensed under GNU GPL v2
For bug reports, read
http://openocd.org/doc/doxygen/bugs.html
none separate
Info : auto-selecting first available session transport "jtag". To override use 'transport select <transport>'.
adapter speed: 25000 kHz
fpga_program
Info : ftdi: if you experience problems at higher adapter clocks, try the command "ftdi_tdo_sample_edge falling"
Info : clock speed 25000 kHz
Info : JTAG tap: xc7.tap tap/device found: 0x0362d093 (mfg: 0x049 (Xilinx), part: 0x362d, ver: 0x0)
loaded file microwatt/openocd/bscan_spi_xc7a35t.bit to pld device 0 in 0s 136459us
Info : JTAG tap: xc7.tap tap/device found: 0x0362d093 (mfg: 0x049 (Xilinx), part: 0x362d, ver: 0x0)
Info : Found flash device 'micron n25q128' (ID 0x0018ba20)
flash 'jtagspi' found at 0x00000000
auto erase enabled
Info : Found flash device 'micron n25q128' (ID 0x0018ba20)
Info : Found flash device 'micron n25q128' (ID 0x0018ba20)
Info : Found flash device 'micron n25q128' (ID 0x0018ba20)
Info : sector 0 took 241 ms
Info : sector 1 took 242 ms
Info : sector 2 took 241 ms
Info : sector 3 took 247 ms
Info : sector 4 took 253 ms
Info : sector 5 took 244 ms
Info : sector 6 took 246 ms
Info : sector 7 took 237 ms
Info : sector 8 took 258 ms
Info : sector 9 took 260 ms
Info : sector 10 took 262 ms
Info : sector 11 took 253 ms
Info : sector 12 took 256 ms
Info : sector 13 took 255 ms
wrote 917504 bytes from file microwatt-fusesoc/build/microwatt_0/arty_a7-35-vivado/microwatt_0.bit in 9.642746s (92.920 KiB/s)
Info : Found flash device 'micron n25q128' (ID 0x0018ba20)
read 907483 bytes from file microwatt-fusesoc/build/microwatt_0/arty_a7-35-vivado/microwatt_0.bit and flash bank 0 at offset 0x00000000 in 0.557387s (1589.944 KiB/s)
contents match
Signed-off-by: Joel Stanley <joel@jms.id.au>
Benjamin Herrenschmidt [Sat, 9 May 2020 01:20:59 +0000 (11:20 +1000)]
litedram: Use 32-bit CSR bus
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 8 May 2020 15:09:26 +0000 (01:09 +1000)]
litedram: Add support for Microwatt-initialized controller
This adds support for initializing the memory controller from microwatt
rather than using a built-in RiscV processor. This might require some
fixes to LiteX and LiteDRAM (they haven't been merged as of this commit
yet).
This is enabled in the shipped generated files and can be changed via
modifying the generator script to pass False to "mw_init"
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 8 May 2020 00:42:01 +0000 (10:42 +1000)]
hello_world: Use new headers and frequency from syscon
This uses the new header files for register definitions and
extracts the core frequency from syscon rather than hard coding it.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Tue, 24 Sep 2019 06:19:29 +0000 (16:19 +1000)]
litedram: Improve sdram init boot messages
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 8 May 2020 00:27:33 +0000 (10:27 +1000)]
Add microwatt_soc.h and io.h include file
This contains C definitions for various Microwatt internal MMIOs
and a set of accessors.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Tue, 24 Sep 2019 12:24:31 +0000 (22:24 +1000)]
syscon: Add syscon registers
These provides some info about the SoC (though it's still somewhat
incomplete and needs more work, see comments).
There's also a control register for selecting DRAM vs. BRAM at 0
(and for soft-resetting the SoC but that isn't wired up yet).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Wed, 11 Sep 2019 09:59:13 +0000 (10:59 +0100)]
fpga: Hookup nexys-video to litedram
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Tue, 10 Sep 2019 17:24:06 +0000 (18:24 +0100)]
fpga: Hookup Arty to litedram
The old toplevel.vhdl becomes top-generic.vhdl, which is to be used
by platforms that do not have a litedram option.
