control tag -1 + patch

Hi,

On 2026-02-23 13:44, Emanuele Rocca wrote:
> Source: scamp
> Version: 2.14.0-1
> Severity: important
> Tags: ftbfs upstream
> Justification: fails to build from source
> User: [email protected]
> Usertags: glibc-2.43
> 
> Hi,
> 
> scamp fails to build from source on arm64 when using glibc 2.43,
> currently in experimental.
> 
> The issue is triggered by glibc 2.43 on arm64 enabling 2MB THP by
> default:
> https://sourceware.org/git/?p=glibc.git;a=commit;h=321e1fc73f53081d92ba357cdd48c56b79292020
> 
> The problem is however not architecture-specific. For example it can be
> reproduced with glibc 2.42 on amd64 by building the package with the
> GLIBC_TUNABLES environment variable set to glibc.malloc.hugetlb=2.
> 
> Successful build with glibc 2.42, currently in sid:
> https://people.debian.org/~ema/glibc-2.43-rebuilds/output-2/scamp_arm64.build
> 
> Logs of a failed build with glibc 2.43 are here:
> https://people.debian.org/~ema/glibc-2.43-rebuilds/output-1/scamp_arm64.build
> 
> Salient part:
> 
> make[4]: Entering directory '/build/reproducible-path/scamp-2.14.0/tests'
> PASS: test_chealpixstore
> PASS: test_chealpix
> ../autoconf/test-driver: line 119:  3109 Aborted                    "$@" >> 
> "$log_file" 2>&1
> FAIL: test_crossid_single_catalog
> ../autoconf/test-driver: line 119:  3110 Aborted                    "$@" >> 
> "$log_file" 2>&1
> FAIL: test_crossid_single_catalog_moving
> [...]
> 
> FAIL: test_crossid_single_catalog
> =================================
> 
> test_crossid_single_catalog: test_crossid_single_catalog.c:87: 
> compare_fields: Assertion `linkedspl[0]->prevsamp == NULL' failed.
> FAIL test_crossid_single_catalog (exit status: 134)
> 
> FAIL: test_crossid_single_catalog_moving
> ========================================
> 
> test_crossid_single_catalog_moving: test_crossid_single_catalog_moving.c:86: 
> compare_fields: Assertion `linkedspl[0]->prevsamp == NULL' failed.
> FAIL test_crossid_single_catalog_moving (exit status: 134)

It happens that the failure is actually reproducible by other means, 
like ASAN, and is due to non-initialized memory in the tests. Using THP 
memory means more chance to get uninitialized memory.

The following patch is enough for me to get the testsuite passing again:

--- scamp-2.14.0.orig/tests/test_crossid_single_catalog.c
+++ scamp-2.14.0/tests/test_crossid_single_catalog.c
@@ -28,6 +28,7 @@ static fieldstruct*
 load_ascii_field(int id)
 {
     fieldstruct *field = malloc(sizeof(fieldstruct));
+    memset(field, 0, sizeof(fieldstruct));
     field->lng = 0;
     field->lat = 1;
     field->fieldindex = id;
--- scamp-2.14.0.orig/tests/test_crossid_single_catalog_moving.c
+++ scamp-2.14.0/tests/test_crossid_single_catalog_moving.c
@@ -27,6 +27,7 @@ static fieldstruct*
 load_ascii_field(int id)
 {
     fieldstruct *field = malloc(sizeof(fieldstruct));
+    memset(field, 0, sizeof(fieldstruct));
     field->lng = 0;
     field->lat = 1;
     field->fieldindex = id;

Regards
Aurelien

-- 
Aurelien Jarno                          GPG: 4096R/1DDD8C9B
[email protected]                     http://aurel32.net

Reply via email to