Commit Graph

36678 Commits

Author SHA1 Message Date
YAMAMOTO Takashi
2b139625bb sim: Add a config to test examples/{module,sotest} with -m32 2020-03-29 14:16:14 +01:00
YAMAMOTO Takashi
4ae8e435d1 sim: Add a config to test examples/{module,sotest} 2020-03-29 14:16:14 +01:00
YAMAMOTO Takashi
4c0783353a tools/testbuild.sh: Make blacklist work on macOS
I couldn't find where "\b" was documented.
I guess it's a zero-width word boundary or something like that.
Anyway, it doesn't seem to work on macOS.
2020-03-29 09:39:11 +01:00
YAMAMOTO Takashi
35e1aa82cb tools/testbuild.sh: Allow spaces in comments 2020-03-28 11:23:42 -05:00
YAMAMOTO Takashi
01c91e8af8 sim: Remove CONFIG_RAW_BINARY=y from configs
Because:

* A raw binary doesn't make sense for sim, where ./nuttx is
  a host OS executable.

* It breaks test builds on macOS, where native objcopy
  is not available.
2020-03-28 10:47:43 -05:00
Brennan Ashton
f787df1d38 Pull the ESP32 binary blobs from a configurable path 2020-03-28 06:58:35 -05:00
Brennan Ashton
47dc68aa45 Use new images 2020-03-28 06:58:35 -05:00
YAMAMOTO Takashi
40d7f83e31 sim: Appease a few prototype warnings 2020-03-28 12:08:57 +01:00
raiden00pl
d217b3a889 arch/arm/src/nrf52: fix coding standard issues 2020-03-27 12:51:54 -06:00
raiden00pl
fef7abb598 arch/arm/src/nrf52/nrf52_i2c.c: protect the I2C transfer with a semaphore 2020-03-27 12:51:54 -06:00
YAMAMOTO Takashi
4f130d4274 README.txt: Mention a few tools for macOS 2020-03-27 02:42:11 -05:00
YAMAMOTO Takashi
f1a68ab444 Revert "Revert "README.txt: add MacOS section for flock installation""
This reverts commit 884fcb605d.
2020-03-27 02:42:11 -05:00
YAMAMOTO Takashi
2f6d458bcf sim/README: Mention macOS 2020-03-27 02:42:11 -05:00
YAMAMOTO Takashi
ebc6627b77 Bump DEFAULT_TASK_STACKSIZE for the sim
The sim demands more stack because:
 * It's often 64-bit
 * It calls host OS libraries
2020-03-27 01:14:10 -05:00
YAMAMOTO Takashi
a493c92826 sim: Use executable memory for the heap
On modern environments, bss is not executable.

Alternatively we can use the ARCH_HAVE_MODULE_TEXT mechanism.
But it's considered overkill for the sim, which is mainly
for development.
2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
748777e3f5 sim: MODULE definitions for macOS 2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
18c574d66a Config.mk: Provide the default MODULECC/MODULELD
In some cases, a separate toolchain is necessary to
build modules.
E.g. the sim on macOS, where the native format is Mach-O.

By default, just use CC/LD, which is appropriate for
the most of platforms.
2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
c62f806581 sim: modlib: Implement some x86_64 relocations
The boilerplate part of the code was copied from:

    libs/libc/machine/risc-v/rv64/arch_elf.c
2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
24c9fa3cd9 Kconfig: Move ELF_64BIT out of !BINFMT_DISABLE
Because it's also used by LIBC_MODLIB.
2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
bb09d564db libelf: Don't truncate 64-bit values 2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
bd7829a44d modlib: Don't truncate 64-bit values 2020-03-27 00:04:36 -05:00
YAMAMOTO Takashi
f8077022bd Introduce CONFIG_DEFAULT_TASK_STACKSIZE
* The appropriate size of stack varies among archs.
  E.g. for 64-bit sim, 2048 is way too small, especially when the task
  happens to use host OS functionalities.
  I plan to allow an arch provide its own default.

* I plan to use this to replace hardcoded "STACKSIZE = 2048" in APPDIR.
2020-03-26 22:30:13 -05:00
Gregory Nutt
44b8f3e467 include/nuttx/sched.h: Move prototype for sched_releasetcb()
Move the prototype for the internal OS from from sched/sched/sched.h to include/nuttx/sched.h.  This was done because binfmt/binfmt/excecmodule.c requires the prototype for sched_releasetcb() and was illegally including sched/sched/sched.h.  That is a blatant violation of the OS modular design and the person that did this should be hung up by their thumbs.  Oh... I did that back in a bad moment in 2014.  Now that is made right.
2020-03-26 14:12:34 -03:00
Gregory Nutt
5c0e8e88b1 Revert "Makefile: move INCDIROPT to common place (#625)"
This reverts commit b9ace36fcc.

