Commit graph

68 commits

Author SHA1 Message Date
Alan Carvalho de Assis
c027e7c3e4 tools: fix stale archive members surviving a Kconfig-driven CSRCS change
During the Toybox port to NuttX, Claude noticed that changes in the
menuconfig weren't taking affect. This issue exists for a long time on
NuttX, in fact BayLibre's presentation from 2017 make jokes about our
building system not been reliable:
https://www.youtube.com/watch?v=XUJK2htXxKw&t=320s

Stale archive members from $(AR)'s additive-only behavior can linger
after Kconfig toggles change which files provide a symbol, causing dead
weight or "multiple definition" link errors on incremental builds.
Fixed by splitting ARCHIVE into two macros: ARCHIVE keeps the original
additive behavior for apps/libapps.a, which many independent
subdirectories contribute to across a build, while the new
ARCHIVE_REBUILD deletes then archives for the far more common case
of a single Makefile building its own self-contained $(OBJS)
- all 39 such call sites now use it.

Assisted-By: Claude Sonnet 5
Signed-off-by: Alan C. Assis <acassis@gmail.com>
2026-07-28 21:26:03 -03:00
Karel Kočí
d61242c319 libs/libm: correct implementation of truncl if long double is double
This reuses implementation of trunc in case long double has same size as
double.

The previous implementation is used only in case x87 80-bit float point
is the long double. In other cases the logic is intentionally replaced
with panic to not provide a wrong result.

Signed-off-by: Karel Kočí <kkoci@elektroline.cz>
2026-07-02 09:02:21 -03:00
Xiang Xiao
c32b683085 !compiler: drop CONFIG_HAVE_LONG_LONG and require long long support
Every compiler supported by NuttX provides the "long long" types,
so the CONFIG_HAVE_LONG_LONG indirection is no longer useful.
Remove the option from include/nuttx/compiler.h and treat
"long long" as unconditionally available across the OS.

In addition to deleting the guard itself, this commit unconditionally
enables the long-long flavored helpers that used to be gated behind
it:

  - libs/libc/fixedmath: drop the soft-emulated b32/ub32 routines
    in lib_fixedmath.c (-261 lines) and trim the matching
    prototypes, Make.defs and CMakeLists.txt entries; keep only
    the long-long backed implementations.
  - include/sys/endian.h, include/strings.h, libs/libc/string
    /lib_ffsll.c, lib_flsll.c: always expose the 64-bit byte-swap
    and ffsll/flsll variants.
  - libs/libm/libm/lib_llround{,f,l}.c: drop the empty stubs.
  - libs/libc/stdlib (atoll, llabs, lldiv, strtoll/ull, rand48,
    strtold), libs/libc/stream (libvsprintf, libvscanf,
    libbsprintf, ultoa_invert), libs/libc/misc (crc64, crc64emac),
    libs/libc/inttypes/strtoimax, libs/libc/lzf, libs/libc/libc.csv,
    libs/libc/string (memset, vikmemcpy): remove the
    #ifdef CONFIG_HAVE_LONG_LONG branches.
  - include/{stddef.h,stdlib.h,fixedmath.h,sys/epoll.h,cxx/cstdlib,
    nuttx/audio/audio.h,nuttx/crc64.h,nuttx/lib/math.h,
    nuttx/lib/math32.h,nuttx/lib/stdbit.h}: same guard cleanup.
  - drivers/note/note_driver.c, fs/spiffs/src/spiffs.h,
    sched/irq/irq_procfs.c: drop their local guards as well.
  - Documentation/applications/netutils/ntpclient/index.rst:
    refresh the documentation snippet.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2026-05-19 16:21:28 +08:00
SPRESENSE
72b67832ea Makefile: Remove make depend files by make distclean
Intermediate files of make depend like .ddc and .dds may remain
when make is interrupted. Remove them using make distclean.

