RFR(L): 8198423: Improve metaspace chunk allocation (was: Proposal for improvements to the metaspace chunk allocator) (original) (raw)

Thomas Stüfe thomas.stuefe at gmail.com
Thu Mar 1 10:36:37 UTC 2018

Hi Coleen,

thanks a lot for the review and the sponsoring offer!

New version (full): http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalescation/2018-03-01/webrev-full/webrev/ incremental: http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalescation/2018-03-01/webrev-incr/webrev/

Please find remarks inline:

On Tue, Feb 27, 2018 at 11:22 PM, <coleen.phillimore at oracle.com> wrote:

Thomas, review comments: http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalesc ation/2018-02-26/webrev/src/hotspot/share/memory/metachunk.hpp.udiff.html +// ChunkIndex (todo: rename?) defines the type of chunk. Chunk types

It's really both, isn't it? The type is the index into the free list or in use lists. The name seems fine. You are right. What I meant was that a lot of code needs to know about the different chunk sizes, but naming it "Index" and adding enum values like "NumberOfFreeLists" we expose implementation details no-one outside of SpaceManager and ChunkManager cares about (namely, the fact that these values are internally used as indices into arrays). A more neutral naming would be something like "enum ChunkTypes { spec,small, .... , NumberOfNonHumongousChunkTypes, NumberOfChunkTypes }.

However, I can leave this out for a possible future cleanup. The change is big enough as it is.

Can you add comments on the #endifs if the #ifdef is more than a couple 2-3 lines above (it's a nit that bothers me).

+#ifdef ASSERT + // A 32bit sentinel for debugging purposes. +#define CHUNKSENTINEL 0x4d4554EF // "MET" +#define CHUNKSENTINELINVALID 0xFEEEEEEF + uint32t sentinel; +#endif + const ChunkIndex chunktype; + const bool isclass; + // Whether the chunk is free (in freelist) or in use by some class loader. bool istaggedfree; +#ifdef ASSERT + ChunkOrigin origin; + int usecount; +#endif + I removed the asserts completely, following your suggestion below that "origin" would be valuable in customer scenarios too. By that logic, the other members are valuable too: the sentinel is valuable when examining memory dumps to see the start of chunks, and the in-use counter is useful too. What do you think?

So, I leave the members in - which, depending what the C++ compiler does to enums and bools, may cost up to 128bit additional header space. I think that is ok. In one of my earlier versions of this patch I hand-crafted the header using chars and bitfields to be as small as possible, but that seemed over-engineered.

However, I left out any automatic verifications accessing these debug members. These are still only done in debug builds.

It seems that if you could move origin and usecount into the ASSERT block above (maybe putting usecount before origin. http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalesc ation/2018-02-26/webrev/src/hotspot/share/memory/metaspace.cpp.udiff.html In takefromcommitted, can the allocation of padding chunks be its own function like addchunkstoaligment() lines 1574-1615? The function is too long now. I moved the padding chunk allocation into an own function as you suggested.

I don't think coalescation is a word in English, at least my dictionary cannot find it. Although it makes sense in the context, just distracting.

I replaced "coalescation" with "chunk merging" throughout the code. Also less of a tongue breaker.

+ // Now check if in the coalescation area there are still life chunks.

"live" chunks I guess. A sentence you won't read often :).

Now that I read it it almost sounded sinister :) Fixed.

In freechunksget() can you handle the Humongous case first? The else for humongous chunk size is buried tons of lines below. Otherwise it might be helpful to the logic to make your addition to this function be a function you call like chunk = splitfromlargerfreechunk();

I did the latter. I moved the splitting of a larger chunk to an own function. This causes a slight logic change: the new function (ChunkManager::split_chunk()) splits an existing large free chunks into n smaller free chunks and adds them all back to the freelist - that includes the chunk we are about to return. That allows us to use the same exit path

To make the test more readable, I also remove the "test-that-free-chunks-are-optimally-merged" verification - which was quite lengthy - from VirtualSpaceNode::verify() to a new function, VirtualSpaceNode::verify_free_chunks_are_ideally_merged().

You might want to keep the origin in product mode if it doesn't add to the chunk footprint. Might help with customer debugging.

See above

Awesome looking test...

Thanks, I was worried it would be too complicated. I changed it a bit because there were sporadic errors. Not a "real" error, just the test itself was faulty. The "metaspaces_in_use" counter was slightly wrong in one corner case.

I've read through most of this and thank you for adding this to at least partially solve the fragmentation problem. The irony is that we templatized the Dictionary from CMS so that we could use it for Metaspace and that has splitting and coalescing but it seems this code makes more sense than adapting that code (if it's even possible).

Well, it helps other metadata use cases too, no.

Thank you for working on this. I'll sponsor this for you. Coleen

Thanks again!

I also updated my jdk-submit branch to include these latest changes; tests are still runnning.

