Tag Archives: private bytes

Finding the largest free block of address space

BuildingsI’ve been seeing problems recently with fragmented virtual address space. During the lifetime of a process, bits and pieces of memory are allocated throughout the 2GB 32-bit address space to such an extent that large contiguous blocks of free space are no longer available. If anything subsequently requires a large block of memory (like, for instance, a somewhat out-of-date version of the GHC runtime), it will fail to get it.

It’s obvious looking at the output from VMmap or windbg’s !address command what the largest contiguous block is, e.g.

0:008> !address -summary
Largest free region: Base 07300000 - Size 63ed0000 (1637184 KB)

But what if you need that number in order to make a decision at run-time? For instance, to decide whether your process is in a fit state to continue, or if it should instead commit hara-kiri. In that case, you need to access the information programmatically. That’s where the immensely useful VirtualQueryEx function comes in…
Continue reading Finding the largest free block of address space

Diagnosing out of memory errors with VMMap

The other day a colleague pointed me to a new tool from Mark Russinovich et al (ex-SysInternals) that turns out to be very useful for diagnosing virtual memory/address space exhaustion issues. I thought I’d describe it here, and give some – hopefully useful – extra information on what it reports.

(I had problems with WordPress choking on this long post, so I’ve split it into 2 parts. This the first part, the second part is here).

Continue reading Diagnosing out of memory errors with VMMap

Programmatically checking memory usage

One of the things that’s useful in a pre-release check is do a regression test on the memory usage of your unmanaged functions. This should help to ensure that the fantastic new data structure you introduced doesn’t cost too much in additional storage for the order-of-magnitude performance improvement you were boasting about.

Like most of my posts, this assumes that it’s not feasible to go through all your source code, and say, replace all instances of new with a version that tracks usage (the approach used by the debug CRT). As well as being logistically infeasible, this also tends to miss allocations that don’t go via new, for example, direct calls to HeapAlloc.

Continue reading Programmatically checking memory usage