Signed-off-by: SPRESENSE <41312067+SPRESENSE@users.noreply.github.com>
2026-02-16 16:27:57 +01:00
buxiasen
f0cb80d297 libm: fix warning maybe-uninitialized
nuttx/libs/libm/openlibm/openlibm/src/k_rem_pio2.c:421:24: error: 'fq' may be used uninitialized [-Werror=maybe-uninitialized]

Signed-off-by: buxiasen <buxiasen@xiaomi.com>
2026-01-19 23:26:14 +08:00
fangxinyong
507056e870 cmake(bugfix):fix arch x86_64 cmake build openlibm warning
Building C object libs/libm/openlibm/CMakeFiles/m.dir/openlibm/src/s_clog.c.obj
In file included from /work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/src/math_private.h:22,
                 from /work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/src/s_clog.c:61:
/work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/src/s_clog.c:78:33: warning: ‘clogl’ alias between functions of incompatible types ‘_Complex long double(_Complex long double)’ and ‘_Complex double(_Complex double)’ [-Wattribute-alias=]
   78 | openlibm_strong_reference(clog, clogl);
      |                                 ^~~~~
/work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/src/cdefs-compat.h:24:50: note: in definition of macro ‘openlibm_strong_reference’
   24 |         OLM_DLLEXPORT extern __typeof (aliassym) aliassym __attribute__ ((__alias__ (#sym)));
      |                                                  ^~~~~~~~
/work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/src/s_clog.c:64:1: note: aliased declaration here
   64 | clog(double complex z)

Building C object libs/libm/openlibm/CMakeFiles/m.dir/openlibm/amd64/fenv.c.obj
In file included from /work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/include/openlibm_fenv.h:8,
                 from /work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/amd64/fenv.c:35:
/work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/include/openlibm_fenv_amd64.h:99:1: warning: declaration of ‘feclearexcept’ shadows a built-in function [-Wshadow]
   99 | feclearexcept(int __excepts)
      | ^~~~~~~~~~~~~
/work/rel-5.0/nuttx/libs/libm/openlibm/openlibm/include/openlibm_fenv_amd64.h:117:1: warning: declaration of ‘fegetexceptflag’ shadows a built-in function [-Wshadow]
  117 | fegetexceptflag(fexcept_t *__flagp, int __excepts)

Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
2026-01-19 23:26:14 +08:00
Carlos Sanchez
23a9ff1196 libm: prevent atanf() yielding NaN for high inputs values.
Without this fix, values of x where x * x + 1 are rounded
down could make asin() argument to be out of range.

Signed-off-by: Carlos Sanchez <carlossanchez@geotab.com>
2026-01-15 16:19:52 -03:00
Petro Karashchenko
8c3106e2a4 libs/libm: fix typo in Kconfig
fix typo in Kconfig

Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2025-12-09 18:03:10 +08:00
fangxinyong
31b5ef357c libm/newlib: Include fenv sources for all ARCHs and extend x86 includes
newlib/libm/fenv/*.c are the non-functional implementation that should
be overridden by an architecture specific implementation in newlib/libm/machine/ARCH.

Make the architecture check for including shared x86 sys headers for x86 and x86_64.

Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
2025-11-30 20:43:58 +08:00
fangxinyong
d956699f75 libm/newlib: Add duplicate filename filtering for source lists
Implement filename-based filtering to exclude duplicate files
from source lists, ensuring ARCH-specific files (ARCH_CSRCS)
take precedence over files in common or other dirs

Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
2025-11-30 20:43:58 +08:00
raiden00pl
7e11edba5d libs: unify Private Types banners
unify Private Types banners according to NuttX coding standard

Signed-off-by: raiden00pl <raiden00@railab.me>
2025-05-28 10:17:15 +08:00
Lars Kruse
3ce85ca54e style: fix spelling in code comments and strings 2025-05-23 10:48:41 +08:00
Tiago Medicci Serrano
5456668ae8 libm/newlib: Add arch-specific source code first to CSRCS
Source code under `newlib/newlib/newlib/libm/machine/$(ARCH)/*.c`
should be added before common source code from `../../common/*.c`.
Take `newlib/newlib/newlib/libm/machine/riscv/s_fma.c` as an
example:

```

double
fma (double x, double y, double z)
{
        double result;
        asm ("fmadd.d %0, %1, %2, %3" : "=f" (result) : "f" (x), "f" (y), "f" (z));
        return result;
}

```

Note that the common `s_fma.c` will be included by the source file
directly. The order of adding the files to CSRCS matters here.

