devel October 2023

devel@uclibc-ng.org

10 participants
8 discussions

[PATCH] setjmp.h: Fix C++ build and avoid duplicate throw declaration

by Marcus Haehnel

From: Marcus Hähnel <marcus.haehnel(a)kernkonzept.com> Commit 21cbb6fe ("unistd.h: put getppid under XOPEN2K8") introduced a new __THROWNL specification for non-leaf throws. It also made use of these in the setjmp.h header. The functions that use this specification (longjmp and siglongjmp) have their extern __libc__* equivalent definition prototype specified using __typeof__ for the internal function signatures. For C++ this copies the throw() specifier, since this is part of the type for C++. The attribute in C is not. This commit explicitly types out the signature for the two functions to be compatible between the C++ and C worlds. An alternative would be to keep the __typeof__ declaration and use the copy attribute. Then the __THROWNL part could be thrown out since the C attribute would be copied and the C++ exception specifier would be part of the signature from __type__. However, since the copy attribute is not supported for all compilers supported by uclibc-ng this is not viable at this time. --- include/setjmp.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/include/setjmp.h b/include/setjmp.h index 27cac9500..41337aa9f 100644 --- a/include/setjmp.h +++ b/include/setjmp.h @@ -108,8 +108,10 @@ __END_DECLS #ifdef _LIBC extern void __longjmp(__jmp_buf __env, int __val) __THROWNL attribute_noreturn; libc_hidden_proto(__longjmp) -extern __typeof(longjmp) __libc_longjmp __THROWNL attribute_noreturn; -extern __typeof(siglongjmp) __libc_siglongjmp __THROWNL attribute_noreturn; +extern void __libc_longjmp(struct __jmp_buf_tag __env[1], int __val) + __THROWNL attribute_noreturn; +extern void __libc_siglongjmp(sigjmp_buf __env, int __val) + __THROWNL attribute_noreturn; extern void _longjmp_unwind(jmp_buf __env, int __val); libc_hidden_proto(_longjmp_unwind) extern int __sigjmp_save(sigjmp_buf __env, int __savemask) attribute_hidden; -- 2.42.0

1 year

[PATCH] arm: elf-fdpic.h: avoid void pointer subtraction

by Ben Wolsieffer

elf-fdpic.h is included by link.h. When a C++ program includes <link.h>, we get the following build failure: <...>/usr/include/bits/elf-fdpic.h: In function ‘void* __reloc_pointer(void*, const elf32_fdpic_loadmap*)’: <...>/usr/include/bits/elf-fdpic.h:94:54: error: invalid use of ‘void’ 94 | unsigned long offset = p - (void*)map->segs[c].p_vaddr; | ^~~~~~~ void pointer addition and subtraction is not allowed in C++ as it has undetermined size, however in C with language extension it is possible because sizeof void is treated as one byte. This patch was previously applied to Blackfin, FR-V and C6x, but not ARM. Signed-off-by: Ben Wolsieffer <ben.wolsieffer(a)hefring.com> --- libc/sysdeps/linux/arm/bits/elf-fdpic.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libc/sysdeps/linux/arm/bits/elf-fdpic.h b/libc/sysdeps/linux/arm/bits/elf-fdpic.h index 3d6db54af..f2ef9aeca 100644 --- a/libc/sysdeps/linux/arm/bits/elf-fdpic.h +++ b/libc/sysdeps/linux/arm/bits/elf-fdpic.h @@ -91,7 +91,7 @@ __reloc_pointer (void *p, /* This should be computed as part of the pointer comparison above, but we want to use the carry in the comparison, so we can't convert it to an integer type beforehand. */ - unsigned long offset = p - (void*)map->segs[c].p_vaddr; + unsigned long offset = (char*)p - (char*)map->segs[c].p_vaddr; /* We only check for one-past-the-end for the last segment, assumed to be the data segment, because other cases are ambiguous in the absence of padding between segments, and -- 2.42.0

