# [FYI PATCH] t/helper/test-lazy-name-hash: fix compilation 7 messages from 2017-12-18 to 2017-12-22. Participants: Stefan Beller, Jonathan Tan, Jeff Hostetler, Junio C Hamano. Thread: https://gitlist.dev/t/47460 ## Stefan Beller, 2017-12-18 21:49 Subject: [FYI PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: <20171218214947.210042-1-sbeller@google.com> URL: https://gitlist.dev/e/20171218214947.210042-1-sbeller%40google.com ``` I was compiling origin/master today with stricter compiler flags today and was greeted by t/helper/test-lazy-init-name-hash.c: In function ‘cmd_main’: t/helper/test-lazy-init-name-hash.c:172:5: error: ‘nr_threads_used’ may be used uninitialized in this function [-Werror=maybe-uninitialized] printf("avg [size %8d] [single %f] %c [multi %f %d]\n", ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nr, ~~~ (double)avg_single/1000000000, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (avg_single < avg_multi ? '<' : '>'), ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (double)avg_multi/1000000000, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nr_threads_used); ~~~~~~~~~~~~~~~~ t/helper/test-lazy-init-name-hash.c:115:6: note: ‘nr_threads_used’ was declared here int nr_threads_used; ^~~~~~~~~~~~~~~ I do not see how we can arrive at that line without having `nr_threads_used` initialized, as we'd have `count > 1` (which asserts that we ran the loop above at least once, such that it *should* be initialized). I do not have time to dive into further analysis. Signed-off-by: Stefan Beller --- t/helper/test-lazy-init-name-hash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/helper/test-lazy-init-name-hash.c b/t/helper/test-lazy-init-name-hash.c index 6368a89345..297fb01d61 100644 --- a/t/helper/test-lazy-init-name-hash.c +++ b/t/helper/test-lazy-init-name-hash.c @@ -112,7 +112,7 @@ static void analyze_run(void) { uint64_t t1s, t1m, t2s, t2m; int cache_nr_limit; - int nr_threads_used; + int nr_threads_used = 0; int i; int nr; -- 2.15.1.620.gb9897f4670-goog ``` ## Jonathan Tan, 2017-12-18 23:43 Subject: Re: [FYI PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: <20171218154310.aa62ea2c9f094b663393556e@google.com> URL: https://gitlist.dev/e/20171218154310.aa62ea2c9f094b663393556e%40google.com In-Reply-To: <20171218214947.210042-1-sbeller@google.com> ``` On Mon, 18 Dec 2017 13:49:47 -0800 Stefan Beller wrote: > I was compiling origin/master today with stricter compiler flags today > and was greeted by > > t/helper/test-lazy-init-name-hash.c: In function ‘cmd_main’: > t/helper/test-lazy-init-name-hash.c:172:5: error: ‘nr_threads_used’ may be used uninitialized in this function [-Werror=maybe-uninitialized] > printf("avg [size %8d] [single %f] %c [multi %f %d]\n", > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr, > ~~~ > (double)avg_single/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (avg_single < avg_multi ? '<' : '>'), > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (double)avg_multi/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr_threads_used); > ~~~~~~~~~~~~~~~~ > t/helper/test-lazy-init-name-hash.c:115:6: note: ‘nr_threads_used’ was declared here > int nr_threads_used; > ^~~~~~~~~~~~~~~ > > I do not see how we can arrive at that line without having `nr_threads_used` > initialized, as we'd have `count > 1` (which asserts that we ran the > loop above at least once, such that it *should* be initialized). Your analysis makes sense. (The compiler probably couldn't detect it because "count" is a static variable, not a local variable.) > --- a/t/helper/test-lazy-init-name-hash.c > +++ b/t/helper/test-lazy-init-name-hash.c > @@ -112,7 +112,7 @@ static void analyze_run(void) > { > uint64_t t1s, t1m, t2s, t2m; > int