linux/tools/lib
Alan Maguire f3c51fe02c libbpf: Btf dedup identical struct test needs check for nested structs/arrays
When examining module BTF, it is common to see core kernel structures
such as sk_buff, net_device duplicated in the module.  After adding
debug messaging to BTF it turned out that much of the problem
was down to the identical struct test failing during deduplication;
sometimes the compiler adds identical structs.  However
it turns out sometimes that type ids of identical struct members
can also differ, even when the containing structs are still identical.

To take an example, for struct sk_buff, debug messaging revealed
that the identical struct matching was failing for the anon
struct "headers"; specifically for the first field:

__u8       __pkt_type_offset[0]; /*   128     0 */

Looking at the code in BTF deduplication, we have code that guards
against the possibility of identical struct definitions, down to
type ids, and identical array definitions.  However in this case
we have a struct which is being defined twice but does not have
identical type ids since each duplicate struct has separate type
ids for the above array member.   A similar problem (though not
observed) could occur for struct-in-struct.

The solution is to make the "identical struct" test check members
not just for matching ids, but to also check if they in turn are
identical structs or arrays.

The results of doing this are quite dramatic (for some modules
at least); I see the number of type ids drop from around 10000
to just over 1000 in one module for example.

For testing use latest pahole or apply [1], otherwise dedups
can fail for the reasons described there.

Also fix return type of btf_dedup_identical_arrays() as
suggested by Andrii to match boolean return type used
elsewhere.

Fixes: efdd3eb801 ("libbpf: Accommodate DWARF/compiler bug with duplicated structs")
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/1666622309-22289-1-git-send-email-alan.maguire@oracle.com

[1] https://lore.kernel.org/bpf/1666364523-9648-1-git-send-email-alan.maguire
2022-10-25 16:16:14 -07:00
..
api perf record: Fix way of handling non-perf-event pollfds 2022-10-04 08:55:19 -03:00
bpf libbpf: Btf dedup identical struct test needs check for nested structs/arrays 2022-10-25 16:16:14 -07:00
perf perf tools: Remove special handling of system-wide evsel 2022-10-06 08:03:53 -03:00
subcmd perf subcmd: Set environment variable "PREFIX" 2022-10-04 08:55:23 -03:00
symbol kallsyms: increase maximum kernel symbol length to 512 2022-09-28 08:56:25 +02:00
thermal tools/lib/thermal: remove unneeded semicolon 2022-05-19 12:11:52 +02:00
traceevent libtraceevent: Add __rel_loc relative location attribute support 2021-12-06 15:37:22 -05:00
argv_split.c tools lib: Move argv_{split,free} from tools/perf/util/ 2019-07-01 22:50:40 -03:00
bitmap.c Bitmap patches for v6.0-rc1 2022-08-07 17:52:35 -07:00
ctype.c
find_bit.c tools: sync find_bit() implementation 2022-09-21 12:21:44 -07:00
hweight.c
list_sort.c tools lib: Adopt list_sort() from the kernel sources 2021-10-20 10:30:59 -03:00
rbtree.c tools/: replace HTTP links with HTTPS ones 2020-08-07 11:33:21 -07:00
slab.c tools: Move gfp.h and slab.h from radix-tree to lib 2022-02-20 08:44:37 +02:00
str_error_r.c
string.c tools lib: Adopt memchr_inv() from kernel 2020-11-27 08:34:52 -03:00
vsprintf.c
zalloc.c tools lib: Adopt zalloc()/zfree() from tools/perf 2019-07-09 10:13:26 -03:00