build system: use thread-safe stdio #21438

maribu · 2025-04-25T06:14:42Z

Contribution description

build system: introduce `bug_%` feature category

This introduces a new feature category to declare which bugs are present
in a given build that we cannot fix in RIOT, but need to work around.
These bugs may be silicon bugs, software bugs in ROM, software bugs in
binary blobs needed for a platform, or bugs in the toolchain.

These features behave the same way as arch_% features: Every provided
bug is always used, so that we can inspect FEATURES_USED to check
which bugs need to be worked around.

build system: use thread-safe stdio

This makes use of the new bug modeling to declare all platforms that
can use newlib and have no reentrancy hooks as affected by the
non-thread-safe stdio bug. (Which is every platform but ESP* and AVR,
the former because the reentrancy hooks are provided, the latter because
we do not and never will support newlib on them.)

Building on that, the mpaland-printf package is used when newlib is used
and the bug is present. This way we can rely on the stdio being
thread-safe on every platform and not causing random crashes at
run time.

sys/newlib: drop workaround for stdio

This drops a workaround that initialized newlib's reentrancy structure
on boot to reduce the chances of crashes when using the non-thread-safe
(unless reentrancy hooks are provided) stdio implementation of newlib.

Now that the newlib stdio implementation is only ever used if it is
thread-safe, we no longer need a workaround that reduces the chance
of crashes on concurrent use of stdio.

Testing procedure

Now the package mpaland-printf should be pulled in for all non-ESP boards when newlib is used. E.g.

for board in hifive1b same54-xpro native32 native64 arduino-mega2560 esp32-mh-et-live-minikit esp8266-esp-12x z1 nucleo-l011k4; do
    printf '\n%s\n===================\n' $board
    echo "Features:"
    make BOARD=$board --no-print-directory -C examples/basic/hello-world info-features-used | grep -E 'newlib|libc'
    echo "Packages:"
    make BOARD=$board --no-print-directory -C examples/basic/hello-world info-packages
done

hifive1b
===================
Features:
bug_newlib_broken_stdio
newlib
Packages:
mpaland-printf

same54-xpro
===================
Features:
bug_newlib_broken_stdio
newlib
Packages:
cmsis
mpaland-printf

native32
===================
Features:
Packages:

native64
===================
Features:
Packages:

arduino-mega2560
===================
Features:
Packages:

esp32-mh-et-live-minikit
===================
Features:
newlib
Packages:
esp32_sdk

esp8266-esp-12x
===================
Features:
newlib
Packages:
esp8266_sdk

z1
===================
Features:
bug_newlib_broken_stdio
Packages:
mpaland-printf

nucleo-l011k4
===================
Features:
bug_newlib_broken_stdio
picolibc
Packages:
cmsis

With this, the test in tests/sys/snprintf now should also pass for each and every board, except for ESP32 boards. The reason is that, since ESP* provides the reentrancy hooks, newlib's stdio is thread-safe on those. So we do not pull in the alternative printf. For ESP8266 boards newlib is configured to use the non-nano stdio, which passes the test. For ESP32 newlib-nano is used, which does not implement all format specifiers and, therefore, fails the test.

Issues/PRs references

Depends on and includes:

pkg/mpaland-printf: fix overriding printf with printf_float #21439

Makefile.include

riot-ci · 2025-04-25T10:19:25Z

Murdock results

✔️ PASSED

4a84f6f tests/sys/snprintf: use printf_long_long

Success	Failures	Total	Runtime
10536	0	10536	20m:02s

Artifacts

Documentation preview

crasbe · 2025-04-25T11:18:09Z

We should probably give this a full build without fast fail, otherwise you'll be chasing your tail for days 😅

maribu · 2025-04-25T11:21:02Z

We should probably give this a full build without fast fail, otherwise you'll be chasing your tail for days 😅

No, actually the memory requirements will go down with this. Unless... both printf variants get linked in into the same app.

See #21439 for a fix of that.