Arty has its own top-arty.vhdl which supports litedram and is now
hooked up
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras [Fri, 24 Apr 2020 00:58:56 +0000 (10:58 +1000)]
MMU: Implement reading of the process table
This adds the PID register and repurposes SPR 720 as the PRTBL
register, which points to the base of the process table. There
doesn't seem to be any point to implementing the partition table given
that we don't have hypervisor mode.
The MMU caches entry 0 of the process table internally (in pgtbl3)
plus the entry indexed by the value in the PID register (pgtbl0).
Both caches are invalidated by a tlbie[l] with RIC=2 or by a move to
PRTBL. The pgtbl0 cache is invalidated by a move to PID. The dTLB
and iTLB are cleared by a move to either PRTBL or PID.
Which of the two page table root pointers is used (pgtbl0 or pgtbl3)
depends on the MSB of the address being translated. Since the segment
checking ensures that address(63) = address(62), this is sufficient to
map quadrants 0 and 3.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 7 May 2020 10:12:46 +0000 (20:12 +1000)]
tests/mmu: Add a test of PTE refetching on permission error
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sun, 3 May 2020 22:57:05 +0000 (08:57 +1000)]
tests/mmu: Add a test for dcbz with translation on
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 7 May 2020 10:02:21 +0000 (20:02 +1000)]
Implement slbia as a dTLB/iTLB flush
Slbia (with IH=7) is used in the Linux kernel to flush the ERATs
(our iTLB/dTLB), so make it do that.
This moves the logic to work out whether to flush a single entry
or the whole TLB from dcache and icache into mmu. We now invalidate
all dTLB and iTLB entries when the AP (actual pagesize) field of
RB is non-zero on a tlbie[l], as well as when IS is non-zero.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 04:23:14 +0000 (14:23 +1000)]
Decode tlbiel as tlbie
The Linux kernel contains tlbiel instructions, which we can treat
identically to tlbie.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 7 May 2020 02:08:43 +0000 (12:08 +1000)]
tests/privileged: Update for instruction translation
Since setting MSR[PR] = 1 forces instruction translation on, we need
to set up translations for the problem state code to use.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Tue, 28 Apr 2020 06:00:00 +0000 (16:00 +1000)]
tests: mmu: Add tests for instruction translation
This adds tests of instruction translation to the mmu test.
This also clears the BSS and improves the linker script.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Tue, 28 Apr 2020 04:54:22 +0000 (14:54 +1000)]
MMU: Do radix page table walks on iTLB misses
This hooks up the connections so that an OP_FETCH_FAILED coming down
to loadstore1 will get sent to the MMU for it to do a radix tree walk
for the instruction address. The MMU then sends the resulting PTE to
the icache module to be installed in the iTLB. If no valid PTE can
be found, the MMU sends an error signal back to loadstore1 which sends
it on to execute1 to generate an ISI.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Mon, 27 Apr 2020 07:43:19 +0000 (17:43 +1000)]
Add TLB to icache
This adds a direct-mapped TLB to the icache, with 64 entries by default.
Execute1 now sends a "virt_mode" signal from MSR[IR] to fetch1 along
with redirects to indicate whether instruction addresses should be
translated through the TLB, and fetch1 sends that on to icache.
Similarly a "priv_mode" signal is sent to indicate the privilege
mode for instruction fetches. This means that changes to MSR[IR]
or MSR[PR] don't take effect until the next redirect, meaning an
isync, rfid, branch, etc.
The icache uses a hash of the effective address (i.e. next instruction
address) to index the TLB. The hash is an XOR of three fields of the
address; with a 64-entry TLB, the fields are bits 12--17, 18--23 and
24--29 of the address. TLB invalidations simply invalidate the
indexed TLB entry without checking the contents.
