github-mirrors/ReonV
1
0
Fork 0
mirror of https://github.com/lcbcFoo/ReonV.git synced 2025-04-24 05:27:07 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions
ReonV/designs/leon3-asic/config.help

1258 lines
47 KiB
Text
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

Prompt for target technology
CONFIG_SYN_INFERRED
Selects the target technology for memory and pads.
The following are available:
- Inferred: Generic FPGA or ASIC targets if your synthesis tool
is capable of inferring RAMs and pads automatically.
- Actel ProAsic/P/3, IGLOO/2, RTG4 and Axcelerator FPGAs
- Aeroflex UT25CRH Rad-Hard 0.25 um CMOS
- Altera: Most Altera FPGA families
- Altera-CycloneIII: Altera Cyclone-III/IV FPGA family
- Altera-Stratix: Altera Stratix FPGA family
- Altera-StratixII: Altera Stratix/Cyclone-II FPGA families
- Altera-StratixIII: Altera Stratix-III FPGA family
- Altera-StratixIV: Altera Stratix-IV FPGA family
- Altera-StratixV: Altera Stratix-V FPGA family
- ATC18: Atmel-Nantes 0.18 um rad-hard CMOS
- Lattice : EC/ECP/XP FPGAs
- Quicklogic : Eclipse/E/II FPGAs
- UMC-0.18 : UMC 0.18 um CMOS with Virtual Silicon libraries
- Xilinx-Spartan/2/3/6: Xilinx Spartan/2/3/6 libraries
- Xilinx-Spartan3E: Xilinx Spartan3E libraries
- Xilinx-Virtex/E: Xilinx Virtex/E libraries
- Xilinx-Virtex2/4/5/6/7: Xilinx Virtex2/4/5/6/7 libraries
Note: Level of technology support depends on type of GRLIB
distribution. A technology may be present in this list while the
tech support files are missing from the GRLIB distribution.
Actel support is only available in commercial and FT distributions.
Additional target technologies are available that are not selectable
via the xconfig tool.
Ram library
CONFIG_MEM_VIRAGE
Select RAM generators for ASIC targets.
Transceiver type
CONFIG_TRANS_GTP0
Select the transceiver type used in your FPGA
Infer ram
CONFIG_SYN_INFER_RAM
Say Y here if you want the synthesis tool to infer your
RAM automatically. Say N to directly instantiate technology-
specific RAM cells for the selected target technology package.
Infer pads
CONFIG_SYN_INFER_PADS
Say Y here if you want the synthesis tool to infer pads.
Say N to directly instantiate technology-specific pads from
the selected target technology package.
No async reset
CONFIG_SYN_NO_ASYNC
Say Y here if you disable asynchronous reset in some of the IP cores.
Might be necessary if the target library does not have cells with
asynchronous set/reset.
Scan support
CONFIG_SYN_SCAN
Say Y here to enable scan support in some cores. This will enable
the scan support generics where available and add logic to make
the design testable using full-scan.
Enable JTAG Boundary Scan
CONFIG_BOUNDSCAN_EN
Say Y to create boundary scan registers along the I/O pad ring and
some control logic to allow I/O control and sampling through the JTAG
TAP.
The supported boundary-scan commands are SAMPLE/PRELOAD, EXTEST,
INTEST, HIGHZ and CLAMP. HIGHZ may not be implemented on all designs
since it requires tri-state ability on all pads.
If the JTAG debug link is enabled, the boundary scan logic connects
to the same TAP controller, otherwise a dedicated TAP controller is
added to the design.
Use Virtex CLKDLL for clock synchronisation
CONFIG_CLK_INFERRED
Certain target technologies include clock generators to scale or
phase-adjust the system and SDRAM clocks. This is currently supported
for Xilinx, Altera and Proasic3 FPGAs. Depending on technology, you
can select to use the Xilinx CKLDLL macro (Virtex, VirtexE, Spartan1/2),
the Xilinx DCM (Virtex-2, Spartan3, Virtex-4), the Xilinx PLLE2 (Virtex-7,
Kintex-7, Artix-4), the Altera ALTDLL (Stratix, Cyclone), or the
Proasic3 PLL. Choose the 'inferred' option to skip a clock generator.
Clock multiplier
CONFIG_CLK_MUL
When using the Xilinx DCM, Xilinx PLLE2 or Altera ALTPLL,
the system clock can be multiplied with a factor of 2 - 32,
and divided by a factor of 1 - 32. This makes it possible to
generate almost any desired processor frequency. When using
the Xilinx CLKDLL generator, the resulting frequency scale f
actor (mul/div) must be one of 1/2, 1 or 2. On Proasic3,
the factor can be 1 - 128.
WARNING: The resulting clock must be within the limits specified
by the target FPGA family.
Clock divider
CONFIG_CLK_DIV
When using the Xilinx DCM, Xilinx PLLE2 or Altera ALTPLL,
the system clock can be multiplied with a factor of 2 - 32,
and divided by a factor of 1 - 32. This makes it possible to
generate almost any desired processor frequency. When using
the Xilinx CLKDLL generator, the resulting frequency scale f
actor (mul/div) must be one of 1/2, 1 or 2. On Proasic3,
the factor can be 1 - 128.
WARNING: The resulting clock must be within the limits specified
by the target FPGA family.
Output clock divider
CONFIG_OCLK_DIV
When using the Proasic3 PLL, the system clock is generated by three
parameters: input clock multiplication, input clock division and
output clock division. Only certain values of these parameters
are allowed, but unfortunately this is not documented by Actel.
To find the correct values, run the Libero Smartgen tool and
insert you desired input and output clock frequencies in the
Static PLL configurator. The mul/div factors can then be read
out from tool.
Output clock divider, 2nd clock
CONFIG_OCLKB_DIV
See help for 'Ouput division factor'. Set this to 0 to disable the
second clock output.
Output clock divider, 3rd clock
CONFIG_OCLKC_DIV
See help for 'Ouput division factor'. Set this to 0 to disable the
third clock output.
