Re: [PATCH] powerpc/kprobes: Remove kretprobe_trampoline_holder.
flat view
From: Thiago Jung Bauermann <hidden>
Date: 2016-04-14 00:43:20
Hello, People seem to be considering patches for next, so this looks like a go= od=20 moment to ping about this one. --=20 []'s Thiago Jung Bauermann IBM Linux Technology Center Am Donnerstag, 31 M=E4rz 2016, 17:10:40 schrieb Thiago Jung Bauermann:
Fixes the following testsuite failure: =20 $ sudo ./perf test -v kallsyms 1: vmlinux symtab matches kallsyms : --- start --- test child forked, pid 12489 Using /proc/kcore for kernel object code Looking at the vmlinux_path (8 entries long) Using /boot/vmlinux for symbols 0xc00000000003d300: diff name v: .kretprobe_trampoline_holder k: kretprobe_trampoline Maps only in vmlinux: c00000000086ca38-c000000000879b6c 87ca38 [kernel].text.unlikely c000000000879b6c-c000000000bf0000 889b6c [kernel].meminit.text c000000000bf0000-c000000000c53264 c00000 [kernel].init.text c000000000c53264-d000000004250000 c63264 [kernel].exit.text d000000004250000-d000000004450000 0 [libcrc32c] d000000004450000-d000000004620000 0 [xfs] d000000004620000-d000000004680000 0 [autofs4] d000000004680000-d0000000046e0000 0 [x_tables] d0000000046e0000-d000000004780000 0 [ip_tables] d000000004780000-d0000000047e0000 0 [rng_core] d0000000047e0000-ffffffffffffffff 0 [pseries_rng] Maps in vmlinux with a different name in kallsyms: Maps only in kallsyms: d000000000000000-f000000000000000 1000000000010000 [kernel.kallsym=
s]
f000000000000000-ffffffffffffffff 3000000000010000 [kernel.kallsym=
s]
test child finished with -1 ---- end ---- vmlinux symtab matches kallsyms: FAILED! =20 The problem is that the kretprobe_trampoline symbol looks like this: =20 $ eu-readelf -s /boot/vmlinux G kretprobe_trampoline 2431: c000000001302368 24 NOTYPE LOCAL DEFAULT 37 kretprobe_trampoline_holder 2432: c00000000003d300 8 FUNC LOC=
AL=20
DEFAULT 1 .kretprobe_trampoline_holder 97543: c00000000003d300=
=20
0 NOTYPE GLOBAL DEFAULT 1 kretprobe_trampoline =20 Its type is NOTYPE, and its size is 0, and this is a problem because symbol-elf.c:dso__load_sym skips function symbols that are not STT_FU=
NC
or STT_GNU_IFUNC (this is determined by elf_sym__is_function). Even if the type is changed to STT_FUNC, when dso__load_sym calls symbols__fixup_duplicate, the kretprobe_trampoline symbol is dropped =
in
favour of .kretprobe_trampoline_holder because the latter has non-zer=
o
size (as determined by choose_best_symbol).
=20
With this patch, all vmlinux symbols match /proc/kallsyms and the
testcase passes.
=20
Commit c1c355ce14c0 ("x86/kprobes: Get rid of
kretprobe_trampoline_holder()") gets rid of kretprobe_trampoline_hold=er
altogether on x86. This commit does the same on powerpc. This change introduces no regressions on the perf and ftracetest testsuite result=
s.
quoted hunk ↗ jump to hunk
=20 Cc: Ananth N Mavinakayanahalli <redacted> Cc: Michael Ellerman <mpe@ellerman.id.au> Reviewed-by: Naveen N. Rao <redacted> Signed-off-by: Thiago Jung Bauermann <redacted> --- arch/powerpc/kernel/kprobes.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) =20diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kpro=
bes.c
quoted hunk ↗ jump to hunk
index 7c053f281406..417c0eadd094 100644--- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c@@ -278,12 +278,11 @@ no_kprobe: * =09- When the probed function returns, this probe * =09=09causes the handlers to fire */ -static void __used kretprobe_trampoline_holder(void) -{ -=09asm volatile(".global kretprobe_trampoline\n" -=09=09=09"kretprobe_trampoline:\n" -=09=09=09"nop\n"); -} +asm(".global kretprobe_trampoline\n" +=09".type kretprobe_trampoline, @function\n" +=09"kretprobe_trampoline:\n" +=09"nop\n" +=09".size kretprobe_trampoline, .-kretprobe_trampoline\n");=20 /* * Called when the probe at kretprobe trampoline is hit