Enoch247 · 2025-04-26T00:16:14Z

in one of your commit messages and in the description of this PR you have every misspelled.

This way we can rely on the stdio being thread-safe on evyer platform and not causing random crashes at run time.

mguetschow

Thanks for working on this!

Unfortunately, something seems to fail on the first board I've tested this with:

$ make -C tests/sys/snprintf BOARD=adafruit-feather-nrf52840-sense clean flash term    
make: Entering directory '/home/mikolai/TUD/Code/RIOT/tests/sys/snprintf'
rm -rf /home/mikolai/TUD/Code/RIOT/tests/sys/snprintf/bin/adafruit-feather-nrf52840-sense/pkg-build/cmsis
rm -rf /home/mikolai/TUD/Code/RIOT/tests/sys/snprintf/bin/adafruit-feather-nrf52840-sense/pkg-build/mpaland-printf
Building application "tests_snprintf" for "adafruit-feather-nrf52840-sense" with CPU "nrf52".

"make" -C /home/mikolai/TUD/Code/RIOT/pkg/cmsis/ 
"make" -C /home/mikolai/TUD/Code/RIOT/pkg/mpaland-printf/ 
"make" -C /home/mikolai/.riot/pkg/mpaland-printf -f /home/mikolai/TUD/Code/RIOT/pkg/mpaland-printf/mpaland-printf.mk
"make" -C /home/mikolai/TUD/Code/RIOT/boards
"make" -C /home/mikolai/TUD/Code/RIOT/boards/common/adafruit-nrf52-bootloader
"make" -C /home/mikolai/TUD/Code/RIOT/boards/common/init
"make" -C /home/mikolai/TUD/Code/RIOT/boards/adafruit-feather-nrf52840-sense
"make" -C /home/mikolai/TUD/Code/RIOT/core
"make" -C /home/mikolai/TUD/Code/RIOT/core/lib
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/cortexm_common
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/cortexm_common/periph
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52/periph
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52/vectors
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf5x_common
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf5x_common/periph
"make" -C /home/mikolai/TUD/Code/RIOT/drivers
"make" -C /home/mikolai/TUD/Code/RIOT/drivers/periph_common
"make" -C /home/mikolai/TUD/Code/RIOT/sys
"make" -C /home/mikolai/TUD/Code/RIOT/sys/auto_init
"make" -C /home/mikolai/TUD/Code/RIOT/sys/auto_init/usb
"make" -C /home/mikolai/TUD/Code/RIOT/sys/div
"make" -C /home/mikolai/TUD/Code/RIOT/sys/event
"make" -C /home/mikolai/TUD/Code/RIOT/sys/fmt
"make" -C /home/mikolai/TUD/Code/RIOT/sys/isrpipe
"make" -C /home/mikolai/TUD/Code/RIOT/sys/libc
"make" -C /home/mikolai/TUD/Code/RIOT/sys/luid
"make" -C /home/mikolai/TUD/Code/RIOT/sys/malloc_thread_safe
"make" -C /home/mikolai/TUD/Code/RIOT/sys/newlib_syscalls_default
"make" -C /home/mikolai/TUD/Code/RIOT/sys/preprocessor
"make" -C /home/mikolai/TUD/Code/RIOT/sys/stdio
"make" -C /home/mikolai/TUD/Code/RIOT/sys/stdio_uart
"make" -C /home/mikolai/TUD/Code/RIOT/sys/test_utils/interactive_sync
"make" -C /home/mikolai/TUD/Code/RIOT/sys/test_utils/print_stack_usage
"make" -C /home/mikolai/TUD/Code/RIOT/sys/tsrb
"make" -C /home/mikolai/TUD/Code/RIOT/sys/usb/usbus
"make" -C /home/mikolai/TUD/Code/RIOT/sys/usb/usbus/cdc/acm
"make" -C /home/mikolai/TUD/Code/RIOT/sys/usb_board_reset
   text    data     bss     dec     hex filename