If the icache detects a TLB miss with virt_mode=1, it will send a
fetch_failed indication through fetch2 to decode1, which will turn it
into a special OP_FETCH_FAILED opcode with unit=LDST. That will get
sent down to loadstore1 which will currently just raise a Instruction
Storage Interrupt (0x400) exception.
One bit in the PTE obtained from the TLB is used to check whether an
instruction access is allowed -- the privilege bit (bit 3). If bit 3
is 1 and priv_mode=0, then a fetch_failed indication is sent down to
fetch2 and to decode1, which generates an OP_FETCH_FAILED. Any PTEs
with PTE bit 0 (EAA[3]) clear or bit 8 (R) clear should not be put
into the iTLB since such PTEs would not allow execution by any
context.
Tlbie operations get sent from mmu to icache over a new connection.
Unfortunately the privileged instruction tests are broken for now.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 23 Apr 2020 05:33:36 +0000 (15:33 +1000)]
tests: Add a test for the MMU radix page table walks
This adds tests to check that the MMU and dTLB are translating
addresses and checking permissions correctly.
We use a simple 2-level radix tree. The radix tree maps 2GB of
address space and has a 1024-entry page directory pointing to
512-entry page table pages.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 7 May 2020 00:17:08 +0000 (10:17 +1000)]
MMU: Remove software-loaded dTLB mode
This removes the hack where the tlbie instruction could be used to
load entries directly into the dTLB, because we don't report the
correct DSISR values for accesses that hit software-loaded dTLB
entries and have privilege or permission errors.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Wed, 6 May 2020 10:21:01 +0000 (20:21 +1000)]
MMU: Refetch PTE on access fault
This is required by the architecture. It means that the error bits
reported in DSISR or SRR1 now come from the permission/RC check done
on the refetched PTE rather than the TLB entry. Unfortunately that
somewhat breaks the software-loaded TLB mode of operation in that
DSISR/SRR1 always report no PTE rather than permission error or
RC failure.
This also restructures the loadstore1 state machine a bit, combining
the FIRST_ACK_WAIT and LAST_ACK_WAIT states into a single state and
the MMU_LOOKUP_1ST and MMU_LOOKUP_LAST states likewise. We now have a
'dwords_done' bit to say whether the first transfer of two (for an
unaligned access) has been done.
The cache paradox error (where a non-cacheable access finds a hit in
the cache) is now the only cause of DSI from the dcache. This should
probably be a machine check rather than DSI in fact.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 23 Apr 2020 11:54:08 +0000 (21:54 +1000)]
MMU: Implement data segment interrupts
A data segment interrupt (DSegI) occurs when an address to be
translated by the MMU is outside the range of the radix tree
or the top two bits of the address (the quadrant) are 01 or 10.
This is detected in a new state of the MMU state machine, and
is sent back to loadstore1 as an error, which sends it on to
execute1 to generate an interrupt to the 0x380 vector.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 23 Apr 2020 05:28:22 +0000 (15:28 +1000)]
MMU: Implement radix page table machinery
This adds the necessary machinery to the MMU for it to do radix page
table walks. The core elements are a shifter that can shift the
address right by between 0 and 47 bits, a mask generator that can
generate a mask of between 5 and 16 bits, a final mask generator,
and new states in the state machine.
(The final mask generator is used for transferring bits of the
original address into the resulting TLB entry when the leaf PTE
corresponds to a page size larger than 4kB.)
The hardware does not implement a partition table or a process table.
Software is expected to load the appropriate process table entry
into a new SPR called PGTBL0, SPR 720. The contents should be
formatted as described in Book III section 5.7.6.2 of the Power ISA
v3.0B. PGTBL0 is set to 0 on hard reset. At present, the top two bits
of the address (the quadrant) are ignored.