System clock multiplier
CONFIG_CLKDLL_1_2
The Xilinx CLKDLL can scale the input clock with a factor of 0.5, 1.0,
or 2.0. Useful when the target board has an oscillator with a too high
(or low) frequency for your design. The divided clock will be used as the
main clock for the whole processor (except PCI and ethernet clocks).
System clock multiplier
CONFIG_DCM_2_3
The Xilinx DCM and Altera ALTDLL can scale the input clock with a large
range of factors. Useful when the target board has an oscillator with a
too high (or low) frequency for your design. The divided clock will
be used as the main clock for the whole processor (except PCI and
ethernet clocks). NOTE: the resulting frequency must be at least
24 MHz or the DCM and ALTDLL might not work.
Enable CLKDLL for PCI clock
CONFIG_PCI_CLKDLL
Say Y here to re-synchronize the PCI clock using a
Virtex BUFGDLL macro. Will improve PCI clock-to-output
delays on the expense of input-setup requirements.
Use PCI clock system clock
CONFIG_PCI_SYSCLK
Say Y here to the PCI clock to generate the system clock.
The PCI clock can be scaled using the DCM or CLKDLL to
generate a suitable processor clock.
External SDRAM clock feedback
CONFIG_CLK_NOFB
Say Y here to disable the external clock feedback to synchronize the
SDRAM clock. This option is necessary if your board or design does not
have an external clock feedback that is connected to the pllref input
of the clock generator.
Number of processors
CONFIG_PROC_NUM
The number of processor cores. The LEON3MP design can accomodate
up to 4 LEON3 processor cores. Use 1 unless you know what you are
doing ...
Force values from example configuration
CONFIG_LEON3_MIN
If you select any other value than Custom-configuration then the
configuration choices will be forced to settings corresponding to
the selected example configuration. The example configurations are
described in the document LEON/GRLIB Design and Configuration Guide
located at doc/guide.pdf.
Due to limitations in the configuration tool it is not possible to
update all choices without first saving the configuration and
restarting the tool. Perform the steps below to set the processor
configuration to example configuration values:
1. Select example configuration (MIN, GP, HP). This will result
in most processor configuration menus being disabled.
2. Go back to the main menu and select Save and Exit
3. Start the xconfig tool again
4. Go into processor configuration and select Custom-configuration.
All configuration values can now be changed again and have been
initialized to example configuration that was selected earlier.
5. Make any changes and then quit xconfig via Save and Exit
Some options set from the example configuration may need to be changed.
One such setting is the FPU multiplier, if the FPU is enabled, where
timing is improved for several FPGA technologies by selecting the
technology specific multiplier and where the DW multiplier is generally
a good choice for ASIC.
Number of SPARC register windows
CONFIG_IU_NWINDOWS
The SPARC architecture (and LEON) allows 2 - 32 register windows.
However, any number except 8 will require that you modify and
recompile your run-time system or kernel. Unless you know what
you are doing, use 8.
SPARC V8 multiply and divide instruction
CONFIG_IU_V8MULDIV
If you say Y here, the SPARC V8 multiply and divide instructions
will be implemented. The instructions are: UMUL, UMULCC, SMUL,
SMULCC, UDIV, UDIVCC, SDIV, SDIVCC. In code containing frequent
integer multiplications and divisions, significant performance
increase can be achieved. Emulated floating-point operations will
also benefit from this option.
By default, the gcc compiler does not emit multiply or divide
instructions and your code must be compiled with -mv8 to see any
performance increase. On the other hand, code compiled with -mv8
will generate an illegal instruction trap when executed on processors
with this option disabled.
The divider consumes approximately 2 kgates, the multiplier 6 kgates.
Multiplier latency
CONFIG_IU_MUL_LATENCY_2
Implementation options for the integer multiplier.
Type Implementation issue-rate/latency
2-clocks 32x32 pipelined multiplier 1/2
4-clocks 16x16 standard multiplier 4/4
5-clocks 16x16 pipelined multiplier 4/5
MAC operation
CONFIG_IU_MUL_MAC
If you say Y here, the SPARC V8e UMAC/SMAC (multiply-accumulate)
instructions will be enabled. The instructions implement a
single-cycle 16x16->32 bits multiply with a 40-bits accumulator.
The details of these instructions can be found in the LEON manual,
This option is only available when 16x16 multiplier is used.
Multiplier structure
CONFIG_IU_MUL_INFERRED
Structure options for the integer multiplier. The multiplier
can be implemented with the following structures:
* Inferred by the synthesis tool
* Generated using Module Generators from NTNU
* Using technology specific netlists (TechSpec)
* Using Synopsys Designware (DW02_mult and DW_mult_pipe)
Branch prediction
CONFIG_IU_BP
Enabling branch prediction will improve performance with
up to 20%, depending on application. The timing and area
overhead are minor, so it is recommended to always enable
this option.
Single vector trapping
CONFIG_IU_SVT
Single-vector trapping is a SPARC V8e option to reduce code-size
in small applications. If enabled, the processor will jump to
the address of trap 0 (tt = 0x00) for all traps. No trap table
is then needed. The trap type is present in %psr.tt and must
be decoded by the O/S. Saves 4 Kbyte of code, but increases
trap and interrupt overhead. Currently, the only O/S supporting
this option is eCos. To enable SVT, the O/S must also set bit 13
in %asr17.
Load latency
CONFIG_IU_LDELAY
Defines the pipeline load delay (= pipeline cycles before the data
from a load instruction is available for the next instruction).
One cycle gives best performance, but might create a critical path
on targets with slow (data) cache memories. A 2-cycle delay can
improve timing but will reduce performance with about 5%.
Reset address
CONFIG_IU_RSTADDR
By default, a SPARC processor starts execution at address 0.
With this option, any 4-kbyte aligned reset start address can be
choosen. Keep at 0 unless you really know what you are doing.
Power-down
CONFIG_PWD
Say Y here to enable the power-down feature of the processor.
Might reduce the maximum frequency slightly on FPGA targets.
For details on the power-down operation, see the LEON3 manual.