1 year

Possible read-beyond string parameter in fnmatch()

by Frank Mehnert

Hi all, in certain cases, fnmatch(pattern, string, flags) reads beyond the end of pattern. This can be triggered by parameters like this: fnmatch(""[A-Z[.", "F", 0); The corresponding code can be found here: https://cgit.uclibc-ng.org/cgi/cgit/uclibc-ng.git/tree/libc/misc/fnmatch/fn… After line 920 is executed, p points to '\0' (the end of the pattern). Then, in line 923, p is unconditionally increased again and the value _after_ the end of the pattern is read (to find out if the pattern has ended). Suggested fix: Just remove line 920. Kind regards, Frank -- Dr.-Ing. Frank Mehnert, frank.mehnert(a)kernkonzept.com, +49-351-41 883 224 Kernkonzept GmbH. Sitz: Dresden. Amtsgericht Dresden, HRB 31129. Geschäftsführer: Dr.-Ing. Michael Hohmuth

1 year

[PATCH] Fix -Warray-parameter warning for __sigsetjmp

by Pavel Kozlov

From: Pavel Kozlov <pavel.kozlov(a)synopsys.com> Fix the [-Warray-parameter=] warning for __sigsetjmp generated by GCC 11 and later GCC versions: | | warning: argument 1 of type 'struct __jmp_buf_tag *' declared as a pointer [-Warray-parameter=] | extern int __sigsetjmp (struct __jmp_buf_tag *__env, int __savemask) __THROWNL; | ... | note: previously declared as an array 'struct __jmp_buf_tag[1]' | extern int __sigsetjmp (struct __jmp_buf_tag __env[1], int __savemask) | Use the same fix as in glibc. The fix is to move the struct __jmp_buf_tag definition to a separate bits/ header so it can be included in pthread.h, to allow to use an array (as in setjmp.h) rather than a pointer in the declaration. Signed-off-by: Pavel Kozlov <pavel.kozlov(a)synopsys.com> --- include/setjmp.h | 16 +-------- libc/sysdeps/linux/common/bits/jmp_buf_tag.h | 37 ++++++++++++++++++++ libpthread/nptl/sysdeps/pthread/pthread.h | 4 +-- 3 files changed, 40 insertions(+), 17 deletions(-) create mode 100644 libc/sysdeps/linux/common/bits/jmp_buf_tag.h diff --git a/include/setjmp.h b/include/setjmp.h index 27cac9500a15..e6f437b93ad9 100644 --- a/include/setjmp.h +++ b/include/setjmp.h @@ -27,21 +27,7 @@ __BEGIN_DECLS #include <bits/setjmp.h> /* Get `__jmp_buf'. */ -#include <bits/sigset.h> /* Get `__sigset_t'. */ - - -/* Calling environment, plus possibly a saved signal mask. */ -struct __jmp_buf_tag - { - /* NOTE: The machine-dependent definitions of `__sigsetjmp' - assume that a `jmp_buf' begins with a `__jmp_buf' and that - `__mask_was_saved' follows it. Do not move these members - or add others before it. */ - __jmp_buf __jmpbuf; /* Calling environment. */ - int __mask_was_saved; /* Saved the signal mask? */ - __sigset_t __saved_mask; /* Saved signal mask. */ - }; - +#include <bits/jmp_buf_tag.h> __BEGIN_NAMESPACE_STD diff --git a/libc/sysdeps/linux/common/bits/jmp_buf_tag.h b/libc/sysdeps/linux/common/bits/jmp_buf_tag.h new file mode 100644 index 000000000000..4e1a4310a7e2 --- /dev/null +++ b/libc/sysdeps/linux/common/bits/jmp_buf_tag.h @@ -0,0 +1,37 @@ +/* Define struct __jmp_buf_tag. + Copyright (C) 1991-2023 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef __jmp_buf_tag_defined +#define __jmp_buf_tag_defined 1 + +#include <bits/setjmp.h> +#include <bits/sigset.h> + +/* Calling environment, plus possibly a saved signal mask. */ +struct __jmp_buf_tag + { + /* NOTE: The machine-dependent definitions of `__sigsetjmp' + assume that a `jmp_buf' begins with a `__jmp_buf' and that + `__mask_was_saved' follows it. Do not move these members + or add others before it. */ + __jmp_buf __jmpbuf; /* Calling environment. */ + int __mask_was_saved; /* Saved the signal mask? */ + __sigset_t __saved_mask; /* Saved signal mask. */ + }; + +#endif \ No newline at end of file diff --git a/libpthread/nptl/sysdeps/pthread/pthread.h b/libpthread/nptl/sysdeps/pthread/pthread.h index 1fba7fca63c9..531e173220d2 100644 --- a/libpthread/nptl/sysdeps/pthread/pthread.h +++ b/libpthread/nptl/sysdeps/pthread/pthread.h @@ -28,6 +28,7 @@ #include <signal.h> #include <bits/pthreadtypes.h> #include <bits/setjmp.h> +#include <bits/jmp_buf_tag.h> #include <bits/wordsize.h> #if defined _LIBC && ( defined IS_IN_libc || !defined NOT_IN_libc ) #include <bits/uClibc_pthread.h> @@ -726,8 +727,7 @@ extern void __pthread_unwind_next (__pthread_unwind_buf_t *__buf) #endif /* Function used in the macros. */ -struct __jmp_buf_tag; -extern int __sigsetjmp (struct __jmp_buf_tag *__env, int __savemask) __THROWNL; +extern int __sigsetjmp (struct __jmp_buf_tag __env[1], int __savemask) __THROWNL; /* Mutex handling. */ -- 2.25.1