There is currently no caching of any step in the translation process
or of the final result, other than the entry created in the dTLB.
That entry is a 4k page entry even if the leaf PTE found in the walk
corresponds to a larger page size.
This implementation can handle almost any page table layout and any
page size. The RTS field (in PGTBL0) can have any value between 0
and 31, corresponding to a total address space size between 2^31
and 2^62 bytes. The RPDS field of PGTBL0 can be any value between
5 and 16, except that a value of 0 is taken to disable radix page
table walking (for use when one is using software loading of TLB
entries). The NLS field of the page directory entries can have any
value between 5 and 16. The minimum page size is 4kB, meaning that
the sum of RPDS and the NLS values of the PDEs found on the path to
a leaf PTE must be less than or equal to RTS + 31 - 12.
The PGTBL0 SPR is in the mmu module; thus this adds a path for
loadstore1 to read and write SPRs in mmu. This adds code in dcache
to service doubleword read requests from the MMU, as well as requests
to write dTLB entries.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Wed, 22 Apr 2020 01:10:56 +0000 (11:10 +1000)]
Add framework for implementing an MMU
This adds a new module to implement an MMU. At the moment it doesn't
do very much. Tlbie instructions now get sent by loadstore1 to mmu,
which sends them to dcache, rather than loadstore1 sending them
directly to dcache. TLB misses from dcache now get sent by loadstore1
to mmu, which currently just returns an error. Loadstore1 then
generates a DSI in response to the error return from mmu.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Mon, 20 Apr 2020 02:43:06 +0000 (12:43 +1000)]
Implement access permission checks
This adds logic to the dcache to check the permissions encoded in
the PTE that it gets from the dTLB. The bits that are checked are:
R must be 1
C must be 1 for a store
EAA(0) - if this is 1, MSR[PR] must be 0
EAA(2) must be 1 for a store
EAA(1) | EAA(2) must be 1 for a load
In addition, ATT(0) is used to indicate a cache-inhibited access.
This now implements DSISR bits 36, 38 and 45.
(Bit numbers above correspond to the ISA, i.e. using big-endian
numbering.)
MSR[PR] is now conveyed to loadstore1 for use in permission checking.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Tue, 7 Apr 2020 06:17:37 +0000 (16:17 +1000)]
Implement data storage interrupts
This adds a path from loadstore1 back to execute1 for reporting
errors, and machinery in execute1 for generating data storage
interrupts at vector 0x300.
If dcache is given two requests in successive cycles and the
first encounters an error (e.g. a TLB miss), it will now cancel
the second request.
Loadstore1 now responds to errors reported by dcache by sending
an exception signal to execute1 and returning to the idle state.
Execute1 then writes SRR0 and SRR1 and jumps to the 0x300 Data
Storage Interrupt vector. DAR and DSISR are held in loadstore1.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Mon, 6 Apr 2020 07:54:45 +0000 (17:54 +1000)]
dcache: Implement data TLB
This adds a TLB to dcache, providing the ability to translate
addresses for loads and stores. No protection mechanism has been
implemented yet. The MSR_DR bit controls whether addresses are
translated through the TLB.
The TLB is a fixed-pagesize, set-associative cache. Currently
the page size is 4kB and the TLB is 2-way set associative with 64
entries per set.
This implements the tlbie instruction. RB bits 10 and 11 control
whether the whole TLB is invalidated (if either bit is 1) or just
a single entry corresponding to the effective page number in bits
12-63 of RB.
As an extension until we get a hardware page table walk, a tlbie
instruction with RB bits 9-11 set to 001 will load an entry into
the TLB. The TLB entry value is in RS in the format of a radix PTE.
Currently there is no proper handling of TLB misses. The load or
store will not be performed but no interrupt is generated.
