kallsyms: Don't let kallsyms_lookup_size_offset() fail on retrieving the first symbol
PATCH 1/1 kallsyms: Don't let kallsyms_lookup_size_offset() fail on retrieving the first symbol
An arm64 kernel configured with
CONFIG_KPROBES=y
CONFIG_KALLSYMS=y
# CONFIG_KALLSYMS_ALL is not set
CONFIG_KALLSYMS_BASE_RELATIVE=y
reports the following kprobe failure:
[ 0.032677] kprobes: failed to populate blacklist: -22 [ 0.033376] Please take care of using kprobes.
It appears that kprobe fails to retrieve the symbol at address 0xffff000010081000, despite this symbol being in System.map:
ffff000010081000 T __exception_text_start
This symbol is part of the first group of aliases in the kallsyms_offsets array (symbol names generated using ugly hacks in scripts/kallsyms.c):
kallsyms_offsets:
.long 0x1000 // do_undefinstr
.long 0x1000 // efi_header_end
.long 0x1000 // _stext
.long 0x1000 // __exception_text_start
.long 0x12b0 // do_cp15instrLooking at the implementation of get_symbol_pos(), it returns the lowest index for aliasing symbols. In this case, it return 0.
But kallsyms_lookup_size_offset() considers 0 as a failure, which is obviously wrong (there is definitely a valid symbol living there). In turn, the kprobe blacklisting stops abruptly, hence the original error.
A CONFIG_KALLSYMS_ALL kernel wouldn't fail as there is always some random symbols at the beginning of this array, which are never looked up via kallsyms_lookup_size_offset.
Fix it by considering that get_symbol_pos() is always successful (which is consistent with the other uses of this function).
--- kernel/kallsyms.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
kernel/kallsyms.c +4 -2
@@ -263,8 +263,10 @@ int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize, {char namebuf[KSYM_NAME_LEN ]; -if (is_ksym_addr (addr)) -return !!get_symbol_pos (addr, symbolsize, offset); +if (is_ksym_addr (addr)) { +get_symbol_pos (addr, symbolsize, offset); +return 1 ; + }return !!module_address_lookup (addr, symbolsize, offset,NULL , namebuf) || !!__bpf_address_lookup (addr, symbolsize, offset, namebuf); }
Signature · 1 line
2.20.1
Quoted 36 lines · On Sat, 24 Aug 2019 14:12:31 +0100 Marc Zyngier <[email protected]> wrote:
An arm64 kernel configured with
CONFIG_KPROBES=y
CONFIG_KALLSYMS=y
#CONFIG_KALLSYMS_ALLis not set
CONFIG_KALLSYMS_BASE_RELATIVE=yreports the following kprobe failure:
[ 0.032677] kprobes: failed to populate blacklist: -22 [ 0.033376] Please take care of using kprobes.It appears that kprobe fails to retrieve the symbol at address
0xffff000010081000, despite this symbol being in System.map:
ffff000010081000T__exception_text_startThis symbol is part of the first group of aliases in the
kallsyms_offsetsarray (symbol names generated using ugly hacks inscripts/kallsyms.c):
kallsyms_offsets:.long 0x1000 // do_undefinstr .long 0x1000 // efi_header_end .long 0x1000 // _stext .long 0x1000 // __exception_text_start .long 0x12b0 // do_cp15instrLooking at the implementation of
get_symbol_pos(), it returns the lowest index for aliasing symbols. In this case, it return 0.But
kallsyms_lookup_size_offset()considers 0 as a failure, which is obviously wrong (there is definitely a valid symbol living there). In turn, the kprobe blacklisting stops abruptly, hence the original error.
Quoted 6 lines
A
CONFIG_KALLSYMS_ALLkernel wouldn't fail as there is always some random symbols at the beginning of this array, which are never looked up viakallsyms_lookup_size_offset.Fix it by considering that
get_symbol_pos()is always successful (which is consistent with the other uses of this function).
Thank you for fixing this issue!
This looks good to me :)
Thank you,
Quoted 32 lines
Fixes ffc5089196446 ("[PATCH] Create kallsyms_lookup_size_offset()")Cc Masami Hiramatsu <[email protected]>Cc Arnaldo Carvalho de Melo <[email protected]>Cc Peter Zijlstra <[email protected]>Cc Will Deacon <[email protected]>Cc Catalin Marinas <[email protected]>Signed-off-by Marc Zyngier <[email protected]>--- kernel/kallsyms.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)kernel/kallsyms.c +4
-2@@ -263,8 +263,10 @@ int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize, {char namebuf[KSYM_NAME_LEN ]; -if (is_ksym_addr (addr)) -return !!get_symbol_pos (addr, symbolsize, offset); +if (is_ksym_addr (addr)) { +get_symbol_pos (addr, symbolsize, offset); +return 1 ; + }return !!module_address_lookup (addr, symbolsize, offset,NULL , namebuf) || !!__bpf_address_lookup (addr, symbolsize, offset, namebuf); }Signature · 1 line
2.20.1
Signature · 1 line
Masami Hiramatsu <[email protected]>
Quoted 43 lines · Masami Hiramatsu
Quoted 40 lines · On Sat, 24 Aug 2019 14:12:31 +0100 Marc Zyngier <[email protected]> wrote:
An arm64 kernel configured with
CONFIG_KPROBES=y
CONFIG_KALLSYMS=y
#CONFIG_KALLSYMS_ALLis not set
CONFIG_KALLSYMS_BASE_RELATIVE=yreports the following kprobe failure:
[ 0.032677] kprobes: failed to populate blacklist: -22 [ 0.033376] Please take care of using kprobes.It appears that kprobe fails to retrieve the symbol at address
0xffff000010081000, despite this symbol being in System.map:
ffff000010081000T__exception_text_startThis symbol is part of the first group of aliases in the
kallsyms_offsetsarray (symbol names generated using ugly hacks inscripts/kallsyms.c):
kallsyms_offsets:.long 0x1000 // do_undefinstr .long 0x1000 // efi_header_end .long 0x1000 // _stext .long 0x1000 // __exception_text_start .long 0x12b0 // do_cp15instrLooking at the implementation of
get_symbol_pos(), it returns the lowest index for aliasing symbols. In this case, it return 0.But
kallsyms_lookup_size_offset()considers 0 as a failure, which is obviously wrong (there is definitely a valid symbol living there). In turn, the kprobe blacklisting stops abruptly, hence the original error.A
CONFIG_KALLSYMS_ALLkernel wouldn't fail as there is always some random symbols at the beginning of this array, which are never looked up viakallsyms_lookup_size_offset.
Quoted 7 lines
Quoted 2 lines
Fix it by considering that
get_symbol_pos()is always successful (which is consistent with the other uses of this function).Thank you for fixing this issue!
This looks good to me :)Reviewed-by Masami Hiramatsu <[email protected]>
There doesn't appear to be an official tree for this file so, unless anybody objects, I'll pick this up via arm64 as it's looking like there are some other fixes on the horizon too.
Will