  21252     128    4524   25904    6530 /home/mikolai/TUD/Code/RIOT/tests/sys/snprintf/bin/adafruit-feather-nrf52840-sense/tests_snprintf.elf
stty -F /dev/ttyACM0 raw ispeed 1200 ospeed 1200 cs8 -cstopb ignpar eol 255 eof 255
sleep 10
[INFO] uf2conv.py not found - fetching it from GitHub now
CC= CFLAGS= make -C /home/mikolai/TUD/Code/RIOT/dist/tools/uf2
[INFO] uf2conv.py successfully fetched!
SoftDevice version S140 6.1.1 found.
/home/mikolai/TUD/Code/RIOT/dist/tools/uf2/uf2conv.py -f 0xADA52840 /home/mikolai/TUD/Code/RIOT/tests/sys/snprintf/bin/adafruit-feather-nrf52840-sense/tests_snprintf.hex --base 0x26000
Converted to uf2, output size: 43008, start address: 0x26000
Flashing /media/mikolai/FTHRSNSBOOT (nRF52840-Feather-Sense)
Wrote 43008 bytes to /media/mikolai/FTHRSNSBOOT/NEW.UF2
sleep 2
/home/mikolai/TUD/Code/RIOT/dist/tools/pyterm/pyterm -p "/dev/ttyACM0" -b "115200" -ln "/tmp/pyterm-mikolai" -rn "2025-08-21_10.12.25-tests_snprintf-adafruit-feather-nrf52840-sense"  
Twisted not available, please install it if you want to use pyterm's JSON capabilities
2025-08-21 10:12:25,059 # Connect to serial port /dev/ttyACM0
Welcome to pyterm!
Type '/exit' to exit.
2025-08-21 10:12:26,061 # Help: Press s to start test, r to print it is ready
r
2025-08-21 10:12:27,896 # READY
s
2025-08-21 10:12:28,776 # START
2025-08-21 10:12:28,782 # main(): This is RIOT! (Version: 2025.10-devel-176-g18dac1-HEAD)
2025-08-21 10:12:28,785 # Testing snprintf() implementation...
2025-08-21 10:12:28,789 # Expected: "16045690984050070327"
2025-08-21 10:12:28,790 # Got:      ""
2025-08-21 10:12:28,793 # Expected: "deadbeefaffe1337"
2025-08-21 10:12:28,795 # Got:      ""
2025-08-21 10:12:28,798 # Expected: "0xdeadbeefaffe1337"
2025-08-21 10:12:28,800 # Got:      ""
2025-08-21 10:12:28,803 # Expected: "DEADBEEFAFFE1337"
2025-08-21 10:12:28,804 # Got:      ""
2025-08-21 10:12:28,808 # Expected: "0XDEADBEEFAFFE1337"
2025-08-21 10:12:28,809 # Got:      ""
2025-08-21 10:12:28,813 # Expected: "1572555756765777411467"
2025-08-21 10:12:28,814 # Got:      ""
2025-08-21 10:12:28,818 # Expected: "01572555756765777411467"
2025-08-21 10:12:28,820 # Got:      ""
2025-08-21 10:12:28,824 # Expected: "-9223372036854775807"
2025-08-21 10:12:28,825 # Got:      ""
2025-08-21 10:12:28,829 # Expected: "-9223372036854775807"
2025-08-21 10:12:28,830 # Got:      ""
2025-08-21 10:12:28,832 # TEST FAILED!
2025-08-21 10:12:28,839 # { "threads": [{ "name": "main", "stack_size": 1536, "stack_used": 440}]}

The test succeeds on current master. I'll try to debug later today.

I've also noted that https://github.com/RIOT-OS/RIOT/blob/18dac11c668e13bf428ac694083efc8ff177c866/tests/sys/snprintf/Makefile.board.dep already pulls in mpaland_printf iff newlib is selected. Should that be dropped within this PR?

core/lib/include/assert.h

mguetschow · 2025-08-21T08:05:14Z

features.yaml

+    help: newlib's stdio is not inherently thread-safe, but depends on the
+          reentrancy hooks to be provided for correct operation. When


What exactly is the reason we cannot provide such reentrancy hooks for newlib in RIOT?

