Re: Predict in ASReml-R leads to 'memory not mapped' crash

From: Butler, David <David.Butler_at_DEEDI.QLD.GOV.AU>
Date: Thu, 4 Feb 2010 12:23:02 +1000

Marc, As a workaround add "pworkspace=nnn" to the call where nnn is the amount of memory to reserve in double precision words (groups of 8 bytes). The default is 8e6 words or 64e6 bytes (approx 64 mb). This may involve a bit of experimentation with increasingly larger values. Eg, "pworkspace=32e6" will reserve approximately 256Mb for prediction. Note that there is a companion "workspace" argument that controls the amount of memory available for the fitting process. My apologies for this but this is a persistent bug that has proved very difficult to pin down. Dave. ________________________________ From: ASReml users discussion group [mailto:ASREML-L@DPI.NSW.GOV.AU] On Behalf Of moragues Sent: Thursday, 4 February 2010 10:23 AM To: ASREML-L@DPI.NSW.GOV.AU Subject: Predict in ASReml-R leads to 'memory not mapped' crash Hi, When I try to predict means of an asreml object using the predict() function, R crashes with the following message: > oats.pv <- predict(,classify="Variety:Nitrogen", + sed=list("Variety:Nitrogen"=T)) asreml(): 3.00 Library: 3.01fe IA32 Run: Wed Feb 3 17:09:36 2010 Equations: 44 (20 dense) Initial update shrinkage factor: 0.316 *** caught segfault *** address 0xc06f3568, cause 'memory not mapped' Traceback: 1: .Call("casr", asr.vec, iarg, dpmv, asr.inter, asr.struc, asr.struc$ainv, asr.glm, asr.predict, constr, mbfList, control, aiopt, asr) 2:, asr.inter, asr.struc, asr.glm, asr.predict$predict, asr.predict, weights, offset, multivariate, ntrt, control, keep, constraints, dpmv, debug) 3: asreml(fixed = yield ~ Variety + Nitrogen + Variety:Nitrogen, random = ~Blocks + Blocks:Wplots, data = oats, predict = prd, R.param = object$R.param, G.param = object$G.param) 4: eval(expr, envir, enclos) 5: eval(newcall, sys.parent()) 6: update.asreml(object, predict = prd, ...) 7: update(object, predict = prd, ...) 8: predict.asreml(, classify = "Variety:Nitrogen", sed = list(`Variety:Nitrogen` = T)) 9: predict(, classify = "Variety:Nitrogen", sed = list(`Variety:Nitrogen` = T)) Possible actions: 1: abort (with core dump, if enabled) 2: normal R exit 3: exit R without saving workspace 4: exit R saving workspace Selection: The example above is using the sample data "oats" that comes with the ASReml-R package but I get the same result with my own data. Normally 'memory not mapped' means that there is not enough main memory available in R, but the dataset is not that big and I can do the analysis in R without any problem. Can anybody help me figuring out this? Thanks, Marc. > sessionInfo() R version 2.10.1 (2009-12-14) i486-pc-linux-gnu locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=C 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 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] lattice_0.18-3 asreml_3.01 loaded via a namespace (and not attached): [1] grid_2.10.1[/code] -------------------- m2f -------------------- Sent using Mail2Forum ( Read this topic online here: -------------------- m2f -------------------- This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation. ********************************DISCLAIMER**************************** The information contained in the above e-mail message or messages (which includes any attachments) is confidential and may be legally privileged. It is intended only for the use of the person or entity to which it is addressed. If you are not the addressee any form of disclosure, copying, modification, distribution or any action taken or omitted in reliance on the information is unauthorised. Opinions contained in the message(s) do not necessarily reflect the opinions of the Queensland Government and its authorities. If you received this communication in error, please notify the sender immediately and delete it from your computer system network.
Received on Thu Feb 04 2010 - 12:23:02 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)