djsundog/circuitpython
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
073b9a5eb8
circuitpython/ports/qemu-arm/Makefile

140 lines
3.5 KiB
Makefile
Raw Normal View History

all: Update Makefiles and others to build with new ports/ dir layout. Also renames "stmhal" to "stm32" in documentation and everywhere else.
2017-09-06 00:09:13 -04:00
include ../../py/mkenv.mk
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
-include mpconfigport.mk
# qstr definitions (must come before including py.mk)
QSTR_DEFS = qstrdefsport.h
ports: Enable error text compression for various ports, but not all. Enabled on: bare-arm, minimal, unix coverage/dev/minimal, stm32, esp32, esp8266, cc3200, teensy, qemu-arm, nrf. Not enabled on others to be able to test the code when the feature is disabled (the default case). Code size change for this commit: bare-arm: -600 -0.906% minimal x86: -308 -0.208% unix x64: +0 +0.000% unix nanbox: +0 +0.000% stm32: -3368 -0.869% PYBV10 cc3200: -1024 -0.558% esp8266: -2512 -0.368% GENERIC esp32: -2876 -0.205% GENERIC[incl -3168(data)] nrf: -1708 -1.173% pca10040 samd: +0 +0.000% ADAFRUIT_ITSYBITSY_M4_EXPRESS
2020-03-04 05:56:44 -05:00
# MicroPython feature configurations
MICROPY_ROM_TEXT_COMPRESSION ?= 1
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
# include py core make definitions
all: Make use of $(TOP) variable in Makefiles, instead of "..". $(TOP) is defined in py/mkenv.mk and should be used to refer to the top level of this repository.
2017-08-10 22:22:19 -04:00
include $(TOP)/py/py.mk
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
qemu-arm: Set default board as mps2-an385 to get more flash for tests. And use Ubuntu bionic for qemu-arm Travic CI job.
2020-03-01 23:03:10 -05:00
BOARD ?= mps2-an385
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
ifeq ($(BOARD),netduino2)
CFLAGS += -mthumb -mcpu=cortex-m3 -mfloat-abi=soft
CFLAGS += -DQEMU_SOC_STM32
LDSCRIPT = stm32.ld
qemu-arm: Use gchelper code to get registers for GC scanning.
2019-02-11 22:48:29 -05:00
SRC_BOARD_O = lib/utils/gchelper_m3.o
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
endif
ifeq ($(BOARD),microbit)
CFLAGS += -mthumb -mcpu=cortex-m0 -mfloat-abi=soft
CFLAGS += -DQEMU_SOC_NRF51
LDSCRIPT = nrf51.ld
QEMU_EXTRA = -global nrf51-soc.flash-size=1048576 -global nrf51-soc.sram-size=262144
qemu-arm: Use gchelper code to get registers for GC scanning.
2019-02-11 22:48:29 -05:00
SRC_BOARD_O = lib/utils/gchelper_m0.o
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
endif
ifeq ($(BOARD),mps2-an385)
CFLAGS += -mthumb -mcpu=cortex-m3 -mfloat-abi=soft
CFLAGS += -DQEMU_SOC_MPS2
LDSCRIPT = mps2.ld
qemu-arm: Use gchelper code to get registers for GC scanning.
2019-02-11 22:48:29 -05:00
SRC_BOARD_O = lib/utils/gchelper_m3.o
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
endif
qemu-arm/Makefile: Allow overriding CROSS_COMPILE from another makefile.
2019-12-18 13:26:55 -05:00
CROSS_COMPILE ?= arm-none-eabi-
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
all: Make netutils.h available to all ports by default. Generally, ports should inherit INC from py.mk, append to it, not overwrite it. TODO: Likely should do the same for other vars too.
2015-10-19 11:22:16 -04:00
INC += -I.
all: Make use of $(TOP) variable in Makefiles, instead of "..". $(TOP) is defined in py/mkenv.mk and should be used to refer to the top level of this repository.
2017-08-10 22:22:19 -04:00
INC += -I$(TOP)
qemu-arm: update makefile (d553be5).
2014-04-21 16:08:15 -04:00
INC += -I$(BUILD)
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
CFLAGS += $(INC) -Wall -Wpointer-arith -Werror -std=gnu99 $(COPT) \
qemu-arm: Enable software floating point support, and float tests. This helps to test floating point code on Cortex-M hardware. As part of this patch the link-time-optimisation was disabled because it wasn't compatible with software FP support. In particular, the linker could not find the __aeabi_f2d, __aeabi_d2f etc functions even though they were provided by lib/libm/math.c.