Although the CMake-based build system does not have the same build
problem of including the a source-file with the same in the wrong
order, this commit also changes the order of inclusion for CMake
too to keep it consistent.
2024-12-04 02:10:41 +08:00
ligd
9026ba6883 libm: fix typo caused compile failed
make[1]: Entering directory '/home/ligd/platform/trunk_merge_all/nuttx/libs/libm'
Makefile:93: *** target file 'context' has both : and :: entries.  Stop.
make[1]: Leaving directory '/home/ligd/platform/trunk_merge_all/nuttx/libs/libm'

Signed-off-by: ligd <liguiding1@xiaomi.com>
2024-11-14 19:38:58 +08:00
wangmingrong1
6dd313939f makefile/build: Delete unnecessary bin and kbin folders
1. Calling CLEAN on these folders did not have any effect
2. bin,kbin will be created during the context construction process

Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
2024-11-10 01:18:26 +08:00
wangmingrong1
f2c46e08c8 Remove all jira and changeid information from the kernel
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
2024-11-06 02:04:03 +08:00
xuxin19
c60b980968 cmake(bugfix):fix robot openlibm cmake build missing source error
Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
2024-10-30 22:14:40 +08:00
Yongrong Wang
4baec5a248 libs/xxx/CMakeLists.txt: add cmake compile support
Add the lib_bzero.c files to the build use it's API
Add the floating point math C files to the build to use it's API

Signed-off-by: Yongrong Wang <wangyongrong@xiaomi.com>
2024-10-16 12:33:54 +08:00
liwenxiang1
3fd75efed9 Revert "libs/libc:Openlibm adds exp10 and exp10f function implementations"
This implementation is from glibc, and the license is LGPL

This reverts commit 862244586976b9388f87dc639225d714dd70e4af.

Signed-off-by: liwenxiang1 <liwenxiang1@xiaomi.com>
2024-10-16 08:03:31 +08:00
liwenxiang1
8d09eaaee0 libs/libc:Openlibm adds exp10 and exp10f function implementations
Signed-off-by: liwenxiang1 <liwenxiang1@xiaomi.com>
2024-10-13 02:31:44 +08:00
liwenxiang1
1c65fbf15a libc/newlib:Declaration of adding depth
Resolve the issue of inability to compile when opening newlib

Signed-off-by: liwenxiang1 <liwenxiang1@xiaomi.com>
2024-10-13 02:30:51 +08:00
xuxingliang
b9adff0e90 libm/newlib: remove -Wno-maybe-uninitialized
Since there's no warning presents now.

Signed-off-by: xuxingliang <xuxingliang@xiaomi.com>
2024-10-11 19:52:52 +08:00
Alin Jerpelea
d5754ec7da libs/libm: migrate to SPDX identifier
Most tools used for compliance and SBOM generation use SPDX identifiers
This change brings us a step closer to an easy SBOM generation.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2024-10-01 12:25:52 +08:00
Petro Karashchenko
f0cbcb95ff nuttx/libm: switch from integer to double constants in pow
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-09-30 08:32:46 +08:00
Petro Karashchenko
776035ee9e nuttx/libm: fix powl wrong if first parameter is negative
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-09-30 08:32:46 +08:00
yanghuatao
7cd38dca6a nuttx/libm: Fix powf wrong if first parameter is negative
Signed-off-by: yanghuatao <yanghuatao@xiaomi.com>
2024-09-28 10:42:36 +08:00
yanghuatao
375178b42a nuttx/libm: Move define __BSD_VISIBLE to openlibm_math.h
Signed-off-by: yanghuatao <yanghuatao@xiaomi.com>
2024-09-27 23:21:01 +08:00
Masayuki Ishikawa
df298c186f Revert "build depend:Revert Make.dep intermediate ddc file"
This reverts commit ddc3119c4e.
2024-09-15 19:29:47 +08:00
xuxin19
997be40003 cmake:bugfix fix sim CMake build break
fix lim CMake config prefix error

Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
2024-09-15 10:29:21 +08:00
xuxin19
5b3275c07b cmake:add libmcs cmake build
Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
2024-09-15 10:29:21 +08:00
xuxin19
21c68b44da cmake:support openlibm cmake build
Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
2024-09-15 10:29:21 +08:00
xuxin19
ddc3119c4e build depend:Revert Make.dep intermediate ddc file
Revert "Parallelize depend file generation"
This reverts commit d5b6ec450f.

parallel depend ddc does not significantly speed up compilation,
intermediately generated .ddc files can cause problems if compilation is interrupted unexpectedly

Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
2024-09-15 10:01:58 +08:00
Petro Karashchenko
c6a7816bed libs/libm/libm: increase number of 'log' iterations before epsilon relaxation
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-09-06 17:40:57 +08:00
Petro Karashchenko
02b1e895e2 libs/libm/libm: apply epsilon relax factor only if epsilon is small
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-09-06 17:40:57 +08:00
Petro Karashchenko
9cfb10069c Revert "libs/libm: fix epsilon relaxation in log and logf"
This reverts commit a2ed0ddd72.
2024-09-06 17:40:57 +08:00
Petro Karashchenko
a2ed0ddd72 libs/libm: fix epsilon relaxation in log and logf
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-09-05 10:14:04 +08:00
Petro Karashchenko
1528b8dcca nuttx: resolve various 'FAR' and 'CODE' issues
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2024-08-26 10:21:03 +08:00
Yingwei Zheng
871cd306c5 libm/copysign: respect signed zero/NaN in copysign 2024-08-01 01:08:19 +08:00
p-szafonimateusz
c932fe3045 libm/newlib: disable optimization for sincos
Disable sincos optimization for all functions in this file,
otherwise gcc would generate infinite calls.
Refer to gcc bug 46926.
-fno-builtin-sin or -fno-builtin-cos can disable sincos optimization,
but these two options do not work inside optimize pragma in-file.
Thus we just enforce -O0 when compiling this file.

Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
2024-06-28 01:26:18 +08:00
p-szafonimateusz
23b2f96c9d libm/libm: disable optimization for sincos
Disable sincos optimization for all functions in this file,
otherwise gcc would generate infinite calls.
Refer to gcc bug 46926.
-fno-builtin-sin or -fno-builtin-cos can disable sincos optimization,
but these two options do not work inside optimize pragma in-file.
Thus we just enforce -O0 when compiling this file.

Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
2024-06-28 01:26:18 +08:00
simbit18
09bfaa7292 fix nxstyle
fix Relative file path does not match actual file.
2024-05-11 01:19:06 +08:00
simbit18
7677f10d3f fix nxstyle
fix Relative file path does not match actual file.
2024-04-16 19:09:12 +08:00
p-szafonimateusz
e967be059a newlib: fix compilation for sincosl
aggresive optimisation can replace occurrences of sinl() and cosl() with
sincosl(), but sincosl() is missing in newlib which causes error. So let's
use custom implementation here.

Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
2024-02-26 23:56:47 +08:00
p-szafonimateusz
5288e063ec newlib: fix support for x86
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
2024-02-26 23:56:47 +08:00
SPRESENSE
6ba30033b3 libm: Fix an issue that public header files are not exported
Fix an issue that math library header files are not exported by make export.
Create symbolic links of libmcs, newlib and openlibm header to nuttx/include.
2023-11-23 16:32:53 +01:00
SPRESENSE
63182d43b9 libm/newlib: Change the download site to https
Change newlib download site from ftp to more secure https.
2023-11-10 16:28:13 +08:00
chao an
5026a96cfa nxstyle: cleanup UTF-8 Unicode to ASCII
Signed-off-by: chao an <anchao@xiaomi.com>
2023-09-18 11:54:17 -04:00
yanghuatao
7ea52505c2 libm/libmcs: Fix clang build download libmcs warning
VELAPLATFO-16865

Fix warnings:
CC:  libmcs/libmcs/libm/common/fenv.c libmcs/libmcs/libm/common/fenv.c:26:15: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
int fegetround()
              ^
               void
Signed-off-by: yanghuatao <yanghuatao@xiaomi.com>
2023-09-18 23:00:38 +08:00
yanghuatao
88193d3ad5 libm/libmcs: Fix clang build libmcs warning
VELAPLATFO-15795