We actually could provide reentrancy hooks (just nobody did so yet). There even is a stale PR somewhere for it.

But that will be a hard sell unless it is opt-in due to the large RAM requirements (per thread!) it has. (The stale PR adds an opt-in module for reentrant newlib, if I recall correctly.)

The main reason people enable reentrancy in newlib is to have malloc() / free() and friends thread-safe, which we already have with a much leaner implementation. The second reason is to have stdio thread-safe.

Reentrancy would also give you thread safety for a number of other standard C lib functions that operate on shared resources, but those are rarely ever used in the context of RIOT. (E.g. our VFS system fully bypasses the standard C lib anyway.) Once we have thread-safe stdio, there will be few use cases that would actually benefit from reentrancy, and some might rather rework the code than to pay the per-thread RAM overhead of newlib's reentrancy mechanism.

I see, thanks for the explanation! Would be great to have this noted down somewhere in the documentation for future reference, but I can't even find pages on neither newlib nor picolibc o.O

Would you be up for writing out some bits in a follow-up PR?

mguetschow · 2025-08-21T08:19:13Z

Ah sorry for the noise, that's probably the removed printf_long_long support.

Should we rather add printf_long_long to the test application or feature-gate test_int64?

maribu · 2025-08-21T08:40:57Z

Ah sorry for the noise, that's probably the removed printf_long_long support.

No, its IMO not noise. A test failing with default configuration is IMO a genuine bug :)

Should we rather add printf_long_long to the test application or feature-gate test_int64?

I did both. There might be users with tiny boards that can benefit from easily being able to disable printf_long_long in the test, but having the larger test coverage by default is probably better.

mguetschow

Thanks, I can confirm the test is passing now for both configurations on the hardware I tested with (nrf52840dk).

tests/sys/snprintf/Makefile

mguetschow · 2025-08-21T10:00:33Z

Please squash :)

mguetschow

sorry, two nits more

core/lib/include/assert.h

- `arch_64bit` depends on long long support (rather than just enabling it by default), as otherwise `%p` will not work correctly ==> add hard dependency - On 32 bit platforms support of printing long long is not almost free ==> do not enable long long on `arch_32bit` by default - The ESP SDK contains binary blobs that already link against newlib and mpaland-printf is not ABI compatible with newlib's stdio, it is only API compatible. ==> mark mpaland-printf as incompatible to ESP MCUs Co-authored-by: crasbe <[email protected]>

This introduces a new feature category to declare which bugs are present in a given build that we cannot fix in RIOT, but need to work around. These bugs may be silicon bugs, software bugs in ROM, software bugs in binary blobs needed for a platform, or bugs in the toolchain. These features behave the same way as `arch_%` features: Every provided bug is always used, so that we can inspect `FEATURES_USED` to check which bugs need to be worked around. Co-authored-by: crasbe <[email protected]>

This makes use of the new bug modeling to declare all platforms that can use newlib and have no reentrancy hooks as affected by the non-thread-safe stdio bug. (Which is every platform but ESP* and AVR, the former because the reentrancy hooks are provided, the latter because we do not and never will support newlib on them.) Building on that, the mpaland-printf package is used when newlib is used and the bug is present. This way we can rely on the stdio being thread-safe on every platform and not causing random crashes at run time. Co-authored-by: mguetschow <[email protected]>

This drops a workaround that initialized newlib's reentrancy structure on boot to reduce the chances of crashes when using the non-thread-safe (unless reentrancy hooks are provided) stdio implementation of newlib. Now that the newlib stdio implementation is only ever used if it is thread-safe, we no longer need a workaround that reduces the chance of crashes on concurrent use of stdio.