Hardware watchpoints
CONFIG_IU_WATCHPOINTS
The processor can have up to 4 hardware watchpoints, allowing to
create both data and instruction breakpoints at any memory location,
also in PROM. Each watchpoint will use approximately 500 gates.
Use 0 to disable the watchpoint function.
Floating-point enable
CONFIG_FPU_ENABLE
Say Y here to enable the floating-point interface for the GRFPU-lite
or GRFPU. Note that no FPUs are provided with the GPL version
of GRLIB. Both FPUs are commercial cores and must be obtained separately.
FPU selection
CONFIG_FPU_GRFPU
Select between Gaisler Research's GRFPU and GRFPU-lite FPUs. All cores
are fully IEEE-754 compatible and support all SPARC FPU instructions.
GRFPU Multiplier
CONFIG_FPU_GRFPU_INFMUL
On FPGA targets choose inferred multiplier. For ASIC implementations
choose between Synopsys Design Ware (DW) multiplier or Module
Generator (ModGen) multiplier. The DW multiplier gives better results
(smaller area and better timing) but requires a DW license.
The ModGen multiplier is part of GRLIB and does not require a license.
The Technology Specific multiplier option selects a pre-designed
multiplier using technology specific macrocells when available, else
an inferred multiplier is used.
Shared GRFPU
CONFIG_FPU_GRFPU_SH
If enabled multiple CPU cores will share one GRFPU.
GRFPC Configuration
CONFIG_FPU_GRFPC0
Configures the GRFPU-LITE controller.
In simple configuration controller executes FP instructions
in parallel with integer instructions. FP operands are fetched
in the register file stage and the result is written in the write
stage. This option uses least area resources.
Data forwarding configuration gives ~ 10 % higher FP performance than
the simple configuration by adding data forwarding between the pipeline
stages.
Non-blocking controller allows FP load and store instructions to
execute in parallel with FP instructions. The performance increase is
~ 20 % for FP applications. This option uses most logic resources and
is suitable for ASIC implementations.
Floating-point netlist
CONFIG_FPU_NETLIST
Say Y here to use a VHDL netlist of the GRFPU-Lite. This is
only available in certain versions of grlib.
Enable Instruction cache
CONFIG_ICACHE_ENABLE
The instruction cache should always be enabled to allow
maximum performance. Some low-end system might want to
save area and disable the cache, but this will reduce
the performance with a factor of 2 - 3.
Enable Data cache
CONFIG_DCACHE_ENABLE
The data cache should always be enabled to allow
maximum performance. Some low-end system might want to
save area and disable the cache, but this will reduce
the performance with a factor of 2 at least.
Instruction cache associativity
CONFIG_ICACHE_ASSO1
The instruction cache can be implemented as a multi-way cache with
1 - 4 ways. Higher associativity usually increases the cache hit
rate and thereby the performance. The downside is higher power
consumption and increased gate-count for tag comparators.
Note that a 1-way cache is effectively a direct-mapped cache.
Instruction cache way size
CONFIG_ICACHE_SZ1
The size of each way in the instuction cache (kbytes). Valid values
are 1 - 64 in binary steps. Note that the full range is only supported
by the generic and virtex2 targets. Most target packages are limited
to 2 - 16 kbyte. Large way size gives higher performance but might
affect the maximum frequency (on ASIC targets). The total instruction
cache size is the number of way multiplied with the way size.
Instruction cache line size
CONFIG_ICACHE_LZ16
The instruction cache line size. Can be set to either 16 or 32
bytes per line. Instruction caches typically benefit from larger
line sizes, but on small caches it migh be better with 16 bytes/line
to limit eviction miss rate.
Instruction cache replacement algorithm
CONFIG_ICACHE_ALGORND
Cache replacement algorithm for caches with 2 - 4 ways. The 'random'
algorithm selects the way to evict randomly. The least-recently-replaced
(LRR) algorithm evicts the way least recently replaced. The least-
recently-used (LRU) algorithm evicts the way least recently accessed.
The random algorithm uses a simple 1- or 2-bit counter to select
the eviction way and has low area overhead. The LRR scheme uses one
extra bit in the tag ram and has therefore also low area overhead.
However, the LRR scheme can only be used with 2-way caches. The LRU
scheme has typically the best performance but also highest area overhead.
A 2-way LRU uses 1 flip-flop per line, a 3-way LRU uses 3 flip-flops
per line, and a 4-way LRU uses 5 flip-flops per line to store the access
history.
Instruction cache locking
CONFIG_ICACHE_LOCK
Say Y here to enable cache locking in the instruction cache.
Locking can be done on cache-line level, but will increase the
width of the tag ram with one bit. If you don't know what
locking is good for, it is safe to say N.
Data cache associativity
CONFIG_DCACHE_ASSO1
The data cache can be implemented as a multi-way cache with
1 - 4 ways. Higher associativity usually increases the cache hit
rate and thereby the performance. The downside is higher power
consumption and increased gate-count for tag comparators.
Note that a 1-way cache is effectively a direct-mapped cache.
If the MMU is enabled then bus snooping is required to avoid
aliasing effects in multi-way caches.
Data cache way size
CONFIG_DCACHE_SZ1
The size of each way in the data cache (kbytes). Valid values are
1 - 64 in binary steps. Note that the full range is only supported
by the generic and virtex2 targets. Most target packages are limited
to 2 - 16 kbyte. A large cache gives higher performance but the
data cache is timing critical an a too large setting might affect
the maximum frequency (on ASIC targets). The total data cache size
is the number of ways multiplied with the way size.
Note that when the MMU is enabled, bus snooping should also be
enabled and the data cache way size should not exceed the MMU
page size.
Data cache line size
CONFIG_DCACHE_LZ16
The data cache line size. Can be set to either 16 or 32 bytes per
line. A smaller line size gives better associativity and higher
cache hit rate, but requires a larger tag memory.
Data cache replacement algorithm
CONFIG_DCACHE_ALGORND
See the explanation for instruction cache replacement algorithm.