1 year

[PATCH] setrlimit/getrlimit: fix prlimit64 syscall use for 32-bit CPUs

by Pavel Kozlov

From: Pavel Kozlov <pavel.kozlov(a)synopsys.com> Commit 95e38b37 ("add support for systems without legacy setrlimit/getrlimit syscalls") has added use of the prlimit64 syscall in getrlimit and setrlimit functions. This change causes memory corruption on getrlimit call for 32-bit CPUs like ARC, as ARC doesn't have ugetrlimit syscall and uses prlimit64. Also, setrlimit has been broken by prlimit64 call on 32-bit CPUs like, i386, ARM, ARC. For the prlimit64 syscall the kernel expects an rlimit struct with 64-bit fields, but on 32-bit CPUs without _FILE_OFFSET_BITS=64 the struct rlimit has 32-bit fields. Add safe implementations of getrlimit, setrlimit, prlimit for 32-bit CPUs with a local struct rlimit64 variable for use in the prlimit64 syscall. For 64-bit CPUs and configurations with _FILE_OFFSET_BITS=64 use getrlimit, setrlimit, prlimit as aliases to getrlimit64, setrlimit64 and prlimit64. Add a new function prlimit64. Tested on aarch64, arm, i386, arc. Fixes: 95e38b37 ("add support for systems without legacy setrlimit/getrlimit syscalls") Signed-off-by: Pavel Kozlov <pavel.kozlov(a)synopsys.com> --- include/sys/resource.h | 4 ++ libc/sysdeps/linux/common/getrlimit.c | 55 ++++++++++++++++++------- libc/sysdeps/linux/common/getrlimit64.c | 26 ++++++++++-- libc/sysdeps/linux/common/prlimit.c | 53 +++++++++++++++++++++--- libc/sysdeps/linux/common/prlimit64.c | 36 ++++++++++++++++ libc/sysdeps/linux/common/setrlimit.c | 41 +++++++++++------- libc/sysdeps/linux/common/setrlimit64.c | 24 +++++++++-- 7 files changed, 197 insertions(+), 42 deletions(-) create mode 100644 libc/sysdeps/linux/common/prlimit64.c diff --git a/include/sys/resource.h b/include/sys/resource.h index 00c63ff0f1e5..e9fac2c656bf 100644 --- a/include/sys/resource.h +++ b/include/sys/resource.h @@ -106,6 +106,10 @@ libc_hidden_proto(setpriority) extern int prlimit (__pid_t __pid, enum __rlimit_resource __resource, const struct rlimit *__new_limit, struct rlimit *__old_limit) __THROW; + +extern int prlimit64 (__pid_t __pid, enum __rlimit_resource __resource, + const struct rlimit64 *__new_limit, + struct rlimit64 *__old_limit) __THROW; #endif __END_DECLS diff --git a/libc/sysdeps/linux/common/getrlimit.c b/libc/sysdeps/linux/common/getrlimit.c index ad3f4a0e494d..46726fcbd94f 100644 --- a/libc/sysdeps/linux/common/getrlimit.c +++ b/libc/sysdeps/linux/common/getrlimit.c @@ -24,21 +24,53 @@ int getrlimit(__rlimit_resource_t resource, struct rlimit *rlimits) { return __syscall_ugetrlimit(resource, rlimits); } +libc_hidden_def(getrlimit) -#else - -# if !defined(__UCLIBC_HANDLE_OLDER_RLIMIT__) +#elif defined(__NR_prlimit64) +/* Use prlimit64 if present, the prlimit64 syscall is free from a back + compatibility stuff for an old getrlimit */ -# if defined(__NR_prlimit64) +# if __WORDSIZE == 32 && !defined(__USE_FILE_OFFSET64) +/* If struct rlimit has 64-bit fields (if __WORDSIZE == 64 or __USE_FILE_OFFSET64 + is defined), then use getrlimit as an alias to getrlimit64, see getrlimit64.c */ int getrlimit(__rlimit_resource_t resource, struct rlimit *rlimits) { - return INLINE_SYSCALL (prlimit64, 4, 0, resource, NULL, rlimits); + struct rlimit64 