2016-11-02 21:28:31 -04:00
-ffunction-sections -fdata-sections
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
#Debugging/Optimization
ifeq ($(DEBUG), 1)
CFLAGS += -g -DPENDSV_DEBUG
COPT = -O0
else
COPT += -Os -DNDEBUG
endif
cortex-m3-qemu: refactor the port. Switch from CodeSourcery to ARM GCC and clean-up some stale files, also copy `main.c` and `mpconfigport.h` from bare-arm.
2014-04-21 16:00:23 -04:00
## With CoudeSourcery it's actually a little different, you just need `-T generic-m-hosted.ld`.
## Although for some reason `$(LD)` will not find that linker script, it works with `$(CC)`.
## It turns out that this is specific to CoudeSourcery, and ARM version of GCC ships something
## else instead and according to the following files, this is what we need to pass to `$(CC).
## - gcc-arm-none-eabi-4_8-2014q1/share/gcc-arm-none-eabi/samples/src/makefile.conf
## - gcc-arm-none-eabi-4_8-2014q1/share/gcc-arm-none-eabi/samples/src/qemu/Makefile
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
LDFLAGS= -T $(LDSCRIPT) --gc-sections -Map=$(@:.elf=.map)
LIBS = $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
SRC_COMMON_C = \
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
startup.c \
uart.c \
qemu-arm: Add basic uos module with generic VFS capabilities.
2017-03-13 06:24:21 -04:00
moduos.c \
qemu-arm: Enable machine module and associated tests.
2017-03-02 00:39:58 -05:00
modmachine.c \
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
SRC_RUN_C = \
main.c \
qemu-arm: Add 'test' target to Makefile to run and verify test suite. Replaces RUN_TEST=1 definition; now "make test" in qemu-arm directory will run tests/basics/ and check that they all succeed. This patch also enables the test on Travis CI.
2015-01-08 19:03:21 -05:00
SRC_TEST_C = \
test_main.c \
qemu-arm: Let tinytest.o be built by standard build rules. This makes sure tinytest.o is rebuilt if any of its dependencies change.
2019-12-16 21:18:08 -05:00
lib/tinytest/tinytest.c \
qemu-arm: fully integrated test suite. This is primarily intended to provide testing of Thumb-specific code within Travis CI as well as if anyone else want to run it locally. As discussed in purposes. This is currently agains an emulated Cortex-M3 core, however in the near future it can extended to support M0, M0+ as well M4 (work in progress exists in sushihangover/qemu). It's probably true that most of the code base can be covered running uPy natively on a POSIX system, however we do have the tiny bit of assembly code. There may exist bugs related to endianness and type aliases, let alone potential standard library or compiler bugs or even architecture-specific optimisations. This could also incorporate lwIP (or other TCP/IP stack) integration as well as SDIO+FATFS drivers. The solution to inline the test cases was chose due to simplicity. It could alternatively be implemented in a number of different way (see #515), but this looked the simplest. Inclusion of tinytest was just to avoid writing boilerplate code for counting failed tests and other utility functions. Currently only a few functions are used, however this could be extended. Checking in the code instead of using submodule was a personal preference, but if people do want the pain of submodules, this can provided. This particular framework is also pretty good if one desires to run unit test on target. The approach with scripts being inlined is probably not quite suited for the size of memory an MCU has, but the tinytest itself should be good, if lower-level C code is to be unit tested.
2014-05-05 12:06:45 -04:00
ports/qemu-arm: Rework "test" target using upytesthelper. The way tinytest was used in qemu-arm test target is that it didn't test much. MicroPython tests are based on matching the test output against reference output, but qemu-arm's implementation didn't do that, it effectively tested just that there was no exception during test execution. "upytesthelper" wrapper was introduce to fix it, so switch test implementation to use it. This requires passing different CFLAGS when building the firmware, so split out test-related parts to Makefile.test.
2017-12-11 13:04:27 -05:00
LIB_SRC_C += $(addprefix lib/,\
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
libc/string0.c \
qemu-arm: Enable software floating point support, and float tests. This helps to test floating point code on Cortex-M hardware. As part of this patch the link-time-optimisation was disabled because it wasn't compatible with software FP support. In particular, the linker could not find the __aeabi_f2d, __aeabi_d2f etc functions even though they were provided by lib/libm/math.c.
2016-11-02 21:28:31 -04:00
libm/math.c \
libm/fmodf.c \
py/modbuiltins: For round() builtin use nearbyint instead of round. The C nearbyint function has exactly the semantics that Python's round() requires, whereas C's round() requires extra steps to handle rounding of numbers half way between integers. So using nearbyint reduces code size and potentially eliminates any source of errors in the handling of half-way numbers. Also, bare-metal implementations of nearbyint can be more efficient than round, so further code size is saved (and efficiency improved). nearbyint is provided in the C99 standard so it should be available on all supported platforms.