Data cache locking
CONFIG_DCACHE_LOCK
Say Y here to enable cache locking in the data cache.
Locking can be done on cache-line level, but will increase the
width of the tag ram with one bit. If you don't know what
locking is good for, it is safe to say N.
Data cache snooping
CONFIG_DCACHE_SNOOP
Say Y here to enable data cache snooping on the AHB bus.
With snooping, AHB writes by other masters to data that is in the
data cache will be automatically detected and the cache line will
be marked as invalid. This simplifies software design in systems
with DMA units or multiple processors accessing the same memory.
It is a requirement (together with separate physical/snoop tags)
to run Linux on more than one CPU.
Depending on the separate snoop tags option, snooping will be
implemented using either dual-port (common tags) or two-port
RAMs (separate tags). A workaround if dual port RAMs are
not available for the target technology is to use separate
snoop tags even if it is not needed otherwise.
Separate snoop tags
CONFIG_DCACHE_SNOOP_SEPTAG
Enable a separate memory to store the data tags used for snooping.
This is necessary when snooping support is wanted in systems
with MMU, typically for SMP systems. In this case, the snoop
tags will contain the physical tag address while the normal
tags contain the virtual tag address.
When separate snoop tags are enabled, the tag RAMs will be implemented
using two-port RAMs instead of dual-port RAMs. A workaround if
dual port RAMs are not available for the target technology is
therefore to use separate snoop tags even if it is not needed.
Note that it is also possible to implement the tags using single port
RAMs with valid bits in flip-flops.
Single-port RAM for separate tags
CONFIG_DCACHE_SNOOP_SP
Use single-port RAM to implement separate physical tags. This leads
to the cache tag valid bits being implemented in flip-flops.
Fixed cacheability map
CONFIG_CACHE_FIXED
If this variable is 0, the cacheable memory regions are defined
by the AHB plug&play information (default). To overriden the
plug&play settings, this variable can be set to indicate which
areas should be cached. The value is treated as a 16-bit hex value
with each bit defining if a 256 Mbyte segment should be cached or not.
The right-most (LSB) bit defines the cacheability of AHB address
0 - 256 MByte, while the left-most bit (MSB) defines AHB address
3840 - 4096 MByte. If the bit is set, the corresponding area is
cacheable. A value of 00F3 defines address 0 - 0x20000000 and
0x40000000 - 0x80000000 as cacheable.
Local data ram
CONFIG_DCACHE_LRAM
Say Y here to add a local ram to the data cache controller.
Accesses to the ram (load/store) will be performed at 0 waitstates
and store data will never be written back to the AHB bus.
Size of local data ram
CONFIG_DCACHE_LRAM_SZ1
Defines the size of the local data ram in Kbytes. Note that most
technology libraries do not support larger rams than 16 Kbyte.
Start address of local data ram
CONFIG_DCACHE_LRSTART
Defines the 8 MSB bits of start address of the local data ram.
By default set to 8f (start address = 0x8f000000), but any value
(except 0) is possible. Note that the local data ram 'shadows'
a 16 Mbyte block of the address space.
MMU enable
CONFIG_MMU_ENABLE
Say Y here to enable the Memory Management Unit.
MMU split icache/dcache table lookaside buffer
CONFIG_MMU_COMBINED
Select "combined" for a combined icache/dcache table lookaside buffer,
"split" for a split icache/dcache table lookaside buffer
MMU tlb replacement scheme
CONFIG_MMU_REPARRAY
Select "LRU" to use the "least recently used" algorithm for TLB
replacement, or "Increment" for a simple incremental replacement
scheme.
Combined i/dcache tlb
CONFIG_MMU_I2
Select the number of entries for the instruction TLB, or the
combined icache/dcache TLB if such is used.
Split tlb, dcache
CONFIG_MMU_D2
Select the number of entries for the dcache TLB.
Fast writebuffer
CONFIG_MMU_FASTWB
Only selectable if split tlb is enabled. In case fast writebuffer is
enabled the tlb hit will be made concurrent to the cache hit. This
leads to higher store performance, but increased power and area.
MMU pagesize
CONFIG_MMU_PAGE_4K
The deafult SPARC V8 SRMMU page size is 4 Kbyte. This limits the
cache way size to 4 Kbyte, and total data cache size to 16 Kbyte,
when the MMU is used. To increase the maximum data cache size,
the MMU pages size can be increased to up 32 Kbyte. This will
give a maximum data cache size of 128 Kbyte.
Note that an MMU page size different than 4 Kbyte will require
a special linux tool-chain if glibc is used. If you don't know
what you are doing, stay with 4 Kbyte.
DSU enable
CONFIG_DSU_ENABLE
The debug support unit (DSU) allows non-intrusive debugging and tracing
of both executed instructions and AHB transfers. If you want to enable
the DSU, say Y here and select the configuration below.
Trace buffer enable
CONFIG_DSU_TRACEBUF
Say Y to enable the trace buffer. The buffer is not necessary for
debugging, only for tracing instructions and data transfers.
Enable instruction tracing
CONFIG_DSU_ITRACE
If you say Y here, an instruction trace buffer will be implemented
in each processor. The trace buffer will trace executed instructions
and their results, and place them in a circular buffer. The buffer
can be read out by any AHB master, and in particular by the debug
communication link.
Size of trace buffer
CONFIG_DSU_ITRACESZ1
Select the buffer size (in kbytes) for the instruction trace buffer.
Each line in the buffer needs 16 bytes. A 128-entry buffer will thus
need 2 kbyte.
Enable two-port instruction trace buffer
CONFIG_DSU_ITRACE_2P
If you say Y here, the instruction trace buffer will be implemented
in each processor with a two-port RAM. It will then be possible
to read the buffer through its second port, while the processor
instructions are being traced.
Enable AHB tracing
CONFIG_DSU_ATRACE
If you say Y here, an AHB trace buffer will be implemented in the
debug support unit processor. The AHB buffer will trace all transfers
on the AHB bus and save them in a circular buffer. The trace buffer
can be read out by any AHB master, and in particular by the debug
communication link.
