[Rd] Difficult debug

Therneau, Terry M., Ph.D. therne@u @end|ng |rom m@yo@edu
Wed Feb 7 21:01:44 CET 2024


  I've hit a roadblock debugging a new update to the survival package.   I do debugging in 
a developement envinment, i.e. I don't create and load a package but rather  source all 
the .R files and dyn.load an .so file, which makes things a bit easier.

   Running with R -d "valgrind --tool=memcheck --leak-check=full" one of my test files 
crashes in simple R code a dozen lines after the  likely culprit has been called, i.e, the 
survfit function for an Aalen-Johansen, containing a .Call to the new C code.     The 
valgrind approach had already allowed me to find a few other (mostly dumb) errors that led 
to an out of bounds access, e.g., the wrong endpoint variable in a for( ) loop.    What 
would others advise as a next step?

Here is the last part of the screen
 > fit2 <- coxph(list(Surv(tstart, tstop, bstat) ~ 1,
+                    c(1:4):5 ~ age / common + shared), id= id, istate=bili4,
+               data=pbc2, ties='breslow', x=TRUE)
 > surv2 <- survfit(fit2, newdata=list(age=50), p0=c(.4, .3, .2, .1, 0))
 > test2 <- mysurv(fit2, pbc2$bili4, p0= 4:0/10, fit2, x0 =50)
==31730== Invalid read of size 8
==31730==    at 0x298A07: Rf_allocVector3 (memory.c:2861)
==31730==    by 0x299B2C: Rf_allocVector (Rinlinedfuns.h:595)
==31730==    by 0x299B2C: R_alloc (memory.c:2330)
==31730==    by 0x3243C6: do_which (summary.c:1152)
==31730==    by 0x23D8EF: bcEval (eval.c:7724)
==31730==    by 0x25731F: Rf_eval (eval.c:1152)
==31730==    by 0x25927D: R_execClosure (eval.c:2362)
==31730==    by 0x25A35A: R_execMethod (eval.c:2535)
==31730==    by 0x887E93F: R_dispatchGeneric (methods_list_dispatch.c:1151)
==31730==    by 0x2A0E72: do_standardGeneric (objects.c:1344)
==31730==    by 0x2577E7: Rf_eval (eval.c:1254)
==31730==    by 0x25927D: R_execClosure (eval.c:2362)
==31730==    by 0x25A01C: applyClosure_core (eval.c:2250)
==31730==  Address 0x10 is not stack'd, malloc'd or (recently) free'd
==31730==

  *** caught segfault ***
address 0x10, cause 'memory not mapped'

Traceback:
  1: which(smap == j)
  2: which(smap == j)
  3: mysurv(fit2, pbc2$bili4, p0 = 4:0/10, fit2, x0 = 50)

The offending call is amost certainly the one to survfit; mysurv() is a local function 
that caculates some things 'by hand'.   It does nothing complex: counts, loops, etc, the 
only non-base action is a call to Matrix::exp near the end, but the which() failure is 
well before that.

The session info just before the offending material:

 > sessionInfo()
R Under development (unstable) (2024-02-07 r85873)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 22.04.3 LTS

Matrix products: default
BLAS:   /usr/local/src/R-devel/lib/libRblas.so
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0

locale:
  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=C
  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
  [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

time zone: America/Chicago
tzcode source: system (glibc)

attached base packages:
[1] splines   stats     graphics  grDevices utils     datasets methods
[8] base

other attached packages:
[1] Matrix_1.6-0

loaded via a namespace (and not attached):
[1] compiler_4.4.0 tools_4.4.0    grid_4.4.0     lattice_0.22-5


---
Footnote.  The impetus for this is realizing that the robust variance for an 
Aalen-Johansen was incorrect when there are case weights for a subject that vary over 
time;  a rare case but will occur with time dependent IPC weights.  Carefully figuring 
this out has been all I did for the last week, leading to a new routine survfitaj.c and 
approx 14 pages of derivation and explanation in the methods.Rnw vignette.   Subjects who 
"change horses in midstream", i.e., swap from one curve to another mid-followup make the 
code more complex.   This arises out of the "extended Kaplan-Meier"; I am not a fan of 
this statistically, but some will use it and expect my code to work.

-- 
Terry M Therneau, PhD
Department of Quantitative Health Sciences
Mayo Clinic
therneau using mayo.edu

"TERR-ree THUR-noh"

	[[alternative HTML version deleted]]



More information about the R-devel mailing list