This is the implementation of parsing the SFrame V3 stack trace information from an .sframe section in an ELF file. It's a continuation of Josh's and Steve's work that can be found here:
https://lore.kernel.org/all/[email protected]/ https://lore.kernel.org/all/[email protected]/ Currently the only way to get a user space stack trace from a stack walk (and not just copying large amount of user stack into the kernel ring buffer) is to use frame pointers. This has a few issues. The biggest one is that compiling frame pointers into every application and library has been shown to cause performance overhead. Another issue is that the format of the frames may not always be consistent between different compilers and some architectures (s390) has no defined format to do a reliable stack walk. The only way to perform user space profiling on these architectures is to copy the user stack into the kernel buffer. SFrame [1] is now supported in binutils (x86-64, ARM64, and s390). There is discussions going on about supporting SFrame in LLVM. SFrame acts more like ORC, and lives in the ELF executable file as its own section. Like ORC it has two tables where the first table is sorted by instruction pointers (IP) and using the current IP and finding it's entry in the first table, it will take you to the second table which will tell you where the return address of the current function is located and then you can use that address to look it up in the first table to find the return address of that function, and so on. This performs a user space stack walk. Now because the .sframe section lives in the ELF file it needs to be faulted into memory when it is used. This means that walking the user space stack requires being in a faultable context. As profilers like perf request a stack trace in interrupt or NMI context, it cannot do the walking when it is requested. Instead it must be deferred until it is safe to fault in user space. One place this is known to be safe is when the task is about to return back to user space. This series makes the deferred unwind user code implement SFrame format V3 and enables it on x86-64. [1]: https://sourceware.org/binutils/wiki/sframe This series applies on top of v7.1-rc4 tag: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git v7.1-rc4 The to be stack-traced user space programs (and libraries) need to be built with the recent SFrame stack trace information format V3, as generated by binutils 2.46+ with assembler option --gsframe-3. Namhyung Kim's related perf tools deferred callchain support can be used for testing ("perf record --call-graph fp,defer" and "perf report/script"). Changes in v15 (see patch notes for details): - Rebase on v7.1-rc4. - New patch to duplicate registered .sframe section data on clone/fork. - Address Sashiko AI review feedback: - Fix sframe end passed to mtree_insert_range(). - Fix outermost frame (FRE without datawords) handling. - Use GFP_KERNEL_ACCOUNT instead of GFP_KERNEL. - Improve text/sframe section start/end validation. - Always use guard(srcu) when accessing struct sframe_section fields. - Validate FDE repetition size for PCTYPE_MASK FDEs to be non-zero to prevent division by zero. - Only add sframe for text that is PT_LOAD in addition to PF_X. - Use pr_debug_once() instead of WARN_ON_ONCE() to prevent user- triggered warning/panic. - Add support for SP/FP-based CFA recovery rules with dereferencing. - Reject FRE control word with reserved_p=1. - x86-64: Fail unwind_user_get_reg() if !user_64bit_mode(). - Validate FDE PC type for supported values (i.e. PCTYPE_INC or PCTYPE_MASK). - Validate FDE function end against text end. - Validate FDE's number of FREs to be less or equal to FDE's function size, as each FRE must cover at least one byte. (Indu) - Validate FRE function offset against FDE repetition size for PCTYPE_MASK. - Change type of struct sframe_fde_internal field fres_num to the one of struct sframe_fda_v3 field fres_num. - Return RC of sframe_init_[cfa_]rule_data() if bad RC. - Normalize error code usage (.sframe is removed for all but ENOENT): ENOENT: No sframe or no FDE for IP found (FDE found but no FRE found is EINVAL) EFAULT: Bad address EINVAL: Invalid input or sframe - Build-time checks for config options: - 64BIT: SFrame V3 only supports 64-bit architectures. - HAVE_EFFICIENT_UNALIGNED_ACCESS: Unaligned access to 16/32-bit SFrame FRE fields and datawords using unsafe_get_user(). (Steven) - Add pr_debug_once() when restoring CFA/FP/RA from an unsupported register number. Changes