Size of trace buffer
CONFIG_DSU_ATRACESZ1
Select the buffer size (in kbytes) for the AHB trace buffer.
Each line in the buffer needs 16 bytes. A 128-entry buffer will thus
need 2 kbyte.
AHB trace buffer filters
CONFIG_DSU_AFILT
If you say Y here, the debug support unit will allow filtering of
AHB trace buffer inputs. Filters allow to select which masters,
slaves and types of accesses that should be saved to the buffer.
AHB statistics outputs
CONFIG_DSU_ASTAT
If you say Y here, the debug support unit will enable its statistics
outputs. These outputs can be connected to a L4STAT unit in order to
count events on the AHB bus. This option will also enable AHB trace
buffer filters.
LEON3FT enable
CONFIG_LEON3FT_EN
Say Y here to use the fault-tolerant LEON3FT core instead of the
standard non-FT LEON3.
IU Register file protection
CONFIG_IUFT_NONE
Select the FT implementation in the LEON3FT integer unit
register file. The options include parity, parity with
sparing, 7-bit BCH, 7-bit BCH with on-the-fly (OTF)
correction, technology-specific and TMR.
FPU Register file protection
CONFIG_FPUFT_NONE
Say Y to enable SEU protection of the FPU register file.
The GRFPU-Lite will be protected with parity, PDMR and TMR.
The GRFPU can use PDMR or TMR only.
Cache memory error injection
CONFIG_RF_ERRINJ
Say Y here to enable error injection in to the IU/FPU regfiles.
Affects only simulation.
Cache memory protection
CONFIG_CACHE_FT_NONE
Enable SEU error-correction in the cache memories.
None - No protection
Parity - Technology agnostic parity protection
SECDEC-BCH - SECDED for cache, slow and large area overhead
TechSpec - Use target technology protection
The recommended setting is Parity. Target technologies
with built in protection on SRAMs can have area savings
with TechSpec. Currently this option should only be used
for Microsemi RTG4.
Cache memory error injection
CONFIG_CACHE_ERRINJ
Say Y here to enable error injection in to the cache memories.
Affects only simulation.
Leon3ft netlist
CONFIG_LEON3_NETLIST
Say Y here to use a VHDL netlist of the LEON3FT. This is
only available in certain versions of grlib.
IU assembly printing
CONFIG_IU_DISAS
Enable printing of executed instructions to the console.
IU assembly printing in netlist
CONFIG_IU_DISAS_NET
Enable printing of executed instructions to the console also
when simulating a netlist. NOTE: with this option enabled, it
will not be possible to pass place&route.
32-bit program counters
CONFIG_DEBUG_PC32
Since the LSB 2 bits of the program counters always are zero, they are
normally not implemented. If you say Y here, the program counters will
be implemented with full 32 bits, making debugging of the VHDL model
much easier. Turn of this option for synthesis or you will be wasting
area.
LEON3 Statistical Module
CONFIG_STAT_ENABLE
Say Y here to enable the LEON3 Statistical Unit. The unit consists
of a number of timers that can count various processor and MMU/FPU
events to allow analysis of processor behavior and performance.
Number of L3S counters
CONFIG_STAT_CNT
Set the number of 32-bit counters in the unit (1 - 32).
Number of L3S accumulative counters
CONFIG_STAT_NMAX
If this value is higher than 0, the core will include functionality
for tracking the longest consecutive time that an event is active or
inactive. The functionality will be available for the nmax first
counters. This value must not exceed the number of available counters
SPARC V8E non-privileged ASI access
CONFIG_NP_ASI
In SPARC-V8E implementations providing for non-privileged ASI access
functions, LOAD and STORE from Alternate space instructions
accessing ASI’s 0016 - 7F16 are privileged instructions. LOAD and
STORE from Alternate space instructions accessing ASI’s 8016 - FF16
are non-privileged instructions. This option enables this non-
privileged behavior.
CONFIG_AHB_DEFMST
Sets the default AHB master (see AMBA 2.0 specification for definition).
Should not be set to a value larger than the number of AHB masters - 1.
For highest processor performance, leave it at 0.
Default AHB master
CONFIG_AHB_RROBIN
Say Y here to enable round-robin arbitration of the AHB bus. A N will
select fixed priority, with the master with the highest bus index having
the highest priority.
Support AHB split-transactions
CONFIG_AHB_SPLIT
Say Y here to enable AHB split-transaction support in the AHB arbiter.
Unless you actually have an AHB slave that can generate AHB split
responses, say N and save some gates.
Enable full PnP decoding
CONFIG_AHB_FPNPEN
Say Y here to enable full decoding of the PnP configuration records in
in the AHB arbiter. When disabled the user-defined registers in the
PnP configuration records are not mapped in the configuration area.
IO area start address
CONFIG_AHB_IOADDR
Selects the MSB adddress (HADDR[31:20]) of the AHB IO area, as defined
in the plug&play extentions of the AMBA bus. Should be kept to FFF
unless you really know what you are doing.
APB bridge address
CONFIG_APB_HADDR
Selects the MSB adddress (HADDR[31:20]) of the APB bridge. Should be
kept at 800 for software compatibility.
AHB monitor
CONFIG_AHB_MON
Say Y to enable the AHB bus monitor. The monitor will check for
illegal AHB transactions during simulation. It has no impact on
synthesis.
Report AHB errors
CONFIG_AHB_MONERR
Print out detected AHB violations on console.
Report AHB warnings
CONFIG_AHB_MONWAR
Print out detected AHB warnings on console.
Write trace to console
CONFIG_AHB_DTRACE
Say yes here to write a trace of all AHB transfers to the
simulator console. Has not impact on final netlist.
DSU enable
CONFIG_DSU_UART
Say Y to enable the AHB uart (serial-to-AHB). This is the most
commonly used debug communication link.
JTAG Enable
CONFIG_DSU_JTAG
Say Y to enable the JTAG debug link (JTAG-to-AHB). Debugging is done
with GRMON through the boards JTAG chain at speeds of up to 800 kbits/s.