rlimits64; + int res = INLINE_SYSCALL (prlimit64, 4, 0, resource, NULL, &rlimits64); + + if (res == 0) { + /* If the syscall succeeds but the values do not fit into a + rlimit structure set EOVERFLOW errno and retrun -1. */ + rlimits->rlim_cur = rlimits64.rlim_cur; + if (rlimits64.rlim_cur != rlimits->rlim_cur) { + if (rlimits64.rlim_cur != RLIM64_INFINITY) { + __set_errno(EOVERFLOW); + return -1; + } + rlimits->rlim_cur = RLIM_INFINITY; + } + + rlimits->rlim_max = rlimits64.rlim_max; + if (rlimits64.rlim_max != rlimits->rlim_max) { + if (rlimits64.rlim_max != RLIM64_INFINITY) { + __set_errno(EOVERFLOW); + return -1; + } + rlimits->rlim_max = RLIM_INFINITY; + } + } + return res; } -# else +libc_hidden_def(getrlimit) +# endif + +#else + +# if !defined(__UCLIBC_HANDLE_OLDER_RLIMIT__) + /* We don't need to wrap getrlimit() */ _syscall2(int, getrlimit, __rlimit_resource_t, resource, struct rlimit *, rlim) -# endif # else @@ -51,11 +83,7 @@ int getrlimit(__rlimit_resource_t resource, struct rlimit *rlimits) { int result; -# if defined(__NR_prlimit64) - result = INLINE_SYSCALL (prlimit64, 4, 0, resource, NULL, rlimits); -# else result = __syscall_getrlimit(resource, rlimits); -# endif if (result == -1) return result; @@ -69,9 +97,6 @@ int getrlimit(__rlimit_resource_t resource, struct rlimit *rlimits) return result; } # endif -#endif -libc_hidden_def(getrlimit) -#if __WORDSIZE == 64 -strong_alias_untyped(getrlimit, getrlimit64) +libc_hidden_def(getrlimit) #endif diff --git a/libc/sysdeps/linux/common/getrlimit64.c b/libc/sysdeps/linux/common/getrlimit64.c index be98098a1ee0..47f1410fb3d9 100644 --- a/libc/sysdeps/linux/common/getrlimit64.c +++ b/libc/sysdeps/linux/common/getrlimit64.c @@ -17,14 +17,31 @@ #include <_lfs_64.h> #include <bits/wordsize.h> +#include <sys/resource.h> +#include <sys/syscall.h> +#include <stddef.h> // needed for NULL to be defined -/* the regular getrlimit will work just fine for 64bit users */ -#if __WORDSIZE == 32 -# include <sys/resource.h> +#if defined(__NR_prlimit64) + +/* the regular prlimit64 will work just fine for 64-bit users */ +int getrlimit64 (__rlimit_resource_t resource, struct rlimit64 *rlimits) +{ + return INLINE_SYSCALL (prlimit64, 4, 0, resource, NULL, rlimits); +} + +# if !defined(__NR_ugetrlimit) && (__WORDSIZE == 64 || defined (__USE_FILE_OFFSET64)) +/* If getrlimit is not implemented through the __NR_ugetrlimit and size of + rlimit_t == rlimit64_t then use getrlimit as an alias to getrlimit64 */ +strong_alias_untyped(getrlimit64, getrlimit) +libc_hidden_def(getrlimit) +# endif + +#else /* Put the soft and hard limits for RESOURCE in *RLIMITS. - Returns 0 if successful, -1 if not (and sets errno). */ + Returns 0 if successful, -1 if not (and sets errno). + The regular getrlimit will work just fine for 64-bit users */ int getrlimit64 (__rlimit_resource_t resource, struct rlimit64 *rlimits) { struct rlimit rlimits32; @@ -44,3 +61,4 @@ int getrlimit64 (__rlimit_resource_t resource, struct rlimit64 *rlimits) return 0; } #endif + diff --git a/libc/sysdeps/linux/common/prlimit.c b/libc/sysdeps/linux/common/prlimit.c index f44dc166492e..f59ade3a379c 100644 --- a/libc/sysdeps/linux/common/prlimit.c +++ b/libc/sysdeps/linux/common/prlimit.c @@ -17,14 +17,57 @@ #include <sys/resource.h> #include <sysdep.h> -#include <bits/kernel-features.h> +#include <stddef.h> // needed for NULL to be defined -#if defined __ASSUME_PRLIMIT64 +#if defined(__NR_prlimit64) && __WORDSIZE == 32 && !defined(__USE_FILE_OFFSET64) int prlimit (__pid_t pid, enum __rlimit_resource resource, - const struct rlimit *new_rlimit, struct rlimit *old_rlimit) + const struct rlimit *new_rlimit, struct rlimit *old_rlimit) { - return INLINE_SYSCALL (prlimit64, 4, pid, resource, new_rlimit, - old_rlimit); + struct rlimit64 new_rlimit64; + struct rlimit64 old_rlimit64; + int res; + + if (new_rlimit != NULL) { + if (new_rlimit->rlim_cur == RLIM_INFINITY) + new_rlimit64.rlim_cur = RLIM64_INFINITY; + else + new_rlimit64.rlim_cur = new_rlimit->rlim_cur; + if (new_rlimit->rlim_max == RLIM_INFINITY) + new_rlimit64.rlim_max = RLIM64_INFINITY; + else + new_rlimit64.rlim_max = new_rlimit->rlim_max; + } + + res = INLINE_SYSCALL (prlimit64, 4, pid, resource, &new_rlimit64, + &old_rlimit64); + + if (res == 0 && old_rlimit != NULL) { + /* If the syscall succeeds but the values do not fit into a + rlimit structure set EOVERFLOW errno and retrun -1. + With current Linux implementation of the prlimit64 syscall, + overflow can't happen. An extra condition has been added to get + the same behavior as in glibc for future potential overflows. */ + old_rlimit->rlim_cur = old_rlimit64.rlim_cur; + if (old_rlimit64.rlim_cur != old_rlimit->rlim_cur) { + if (new_rlimit == NULL && + old_rlimit64.rlim_cur != RLIM64_INFINITY) { + __set_errno(EOVERFLOW); + return -1; + } + old_rlimit->rlim_cur = RLIM_INFINITY; + } + old_rlimit->rlim_max = old_rlimit64.rlim_max; + if (old_rlimit64.rlim_max != old_rlimit->rlim_max) { + if (new_rlimit == NULL && + old_rlimit64.rlim_max != RLIM64_INFINITY) { + __set_errno(EOVERFLOW); + return -1; + } + old_rlimit->rlim_cur = RLIM_INFINITY; + } + } + + return res; } #endif diff --git a/libc/sysdeps/linux/common/prlimit64.c b/libc/sysdeps/linux/common/prlimit64.c new file mode 100644 index 000000000000..6f57b939eeca --- /dev/null +++ b/libc/sysdeps/linux/common/prlimit64.c @@ -0,0 +1,36 @@ +/* Copyright (C) 2023 uClibc-ng + * An prlimit64() - get/set resource limits Linux specific syscall. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Library General Public License for more details. + * + * You should have received a copy of the GNU Library General Public + * License along with this library; if not, see + * <http://www.gnu.org/licenses/>. + */ + +#include <sys/resource.h> +#include <sysdep.h> + +#if defined(__NR_prlimit64) + +int +prlimit64 (__pid_t pid, enum __rlimit_resource resource, + const struct rlimit64 *new_rlimit, struct rlimit64 *old_rlimit) +{ + return INLINE_SYSCALL (prlimit64, 4, pid, resource, new_rlimit, + old_rlimit); +} + +# if __WORDSIZE == 64 || defined (__USE_FILE_OFFSET64) +strong_alias_untyped(prlimit64, prlimit) +# endif + +#endif \ No newline at end of file diff --git a/libc/sysdeps/linux/common/setrlimit.c b/libc/sysdeps/linux/common/setrlimit.c index 8381afc617fd..9c6707235f3e 100644 --- a/libc/sysdeps/linux/common/setrlimit.c +++ b/libc/sysdeps/linux/common/setrlimit.c @@ -23,21 +23,41 @@ int setrlimit(__rlimit_resource_t resource, const struct rlimit *rlimits) { return __syscall_usetrlimit(resource, rlimits); } +libc_hidden_def(setrlimit) -#else +#elif defined(__NR_prlimit64) -# if !defined(__UCLIBC_HANDLE_OLDER_RLIMIT__) +/* Use prlimit64 if present, the prlimit64 syscall is free from a back + compatibility stuff for setrlimit */ -# if defined(__NR_prlimit64) + # if __WORDSIZE == 32 && !defined(__USE_FILE_OFFSET64) +/* If struct rlimit has 64-bit fields (if __WORDSIZE == 64 or __USE_FILE_OFFSET64 + is defined), then use setrlimit as an alias to setrlimit64, see setrlimit64.c */ int setrlimit(__rlimit_resource_t resource, const struct rlimit *rlimits) { - return INLINE_SYSCALL (prlimit64, 4, 0, resource, rlimits, NULL); + struct rlimit64 rlimits64; + + if (rlimits->rlim_cur == RLIM_INFINITY) + rlimits64.rlim_cur = RLIM64_INFINITY; + else + rlimits64.rlim_cur = rlimits->rlim_cur; + if (rlimits->rlim_max == RLIM_INFINITY) + rlimits64.rlim_max = RLIM64_INFINITY; + else + rlimits64.rlim_max = rlimits->rlim_max; + + return INLINE_SYSCALL (prlimit64, 4, 0, resource, &rlimits64, NULL); } -# else +libc_hidden_def(setrlimit) +# endif + +#else + +# if !defined(__UCLIBC_HANDLE_OLDER_RLIMIT__) + /* We don't need to wrap setrlimit() */ _syscall2(int, setrlimit, __rlimit_resource_t, resource, const struct rlimit *, rlim) -# endif # else @@ -66,16 +86,9 @@ int setrlimit(__rlimit_resource_t resource, const struct rlimit *rlimits) RLIM_INFINITY >> 1); rlimits_small.rlim_max = MIN((unsigned long int) rlimits->rlim_max, RLIM_INFINITY >> 1); -# if defined(__NR_prlimit64) - return INLINE_SYSCALL (prlimit64, 4, 0, resource, &rlimits_small, NULL); -# else return __syscall_setrlimit(resource, &rlimits_small); -# endif } # endif -#endif -libc_hidden_def(setrlimit) -#if __WORDSIZE == 64 -strong_alias_untyped(setrlimit, setrlimit64) +libc_hidden_def(setrlimit) #endif diff --git a/libc/sysdeps/linux/common/setrlimit64.c b/libc/sysdeps/linux/common/setrlimit64.c index fee14f4ad4b2..3446c58fee12 100644 --- a/libc/sysdeps/linux/common/setrlimit64.c +++ b/libc/sysdeps/linux/common/setrlimit64.c @@ -17,15 +17,31 @@ #include <_lfs_64.h> #include <bits/wordsize.h> +#include <sys/resource.h> +#include <sys/syscall.h> +#include <stddef.h> // needed for NULL to be defined -/* the regular setrlimit will work just fine for 64bit users */ -#if __WORDSIZE == 32 -# include <sys/resource.h> +#if defined(__NR_prlimit64) + +int setrlimit64 (__rlimit_resource_t resource, const struct rlimit64 *rlimits) +{ + return INLINE_SYSCALL (prlimit64, 4, 0, resource, rlimits, NULL); +} + +# if !defined(__NR_usetrlimit) && (__WORDSIZE == 64 || defined (__USE_FILE_OFFSET64)) +/* If setrlimit is not implemented through the __NR_usetrlimit and size of + rlimit_t == rlimit64_t then use setrlimit as an alias to setrlimit64 */ +strong_alias_untyped(setrlimit64, setrlimit) +libc_hidden_def(setrlimit) +# endif + +#else /* Set the soft and hard limits for RESOURCE to *RLIMITS. Only the super-user can increase hard limits. - Return 0 if successful, -1 if not (and sets errno). */ + Return 0 if successful, -1 if not (and sets errno). + The regular setrlimit will work just fine for 64bit users */ int setrlimit64 (__rlimit_resource_t resource, const struct rlimit64 *rlimits) { struct rlimit rlimits32; -- 2.25.1

