This is an automated email from the ASF dual-hosted git repository.
xiaoxiang781216 pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/nuttx.git
from d430dfc89cb ci/membrowse: Track the mps2-an500 and qemu-armv7a citest
configurations.
new d84bdda6314 lib/math32: fix div64_const not compatible for pre-process
macro
new 0e52f1da3bd include/nuttx: fix compile error.
new ade88c9da2f lib/math32: Fix coverity of divide 0.
new d8476d6b727 lib/math32: fix undefined shift in div64_const32
new c13761842af lib/math32: widen div64_const() base to 64-bit
new 3cfd91351af lib/math32: use plain division in div64_const() with
CONFIG_DEBUG_NOOPT
new 9edc61f7ee0 lib/math32: expose the UMULH primitive as up_umulh64()
new 73b77ae69a2 lib/math32: replace typeof with sizeof() in the div_const
macros
new e695570d658 lib/math32: implement rounding div_const variants as
div_const wrappers
new e411a17d7f3 compiler: define is_constexpr() per compiler capability
new e1a84eb886d lib/math32: dispatch div_const() on compile-time constness
new 1e1be386717 arch/int128_t: add _int128_t/_uint128_t base types
new 740127fba21 include/math32.h: use arch _uint128_t for math32.h
compatible
new dc7cc0d6703 lib/math32: make FLS() width-generic and 64-bit safe on
all hosts
new fcfb257d211 arch/avr: wire optimization level into the avr32 toolchain
flags
new dd1b51fe837 arch: wire optimization level into toolchains missing it
[skip-cmake-parity]
new c09eda91295 arch/esp32: print tick conversions at their full width
new a821a92637c lib/math32: express log2ceil() as fls(n - 1)
new fae6b3291da lib/math32: fix stale invdiv and log2 comments
new 6452e0e40eb lib/math32: declare up_umulh64() in math32.h with a weak
fallback
new 6318b445cfe libs/libc/machine/arm: implement up_umulh64() in UMULL
assembly
new 3b05ebeb10a lib/math32: replace the udiv_soft loop with the
s-parameterized invdiv_udiv128
new 24191387206 lib/math32: unify the invdiv init behind
invdiv_init_param()
new 71997700731 lib/math32: implement div64_const32() with
invdiv_init_param()
new 27bc54d1a10 lib/math32: give the invdiv initializer's constants an
unsigned type
The 25 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
arch/arm64/include/types.h | 6 +
arch/avr/src/avr32/Toolchain.defs | 8 +-
arch/risc-v/include/types.h | 6 +
.../src/esp32c3-legacy/esp32c3_ble_adapter.c | 2 +-
arch/risc-v/src/esp32c3/esp_ble_adapter.c | 2 +-
arch/sim/include/types.h | 6 +
arch/tricore/include/types.h | 6 +
arch/tricore/src/common/tricore_perf.c | 2 +-
arch/x86/src/common/Toolchain.defs | 6 +-
arch/x86_64/include/intel64/types.h | 6 +
arch/xtensa/include/types.h | 6 +
arch/xtensa/src/esp32/esp32_ble_adapter.c | 2 +-
arch/xtensa/src/esp32s3/esp32s3_ble_adapter.c | 2 +-
boards/hc/m9s12/demo9s12ne64/scripts/Make.defs | 8 +-
boards/hc/m9s12/ne64badge/scripts/Make.defs | 8 +-
boards/renesas/m16c/skp16c26/scripts/Make.defs | 8 +-
.../renesas/rx65n/rx65n-grrose/scripts/Make.defs | 8 +-
.../renesas/rx65n/rx65n-rsk1mb/scripts/Make.defs | 8 +-
.../renesas/rx65n/rx65n-rsk2mb/scripts/Make.defs | 8 +-
boards/renesas/rx65n/rx65n/scripts/Make.defs | 8 +-
boards/renesas/sh1/us7032evb1/scripts/Make.defs | 8 +-
include/nuttx/compiler.h | 4 +-
include/nuttx/lib/math32.h | 535 ++++++++++-----------
include/nuttx/timers/oneshot.h | 2 +-
libs/libc/machine/CMakeLists.txt | 6 +
libs/libc/machine/Make.defs | 7 +
libs/libc/machine/arm/CMakeLists.txt | 1 +
libs/libc/machine/arm/Make.defs | 1 +
.../machine/arm/arch_umulh64.c} | 89 ++--
.../libc/machine/up_umulh64.c | 42 +-
30 files changed, 446 insertions(+), 365 deletions(-)
copy libs/{libnx/nxmu/nx_disconnect.c => libc/machine/arm/arch_umulh64.c} (50%)
copy boards/arm/s32k1xx/s32k144evb/src/s32k1xx_uid.c =>
libs/libc/machine/up_umulh64.c (63%)