在 2022/10/28 4:34, Andrii Nakryiko 写道:
On Thu, Oct 27, 2022 at 3:03 AM Wang Yufen <wangyufen@xxxxxxxxxx> wrote:kmemleak reports this issue: unreferenced object 0xffff88817139d000 (size 2048): comm "test_progs", pid 33246, jiffies 4307381979 (age 45851.820s) hex dump (first 32 bytes): 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<0000000045f075f0>] kmalloc_trace+0x27/0xa0 [<0000000098b7c90a>] __check_func_call+0x316/0x1230 [<00000000b4c3c403>] check_helper_call+0x172e/0x4700 [<00000000aa3875b7>] do_check+0x21d8/0x45e0 [<000000001147357b>] do_check_common+0x767/0xaf0 [<00000000b5a595b4>] bpf_check+0x43e3/0x5bc0 [<0000000011e391b1>] bpf_prog_load+0xf26/0x1940 [<0000000007f765c0>] __sys_bpf+0xd2c/0x3650 [<00000000839815d6>] __x64_sys_bpf+0x75/0xc0 [<00000000946ee250>] do_syscall_64+0x3b/0x90 [<0000000000506b7f>] entry_SYSCALL_64_after_hwframe+0x63/0xcd The root case here is: In function prepare_func_exit(), the callee is not released in the abnormal scenario after "state->curframe--;". In addition, function __check_func_call() has a similar problem. In the abnormal scenario before "state->curframe++;", the callee is alse not released.For prepare_func_exit, wouldn't it be correct and cleaner to just move state->curframe--; to the very bottom of the function, right when we free callee and reset frame[] pointer to NULL?
Yes, that't better. will change and test in v2.
For __check_func_call, please use err_out label name to disambiguate it from the "err" variable.
I got it. will change in v2.
Fixes: 69c087ba6225 ("bpf: Add bpf_for_each_map_elem() helper") Fixes: fd978bf7fd31 ("bpf: Add reference tracking to verifier") Signed-off-by: Wang Yufen <wangyufen@xxxxxxxxxx> --- kernel/bpf/verifier.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-)[...]