That way users can rely on elimation of dead code from the optimizer instead of having to use preprocessor conditionals when feature-gating assert implementations behind `!NDEBUG`. Co-authored-by: benpicco <[email protected]> Co-authored-by: crasbe <[email protected]> Co-authored-by: mguetschow <[email protected]>

This avoids inconsistent output when external code gets linked in that directly links against newlib's assert implementation (e.g. binary blobs or packages that do not add `core/lib/include` to the include paths). This also greatly benefits wrapping printf, as newlib's `__assert_func()` directly links to internal printf functions of newlib. Co-authored-by: crasbe <[email protected]>

This is required to support printing 64 bit numbers, as is done in the test. In addition that 64 bit formatting test is feature gated, so that users can simply comment out the `USEMODULE += printf_long_long` to fit tiny boards. Co-authored-by: mguetschow <[email protected]>

maribu · 2025-08-21T16:01:46Z

Yay 🎉

Thanks everyone for bearing with me on this one ❤️

mguetschow · 2025-08-22T08:19:56Z

This resulted in some failing nightly tests, as mpaland-printf prints %p without leading 0x. See the unmerged PR over at mpaland/printf#90.

As the printing format is indeed implementation-defined, the respective tests should probably be adapted to accept output both with and without 0x.

mguetschow · 2025-08-22T09:11:36Z

Another problem, at least on my machine:

$ make -C tests/sys/heap_cmd BOARD=nrf52840dk all
make: Entering directory '/home/mikolai/TUD/Code/RIOT/tests/sys/heap_cmd'
Building application "tests_heap_cmd" for "nrf52840dk" with CPU "nrf52".

