On Sat, 10 May 2025 16:48:16 +0200
Lorenzo Sostero <lorenzo.sost...@unibs.it> wrote:
> Flavor: r-devel-windows-x86_64
> Check: tests, Result: ERROR
> Running 'testthat.R'
> Running the tests in 'tests/testthat.R' failed.
> Complete output:
> > library(testthat)
> > library(EmiR)
> > test_check("EmiR")
> 0% 10 20 30 40 50 60 70 80 90 100%
> [----|----|----|----|----|----|----|----|----|----|
The process crashes trying to access the contents of a zero-length
vector:
Thread 1 received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1564.0x634]
0x00007ff81e1409df in PSPopulation::init (this=this@entry=0xf243ff59d0)
(gdb) info locals
j = 0
i = 0
progress_bar = <optimized out>
delta = 10.4
pop_size = <optimized out>
d = 4
alpha = {<Rcpp::PreserveStorage<Rcpp::Vector<14,
Rcpp::PreserveStorage>>> = {
data = 0x2d9ce6732e8, token = 0x2d9ce66d5c0}, <...>,
cache = {start = 0x1, size = 0}}
(gdb) frame 1
#1 0x00007ff81e0f7b4c in ____C_A_T_C_H____T_E_S_T____0 ()
at test-example.cpp:106
106 pop_noseed1.init();
(gdb) p (R_xlen_t)XLENGTH(pop_noseed1.m_config.m_alpha.data)
$7 = 0
(gdb) p (double*)REAL(pop_noseed1.m_config.m_alpha.data)
$8 = (double *) 0x1
This should be reproducible on R-4.5.0 or R-devel without extra
configuration flags. Previously this used to read some other memory
that didn't belong to the vector; now the data pointers of such vectors
are invalid, so dereferencing them causes a crash:
https://stat.ethz.ch/pipermail/r-devel/2025-April/083994.html
For your future questions: when asking for help with code, it's best to
provide a link to the code in question. This time, it was possible to
find the package at <https://cran.r-project.org/incoming/archive/>.
--
Best regards,
Ivan
______________________________________________
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
