| Message ID | 1643645554-28723-1-git-send-email-alan.maguire@oracle.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | libbpf: name-based u[ret]probe attach | expand |
Hi Alan, On 1/31/22 5:12 PM, Alan Maguire wrote: > This patch series is a refinement of the RFC patchset [1], focusing > on support for attach by name for uprobes and uretprobes. v3 > because there was an earlier RFC [2]. > > Currently attach for such probes is done by determining the offset > manually, so the aim is to try and mimic the simplicity of kprobe > attach, making use of uprobe opts to specify a name string. > Patch 1 adds the "func_name" option to allow uprobe attach by > name; the mechanics are described there. > > Having name-based support allows us to support auto-attach for > uprobes; patch 2 adds auto-attach support while attempting > to handle backwards-compatibility issues that arise. The format > supported is > > u[ret]probe//path/2/binary:[raw_offset|function[+offset]] > > For example, to attach to libc malloc: > > SEC("uprobe//usr/lib64/libc.so.6:malloc") > > Patch 3 introduces a helper function to trace_helpers, allowing > us to retrieve the path to a library by reading /proc/self/maps. > > Finally patch 4 add tests to the attach_probe selftests covering > attach by name, auto-attach and auto-attach failure. Looks like the selftest in the series fails the BPF CI (test_progs & test_progs-no_alu32): https://github.com/kernel-patches/bpf/runs/5012260907?check_suite_focus=true [...] test_attach_probe:PASS:uprobe_offset 0 nsec test_attach_probe:PASS:ref_ctr_offset 0 nsec test_attach_probe:PASS:skel_open 0 nsec test_attach_probe:PASS:check_bss 0 nsec test_attach_probe:PASS:attach_kprobe 0 nsec test_attach_probe:PASS:attach_kretprobe 0 nsec test_attach_probe:PASS:uprobe_ref_ctr_before 0 nsec test_attach_probe:PASS:attach_uprobe 0 nsec test_attach_probe:PASS:uprobe_ref_ctr_after 0 nsec test_attach_probe:PASS:attach_uretprobe 0 nsec test_attach_probe:PASS:auto-attach should fail for old-style name 0 nsec test_attach_probe:PASS:attach_uprobe_byname 0 nsec test_attach_probe:PASS:attach_uretprobe_byname 0 nsec test_attach_probe:PASS:get path to libc 0 nsec test_attach_probe:PASS:find libc path in /proc/self/maps 0 nsec libbpf: failed to open 7f55b225c000-7f55b2282000 r--p 00000000 fe:00 3381 /usr/lib/libc-2.32.so: No such file or directory test_attach_probe:FAIL:attach_uprobe_byname2 unexpected error: -2 test_attach_probe:PASS:uprobe_ref_ctr_cleanup 0 nsec #4 attach_probe:FAIL [...] Thanks, Daniel
This patch series is a refinement of the RFC patchset [1], focusing on support for attach by name for uprobes and uretprobes. v3 because there was an earlier RFC [2]. Currently attach for such probes is done by determining the offset manually, so the aim is to try and mimic the simplicity of kprobe attach, making use of uprobe opts to specify a name string. Patch 1 adds the "func_name" option to allow uprobe attach by name; the mechanics are described there. Having name-based support allows us to support auto-attach for uprobes; patch 2 adds auto-attach support while attempting to handle backwards-compatibility issues that arise. The format supported is u[ret]probe//path/2/binary:[raw_offset|function[+offset]] For example, to attach to libc malloc: SEC("uprobe//usr/lib64/libc.so.6:malloc") Patch 3 introduces a helper function to trace_helpers, allowing us to retrieve the path to a library by reading /proc/self/maps. Finally patch 4 add tests to the attach_probe selftests covering attach by name, auto-attach and auto-attach failure. Changes since RFC [1]: - used "long" for addresses instead of ssize_t (Andrii, patch 1). - used gelf_ interfaces to avoid assumptions about 64-bit binaries (Andrii, patch 1) - clarified string matching in symbol table lookups (Andrii, patch 1) - added support for specification of shared object functions in a non-shared object binary. This approach instruments the Procedure Linking Table (PLT) - malloc@PLT. - changed logic in symbol search to check dynamic symbol table first, then fall back to symbol table (Andrii, patch 1). - modified auto-attach string to require "/" separator prior to path prefix i.e. uprobe//path/to/binary (Andrii, patch 2) - modified auto-attach string to use ':' separator (Andrii, patch 2) - modified auto-attach to support raw offset (Andrii, patch 2) - modified skeleton attach to interpret -ESRCH errors as a non-fatal "unable to auto-attach" (Andrii suggested -EOPNOTSUPP but my concern was it might collide with other instances where that value is returned and reflects a failure to attach a to-be-expected attachment rather than skip a program that does not present an auto-attachable section name. Admittedly -EOPNOTSUPP seems a more natural value here). - moved library path retrieval code to trace_helpers (Andrii, patch 3) [1] https://lore.kernel.org/bpf/1642678950-19584-1-git-send-email-alan.maguire@oracle.com/ [2] https://lore.kernel.org/bpf/1642004329-23514-1-git-send-email-alan.maguire@oracle.com/ Alan Maguire (4): libbpf: support function name-based attach uprobes libbpf: add auto-attach for uprobes based on section name selftests/bpf: add get_lib_path() helper selftests/bpf: add tests for u[ret]probe attach by name tools/lib/bpf/libbpf.c | 327 ++++++++++++++++++++- tools/lib/bpf/libbpf.h | 10 +- .../selftests/bpf/prog_tests/attach_probe.c | 89 +++++- .../selftests/bpf/progs/test_attach_probe.c | 37 +++ tools/testing/selftests/bpf/trace_helpers.c | 17 ++ tools/testing/selftests/bpf/trace_helpers.h | 2 + 6 files changed, 475 insertions(+), 7 deletions(-)