RFR: JDK-8317683: Add JIT memory statistics

[Vladimir Kozlov]

On Fri, 6 Oct 2023 17:31:21 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

> This proposal adds a way to monitor C1/C2 memory usage per compilation. 
> 
> We are flying a bit blind there: NMT is okay for initial triaging but no help when digging deeper into compiler footprint. NMT shows how much native memory the JIT uses but lumps together footprints from concurrent compilations, making its peak values arbitrary. And we have no way to associate the NMT numbers with individual allocations.
> 
> The statistic added by this patch shows footprint *per compilation*. It measures memory spikes per thread for a single compilation. It then presents them with additional information, e.g., the node count (for C2). 
> 
> Example printout:
> 
> 
> 112703:
> Compilation memory statistics
> 
> Legend:
>   total  : memory allocated via arenas while compiling
>   NA     : ...how much in node arenas (if c2)
>   RA     : ...how much in resource areas
>   #nodes : ...how many nodes (if c2)
>   time   : time of last compilation (sec)
>   type   : compiler type
>   #rc    : how often recompiled
>   thread : compiler thread
> 
> total     NA        RA        #nodes  time    type  #rc thread              method
> 30273336  6383040   18027000  18257   1,979   c2    2   0x00007f6f0c136460  java/lang/ClassLoader.loadClass((Ljava/lang/String;)Ljava/lang/Class;) 
> 27690456  5892160   16211800  15179   3,643   c2    2   0x00007f6f0c0433e0  org/springframework/cache/interceptor/CacheOperationSourcePointcut.matches((Ljava/lang/reflect/Method;Ljava/lang/Class;)Z) 
> ...
> ...
> 
> 
> [full example printout here](https://github.com/openjdk/jdk/files/12853477/compiler-statistic-petclinic.txt)
> 
> #### Usage
> 
> The statistic needs to be enabled with a new diagnostic switch (`-XX:+CompilationMemStat`) and then can be printed with a new jcmd:
> 
> `jcmd xxx Compiler.memory`
> 
> You can specify a cutoff memory threshold to filter only the most expensive compilations, and switch between human-readable format:
> 
> `jcmd xxx Compiler.memory -s=10m -H`
> 
> The statistic can also be printed at the end of VM life (similar to metaspace- or NMT-statistics):
> 
> `-XX:+PrintCompilationMemStatAtExit`
> 
> #### Technical details
> 
> The patch exploits the fact that compilers use arenas almost exclusively. It hooks into `Arena::grow()` to track allocations. That makes memory tracking very cheap at an acceptable loss in fidelity (we only track arena chunk allocations, not individual Arena allocations).
> 
> Peak values are tracked per compiler thread. That works since all arenas used during compilation are associated with a compiler thread and never change that associat...

I think you should allow to collect and print memory stats per method using `CompileCommand=option,<method>,PrintMemStat`
Also use `DirectivesStack::getMatchingDirective(method, comp);` to find when you need to collect memory statistic.

src/hotspot/share/compiler/compiler_globals.hpp line 386:

> 384:                                                                             \
> 385:   product(bool, PrintCompilationMemStatAtExit, false, DIAGNOSTIC,           \
> 386:           "Print compilation memory statistics at VM exit")                 \

Is there reason to have 2 flags? Also this is compilers flags so you don't need `Compilation` word.
I think `PrintMemStat` flag should be enough. JCMD can work through Directives and check I suggested.

-------------

PR Review: https://git.openjdk.org/jdk/pull/16076#pullrequestreview-1668459461
PR Review Comment: https://git.openjdk.org/jdk/pull/16076#discussion_r1353042330