prasun3

I am seeing this issue as well. Looking at the code, I found that this error being hit in pt_insn_event ``` /* We must currently process an event. */ if...

I'm not sure what the fix or even the root cause is. Hoping this provides some insight to others more familiar with the code.

That is right. We see pretty large trace sizes and run into disk space issues.

> Or you would prefer this feature as filed to trace a certain number of instructions without regard to any corresponding application phases or code boundaries? We would prefer this...

Reading some of the stuff had me wondering: with the bbdup mechanism would it possible to implement switching between different tracers too? For example, we could collect an "L0_filter" trace...

Could thread preemption cause large delays? Or would that be rare since we are going into the kernel for IO frequently? For `-L0_filter` traces we could have larger delays since...

I was also trying to use the timestamps to "visualize" the traces but we no longer have the information on gaps. Suppose a thread tid0 was running on CPUx and...

Noting two use cases which could be handled differently: - DGC code where there is no binary: in this case the opcodes need to be traced at runtime - Using...

Purely software based approach (like some prior academic work) not being considered? https://sorav.compiler.ai/pubs/btkernel.pdf http://www.cs.toronto.edu/~peter/feiner_asplos_2012.pdf

Yes, for offline traces we will have separate files. I have been focusing on offline traces and hadn't thought about online trace. I think the marker idea sounds good --...