2017-03-23 19:40:25 -04:00
libm/nearbyintf.c \
qemu-arm: Enable software floating point support, and float tests. This helps to test floating point code on Cortex-M hardware. As part of this patch the link-time-optimisation was disabled because it wasn't compatible with software FP support. In particular, the linker could not find the __aeabi_f2d, __aeabi_d2f etc functions even though they were provided by lib/libm/math.c.
2016-11-02 21:28:31 -04:00
libm/ef_sqrt.c \
libm/kf_rem_pio2.c \
libm/kf_sin.c \
libm/kf_cos.c \
libm/kf_tan.c \
libm/ef_rem_pio2.c \
libm/sf_sin.c \
libm/sf_cos.c \
libm/sf_tan.c \
libm/sf_frexp.c \
libm/sf_modf.c \
libm/sf_ldexp.c \
libm/asinfacosf.c \
libm/atanf.c \
libm/atan2f.c \
ports: Add lib/libm/roundf.c to bare-metal Makefile's. This function is not used by the core but having it as part of the build allows it to be used by user C modules, or board extensions. The linker won't include it in the final firmware if it remains unused.
2020-03-24 03:11:36 -04:00
libm/roundf.c \
stmhal: Move pybstdio.c to lib/utils/sys_stdio_mphal.c for common use. It provides sys.stdin, sys.stdout, sys.stderr for bare-metal targets based on mp_hal functions.
2017-06-15 04:55:34 -04:00
utils/sys_stdio_mphal.c \
qemu-arm: Enable lots of extmods and enable tests for them. The qemu-arm port is used for testing of ARM Thumb architecture on a desktop so should have many features enabled.
2016-10-13 09:08:19 -04:00
)
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
OBJ_COMMON =
OBJ_COMMON += $(PY_O)
OBJ_COMMON += $(addprefix $(BUILD)/, $(SRC_COMMON_C:.c=.o))
qemu-arm: Use gchelper code to get registers for GC scanning.
2019-02-11 22:48:29 -05:00
OBJ_COMMON += $(addprefix $(BUILD)/, $(SRC_BOARD_O))
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
OBJ_COMMON += $(addprefix $(BUILD)/, $(LIB_SRC_C:.c=.o))
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
OBJ_RUN =
OBJ_RUN += $(addprefix $(BUILD)/, $(SRC_RUN_C:.c=.o))
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
qemu-arm: Add ldscript dependency in the final firmware.elf target. So that the target is rebuilt if the linker script changes.
2019-11-11 05:51:23 -05:00
ALL_OBJ_RUN = $(OBJ_COMMON) $(OBJ_RUN)
qemu-arm: Add 'test' target to Makefile to run and verify test suite. Replaces RUN_TEST=1 definition; now "make test" in qemu-arm directory will run tests/basics/ and check that they all succeed. This patch also enables the test on Travis CI.
2015-01-08 19:03:21 -05:00
OBJ_TEST =
OBJ_TEST += $(addprefix $(BUILD)/, $(SRC_TEST_C:.c=.o))
qemu-arm: fully integrated test suite. This is primarily intended to provide testing of Thumb-specific code within Travis CI as well as if anyone else want to run it locally. As discussed in purposes. This is currently agains an emulated Cortex-M3 core, however in the near future it can extended to support M0, M0+ as well M4 (work in progress exists in sushihangover/qemu). It's probably true that most of the code base can be covered running uPy natively on a POSIX system, however we do have the tiny bit of assembly code. There may exist bugs related to endianness and type aliases, let alone potential standard library or compiler bugs or even architecture-specific optimisations. This could also incorporate lwIP (or other TCP/IP stack) integration as well as SDIO+FATFS drivers. The solution to inline the test cases was chose due to simplicity. It could alternatively be implemented in a number of different way (see #515), but this looked the simplest. Inclusion of tinytest was just to avoid writing boilerplate code for counting failed tests and other utility functions. Currently only a few functions are used, however this could be extended. Checking in the code instead of using submodule was a personal preference, but if people do want the pain of submodules, this can provided. This particular framework is also pretty good if one desires to run unit test on target. The approach with scripts being inlined is probably not quite suited for the size of memory an MCU has, but the tinytest itself should be good, if lower-level C code is to be unit tested.
2014-05-05 12:06:45 -04:00
qemu-arm: Add ldscript dependency in the final firmware.elf target. So that the target is rebuilt if the linker script changes.
2019-11-11 05:51:23 -05:00
ALL_OBJ_TEST = $(OBJ_COMMON) $(OBJ_TEST)
qemu-arm/Makefile: Adjust object-file lists to get correct dependencies. With this change both the "run" and "test" targets can be built with -j passed to make.