"make" -C /home/mikolai/TUD/Code/RIOT/pkg/cmsis/ 
"make" -C /home/mikolai/TUD/Code/RIOT/pkg/mpaland-printf/ 
"make" -C /home/mikolai/.riot/pkg/mpaland-printf -f /home/mikolai/TUD/Code/RIOT/pkg/mpaland-printf/mpaland-printf.mk
"make" -C /home/mikolai/TUD/Code/RIOT/boards
"make" -C /home/mikolai/TUD/Code/RIOT/boards/common/init
"make" -C /home/mikolai/TUD/Code/RIOT/boards/nrf52840dk
"make" -C /home/mikolai/TUD/Code/RIOT/boards/common/nrf52xxxdk
"make" -C /home/mikolai/TUD/Code/RIOT/core
"make" -C /home/mikolai/TUD/Code/RIOT/core/lib
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/cortexm_common
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/cortexm_common/periph
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52/periph
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf52/vectors
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf5x_common
"make" -C /home/mikolai/TUD/Code/RIOT/cpu/nrf5x_common/periph
"make" -C /home/mikolai/TUD/Code/RIOT/drivers
"make" -C /home/mikolai/TUD/Code/RIOT/drivers/periph_common
"make" -C /home/mikolai/TUD/Code/RIOT/sys
"make" -C /home/mikolai/TUD/Code/RIOT/sys/auto_init
"make" -C /home/mikolai/TUD/Code/RIOT/sys/div
"make" -C /home/mikolai/TUD/Code/RIOT/sys/isrpipe
"make" -C /home/mikolai/TUD/Code/RIOT/sys/libc
"make" -C /home/mikolai/TUD/Code/RIOT/sys/malloc_thread_safe
"make" -C /home/mikolai/TUD/Code/RIOT/sys/newlib_syscalls_default
"make" -C /home/mikolai/TUD/Code/RIOT/sys/preprocessor
"make" -C /home/mikolai/TUD/Code/RIOT/sys/ps
"make" -C /home/mikolai/TUD/Code/RIOT/sys/shell
"make" -C /home/mikolai/TUD/Code/RIOT/sys/shell/cmds
"make" -C /home/mikolai/TUD/Code/RIOT/sys/stdio
"make" -C /home/mikolai/TUD/Code/RIOT/sys/stdio_uart
"make" -C /home/mikolai/TUD/Code/RIOT/sys/test_utils/interactive_sync
"make" -C /home/mikolai/TUD/Code/RIOT/sys/test_utils/print_stack_usage
"make" -C /home/mikolai/TUD/Code/RIOT/sys/tsrb
/usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: /usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/lib/thumb/v7e-m+fp/hard/libc_nano.a(lib_a-mstats.o): in function `_mstats_r':
/home/pere/src/newlib-salsa/build_nano/arm-none-eabi/thumb/v7e-m+fp/hard/newlib/libc/stdlib/../../../../../../../../newlib/libc/stdlib/mstats.c:121: undefined reference to `__wrap_fiprintf'
/usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: /home/pere/src/newlib-salsa/build_nano/arm-none-eabi/thumb/v7e-m+fp/hard/newlib/libc/stdlib/../../../../../../../../newlib/libc/stdlib/mstats.c:121: undefined reference to `__wrap_fiprintf'
/usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: /usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/lib/thumb/v7e-m+fp/hard/libc_nano.a(lib_a-nano-mallstatsr.o): in function `_malloc_stats_r':
/home/pere/src/newlib-salsa/build_nano/arm-none-eabi/thumb/v7e-m+fp/hard/newlib/libc/stdlib/../../../../../../../../newlib/libc/stdlib/nano-mallocr.c:523: undefined reference to `__wrap_fiprintf'
/usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: /home/pere/src/newlib-salsa/build_nano/arm-none-eabi/thumb/v7e-m+fp/hard/newlib/libc/stdlib/../../../../../../../../newlib/libc/stdlib/nano-mallocr.c:525: undefined reference to `__wrap_fiprintf'
/usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: /home/pere/src/newlib-salsa/build_nano/arm-none-eabi/thumb/v7e-m+fp/hard/newlib/libc/stdlib/../../../../../../../../newlib/libc/stdlib/nano-mallocr.c:527: undefined reference to `__wrap_fiprintf'
collect2: error: ld returned 1 exit status
make: *** [/home/mikolai/TUD/Code/RIOT/tests/sys/heap_cmd/../../../Makefile.include:755: /home/mikolai/TUD/Code/RIOT/tests/sys/heap_cmd/bin/nrf52840dk/tests_heap_cmd.elf] Error 1
make: Leaving directory '/home/mikolai/TUD/Code/RIOT/tests/sys/heap_cmd'

RIOT/pkg/mpaland-printf/Makefile.include

Lines 16 to 25 in c0cc617

    
           # By wrapping the newlib only `_printf_common` symbol to the undefined 
        
           # `__wrap__printf_common` symbol, linking will fail if any reference to 
        
           # `_printf_common` remains. Since `_printf_common` is only used by newlib's 
        
           # stdio, we will catch any instance of both mpaland-printf and newlib's stdio 
        
           # being both linked in. 
        
           LINKFLAGS += -Wl,-wrap=_printf_common 
        
           LINKFLAGS += -Wl,-wrap=_fprintf_r 
        
           LINKFLAGS += -Wl,-wrap=_printf_r 
        
           LINKFLAGS += -Wl,-wrap=iprintf 
        
           LINKFLAGS += -Wl,-wrap=fiprintf

suggests that this is failing on purpose, but I'm not sure what to make out of this. Apparently that also didn't happen on CI for some reason (maybe it was not properly rebuilding all modules for the test?).

mguetschow · 2025-08-22T09:23:38Z

As the printing format is indeed implementation-defined, the respective tests should probably be adapted to accept output both with and without 0x.

