Numpy 1.1.0 (+ PIL 1.1.6) crashes on large datasets
Hi, I've managed to crash numpy+PIL when feeding it rather large images. Please see the URL for a test image, script, and gdb stack trace. This crashes on my box (Windows XP SP3) as well as on a linux box (the gdb trace I've been provided with) and a Mac. Windows reports the crash to happen in "multiarray.pyd"; the stack trace mentions the equivalent file. Unfortunately, I don't know how to fix this. Can I help somehow? -- Chris [1] http://cracki.ath.cx:10081/pub/numpy-pil-crash/
I've managed to crash numpy+PIL when feeding it rather large images. Please see the URL for a test image, script, and gdb stack trace. This crashes on my box (Windows XP SP3) as well as on a linux box (the gdb trace I've been provided with) and a Mac. Windows reports the crash to happen in "multiarray.pyd"; the stack trace mentions the equivalent file. Unfortunately, I don't know how to fix this. Can I help somehow?
-- Chris
Hmm... I've opened this file with my homebrew image-IO tools, and I cannot provoke a segfault. (These tools are derived from PIL, but with many bugs related to the array interface fixed. I had submitted patches to the PIL mailing list, which, as usual, languished.) I wonder if the issue is with how the PIL is providing the buffer interface to numpy? Can you get the crash if you get the array into numpy through the image's tostring (or whatever) method, and then use numpy.fromstring? Zach PS. This is with a recent SVN version of numpy, on OS X 10.5.4.
participants (2)
-
Christoph Rackwitz
-
Zachary Pincus