2017-05-26 08:51:14 -04:00
# All object files, needed to get dependencies correct
OBJ = $(OBJ_COMMON) $(OBJ_RUN) $(OBJ_TEST)
qemu-arm: Enable lots of extmods and enable tests for them. The qemu-arm port is used for testing of ARM Thumb architecture on a desktop so should have many features enabled.
2016-10-13 09:08:19 -04:00
# List of sources for qstr extraction
qemu-arm/Makefile: Include relevant sources in list for qstr extraction.
2017-06-15 06:02:43 -04:00
SRC_QSTR += $(SRC_COMMON_C) $(SRC_RUN_C) $(LIB_SRC_C)
qemu-arm: Enable lots of extmods and enable tests for them. The qemu-arm port is used for testing of ARM Thumb architecture on a desktop so should have many features enabled.
2016-10-13 09:08:19 -04:00
qemu-arm: Convert to use FROZEN_MANIFEST to specify frozen code.
2019-10-10 08:49:37 -04:00
ifneq ($(FROZEN_MANIFEST),)
CFLAGS += -DMICROPY_MODULE_FROZEN_STR
qemu-arm: Add testing of frozen native modules. - Split 'qemu-arm' from 'unix' for generating tests. - Add frozen module to the qemu-arm test build. - Add test that reproduces the requirement to half-word align native function data.
2019-08-18 20:59:27 -04:00
CFLAGS += -DMICROPY_MODULE_FROZEN_MPY
qemu-arm: Convert to use FROZEN_MANIFEST to specify frozen code.
2019-10-10 08:49:37 -04:00
CFLAGS += -DMICROPY_QSTR_EXTRA_POOL=mp_qstr_frozen_const_pool
qemu-arm: Add testing of frozen native modules. - Split 'qemu-arm' from 'unix' for generating tests. - Add frozen module to the qemu-arm test build. - Add test that reproduces the requirement to half-word align native function data.
2019-08-18 20:59:27 -04:00
MPY_CROSS_FLAGS += -march=armv7m
endif
cortex-m3-qemu: refactor the port. Switch from CodeSourcery to ARM GCC and clean-up some stale files, also copy `main.c` and `mpconfigport.h` from bare-arm.
2014-04-21 16:00:23 -04:00
all: run
qemu-arm: Add 'test' target to Makefile to run and verify test suite. Replaces RUN_TEST=1 definition; now "make test" in qemu-arm directory will run tests/basics/ and check that they all succeed. This patch also enables the test on Travis CI.
2015-01-08 19:03:21 -05:00
run: $(BUILD)/firmware.elf
qemu-arm: Rework to run bare-metal on boards with Cortex-M CPUs. Adds support for 3 Cortex-M boards, selectable via "BOARD" in the Makefile: - microbit, Cortex-M0 via nRF51822 - netduino2, Cortex-M3 via STM32F205 - mps2-an385, Cortex-M3 via FPGA netduino2 is the default board because it's supported by older qemu versions (down to at least 2.5.0).
2018-09-23 22:09:28 -04:00
qemu-system-arm -machine $(BOARD) $(QEMU_EXTRA) -nographic -monitor null -semihosting -kernel $<
qemu-arm: Add 'test' target to Makefile to run and verify test suite. Replaces RUN_TEST=1 definition; now "make test" in qemu-arm directory will run tests/basics/ and check that they all succeed. This patch also enables the test on Travis CI.
2015-01-08 19:03:21 -05:00
cortex-m3-qemu: refactor the port. Switch from CodeSourcery to ARM GCC and clean-up some stale files, also copy `main.c` and `mpconfigport.h` from bare-arm.
2014-04-21 16:00:23 -04:00
## `$(LD)` doesn't seem to like `--specs` for some reason, but we can just use `$(CC)` here.
qemu-arm: Add ldscript dependency in the final firmware.elf target. So that the target is rebuilt if the linker script changes.
2019-11-11 05:51:23 -05:00
$(BUILD)/firmware.elf: $(LDSCRIPT) $(ALL_OBJ_RUN)
$(Q)$(LD) $(LDFLAGS) -o $@ $(ALL_OBJ_RUN) $(LIBS)
cortex-m3-qemu: a working port
2014-04-16 12:27:23 -04:00
$(Q)$(SIZE) $@
all: Make use of $(TOP) variable in Makefiles, instead of "..". $(TOP) is defined in py/mkenv.mk and should be used to refer to the top level of this repository.
2017-08-10 22:22:19 -04:00
include $(TOP)/py/mkrules.mk
Reference in New Issue Copy Permalink