There we go: #21677

maribu requested review from kaspar030, gschorcht and bergzand as code owners April 25, 2025 06:14

crasbe added CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR Type: enhancement The issue suggests enhanceable parts / The PR enhances parts of the codebase / documentation labels Apr 25, 2025

crasbe reviewed Apr 25, 2025

View reviewed changes

Makefile.include Outdated Show resolved Hide resolved

maribu force-pushed the sys/newlib/fix-stdio branch from 9a524a2 to f058796 Compare April 25, 2025 11:09

github-actions bot added the Area: pkg Area: External package ports label Apr 25, 2025

maribu added State: waiting for other PR State: The PR requires another PR to be merged first and removed CI: full build disable CI build filter CI: no fast fail don't abort PR build after first error labels Apr 25, 2025

maribu force-pushed the sys/newlib/fix-stdio branch from f058796 to 31f945b Compare April 25, 2025 11:24

Enoch247 approved these changes Apr 26, 2025

View reviewed changes

maribu force-pushed the sys/newlib/fix-stdio branch 2 times, most recently from 473d6ca to 933acf3 Compare April 27, 2025 07:03

github-actions bot removed the Area: pkg Area: External package ports label Apr 27, 2025

github-actions bot added Platform: ARM Platform: This PR/issue effects ARM-based platforms Platform: RISC-V Platform: This PR/issue effects RISC-V-based platforms labels Aug 20, 2025

mguetschow requested changes Aug 21, 2025

View reviewed changes

maribu requested review from leandrolanzieri, aabadie and MichelRottleuthner as code owners August 21, 2025 08:37

github-actions bot added the Area: tests Area: tests and testing framework label Aug 21, 2025

mguetschow approved these changes Aug 21, 2025

View reviewed changes

tests/sys/snprintf/Makefile Outdated Show resolved Hide resolved

mguetschow reviewed Aug 21, 2025

View reviewed changes

core/lib/include/assert.h Outdated Show resolved Hide resolved

core/lib/include/assert.h Outdated Show resolved Hide resolved

maribu and others added 7 commits August 21, 2025 12:06

maribu force-pushed the sys/newlib/fix-stdio branch from ecdd4ea to 4a84f6f Compare August 21, 2025 10:08

maribu added this pull request to the merge queue Aug 21, 2025

Merged via the queue into RIOT-OS:master with commit e26321e Aug 21, 2025
25 checks passed

maribu deleted the sys/newlib/fix-stdio branch August 21, 2025 16:01

This was referenced Aug 22, 2025

makefiles/boot/riotboot: pass BUILD_DIR to recursive Make call #21676

Merged

tests: accept %p printing in uppercase and without leading 0x #21677

Merged

		help: newlib's stdio is not inherently thread-safe, but depends on the
		reentrancy hooks to be provided for correct operation. When

build system: use thread-safe stdio #21438

build system: use thread-safe stdio #21438

Uh oh!

Conversation

maribu commented Apr 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Contribution description

build system: introduce bug_% feature category

build system: use thread-safe stdio

sys/newlib: drop workaround for stdio

Testing procedure

Issues/PRs references

Uh oh!

Uh oh!

riot-ci commented Apr 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Murdock results

Artifacts

Uh oh!

crasbe commented Apr 25, 2025

Uh oh!

maribu commented Apr 25, 2025

Uh oh!

Enoch247 commented Apr 26, 2025

Uh oh!

mguetschow left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

mguetschow Aug 21, 2025

Choose a reason for hiding this comment

Uh oh!

maribu Aug 21, 2025

Choose a reason for hiding this comment

Uh oh!

mguetschow Aug 21, 2025

Choose a reason for hiding this comment

Uh oh!

mguetschow commented Aug 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

maribu commented Aug 21, 2025

Uh oh!

mguetschow left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

mguetschow commented Aug 21, 2025

Uh oh!

mguetschow left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

maribu commented Aug 21, 2025

Uh oh!

mguetschow commented Aug 22, 2025

Uh oh!

mguetschow commented Aug 22, 2025

Uh oh!

mguetschow commented Aug 22, 2025

Uh oh!

Uh oh!

maribu commented Apr 25, 2025 •

edited

Loading

build system: introduce `bug_%` feature category

riot-ci commented Apr 25, 2025 •

edited

Loading

mguetschow commented Aug 21, 2025 •

edited

Loading