In order to make timing at 100MHz on the Arty A7-100, we compare
the real address from each way of the TLB with the tag from each way
of the cache in parallel (requiring # TLB ways * # cache ways
comparators). Then the result is selected based on which way hit in
the TLB. That avoids a timing path going through the TLB EA
comparators, the multiplexer that selects the RA, and the cache tag
comparators.
The hack where addresses of the form 0xc------- are marked as
cache-inhibited is kept for now but restricted to real-mode accesses.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Wed, 22 Apr 2020 06:53:39 +0000 (16:53 +1000)]
Pass mtspr/mfspr to MMU-related SPRs down to loadstore1
This arranges for some mfspr and mtspr to get sent to loadstore1
instead of being handled in execute1. In particular, DAR and DSISR
are handled this way. They are therefore "slow" SPRs.
While we're at it, fix the spelling of HEIR and remove mention of
DAR and DSISR from the comments in execute1.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 03:31:07 +0000 (13:31 +1000)]
mw_debug: Add support for reading GSPRs and writing memory
This adds a "gpr" command for reading 1 or more GPRs/fast SPRs,
and a "mw" command for writing an 8-byte value to memory. It also
adds an "icreset" command for resetting the instruction cache
and fixes the "creset" command to actually reset the core instead
of starting it. The MSR is now printed along with the NIA in the
status information.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 03:26:30 +0000 (13:26 +1000)]
debug: Provide a way to examine GPRs, fast SPRs and MSR
This provides commands on the debug interface to read the value of
the MSR or any of the 64 GSPR register file entries. The GSPR values
are read using the B port of the register file in a cycle when
decode2 is not using it.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 10:08:10 +0000 (20:08 +1000)]
Improve architectural compliance of mfspr and mtspr
Mfspr from an unimplemented SPR should be a no-op in privileged state,
so in this case we need to write back whatever was previously in the
destination register. For problem state, both mtspr and mfspr to
unimplemented SPRs should cause a program interrupt.
There are special cases in the architecture for SPRs 0, 4 5 and 6
which we still don't implement.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 03:45:39 +0000 (13:45 +1000)]
decode1: Implement eieio as a nop
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Sat, 2 May 2020 03:28:19 +0000 (13:28 +1000)]
Implement the extswsli instruction
This mainly required the addition of an entry to the opcode 31 decode
table and a 32-bit sign-extender in the rotator.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Benjamin Herrenschmidt [Tue, 10 Sep 2019 17:05:56 +0000 (18:05 +0100)]
soc: Add DRAM address decoding
Still not attached to any board
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Thu, 7 May 2020 12:09:59 +0000 (22:09 +1000)]
core: Add alternate reset address
An external signal can control whether the core will start
executing at the standard or the alternate reset address.
This will be used when litedram is initialized by microwatt
itself, to route the reset to the built-in init code secondary
block RAM.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Tue, 10 Sep 2019 17:05:32 +0000 (18:05 +0100)]
litedram: Add basic support for LiteX LiteDRAM
This comes in two parts:
- A generator script which uses LiteX to generate litedram cores
along with their init files for various boards (currently Arty and
Nexys-video). This comes with configs for arty and nexys_video.
- A fusesoc "generator" which uses pre-generated litedram cores
The generation process is manual on purpose. This include pre-generated
cores for the two above boards.
This is done so that one doesn't have to install LiteX to build
microwatt. In addition, the generator script or wrapper vhdl tend to
break when LiteX changes significantly which happens.
This is still rather standalone and hasn't been plumbed into the SoC
or the FPGA toplevel files yet.
At this point LiteDRAM self-initializes using a built-in VexRiscv
"Minimum" core obtained from LiteX and included in this commit. There
is some plumbing to generate and cores that are initialized by Microwatt
directly but this isn't working yet and so isn't enabled yet.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Benjamin Herrenschmidt [Fri, 8 May 2020 01:36:37 +0000 (11:36 +1000)]
core: Improve core reset
The icache would still spit out an instruction which could
cause a 0x700 instead of a reset.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>