cache_nr_limit; > - int nr_threads_used; > + int nr_threads_used = 0; > int i; > int nr; I agree that this is probably the best way to fix it. Another way might be to omit printing out the number of threads used in the printf that prints the average statistics. The best way is probably to not use so many global variables, but that is out of scope of this change. ``` ## Jeff Hostetler, 2017-12-19 14:19 Subject: Re: [FYI PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: <39238da4-c2b8-050d-51e3-675f5515776e@jeffhostetler.com> URL: https://gitlist.dev/e/39238da4-c2b8-050d-51e3-675f5515776e%40jeffhostetler.com In-Reply-To: <20171218214947.210042-1-sbeller@google.com> ``` On 12/18/2017 4:49 PM, Stefan Beller wrote: > I was compiling origin/master today with stricter compiler flags today > and was greeted by > > t/helper/test-lazy-init-name-hash.c: In function ‘cmd_main’: > t/helper/test-lazy-init-name-hash.c:172:5: error: ‘nr_threads_used’ may be used uninitialized in this function [-Werror=maybe-uninitialized] > printf("avg [size %8d] [single %f] %c [multi %f %d]\n", > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr, > ~~~ > (double)avg_single/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (avg_single < avg_multi ? '<' : '>'), > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (double)avg_multi/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr_threads_used); > ~~~~~~~~~~~~~~~~ > t/helper/test-lazy-init-name-hash.c:115:6: note: ‘nr_threads_used’ was declared here > int nr_threads_used; > ^~~~~~~~~~~~~~~ > > I do not see how we can arrive at that line without having `nr_threads_used` > initialized, as we'd have `count > 1` (which asserts that we ran the > loop above at least once, such that it *should* be initialized). > > I do not have time to dive into further analysis. > > Signed-off-by: Stefan Beller > --- > t/helper/test-lazy-init-name-hash.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/t/helper/test-lazy-init-name-hash.c b/t/helper/test-lazy-init-name-hash.c > index 6368a89345..297fb01d61 100644 > --- a/t/helper/test-lazy-init-name-hash.c > +++ b/t/helper/test-lazy-init-name-hash.c > @@ -112,7 +112,7 @@ static void analyze_run(void) > { > uint64_t t1s, t1m, t2s, t2m; > int cache_nr_limit; > - int nr_threads_used; > + int nr_threads_used = 0; > int i; > int nr; Agreed. It should not be a problem. Explicitly initializing it is fine. Signed-off-by: Jeff Hostetler ``` ## Stefan Beller, 2017-12-20 22:24 Subject: [PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: <20171220222416.130872-1-sbeller@google.com> URL: https://gitlist.dev/e/20171220222416.130872-1-sbeller%40google.com In-Reply-To: <39238da4-c2b8-050d-51e3-675f5515776e@jeffhostetler.com> ``` I was compiling origin/master today with the DEVELOPER compiler flags today and was greeted by t/helper/test-lazy-init-name-hash.c: In function ‘cmd_main’: t/helper/test-lazy-init-name-hash.c:172:5: error: ‘nr_threads_used’ may be used uninitialized in this function [-Werror=maybe-uninitialized] printf("avg [size %8d] [single %f] %c [multi %f %d]\n", ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nr, ~~~ (double)avg_single/1000000000, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (avg_single < avg_multi ? '<' : '>'), ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (double)avg_multi/1000000000, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nr_threads_used); ~~~~~~~~~~~~~~~~ t/helper/test-lazy-init-name-hash.c:115:6: note: ‘nr_threads_used’ was declared here int nr_threads_used; ^~~~~~~~~~~~~~~ Fix this issue by assigning 0 to 'nr_threads_used'. Acked-by: Jonathan Tan Signed-off-by: Jeff Hostetler Signed-off-by: Stefan Beller --- Slightly reworded the commit message. I'd really like