Pierre le Riche
Pierre le Riche
_But why did I not need this in the past (old FastMM5)? Simple EnterDebugMode did the job?_ My guess would be that you had some of the FastMM4 defines set...
Hi Vladimir, Thank you for the feedback. With regards to the scenario where FastMM5 is slower, I am considering adding support for having an arena affinity per thread, so that...
I suspect it might be due to cache thrashing. Could you please try forcing 64-byte alignment by calling FastMM_EnterMinimumAddressAlignment(maa64Bytes)? If that improves it then an arena affinity per thread will...
Hi Stephane, Thank you for the suggestion. It is something that has been requested before, so clearly there is a demand for it. There is also a bit of functional...
This is a known issue: https://quality.embarcadero.com/browse/RSP-16796
Hi, If the A/V is handled then it is most likely the stack tracing code, and is nothing to worry about. The stack tracing code maintains a map of memory...
Hi Shane, Unfortunately C++ Builder is not my area of expertise, so I relied on others to implement and maintain the C++ Builder support in version 4. I might take...
Hi Olivier, Unfortunately there is no such option currently. Preference is given to the reuse blocks that are "hot" in the cache - also in debug mode. When FreeMem is...
I'll see what I can do.
Hi Primoz, Thanks for the contribution. Just two issues I have picked up (please correct me if I am wrong): 1) It appears that the new Start/StopRegistering... calls are also...