Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] ddebug_proc_start() fails to return EPILOGUE_TOKEN when resuming sequence iteration at the epilogue index, which violates seq_file semantics and causes the epilogue to be silently omitted if a read boundary falls exactly before it. --
commit c254572f4a41eba87d7288c5cb6c6ef71571d0a3 Author: Jim Cromie <[email protected]> dyndbg: add epilogue to dynamic_debug/control file This commit adds an epilogue to the control-file to allow display of statistics without disturbing the header. It introduces an EPILOGUE_TOKEN yielded at the end of seq_file iteration. > diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c > index b7ccf471b5efe..d098afe8d340f 100644 > --- a/lib/dynamic_debug.c > +++ b/lib/dynamic_debug.c [ ... ] > static void *ddebug_proc_next(struct seq_file *m, void *p, loff_t *pos) > { > struct ddebug_iter *iter = m->private; > struct _ddebug *dp; > > + (*pos)++; > + > + if (p == EPILOGUE_TOKEN) > + return NULL; > + > if (p == SEQ_START_TOKEN) > dp = ddebug_iter_first(iter); > else > dp = ddebug_iter_next(iter); > - ++*pos; > - return dp; > + > + if (dp) > + return dp; > + > + return EPILOGUE_TOKEN; [Severity: Medium] Since ddebug_proc_next() now yields EPILOGUE_TOKEN as an additional sequence element, does ddebug_proc_start() also need to be updated to handle resuming iteration at the epilogue index? If a read boundary falls exactly before the epilogue (where pos equals the total number of items plus one), ddebug_proc_start() will iterate until dp is NULL, and then return NULL instead of EPILOGUE_TOKEN: lib/dynamic_debug.c:ddebug_proc_start() { ... dp = ddebug_iter_first(iter); while (dp != NULL && --n > 0) dp = ddebug_iter_next(iter); return dp; } Will this cause the epilogue to be silently omitted in that scenario? -- Sashiko AI review ยท https://sashiko.dev/#/patchset/[email protected]?part=40
