Driver is invalidating malloc'd memory 4.0RC2 Win32

Accelerated Computing CUDA CUDA Programming and Performance
1

This bug is erratic so an isolated case is not possible, but clearly there is a big problem with unified memory in the driver.

If I malloc a block using regular c malloc, it runs fins for a while then all of a sudden, writing to the middle of that block errors out that it is an invalid address. When I view memory, sure enough there is a big block right in the middle of the block that is invalid. Not garbage, but completely invalid address so even the debugger can’t view it. This is right in the middle of a regular host memory block via malloc.

I am using lots of page locked memory but shrank that down to no avail. All cudaHostMalloc are fine, as are cudaMallocs, and run fine. No kernel errors, no additional mallocs.

Any ideas anyone?

2

I am very hesitant to blame this on the CUDA driver without a repro case. (in other words, the problem is almost certainly in your code–we’ve never seen any sort of issues like this with unified addressing, and we’ve been using it for quite a while)

also, unified addressing is only active on TCC, XP, and Linux on 64-bit OSes and Fermi. if one of those is missing, you’re not using unified addressing at all