The TAP controller can be implemented in custom macros on
Altera, Actel Proasic/3 and Xilinx devices. The commercial
GRLIB also includes a generic TAP controller in VHDL.
Supported JTAG cables are Xilinx Parallel Cable III and IV,
Xilinx Platform cables (USB), and Altera parallel and USB cables,
Amontech JTAG key, various FTDI chip based USB/JTAG devices, and
Actel Flash Pro 3/4 cable.
Ethernet DSU enable
CONFIG_DSU_ETH
Say Y to enable the Ethernet Debug Communication Link (EDCL). The link
provides a DSU gateway between ethernet and the AHB bus. Debugging is
done at 10 or 100 Mbit/s, using the GRMON debug monitor. You must
enable the GRETH Ethernet MAC for this option to become active.
Size of EDCL trace buffer
CONFIG_DSU_ETHSZ1
Select the buffer size (in kbytes) for the EDCL. 1 or 2 kbyte is
usually enough, while a larger buffer will increase the transfer rate.
When operating at 100 Mbit, use a buffer size of at least 8 kbyte for
maximum throughput.
MSB IP address
CONFIG_DSU_IPMSB
Set the MSB 16 bits of the IP address of the EDCL.
LSB IP address
CONFIG_DSU_IPLSB
Set the LSB 16 bits of the IP address of the EDCL.
MSB ethernet address
CONFIG_DSU_ETHMSB
Set the MSB 24 bits of the ethernet address of the EDCL.
LSB ethernet address
CONFIG_DSU_ETHLSB
Set the LSB 24 bits of the ethernet address of the EDCL.
Programmable MAC/IP address
CONFIG_DSU_ETH_PROG
Say Y to make the LSB 4 bits of the EDCL MAC and IP address
configurable using the ethi.edcladdr inputs.
Leon2 memory controller
CONFIG_MCTRL_LEON2
Say Y here to enable the LEON2 memory controller. The controller
can access PROM, I/O, SRAM and SDRAM. The bus width for PROM
and SRAM is programmable to 8-, 16- or 32-bits.
8-bit memory support
CONFIG_MCTRL_8BIT
If you say Y here, the PROM/SRAM memory controller will support
8-bit mode, i.e. operate from 8-bit devices as if they were 32-bit.
Say N to save a few hundred gates.
16-bit memory support
CONFIG_MCTRL_16BIT
If you say Y here, the PROM/SRAM memory controller will support
16-bit mode, i.e. operate from 16-bit devices as if they were 32-bit.
Say N to save a few hundred gates.
Write strobe feedback
CONFIG_MCTRL_WFB
If you say Y here, the PROM/SRAM write strobes (WRITEN, WEN) will
be used to enable the data bus drivers during write cycles. This
will guarantee that the data is still valid on the rising edge of
the write strobe. If you say N, the write strobes and the data bus
drivers will be clocked on the rising edge, potentially creating
a hold time problem in external memory or I/O. However, in all
practical cases, there is enough capacitance in the data bus lines
to keep the value stable for a few (many?) nano-seconds after the
buffers have been disabled, making it safe to say N and remove a
combinational path in the netlist that might be difficult to
analyze.
Write strobe feedback
CONFIG_MCTRL_5CS
If you say Y here, the 5th (RAMSN[4]) SRAM chip select signal will
be enabled. If you don't intend to use it, say N and save some gates.
SDRAM controller enable
CONFIG_MCTRL_SDRAM
Say Y here to enabled the PC100/PC133 SDRAM controller. If you don't
intend to use SDRAM, say N and save about 1 kgates.
SDRAM controller inverted clock
CONFIG_MCTRL_SDRAM_INVCLK
If you say Y here, the SDRAM controller output signals will be delayed
with 1/2 clock in respect to the SDRAM clock. This will allow the used
of an SDRAM clock which in not strictly in phase with the internal
clock. This option will limit the SDRAM frequency to 40 - 50 MHz.
On FPGA targets without SDRAM clock synchronizations through PLL/DLL,
say Y. On ASIC targets, say N and tell your foundry to balance the
SDRAM clock output.
SDRAM separate address buses
CONFIG_MCTRL_SDRAM_SEPBUS
Say Y here if your SDRAM is connected through separate address
and data buses (SA & SD). This is the case on the GR-CPCI-XC2V6000
board, but not on the GR-PCI-XC2V3000 or Avnet XCV1500E boards.
64-bit data bus
CONFIG_MCTRL_SDRAM_BUS64
Say Y here to enable 64-bit SDRAM data bus.
Page burst enable
CONFIG_MCTRL_PAGE
Say Y here to enable SDRAM page burst operation. This will implement
read operations using page bursts rather than 8-word bursts and save
about 500 gates (100 LUTs). Note that not all SDRAM supports page
burst, so use this option with care.
Programmable page burst enable
CONFIG_MCTRL_PROGPAGE
Say Y here to enable programmable SDRAM page burst operation. This
will allow to dynamically enable/disable page burst by setting
bit 17 in MCFG2.
AHB status register
CONFIG_AHBSTAT_ENABLE
Say Y here to enable the AHB status register (AHBSTAT IP).
The register will latch the AHB address and master index when
an error response is returned by any AHB slave.
SDRAM separate address buses
CONFIG_AHBSTAT_NFTSLV
The AHB status register can also latch the AHB address on an external
input. Select here how many of such inputs are required.
Spacewire link
CONFIG_SPW_ENABLE
Say Y here to enable one or more Spacewire serial links. The links
are based on the GRSPW core from Gaisler Research.
Number of spacewire links
CONFIG_SPW_NUM
Select the number of links to implement. Each link will be a
separate AHB master and APB slave for configuration.
AHB FIFO depth
CONFIG_SPW_AHBFIFO4
Select the AHB FIFO depth (in 32-bit words).
RX FIFO depth
CONFIG_SPW_RXFIFO16
Select the receiver FIFO depth (in bytes).
