Messages in this thread | | | Subject | Re: [PATCH v2 1/3] perf util: Move block_pair_cmp to block-info | From | "Jin, Yao" <> | Date | Tue, 7 Jan 2020 21:57:56 +0800 |
| |
On 1/7/2020 5:57 PM, Jiri Olsa wrote: > On Tue, Jan 07, 2020 at 03:45:23AM +0800, Jin Yao wrote: >> block_pair_cmp() is a function which is used to compare >> two blocks. Moving it from builtin-diff.c to block-info.c >> to let it be used by other builtins. >> >> In block_pair_cmp, there is a minor change. It checks valid >> for map, dso and sym first. If they are invalid, we will not >> compare the address because the address might not make sense. > > please separate the change as well, it's hard to track > what you did when the whole function is moved >
Got it, thanks! I will separate this too.
>> >> v2: >> --- >> New patch created in v2 >> >> Signed-off-by: Jin Yao <yao.jin@linux.intel.com> >> --- >> tools/perf/builtin-diff.c | 17 ----------------- >> tools/perf/util/block-info.c | 23 +++++++++++++++++++++++ >> tools/perf/util/block-info.h | 2 ++ >> 3 files changed, 25 insertions(+), 17 deletions(-) >> >> diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c >> index f8b6ae557d8b..5ff1e21082cb 100644 >> --- a/tools/perf/builtin-diff.c >> +++ b/tools/perf/builtin-diff.c >> @@ -572,23 +572,6 @@ static void init_block_hist(struct block_hist *bh) >> bh->valid = true; >> } >> >> -static int block_pair_cmp(struct hist_entry *a, struct hist_entry *b) >> -{ >> - struct block_info *bi_a = a->block_info; >> - struct block_info *bi_b = b->block_info; >> - int cmp; >> - >> - if (!bi_a->sym || !bi_b->sym) >> - return -1; >> - >> - cmp = strcmp(bi_a->sym->name, bi_b->sym->name); >> - >> - if ((!cmp) && (bi_a->start == bi_b->start) && (bi_a->end == bi_b->end)) >> - return 0; >> - >> - return -1; >> -} >> - >> static struct hist_entry *get_block_pair(struct hist_entry *he, >> struct hists *hists_pair) >> { >> diff --git a/tools/perf/util/block-info.c b/tools/perf/util/block-info.c >> index c4b030bf6ec2..18a445938681 100644 >> --- a/tools/perf/util/block-info.c >> +++ b/tools/perf/util/block-info.c >> @@ -475,3 +475,26 @@ float block_info__total_cycles_percent(struct hist_entry *he) >> >> return 0.0; >> } >> + >> +int block_pair_cmp(struct hist_entry *pair, struct hist_entry *he) >> +{ >> + struct block_info *bi_p = pair->block_info; >> + struct block_info *bi_h = he->block_info; >> + struct map_symbol *ms_p = &pair->ms; >> + struct map_symbol *ms_h = &he->ms; >> + int cmp; >> + >> + if (!ms_p->map || !ms_p->map->dso || !ms_p->sym || >> + !ms_h->map || !ms_h->map->dso || !ms_h->sym) { >> + return -1; >> + } >> + >> + cmp = strcmp(ms_p->sym->name, ms_h->sym->name); >> + if (cmp) >> + return -1; > > should this return cmp? also you don't mention this change in the changelog >
Yes, return cmp should be OK.
It's changed from "strcmp(bi_a->sym->name, bi_b->sym->name)" to "strcmp(ms_p->sym->name, ms_h->sym->name)" is because we don't need an additional checking for bi_a->sym and bi_b->sym.
If we use "strcmp(bi_a->sym->name, bi_b->sym->name)" here, I think we'd better check the sym first. I will mention that in changelog.
Thanks Jin Yao
> thanks, > jirka >
| |