On 3/2/20 11:42 PM, Andrew Morton wrote: > On Mon, 02 Mar 2020 14:31:48 -0500 Qian Cai <cai@xxxxxx> wrote: > >> On Mon, 2020-03-02 at 13:20 -0500, Qian Cai wrote: >>> This patch spams the console like crazy while reading sysfs, >>> >>> # dmesg | grep 'buggy seq_file' | wc -l >>> 4204 >>> >>> [ 9505.321981] LTP: starting read_all_proc (read_all -d /proc -q -r 10) >>> [ 9508.222934] buggy seq_file .next function xt_match_seq_next [x_tables] did >>> not updated position index >>> [ 9508.223319] buggy seq_file .next function xt_match_seq_next [x_tables] did >>> not updated position index >>> [ 9508.223654] buggy seq_file .next function xt_match_seq_next [x_tables] did >>> not updated position index >>> [ 9508.223994] buggy seq_file .next function xt_match_seq_next [x_tables] did >>> not updated position index >>> [ 9508.224337] buggy seq_file .next function xt_match_seq_next [x_tables] did >>> not updated position index It should be fixed by following patch-set submitted to Netfilter-Devel mailing list [PATCH v2 0/4] netfilter: seq_file .next functions should increase position index https://lore.kernel.org/netfilter-devel/497a82c1-7b6a-adf4-a4ce-df46fe436aae@xxxxxxxxxxxxx/T/ >>>> --- a/fs/seq_file.c~seq_read-info-message-about-buggy-next-functions >>>> +++ a/fs/seq_file.c >>>> @@ -256,9 +256,12 @@ Fill: >>>> loff_t pos = m->index; >>>> >>>> p = m->op->next(m, p, &m->index); >>>> - if (pos == m->index) >>>> - /* Buggy ->next function */ >>>> + if (pos == m->index) { >>>> + pr_info("buggy seq_file .next function %ps " >>>> + "did not updated position index\n", >>>> + m->op->next); >> >> This? >> >> s/pr_info/pr_info_ratelimited/ >> > > Fair enough - I made that change. >