1 year

[PATCH 1/4] fstatat64: define it as a wrapper of statx if the kernel does not support fstatat64 syscall

by yann＠sionneau.net

From: Yann Sionneau <ysionneau(a)kalray.eu> Define fstatat64 as a wrapper of statx if the kernel does not support fstatat64 syscall This is the case for non-legacy architectures that don't define __ARCH_WANT_NEW_STAT in their linux arch/xxx/include/asm/unistd.h Signed-off-by: Yann Sionneau <ysionneau(a)kalray.eu> --- libc/sysdeps/linux/common/fstatat64.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/libc/sysdeps/linux/common/fstatat64.c b/libc/sysdeps/linux/common/fstatat64.c index 711521a6a..836ed4114 100644 --- a/libc/sysdeps/linux/common/fstatat64.c +++ b/libc/sysdeps/linux/common/fstatat64.c @@ -43,5 +43,28 @@ int fstatat64(int fd, const char *file, struct stat64 *buf, int flag) } libc_hidden_def(fstatat64) #else + +#if defined(__NR_statx) && defined(__UCLIBC_HAVE_STATX__) +# include <sys/stat.h> +# include <statx_cp.h> +# include <fcntl.h> // for AT_NO_AUTOMOUNT + +int fstatat64(int fd, const char *file, struct stat64 *buf, int flag) +{ + struct statx tmp; + + int r = INLINE_SYSCALL(statx, 5, fd, file, AT_NO_AUTOMOUNT | flag, + STATX_BASIC_STATS, &tmp); + + if (r != 0) + return r; + + __cp_stat_statx ((struct stat *)buf, &tmp); + + return 0; +} +libc_hidden_def(fstatat64) +#endif + /* should add emulation with fstat64() and /proc/self/fd/ ... */ #endif -- 2.42.0