RMAP protocol
CONFIG_SPW_RMAP
Enable hardware target support for the RMAP protocol (
draft C for GRSPW1 and ECSS-E-ST-50-11C Draft V1.3
for GRSPW2).
RMAP Buffer depth
CONFIG_SPW_RMAPBUF2
Select the size of the RMAP buffer (in bytes).
RMAP CRC
CONFIG_SPW_RMAPCRC
Enable hardware calculation of the RMAP CRC checksum. RMAP CRC
is always enabled when the RMAP hardware target is enabled so this
parameter will have no effect in that case.
Rx unaligned
CONFIG_SPW_RXUNAL
Enable support for byte writes used for non word-aligned
receiver buffer addresses. Without this enabled data will
still be written at the correct location but complete words
will always be written so data outside the intended boundaries
might be overwritten.
Netlists
CONFIG_SPW_NETLIST
Use the netlist version of GRSPW. This option is required if
you have not licensed the source code of the Spacewire core.
Currently only supported for Xilinx and Actel FPGAs.
The AHB/RX FIFO sizes should be set to 16 word/byte. If the
RMAP is enabled, set the buffer size to 128 bytes. Other netlist
configurations can be requested from Gaisler Research.
Spacewire FT
CONFIG_SPW_FT
Say Y here to implement the Spacewire block rams with fault-tolerance
against SEU errors.
Spacewire core
CONFIG_SPW_GRSPW1
Select to use GRSPW1 core or GRSPW2 core.
DMA channels
CONFIG_SPW_DMACHAN
Set the number of DMA channels for the GRSPW2 core.
Set to 1 for now ...
Ports
CONFIG_SPW_PORTS
Set the number of SpaceWire ports for the GRSPW2 core
Same clock for SpaceWire receiver and transmitter
CONFIG_SPW_RTSAME
Say Y here if the same clock is connected to both the receiver
and transmitter in the GRSPW2 core. This will remove two
asynchronous resets and some synchronization logic. This is only
applicable for the SDR and DDR inputs modes.
Receiver clock type
CONFIG_SPW_RX_SDR
Selects the input clocking scheme for the GRSPW2. SDR means that the
core samples data and strobe using single data rate registers at the
receiver clock frequency. DDR is the same except DDR registers are used.
Xor selects the traditional self clocking scheme using a xor gate.
Aeroflex sets the receiver in a mode compatible with the Aeroflex
SpaceWire transceiver.
Receiver clock type
CONFIG_SPW_TX_SDR
Selects the output clocking scheme for the GRSPW2. SDR means that the
core transmits data and strobe using single data rate registers at the
transmitter clock frequency. DDR is the same except DDR registers are used.
Aeroflex sets the transmitter in a mode compatible with the Aeroflex
SpaceWire transceiver.
Gaisler Ethernet MAC enable
CONFIG_GRETH_ENABLE
Say Y here to enable the Gaisler Research Ethernet MAC . The MAC has
one AHB master interface to read and write packets to memory, and one
APB slave interface for accessing the control registers.
Gaisler Ethernet 1G MAC enable
CONFIG_GRETH_GIGA
Say Y here to enable the Gaisler Research 1000 Mbit Ethernet MAC .
The 1G MAC is only available in the commercial version of GRLIB,
so do NOT enable it if you are using the GPL version.
CONFIG_GRETH_FIFO4
Set the depth of the receive and transmit FIFOs in the MAC core.
The MAC core will perform AHB burst read/writes with half the
size of the FIFO depth.
CONFIG_GRETH_FT
Enable fault-tolerance (parity with sparing) for internal RAM
in Ethernet core. Note that FT for EDCL buffer is enabled via
separate setting.
CONFIG_GRETH_EDCLFT
Enable protection for EDCL buffer. This is typically not used as
the EDCL debug link is not used in harsh environments.
SPI memory controller
CONFIG_SPIMCTRL
Say Y here to enable a simple SPI memory controller.
The controller maps a SPI memory device into AMBA address space and
also has a simple interface that allows sending commands directly
to the SPI device.
Read command
CONFIG_SPIMCTRL_READCMD
Read instruction for SPI memory device (hex).
Dummy byte
CONFIG_SPIMCTRL_DUMMYBYTE
Output dummy byte after address when issuing read instruction.
Dual output
CONFIG_SPIMCTRL_DUALOUTPUT
Memory device supports dual output when reading data.
Address offset
CONFIG_SPIMCTRL_OFFSET
Offset that will be added by core on SPI memory address (hex).
Clock scaler
CONFIG_SPIMCTRL_SCALER
Selects the divisor used when dividing the system clock to produce
the memory device clock. The divisor used is two to the power of the
specified value. This value must be at least 1.
Alternate clock scaler
CONFIG_SPIMCTRL_ASCALER
Selects the divisor used when dividing the system clock to produce
the alternate memory device clock. If the selected memory device is
a SD Card this clock will be used during card initialization. The
divisor used is two to the power of the specified value. This
value must be at least 1.
Power-up cnt
CONFIG_SPIMCTRL_PWRUPCNT
Number of system clock cycles to wait before issuing first command.
Gaisler Research SPI controller
CONFIG_SPICTRL_ENABLE
Say Y here to enable the SPI controller(s)
CONFIG_SPICTRL_NUM
Number of SPI controllers to implement in design. Note that most
template designs are limited to one SPI controller.
Configuration options made here in xconfig will apply to all
implemented SPI controllers.
CONFIG_SPICTRL_MAXWLEN
0: Core will support lengths up to 32 bits
1-2: Illegal values
3-15: Maximum word length will be value+1 (4-16)
CONFIG_SPICTRL_SYNCRAM
Say Y here to use SYNCRAM_2P components for the core's receive
and transmit queues. This is the recommended setting, particularly
if the core is implemented with support for automatic mode.
CONFIG_SPICTRL_FT
Fault-tolerance for internal buffers. Only applicable if core
buffers are implemented with SYNCRAM components.