This change was added by PR 625 but has a serious logic flaw.  It removes all occurrences of INCDIROPT and replaces it with a definition in tools/Config.mk:

    else ifeq ($(WINTOOL),y)
      DEFINE = "$(TOPDIR)/tools/define.sh"
      INCDIR = "$(TOPDIR)/tools/incdir.sh" -w

This logic flaw is the Config.mk is included in all Make.defs files BEFORE WINTOOL is defined.  As a result, the definition is wrong in many places when building under Cygwin with a Windows native toolchain.
2020-03-26 08:50:29 -07:00
Xiang Xiao
b9ace36fcc
Makefile: move INCDIROPT to common place (#625) 2020-03-26 08:09:59 -06:00
YAMAMOTO Takashi
8ad9d1e0b4 libelf: Appease nxstyle complaints
Also, fix a minor typo.
2020-03-26 02:51:34 -05:00
YAMAMOTO Takashi
d6d03d64d5 sim: Hide strerror/strerror_r 2020-03-25 21:35:17 -06:00
YAMAMOTO Takashi
27aa16111c sim: hostfs: Stop assuming CONFIG_NAME_MAX=32 2020-03-25 20:31:32 -06:00
YAMAMOTO Takashi
f89ccf873e sim: host_readdir: Ensure NUL termination when truncating d_name 2020-03-25 20:31:32 -06:00
YAMAMOTO Takashi
c65bdde5ac include/nuttx/fs/hostfs.h: Appease nxstyle errors 2020-03-25 20:31:32 -06:00
Andrey Zabolotnyi
3ff956dc00
stm32h7: support SDRAM via FMC peripherial (#459)
* stm32h7: Add support for SDRAM connected to FMC

* stm32h7: nxstyle fixes in stm32_fmc.h and stm32h7x3xx_rcc.{c,h}
2020-03-25 08:34:15 -07:00
YAMAMOTO Takashi
2991836ee8 sim: Appease nxstyle errors 2020-03-25 00:09:21 -05:00
YAMAMOTO Takashi
f19355132a sim: Mention vpnkit docker image 2020-03-24 20:49:19 -05:00
Xiang Xiao
3f860280e5 arm/common: Fix nxstyle issue in arch/arm/src/common/up_checkstack.c
and arch/arm/src/common/up_initialize.c
2020-03-24 10:50:02 +00:00
Xiang Xiao
bc29b25983 arm/imx6: compile up_hostfs.c when CONFIG_ARM_SEMIHOSTING_HOSTFS equal true 2020-03-24 10:49:59 +00:00
Xiang Xiao
d5905d5963 arch/arm: Fix linker error: undefined reference to `g_intstackalloc'
when CONFIG_ARCH_INTERRUPTSTACK, CONFIG_SMP and CONFIG_STACK_COLORATION are true
2020-03-24 10:49:53 +00:00
liuhaitao
272a1460c5 Revert "tools/testbuild.sh: suppress lots of stdout log from configure.sh"
f6fc87bc36 commit silence configure.sh
make output defautly, so revert it.

This reverts commit f0267aff33.
2020-03-23 22:04:27 -05:00
Ouss4
f6fc87bc36 tools/sethost.sh & tools/configure.c: When no make argument is passed, silence the make output. 2020-03-23 19:56:11 -05:00
YAMAMOTO Takashi
9fdda58eca sim: Hide putenv 2020-03-23 18:17:58 -06:00
YAMAMOTO Takashi
2e933102e1 sim: Hide getopt related globals
Right now this is rather theorethic as we don't actually
use getopt though.
2020-03-23 14:38:17 -05:00
YAMAMOTO Takashi
702d82d052 vpnkit: Add a config to specify the socket 2020-03-23 14:37:56 -05:00
YAMAMOTO Takashi
12a97acaed vpnkit: Make this less verbose on connection errors 2020-03-23 14:37:28 -05:00
Xiang Xiao
e4c8f5d767 esp32: Guard the binary generation with CONFIG_ESP32CORE_BINARY 2020-03-23 19:19:43 +00:00
Xiang Xiao
d2e6378322 Move CXD56_BINARY to boards/arm/cxd56xx/spresense/Kconfig 2020-03-23 19:19:40 +00:00
Xiang Xiao
e1385b1074 tools/Config.mk: CATFILE clear the existing file
To simplify the caller usage
2020-03-23 12:23:05 -06:00
Xiang Xiao
aaa32156df tools/Config.mk: implement a simple version of TESTANDREPLACEFILE for Windows 2020-03-23 12:22:50 -06:00
Peter van der Perk
1848e83257 Added S32K1XX FlexTimer register definitions 2020-03-23 12:34:28 -05:00
YAMAMOTO Takashi
1ffa009c8b Revert "Don't generate .depend anymore"
This reverts commit 79af7fbf4e.

Because:

* btashton reported some issues in local builds:

  https://github.com/apache/incubator-nuttx/pull/603#issuecomment-602264860

* this might be related to the current CI breakage:

  > /bin/sh: 1: /__w/incubator-nuttx/incubator-nuttx/nuttx/tools/mkdeps: not found
2020-03-22 23:07:29 -05:00
Xiang Xiao
79af7fbf4e
Don't generate .depend anymore 2020-03-22 18:15:29 +00:00
raiden00pl
1b4425ab5d boards/arm/xmc4: add initial support for the Infineon XMC4700 Relax board 2020-03-22 08:29:51 -06:00