Messages in this thread | | | From | Namhyung Kim <> | Date | Tue, 4 Nov 2014 00:01:26 +0900 | Subject | Re: [PATCH 1/8] perf tools: Preparation for compressed kernel module support |
| |
Hi Jiri,
On Mon, Nov 3, 2014 at 10:51 PM, Jiri Olsa <jolsa@redhat.com> wrote: > On Mon, Nov 03, 2014 at 04:27:51PM +0900, Namhyung Kim wrote: > > SNIP > >> +static int decompress_dummy(const char *input __maybe_unused, >> + int output __maybe_unused) >> +{ >> + return -1; >> +} >> + >> +static const struct { >> + const char *fmt; >> + int (*decompress)(const char *input, int output); >> +} compressions[] = { >> + { "gz", decompress_dummy }, >> +}; >> + >> +bool is_supported_compression(const char *ext) >> +{ >> + unsigned i; >> + >> + for (i = 0; i < ARRAY_SIZE(compressions); i++) { >> + if (!strcmp(ext, compressions[i].fmt)) >> + return true; >> + } >> + return false; >> +} >> + >> +bool is_kmodule_extension(const char *ext) >> +{ >> + if (!strcmp(ext, "ko")) >> + return true; >> + >> + if (ext[2] == '\0' || (ext[2] == '.' && is_supported_compression(ext+3))) >> + return true; > > I got stuck in here.. so any 2 chars string is kernel module extension?
Oops, the first check should be:
if (strncmp(ext, "ko", 2)) return false;
> plus how do u know ext[3] (or [2]) is valid memory?
Hmm.. if it ends at ext[0] or ext[1], it should be checked in the first check above. And I check ext[2] first, and then access ext[3] and so on.
Thanks, Namhyung
| |