CAN interface enable
CONFIG_CAN_ENABLE
Say Y here to enable one or more CAN cores. The cores has one
AHB slave interface for accessing the control registers. The CAN core
is register-compatible with the SAJ1000 core from Philips, with a
few exceptions. See the GRLIP IP manual for details.
CONFIG_CAN_NUM
Number of CAN cores. The module allows up to 8 independent
CAN cores to be implemented.
CAN register address
CONFIG_CANIO
The control registers of each CAN core occupies 256 bytes, and
address space needed for the full module is thus 2 Kbyte. The cores
are mapped in the AHB bus I/O area (0xFFF00000 - 0xFFFFF000).
This setting defines at which address in the I/O area the registers
appear (HADDR[19:8]).
CAN interrupt
CONFIG_CANIRQ
Defines which interrupt number the CAN core will generate.
CAN interrupt
CONFIG_CANSEPIRQ
Say Y here to assign an individual interrupt to each CAN core,
starting from the base interrupt number. If set to N, all
CAN cores will generate the same interrupt.
CAN FT memories
CONFIG_CAN_FT
If you say Y here, the CAN FIFOs will be implemented using
SEU protected RAM blocks. Only applicable to the FT version
of grlib.
CAN Synchronous reset
CONFIG_CAN_SYNCRST
If you say Y here, the CAN core will be implemented with
synchronous reset rather than asynchronous. This is needed
when the target library does not implement registers with
async reset. Unless you know what you are doing, say N.
UART1 enable
CONFIG_UART1_ENABLE
Say Y here to enable UART1, or the console UART. This is needed to
get any print-out from LEON3 systems regardless of operating system.
UART1 FIFO
CONFIG_UA1_FIFO1
The UART has configurable transmitt and receive FIFO's, which can
be set to 1 - 32 bytes. Use 1 for minimum area, or 8 - 32 for
maximum throughput.
UART2 enable
CONFIG_UART2_ENABLE
Say Y here to enable UART2, or the secondary UART. This UART can be
used to connect a second console (uClinux) or to control external
equipment.
UART2 FIFO
CONFIG_UA2_FIFO1
The UART has configurable transmitt and receive FIFO's, which can
be set to 1 - 32 bytes. Use 1 for minimum area, or 8 - 32 for
maximum throughput.
LEON3 interrupt controller
CONFIG_IRQ3_ENABLE
Say Y here to enable the LEON3 interrupt controller. This is needed
if you want to be able to receive interrupts. Operating systems like
Linux, RTEMS and eCos needs this option to be enabled. If you intend
to use the Bare-C run-time and not use interrupts, you could disable
the interrupt controller and save about 500 gates.
LEON3 interrupt controller broadcast
CONFIG_IRQ3_BROADCAST_ENABLE
If enabled the broadcast register is used to determine which
interrupt should be sent to all cpus instead of just the first
one that consumes it.
Secondary interrupts
CONFIG_IRQ3_SEC
The interrupt controller handles 15 interrupts by default (1 - 15).
These correspond to the 15 SPARC asyncronous traps (0x11 - 0x1F),
and AMBA interrupts 1 - 15. This option will enable 16 additional
(secondary) interrupts, corresponding to AMBA interrupts 16 - 31.
The secondary interrupts will be multiplexed onto one of the first
15 interrupts. The total number of handled interrupts can then
be up to 30 (14 primary and 16 secondary).
Number of interrupts
CONFIG_IRQ3_NSEC
Defines which of the first 15 interrupts should be used for the
secondary (16 - 31) interrupts. Interrupt 15 should be avoided
since it is not maskable by the processor.
Timer module enable
CONFIG_GPT_ENABLE
Say Y here to enable the Modular Timer Unit. The timer unit consists
of one common scaler and up to 7 independent timers. The timer unit
is needed for Linux, RTEMS, eCos and the Bare-C run-times.
Timer module enable
CONFIG_GPT_NTIM
Set the number of timers in the timer unit (1 - 7).
Scaler width
CONFIG_GPT_SW
Set the width if the common pre-scaler (2 - 16 bits). The scaler
is used to divide the system clock down to 1 MHz, so 8 bits should
be sufficient for most implementations (allows clocks up to 256 MHz).
Timer width
CONFIG_GPT_TW
Set the width if the timers (2 - 32 bits). 32 bits is recommended
for the Bare-C run-time, lower values (e.g. 16 bits) can work with
RTEMS and Linux.
Timer Interrupt
CONFIG_GPT_IRQ
Set the interrupt number for the first timer. Remaining timers will
have incrementing interrupts, unless the separate-interrupts option
below is disabled.
Watchdog enable
CONFIG_GPT_WDOGEN
Say Y here to enable the watchdog functionality in the timer unit.
Watchdog time-out value
CONFIG_GPT_WDOG
This value will be loaded in the watchdog timer at reset.
GPIO port
CONFIG_GRGPIO_ENABLE
Say Y here to enable a general purpose I/O port. The port can be
configured from 1 - 32 bits, whith each port signal individually
programmable as input or output. The port signals can also serve
as interrupt inputs.
GPIO port witdth
CONFIG_GRGPIO_WIDTH
Number of bits in the I/O port. Must be in the range of 1 - 32.
GPIO interrupt mask
CONFIG_GRGPIO_IMASK
The I/O port interrupt mask defines which bits in the I/O port
should be able to create an interrupt.
OpenCores I2C Master
CONFIG_I2C_ENABLE
Say Y here to enable the OpenCores I2C master.
UART debugging
CONFIG_DEBUG_UART
During simulation, the output from the UARTs is printed on the
simulator console. Since the ratio between the system clock and
UART baud-rate is quite high, simulating UART output will be very
slow. If you say Y here, the UARTs will print a character as soon
as it is stored in the transmitter data register. The transmitter
ready flag will be permanently set, speeding up simulation. However,
the output on the UART tx line will be garbled. Has not impact on
synthesis, but will cause the LEON test bench to fail.
FPU register tracing
CONFIG_DEBUG_FPURF
If you say Y here, all writes to the floating-point unit register file
will be printed on the simulator console.
Reference in a new issue View git blame Copy permalink