Kind Regards, Thomas

On 2/26/18 9:20 AM, Thomas Stüfe wrote:

Hi all,

I know this patch is a bit larger, but may I please have reviews and/or other input? Issue: https://bugs.openjdk.java.net/browse/JDK-8198423 Latest version: http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalesc ation/2018-02-26/webrev/ For those who followed the mail thread, this is the incremental diff to the last changes (included feedback Goetz gave me on- and off-list): http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalesc ation/2018-02-26/webrev-incr/webrev/ Thank you! Kind Regards, Thomas Stuefe

On Thu, Feb 8, 2018 at 12:58 PM, Thomas Stüfe <thomas.stuefe at gmail.com> wrote: Hi, We would like to contribute a patch developed at SAP which has been live in our VM for some time. It improves the metaspace chunk allocation: reduces fragmentation and raises the chance of reusing free metaspace chunks. The patch: http://cr.openjdk.java.net/~stuefe/webrevs/metaspace-coalesc ation/2018-02-05--2/webrev/ In very short, this patch helps with a number of pathological cases where metaspace chunks are free but cannot be reused because they are of the wrong size. For example, the metaspace freelist could be full of small chunks, which would not be reusable if we need larger chunks. So, we could get metaspace OOMs even in situations where the metaspace was far from exhausted. Our patch adds the ability to split and merge metaspace chunks dynamically and thus remove the "size-lock-in" problem. Note that there have been other attempts to get a grip on this problem, see e.g. "SpaceManager::getsmallchunksandallocate()". But arguably our patch attempts a more complete solution. In 2016 I discussed the idea for this patch with some folks off-list, among them Jon Matsimutso. He then did advice me to create a JEP. So I did: [1]. However, meanwhile changes to the JEP process were discussed [2], and I am not sure anymore this patch needs even needs a JEP. It may be moderately complex and hence carries the risk inherent in any patch, but its effects would not be externally visible (if you discount seeing fewer metaspace OOMs). So, I'd prefer to handle this as a simple RFE. -- How this patch works: 1) When a class loader dies, its metaspace chunks are freed and returned to the freelist for reuse by the next class loader. With the patch, upon returning a chunk to the freelist, an attempt is made to merge it with its neighboring chunks - should they happen to be free too - to form a larger chunk. Which then is placed in the free list. As a result, the freelist should be populated by larger chunks at the expense of smaller chunks. In other words, all free chunks should always be as "coalesced as possible". 2) When a class loader needs a new chunk and a chunk of the requested size cannot be found in the free list, before carving out a new chunk from the virtual space, we first check if there is a larger chunk in the free list. If there is, that larger chunk is chopped up into n smaller chunks. One of them is returned to the callers, the others are re-added to the freelist. (1) and (2) together have the effect of removing the size-lock-in for chunks. If fragmentation allows it, small chunks are dynamically combined to form larger chunks, and larger chunks are split on demand. -- What this patch does not: This is not a rewrite of the chunk allocator - most of the mechanisms stay intact. Specifically, chunk sizes remain unchanged, and so do chunk allocation processes (when do which class loaders get handed which chunk size). Almost everthing this patch does affects only internal workings of the ChunkManager. Also note that I refrained from doing any cleanups, since I wanted reviewers to be able to gauge this patch without filtering noise. Unfortunately this patch adds some complexity. But there are many future opportunities for code cleanup and simplification, some of which we already discussed in existing RFEs ([3], [4]). All of them are out of the scope for this particular patch. -- Details: Before the patch, the following rules held: - All chunk sizes are multiples of the smallest chunk size ("specialized chunks") - All chunk sizes of larger chunks are also clean multiples of the next smaller chunk size (e.g. for class space, the ratio of specialized/small/medium chunks is 1:2:32) - All chunk start addresses are aligned to the smallest chunk size (more or less accidentally, see metaspacereservealignment). The patch makes the last rule explicit and more strict: - All (non-humongous) chunk start addresses are now aligned to their own chunk size. So, e.g. medium chunks are allocated at addresses which are a multiple of medium chunk size. This rule is not extended to humongous chunks, whose start addresses continue to be aligned to the smallest chunk size. The reason for this new alignment rule is that it makes it cheap both to find chunk predecessors of a chunk and to check which chunks are free. When a class loader dies and its chunk is returned to the freelist, all we have is its address. In order to merge it with its neighbors to form a larger chunk, we need to find those neighbors, including those preceding the returned chunk. Prior to this patch that was not easy - one would have to iterate chunks starting at the beginning of the VirtualSpaceNode. But due to the new alignment rule, we now know where the prospective larger chunk must start - at the next lower larger-chunk-size-aligned boundary. We also know that currently a smaller chunk must start there (*). In order to check the free-ness of chunks quickly, each VirtualSpaceNode now keeps a bitmap which describes its occupancy. One bit in this bitmap corresponds to a range the size of the smallest chunk size and starting at an address aligned to the smallest chunk size. Because of the alignment rules above, such a range belongs to one single chunk. The bit is 1 if the associated chunk is in use by a class loader, 0 if it is free. When we have calculated the address range a prospective larger chunk would span, we now need to check if all chunks in that range are free. Only then we can merge them. We do that by querying the bitmap. Note that