1 year

[PATCH] libpthread/nptl: make default stack size configurable

by Ben Wolsieffer

Threads currently have 2-4 MiB stacks by default (depending on the platform). This is fine on MMU platforms, where this stack space is not actually allocated until it is used, but tends to waste a large amount of memory on no-MMU platforms. This patch adds a PTHREADS_STACK_DEFAULT_SIZE Kconfig option that allows the user to override the default stack size at build time. This allows the user to select a reasonable default stack size for the software that runs on their system, and avoids the need to patch every package to add calls to pthread_attr_setstacksize(). An alternative to this patch would be to change the hardcoded default stack size on no-MMU platforms, but it is difficult to choose an appropriate value because the minimum required stack depends on the software in use. This would also be a breaking change. It may also be desirable to completely move the default stack size configuration to Kconfig, with conditional defaults for each platform. This would avoid special casing zero as the default value. Signed-off-by: Ben Wolsieffer <ben.wolsieffer(a)hefring.com> --- extra/Configs/Config.in | 8 ++++++++ libpthread/nptl/init.c | 20 +++++++++++++------- 2 files changed, 21 insertions(+), 7 deletions(-) diff --git a/extra/Configs/Config.in b/extra/Configs/Config.in index 6bbb6f572..ecb55ef3f 100644 --- a/extra/Configs/Config.in +++ b/extra/Configs/Config.in @@ -628,6 +628,14 @@ config PTHREADS_DEBUG_SUPPORT If you are doing development and want to debug applications using uClibc's pthread library, answer Y. Otherwise, answer N. +config PTHREADS_STACK_DEFAULT_SIZE + int "Default thread stack size" + default 0 + help + Set the default thread stack size. If set to zero, an architecture + specific default is used. This option is useful on MMU-less systems + where the stack size is fixed and the default stack size may be + excessively large and waste memory. config UCLIBC_HAS_SYSLOG bool "Syslog support" diff --git a/libpthread/nptl/init.c b/libpthread/nptl/init.c index 5d25ded7d..f5c144e7b 100644 --- a/libpthread/nptl/init.c +++ b/libpthread/nptl/init.c @@ -35,6 +35,12 @@ #include <bits/kernel-features.h> #include <stdio.h> +#if __PTHREADS_STACK_DEFAULT_SIZE__ +#define PTHREAD_STACK_DEFAULT_SIZE __PTHREADS_STACK_DEFAULT_SIZE__ +#else +#define PTHREAD_STACK_DEFAULT_SIZE ARCH_STACK_DEFAULT_SIZE +#endif + /* Size and alignment of static TLS block. */ size_t __static_tls_size; size_t __static_tls_align_m1; @@ -278,17 +284,17 @@ __pthread_initialize_minimal_internal (void) struct rlimit limit; if (getrlimit (RLIMIT_STACK, &limit) != 0 || limit.rlim_cur == RLIM_INFINITY) - /* The system limit is not usable. Use an architecture-specific - default. */ - limit.rlim_cur = ARCH_STACK_DEFAULT_SIZE; - else if (limit.rlim_cur < PTHREAD_STACK_MIN) + /* The system limit is not usable. Use a user-specified or + architecture-specific default. */ + limit.rlim_cur = PTHREAD_STACK_DEFAULT_SIZE; + if (limit.rlim_cur < PTHREAD_STACK_MIN) /* The system limit is unusably small. Use the minimal size acceptable. */ limit.rlim_cur = PTHREAD_STACK_MIN; - /* Do not exceed architecture specific default */ - if (limit.rlim_cur > ARCH_STACK_DEFAULT_SIZE) - limit.rlim_cur = ARCH_STACK_DEFAULT_SIZE; + /* Do not exceed the user-specified or architecture-specific default */ + if (limit.rlim_cur > PTHREAD_STACK_DEFAULT_SIZE) + limit.rlim_cur = PTHREAD_STACK_DEFAULT_SIZE; /* Make sure it meets the minimum size that allocate_stack (allocatestack.c) will demand, which depends on the page size. */ -- 2.42.0

1 year

Issue: Possible buffer overflow in _vfprintf.c (out-of-bounds access)

by Diego Dias

Dear uclibc-ng developers, We have run a static analysis tool (Klocwork) in uclibc and one of its checkers (ABV.GENERAL) indicates a potential buffer overflow in uclibc-ng/src/master/libc/stdio/_vfprintf.c:1045 The problem occurs as an out-of-bounds access to array 'argtype', which is a member of 'ppfs_t'. This array has length 'MAX_ARGS'. According to the static analysis tool, the array can be accessed using index 'n' of value '-1' and '9..254' in the conditional shown below: // File: uclibc-ng/src/master/libc/stdio/_vfprintf.c:1045 if (_is_equal_or_bigger_arg(ppfs->argtype[n], argtype[i])) { ppfs->argtype[n] = argtype[i]; } Triggering an out-of-bounds access for 'n=-1' is relatively simply when using printf or similar functions (e.g. vfprintf). Such out-of-bounds access occurs when positional arguments are specified, as in the following statement: printf("%1$s", "Hello world!"); Although Klocwork claims that the array might be accessed using indexes '9..254', we were not able to trigger an out-of-bounds access for indexes in this range. Kind regards, Diego Dias -- Diego M. Dias, Systems Verification Engineer at Kernkonzept, diego.dias(a)kernkonzept.com Phone: +49 351 41883231 Kernkonzept GmbH at Dresden, Germany, HRB 31129, CEO Dr.-Ing. Michael Hohmuth

1 year

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

devel October 2023