Re: [Buildroot] [PATCH 1/6] configs/stm32f429_disco_xip: fix boot failure after uClibc bump
Julien Olivain via buildroot <[email protected]>
| Newsgroups | net.busybox.buildroot |
|---|---|
| Message-ID | <[email protected]> |
Hi Waldemar, On 24/07/2026 05:11, Waldemar Brodkorb wrote: > Hi Julien, > Julien Olivain via buildroot wrote, > >> Hi Dario, >> >> On 21/07/2026 20:42, Dario Binacchi wrote: >> > Since uClibc was bumped to 1.0.58 (c7fef3704c70, "package/uclibc: >> > bump to 1.0.58"), the board fails to boot: >> > >> > sh: out of memory >> > Kernel panic - not syncing: Attempted to kill init! >> > exitcode=0x00000100 >> > CPU: 0 PID: 1 Comm: init Not tainted 6.1.167 #2 >> > Hardware name: STM32 (Device Tree Support) >> > Function entered at [<0800ea1e>] from [<0800de7b>] >> > Function entered at [<0800de7b>] from [<080e4991>] >> > Function entered at [<080e4991>] from [<080e2d19>] >> > Function entered at [<080e2d19>] from [<080125eb>] >> > Function entered at [<080125eb>] from [<08012b09>] >> > Function entered at [<08012b09>] from [<08012b69>] >> > ---[ end Kernel panic - not syncing: Attempted to kill init! >> > exitcode=0x00000100 ]--- >> > >> > The STM32F429 has no MMU, and uClibc's standard malloc() implementation >> > returns NULL for allocations performed during early init on this target, >> > causing init to abort immediately. >> > >> > Fix this by switching to MALLOC_SIMPLE, a plain mmap-based allocator >> > meant for small/noMMU systems, via a uClibc config fragment. The >> > fragment is placed under board/stmicroelectronics/common/stm32f4xx/, >> > rather than under the board-specific directory, so that it can be >> > shared with the STM32F469, another Cortex-M4/noMMU board affected by >> > the same issue, which will be fixed in a subsequent patch. >> >> For info, the issue may have been introduced by uclibc commit: >> https://github.com/wbx-github/uclibc-ng/commit/487af14988c3c923aa661f204642597cff356a83 >> >> There is an upstream fix, not released yet, which seems to fix the >> issue: >> https://github.com/wbx-github/uclibc-ng/commit/7730f0d05a51cd89a7123bd696e4638c3c035d7b >> >> You could possibly drop this malloc config change after the next >> uclibc >> bump. >> >> > Signed-off-by: Dario Binacchi <[email protected]> >> >> In the meantime, I applied this series to master, thanks. > > I sent following patch to avoid any breakage before the next > release: > https://patchwork.ozlabs.org/project/buildroot/patch/[email protected]/ > > But it was not accepted, yet. Xtensa and M68k noMMU is also > affected. Thanks for pointing that out. Dario, I have merged the uclibc fix in: https://gitlab.com/buildroot.org/buildroot/-/commit/35d3927e85645d0dbfe608bd5954bfa6704200e8 So you should be able to revert those malloc uclibc config changes. > best regards > Waldemar Best regards, Julien. _______________________________________________ buildroot mailing list [email protected] https://lists.buildroot.org/mailman/listinfo/buildroot