Fix warnings:
1 warning generated.
CC:  libmcs/libmcs/libm/mathd/lrintd.c libmcs/libmcs/libm/mathd/lrintd.c:141:20: warning: implicit conversion from 'unsigned long long' to 'long' changes value from 9223372036854775808 to 0 [-Wconstant-conversion]
            return __MIN_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:50:24: note: expanded from macro '__MIN_LONG'
    #define __MIN_LONG 0x8000000000000000L
                       ^~~~~~~~~~~~~~~~~~~
libmcs/libmcs/libm/mathd/lrintd.c:144:20: warning: implicit conversion from 'long long' to 'long' changes value from 9223372036854775807 to -1 [-Wconstant-conversion]
            return __MAX_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:49:24: note: expanded from macro '__MAX_LONG'
    #define __MAX_LONG 0x7FFFFFFFFFFFFFFFL
                       ^~~~~~~~~~~~~~~~~~~
2 warnings generated.
CC:  libmcs/libmcs/libm/mathd/lroundd.c libmcs/libmcs/libm/mathd/lroundd.c:123:20: warning: implicit conversion from 'unsigned long long' to 'long' changes value from 9223372036854775808 to 0 [-Wconstant-conversion]
            return __MIN_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:50:24: note: expanded from macro '__MIN_LONG'
    #define __MIN_LONG 0x8000000000000000L
                       ^~~~~~~~~~~~~~~~~~~
libmcs/libmcs/libm/mathd/lroundd.c:126:20: warning: implicit conversion from 'long long' to 'long' changes value from 9223372036854775807 to -1 [-Wconstant-conversion]
            return __MAX_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:49:24: note: expanded from macro '__MAX_LONG'
    #define __MAX_LONG 0x7FFFFFFFFFFFFFFFL
                       ^~~~~~~~~~~~~~~~~~~
2 warnings generated.
CC:  libmcs/libmcs/libm/mathf/lrintf.c libmcs/libmcs/libm/mathf/lrintf.c:67:20: warning: implicit conversion from 'unsigned long long' to 'long' changes value from 9223372036854775808 to 0 [-Wconstant-conversion]
            return __MIN_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:50:24: note: expanded from macro '__MIN_LONG'
    #define __MIN_LONG 0x8000000000000000L
                       ^~~~~~~~~~~~~~~~~~~
libmcs/libmcs/libm/mathf/lrintf.c:70:20: warning: implicit conversion from 'long long' to 'long' changes value from 9223372036854775807 to -1 [-Wconstant-conversion]
            return __MAX_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:49:24: note: expanded from macro '__MAX_LONG'
    #define __MAX_LONG 0x7FFFFFFFFFFFFFFFL
                       ^~~~~~~~~~~~~~~~~~~
2 warnings generated.
CC:  libmcs/libmcs/libm/mathf/lroundf.c libmcs/libmcs/libm/mathf/lroundf.c:32:20: warning: implicit conversion from 'unsigned long long' to 'long' changes value from 9223372036854775808 to 0 [-Wconstant-conversion]
            return __MIN_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:50:24: note: expanded from macro '__MIN_LONG'
    #define __MIN_LONG 0x8000000000000000L
                       ^~~~~~~~~~~~~~~~~~~
libmcs/libmcs/libm/mathf/lroundf.c:34:20: warning: implicit conversion from 'long long' to 'long' changes value from 9223372036854775807 to -1 [-Wconstant-conversion]
            return __MAX_LONG;
            ~~~~~~ ^~~~~~~~~~
/mnt/yang/vela_keystore_sim/nuttx/libs/libm/libmcs/libmcs/libm/include/internal_config.h:49:24: note: expanded from macro '__MAX_LONG'
    #define __MAX_LONG 0x7FFFFFFFFFFFFFFFL

Signed-off-by: yanghuatao <yanghuatao@xiaomi.com>
2023-09-18 21:02:39 +08:00
Roy Feng
e90d58c3da esp32: Use newlib-esp32
When we are using newlib as libm, we can consider newlib-esp32 for esp32
hardware. An it contains some optimization for esp32 hardware.
Besides, two small issues has been fixed.
1. add the tar ball to .gitignore
2. Fix the some build warnings
2023-09-14 19:33:50 +08:00