this patch to be included such that I can compile git with the DEVELOPER_CFLAGS flags. t/helper/test-lazy-init-name-hash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/helper/test-lazy-init-name-hash.c b/t/helper/test-lazy-init-name-hash.c index 6368a89345..297fb01d61 100644 --- a/t/helper/test-lazy-init-name-hash.c +++ b/t/helper/test-lazy-init-name-hash.c @@ -112,7 +112,7 @@ static void analyze_run(void) { uint64_t t1s, t1m, t2s, t2m; int cache_nr_limit; - int nr_threads_used; + int nr_threads_used = 0; int i; int nr; -- 2.15.1.620.gb9897f4670-goog ``` ## Junio C Hamano, 2017-12-22 18:15 Subject: Re: [PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: URL: https://gitlist.dev/e/xmqq7etetyqo.fsf%40gitster.mtv.corp.google.com In-Reply-To: <20171220222416.130872-1-sbeller@google.com> ``` Stefan Beller writes: > I was compiling origin/master today with the DEVELOPER compiler flags > today and was greeted by > > t/helper/test-lazy-init-name-hash.c: In function ‘cmd_main’: > t/helper/test-lazy-init-name-hash.c:172:5: error: ‘nr_threads_used’ may be used uninitialized in this function [-Werror=maybe-uninitialized] > printf("avg [size %8d] [single %f] %c [multi %f %d]\n", > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr, > ~~~ > (double)avg_single/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (avg_single < avg_multi ? '<' : '>'), > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > (double)avg_multi/1000000000, > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > nr_threads_used); > ~~~~~~~~~~~~~~~~ > t/helper/test-lazy-init-name-hash.c:115:6: note: ‘nr_threads_used’ was declared here > int nr_threads_used; > ^~~~~~~~~~~~~~~ > > Fix this issue by assigning 0 to 'nr_threads_used'. > > Acked-by: Jonathan Tan > Signed-off-by: Jeff Hostetler > Signed-off-by: Stefan Beller > --- > > Slightly reworded the commit message. I'd really like this patch to be included > such that I can compile git with the DEVELOPER_CFLAGS flags. Heh; I do not think there particularly is much difference between stricter flags and DEVELOPER flags, but I would rather not lose the removal of duplicated 'today' I did while I queued the previous one ;-) > > t/helper/test-lazy-init-name-hash.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/t/helper/test-lazy-init-name-hash.c b/t/helper/test-lazy-init-name-hash.c > index 6368a89345..297fb01d61 100644 > --- a/t/helper/test-lazy-init-name-hash.c > +++ b/t/helper/test-lazy-init-name-hash.c > @@ -112,7 +112,7 @@ static void analyze_run(void) > { > uint64_t t1s, t1m, t2s, t2m; > int cache_nr_limit; > - int nr_threads_used; > + int nr_threads_used = 0; > int i; > int nr; ``` ## Stefan Beller, 2017-12-22 18:17 Subject: Re: [PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: URL: https://gitlist.dev/e/CAGZ79kZ9rWYwzrG%2BfJu2neYVfHSgoR5TrkSmNf_EJeUm46QUvg%40mail.gmail.com In-Reply-To: ``` > Heh; I do not think there particularly is much difference between > stricter flags and DEVELOPER flags, but I would rather not lose the > removal of duplicated 'today' I did while I queued the previous one > ;-) I did not realize it was queued anywhere, please ignore then. ``` ## Junio C Hamano, 2017-12-22 20:02 Subject: Re: [PATCH] t/helper/test-lazy-name-hash: fix compilation Message-ID: URL: https://gitlist.dev/e/xmqqtvwir0nz.fsf%40gitster.mtv.corp.google.com In-Reply-To: ``` Stefan Beller writes: >> Heh; I do not think there particularly is much difference between >> stricter flags and DEVELOPER flags, but I would rather not lose the >> removal of duplicated 'today' I did while I queued the previous one >> ;-) > > > I did not realize it was queued anywhere, please ignore then. I just did s/stricter/DEVELOPER/ on the one that have been queued. Thanks. ```