kernel/git/next/linux-next.git/tools/perf/util/unwind-libdw.h, branch master

perf unwind-libdw: Wire up e_flags for CSKY

2026-01-26T21:31:27+00:00

Wire up the e_flags now it can be read for a thread. The e_flags encode the CSKY ABI level and this can impact which perf registers need setting up for unwinding. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Guo Ren Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Swapnil Sapkal Cc: Tianyou Li Signed-off-by: Arnaldo Carvalho de Melo

perf unwind-libdw: Don't discard loaded ELF/DWARF after every unwind

2026-01-20T18:43:56+00:00

The unwind-libdw dwfl has ELF binaries associated with mmap addresses. Experimenting with using the per dso dwfl it is required to alter the address to be 0 based variant. Unfortunately libdwfl doesn't allow a single unwind and then an update to the return address to be 0 based as there are assertions that registers aren't updated once an unwind has started, etc. As removing the dwfl didn't prove possible, an alternative is to just not discard the dwfl when the unwind ends. The dwfl is valid for a process unless a dso is loaded at the same address as a previous one. So keep the dwfl with the maps, invalidate it if a map is removed (in case a new map replaces it) and recycle the dwfl in the unwinding code. A wrinkly in the implementation of this is that the attached thread argument is remembered by the dwfl and so it needs to be a pointer to memory that also persists with the dwfl (struct dwfl_ui_thread_info in the code). Recording 10 seconds of system wide data with --call-graph=dwarf and then processing with perf report shows a total runtime improvement from 41.583s to 2.279s (an 18x speedup). Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo

perf dwarf-regs: Add MIPS perf to DWARF register number mapping functions

2026-01-20T18:43:56+00:00

Despite an unused function declaration, there was no unwind-libdw for MIPS but there is a perf_regs.h and a libdw implementation. Fill in the pieces so hopefully MIPS unwinding with libdw works. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo

perf dwarf-regs: Add S390 perf to DWARF register number mapping functions

2026-01-20T18:43:56+00:00

These functions allow the generic initial register state code in unwind-libdw to be used. Now the non-generic code in unwind-libdw has no uses remove it. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo

perf dwarf-regs: Add RISC-V perf to DWARF register number mapping functions

2026-01-20T18:43:56+00:00

These functions allow the generic initial register state code in unwind-libdw to be used. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo

perf dwarf-regs: Add powerpc perf to DWARF register number mapping functions

2026-01-20T18:43:56+00:00

These functions allow the generic initial register state code in unwind-libdw to be used. Note, the link register was being coped to DWARF register 65 that the SysV ABI spec claims is FPSCR. It is corrected here to 108, but this is unlikely to matter as FPSCR has little to no impact on unwinding. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo

perf dwarf-regs: Add loongarch perf to DWARF register number mapping functions

2026-01-20T18:43:56+00:00

perf dwarf-regs: Add csky perf to dwarf register number mapping functions

2026-01-20T18:43:56+00:00

perf dwarf-regs: Add ARM perf to dwarf register number mapping functions

2026-01-20T18:43:56+00:00

perf dwarf-regs: Add get_dwarf_regnum_for_perf_regnum() and use for x86 unwinding

2026-01-20T18:43:56+00:00

Add a utility to map a perf register number to a DWARF register number for a particular ELF machine type. Create a generic unwind-libdw initial register initialization routine that uses this function and thereby avoids arch specific initialization. The unwind-libdw code does: 1) compute the maximum DWARF register from the set of sampled user registers, 2) allocates a set of DWARF registers, 3) copies the sample registers into the appropriate DWARF registers. This generic solution is initially implemented for use with x86 as only get_dwarf_regnum_for_perf_regnum() support for x86 is currently present. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Haibo Xu Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: Mark Wielaard Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Thomas Falcon Cc: Will Deacon Signed-off-by: Arnaldo Carvalho de Melo