the most common use case here is forming medium chunks from smaller chunks. With the new alignment rules, the bitmap portion covering a medium chunk now always happens to be 16- or 32bit in size and is 16- or 32bit aligned, so reading the bitmap in many cases becomes a simple 16- or 32bit load. If the range is free, only then we need to iterate the chunks in that range: pull them from the freelist, combine them to one new larger chunk, re-add that one to the freelist. (*) Humongous chunks make this a bit more complicated. Since the new alignment rule does not extend to them, a humongous chunk could still straddle the lower or upper boundary of the prospective larger chunk. So I gave the occupancy map a second layer, which is used to mark the start of chunks. An alternative approach could have been to make humongous chunks size and start address always a multiple of the largest non-humongous chunk size (medium chunks). That would have caused a bit of waste per humongous chunk (<64K) in exchange for simpler coding and a simpler occupancy map. -- The patch shows its best results in scenarios where a lot of smallish class loaders are alive simultaneously. When dying, they leave continuous expanses of metaspace covered in small chunks, which can be merged nicely. However, if class loader life times vary more, we have more interleaving of dead and alive small chunks, and hence chunk merging does not work as well as it could. For an example of a pathological case like this see example program: [5] Executed like this: "java -XX:CompressedClassSpaceSize=10M -cp test3 test3.Example2" the test will load 3000 small classes in separate class loaders, then throw them away and start loading large classes. The small classes will have flooded the metaspace with small chunks, which are unusable for the large classes. When executing with the rather limited CompressedClassSpaceSize=10M, we will run into an OOM after loading about 800 large classes, having used only 40% of the class space, the rest is wasted to unused small chunks. However, with our patch the example program will manage to allocate ~2900 large classes before running into an OOM, and class space will show almost no waste. Do demonstrate this, add -Xlog:gc+metaspace+freelist. After running into an OOM, statistics and an ASCII representation of the class space will be shown. The unpatched version will show large expanses of unused small chunks, the patched variant will show almost no waste. Note that the patch could be made more effective with a different size ratio between small and medium chunks: in class space, that ratio is 1:16, so 16 small chunks must happen to be free to form one larger chunk. With a smaller ratio the chance for coalescation would be larger. So there may be room for future improvement here: Since we now can merge and split chunks on demand, we could introduce more chunk sizes. Potentially arriving at a buddy-ish allocator style where we drop hard-wired chunk sizes for a dynamic model where the ratio between chunk sizes is always 1:2 and we could in theory have no limit to the chunk size? But this is just a thought and well out of the scope of this patch. -- What does this patch cost (memory): - the occupancy bitmap adds 1 byte per 4K metaspace. - MetaChunk headers get larger, since we add an enum and two bools to it. Depending on what the c++ compiler does with that, chunk headers grow by one or two MetaWords, reducing the payload size by that amount. - The new alignment rules mean we may need to create padding chunks to precede larger chunks. But since these padding chunks are added to the freelist, they should be used up before the need for new padding chunks arises. So, the maximally possible number of unused padding chunks should be limited by design to about 64K. The expectation is that the memory savings by this patch far outweighs its added memory costs. .. (performance): We did not see measurable drops in standard benchmarks raising over the normal noise. I also measured times for a program which stresses metaspace chunk coalescation, with the same result. I am open to suggestions what else I should measure, and/or independent measurements. -- Other details: I removed SpaceManager::getsmallchunkandallocate() to reduce complexity somewhat, because it was made mostly obsolete by this patch: since small chunks are combined to larger chunks upon return to the freelist, in theory we should not have that many free small chunks anymore anyway. However, there may be still cases where we could benefit from this workaround, so I am asking your opinion on this one. About tests: There were two native tests - ChunkManagerReturnTest and TestVirtualSpaceNode (the former was added by me last year) - which did not make much sense anymore, since they relied heavily on internal behavior which was made unpredictable with this patch. To make up for these lost tests, I added a new gtest which attempts to stress the many combinations of allocation pattern but does so from a layer above the old tests. It now uses Metaspace::allocate() and friends. By using that point as entry for tests, I am less dependent on implementation internals and still cover a lot of scenarios. -- Review pointers: Good points to start are - ChunkManager::returnsinglechunk() - specifically, ChunkManager::attempttocoalescearoundchunk() - here we merge chunks upon return to the free list - ChunkManager::freechunksget(): Here we now split large chunks into smaller chunks on demand - VirtualSpaceNode::takefromcommitted() : chunks are allocated according to align rules now, padding chunks are handles - The OccupancyMap class is the helper class implementing the new occupancy bitmap The rest is mostly chaff: helper functions, added tests and verifications. -- Thanks and Best Regards, Thomas [1] https://bugs.openjdk.java.net/browse/JDK-8166690 [2] http://mail.openjdk.java.net/pipermail/jdk-dev/2017-November /000128.html [3] https://bugs.openjdk.java.net/browse/JDK-8185034 [4] https://bugs.openjdk.java.net/browse/JDK-8176808 [5] https://bugs.openjdk.java.net/secure/attachment/63532/test3.zip

More information about the hotspot-dev mailing list