Linus TorvaldsThis creates a simple specialized object allocator for basic objects.
This avoids wasting space with malloc overhead (metadata and extra alignment), since the specialized allocator knows the alignment, and that objects, once allocated, are never freed.
It also allows us to track some basic statistics about object allocations. For example, for the mozilla import, it shows object usage as follows:
blobs: 627629 (14710 kB)
trees: 1119035 (34969 kB)
commits: 196423 (8440 kB)
tags: 1336 (46 kB)and the simpler allocator shaves off about 2.5% off the memory footprint off a "git-rev-list --all --objects", and is a bit faster too.
[ Side note: this concludes the series of "save memory in object storage". The thing is, there simply isn't much more to be saved on the objects.
Doing "git-rev-list --all --objects" on the mozilla archive has a final total RSS of 131498 pages for me: that's about 513MB. Of that, the object overhead is now just 56MB, the rest is going somewhere else (put another way: the fact that this patch shaves off 2.5% of the total memory overhead, considering that objects are now not much more than 10% of the total shows how big the wasted space really was: this makes object allocations much more memory- and time-efficient).
I haven't looked at where the rest is, but I suspect the bulk of it is just the pack-file loading. It may be that we should pack the tree objects separately from the blob objects: for git-rev-list --objects, we don't actually ever need to even look at the blobs, but since trees and blobs are interspersed in the pack-file, we end up not being dense in the tree accesses, so we end up looking at more pages than we strictly need to.
So with a 535MB pack-file, it's entirely possible - even likely - that most of the remaining RSS is just the mmap of the pack-file itself. We don't need to map in _all_ of it, but we do end up mapping a fair amount. ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org> ---
Makefile | 2 +- alloc.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++++ blob.c | 2 +- cache.h | 11 +++++++++++ commit.c | 2 +- tag.c | 2 +- tree.c | 2 +- 7 files changed, 67 insertions(+), 5 deletions(-)
diff --git a/Makefile b/Makefile index ea8cd28..0887945 100644 --- a/Makefile +++ b/Makefile @@ -216,7 +216,7 @@ LIB_OBJS = \ server-info.o setup.o sha1_file.o sha1_name.o strbuf.o \ tag.o tree.o usage.o config.o environment.o ctype.o copy.o \ fetch-clone.o revision.o pager.o tree-walk.o xdiff-interface.o \ - $(DIFF_OBJS) + alloc.o $(DIFF_OBJS) BUILTIN_OBJS = \ builtin-log.o builtin-help.o builtin-count.o builtin-diff.o builtin-push.o \ diff --git a/alloc.c b/alloc.c new file mode 100644 index 0000000..65ec0cc --- /dev/null +++ b/alloc.c @@ -0,0 +1,51 @@ +/* + * alloc.c - specialized allocator for internal objects + * + * Copyright (C) 2006 Linus Torvalds + * + * The standard malloc/free wastes too much space for objects, partly because + * it maintains all the allocation infrastructure (which isn't needed, since + * we never free an object descriptor anyway), but even more because it ends + * up with maximal alignment because it doesn't know what the object alignment + * for the new allocation is. + */ +#include "cache.h" +#include "object.h" +#include "blob.h" +#include "tree.h" +#include "commit.h" +#include "tag.h" + +#define BLOCKING 1024 + +#define DEFINE_ALLOCATOR(name) \ +static unsigned int name##_allocs; \ +struct name *alloc_##name##_node(void) \ +{ \ + static int nr; \ + static struct name *block; \ + \ + if (!nr) { \ + nr = BLOCKING; \ + block = xcalloc(BLOCKING, sizeof(struct name)); \ + } \ + nr--; \ + name##_allocs++; \ + return block++; \ +} + +DEFINE_ALLOCATOR(blob); +DEFINE_ALLOCATOR(tree); +DEFINE_ALLOCATOR(commit); +DEFINE_ALLOCATOR(tag); + +#define REPORT(name) \ + fprintf(stderr, "%10s: %8u (%zu kB)\n", #name, name##_allocs, name##_allocs*sizeof(struct name) >> 10) + +void alloc_report(void) +{ + REPORT(blob); + REPORT(tree); + REPORT(commit); + REPORT(tag); +} diff --git a/blob.c b/blob.c index 7377008..496f270 100644 --- a/blob.c +++ b/blob.c @@ -8,7 +8,7 @@ struct blob *lookup_blob(const unsigned { struct object *obj = lookup_object(sha1); if (!obj) { - struct blob *ret = xcalloc(1, sizeof(struct blob)); + struct blob *ret = alloc_blob_node(); created_object(sha1, &ret->object); ret->object.type = TYPE_BLOB; return ret; diff --git a/cache.h b/cache.h index 7fcb6d4..eaa5c0c 100644 --- a/cache.h +++ b/cache.h @@ -384,4 +384,15 @@ extern void setup_pager(void); int decode_85(char *dst, char *line, int linelen); void encode_85(char *buf, unsigned char *data, int bytes); +/* alloc.c */ +struct blob; +struct tree; +struct commit; +struct tag; +extern struct blob *alloc_blob_node(void); +extern struct tree *alloc_tree_node(void); +extern struct commit *alloc_commit_node(void); +extern struct tag *alloc_tag_node(void); +extern void alloc_report(void); + #endif /* CACHE_H */ diff --git a/commit.c b/commit.c index 5914200..0fa1198 100644 --- a/commit.c +++ b/commit.c @@ -84,7 +84,7 @@ struct commit *lookup_commit(const unsig { struct object *obj = lookup_object(sha1); if (!obj) { - struct commit *ret = xcalloc(1, sizeof(struct commit)); + struct commit *ret = alloc_commit_node(); created_object(sha1, &ret->object); ret->object.type = TYPE_COMMIT; return ret; diff --git a/tag.c b/tag.c index 9191333..5f70a5b 100644 --- a/tag.c +++ b/tag.c @@ -19,7 +19,7 @@ struct tag *lookup_tag(const unsigned ch { struct object *obj = lookup_object(sha1); if (!obj) { - struct tag *ret = xcalloc(1, sizeof(struct tag)); + struct tag *ret = alloc_tag_node(); created_object(sha1, &ret->object); ret->object.type = TYPE_TAG; return ret; diff --git a/tree.c b/tree.c index 64422fd..1023655 100644 --- a/tree.c +++ b/tree.c @@ -129,7 +129,7 @@ struct tree *lookup_tree(const unsigned { struct object *obj = lookup_object(sha1); if (!obj) { - struct tree *ret = xcalloc(1, sizeof(struct tree)); + struct tree *ret = alloc_tree_node(); created_object(sha1, &ret->object); ret->object.type = TYPE_TREE; return ret;