in v14 (see patch notes for details): - Rebase on v7.1-rc2. - Correct SFRAME_V3_FDE_TYPE_MASK value. - Fix FDE function start address check in __read_fde(). - Rename SFrame V3 definitions accoring to final specification. (Indu) - Improve comments on why UNWIND_USER_RULE_CFA_OFFSET is not implemented. (Mark Rutland) - Add/update/improve sframe debug messages. - Add generic and arch-specific unwind_user.h to MAINTAINERS. - Add arch-specific unwind_user_sframe.h to MAINTAINERS. Changes in v13 (see patch notes for details): - Add support for SFrame V3, including its new flexible FDEs. SFrame V2 is not supported. Changes in v12 (see patch notes for details): - Adjust to Peter's latest undwind user enhancements. - Simplify logic by using an internal SFrame FDE representation, whose FDE function start address field is an address instead of a PC-relative offset (from FDE). - Rename struct sframe_fre to sframe_fre_internal to align with struct sframe_fde_internal. - Remove unused pt_regs from unwind_user_next_common() and its callers. (Peter) - Simplify unwind_user_next_sframe(). (Peter) - Fix a few checkpatch errors and warnings. - Minor cleanups (e.g. move includes, fix indentation). Changes in v11: - Support for SFrame V2 PC-relative FDE function start address. - Support for SFrame V2 representing RA undefined as indication for outermost frames. Patch 1 (new in v14), as a preparatory cleanup, adds the generic and arch-specific unwind_user.h to MAINTAINERS. Patches 2, 5, 12, and 19 have been updated to exclusively support the latest SFrame V3 stack trace information format, that is generated by binutils 2.46+. Old SFrame V2 sections get rejected with dynamic debug message "bad/unsupported sframe header". Patches 8 and 9 add support to unwind user (sframe) for outermost frames. Patches 13-16 add support to unwind user (sframe) for the new SFrame V3 flexible FDEs. Patch 17 improves the performance of searching the SFrame FRE for an IP. Patch 18 (new in v15) duplicates registered .sframe section data on clone/fork from the parent to the child process. Patch 20 is for test purposes only and will get replaced by a new syscall, that Steven is working on: [RFC][PATCH] unwind: Add stacktrace_setup system call https://lore.kernel.org/all/[email protected]/ Regards, Jens Jens Remus (9): unwind_user: Add generic and arch-specific headers to MAINTAINERS unwind_user: Stop when reaching an outermost frame unwind_user/sframe: Add support for outermost frame indication unwind_user: Enable archs that pass RA in a register unwind_user: Flexible FP/RA recovery rules unwind_user: Flexible CFA recovery rules unwind_user/sframe: Add support for SFrame V3 flexible FDEs unwind_user/sframe: Separate reading of FRE from reading of FRE data words unwind_user/sframe: Duplicate registered .sframe section data on clone/fork Josh Poimboeuf (11): unwind_user/sframe: Add support for reading .sframe headers unwind_user/sframe: Store .sframe section data in per-mm maple tree x86/uaccess: Add unsafe_copy_from_user() implementation unwind_user/sframe: Add support for reading .sframe contents unwind_user/sframe: Detect .sframe sections in executables unwind_user/sframe: Wire up unwind_user to sframe unwind_user/sframe: Remove .sframe section on detected corruption unwind_user/sframe: Show file name in debug output unwind_user/sframe: Add .sframe validation option unwind_user/sframe/x86: Enable sframe unwinding on x86 unwind_user/sframe: Add prctl() interface for registering .sframe sections MAINTAINERS | 4 + arch/Kconfig | 23 + arch/x86/Kconfig | 1 + arch/x86/include/asm/mmu.h | 2 +- arch/x86/include/asm/uaccess.h | 39 +- arch/x86/include/asm/unwind_user.h | 74 +- arch/x86/include/asm/unwind_user_sframe.h | 12 + fs/binfmt_elf.c | 48 +- include/linux/mm_types.h | 3 + include/linux/sframe.h | 65 ++ include/linux/unwind_user.h | 20 + include/linux/unwind_user_types.h | 50 +- include/uapi/linux/elf.h | 1 + include/uapi/linux/prctl.h | 4 + kernel/fork.c | 10 + kernel/sys.c | 9 + kernel/unwind/Makefile | 3 +- kernel/unwind/sframe.c | 949 ++++++++++++++++++++++ kernel/unwind/sframe.h | 88 ++ kernel/unwind/sframe_debug.h | 75 ++ kernel/unwind/user.c | 133 ++- mm/init-mm.c | 2 + mm/mmap.c | 9 + 23 files changed, 1586 insertions(+), 38 deletions(-) create mode 100644 arch/x86/include/asm/unwind_user_sframe.h create mode 100644 include/linux/sframe.h create mode 100644 kernel/unwind/sframe.c create mode 100644 kernel/unwind/sframe.h create mode 100644 kernel/unwind/sframe_debug.h -- 2.51.0
