py: Improve allocation policy of qstr data.

Previous to this patch all interned strings lived in their own malloc'd
chunk.  On average this wastes N/2 bytes per interned string, where N is
the number-of-bytes for a quanta of the memory allocator (16 bytes on 32
bit archs).

With this patch interned strings are concatenated into the same malloc'd
chunk when possible.  Such chunks are enlarged inplace when possible,
and shrunk to fit when a new chunk is needed.

RAM savings with this patch are highly varied, but should always show an
improvement (unless only 3 or 4 strings are interned).  New version
typically uses about 70% of previous memory for the qstr data, and can
lead to savings of around 10% of total memory footprint of a running
script.

Costs about 120 bytes code size on Thumb2 archs (depends on how many
calls to gc_realloc are made).

1 files changed, 7 insertions, 4 deletions

diff --git a/py/malloc.c b/py/malloc.c
index f8728c596..0aecbd71d 100644
--- a/py/malloc.c
+++ b/py/malloc.c
@@ -56,7 +56,10 @@
 #define malloc(b) gc_alloc((b), false)
 #define malloc_with_finaliser(b) gc_alloc((b), true)
 #define free gc_free
-#define realloc gc_realloc
+#define realloc(ptr, n) gc_realloc(ptr, n, true)
+#define realloc_ext(ptr, n, mv) gc_realloc(ptr, n, mv)
+#else
+#define realloc_ext(ptr, n, mv) realloc(ptr, n)
 #endif // MICROPY_ENABLE_GC
 
 void *m_malloc(size_t num_bytes) {
@@ -134,11 +137,11 @@ void *m_realloc(void *ptr, size_t new_num_bytes) {
 }
 
 #if MICROPY_MALLOC_USES_ALLOCATED_SIZE
-void *m_realloc_maybe(void *ptr, size_t old_num_bytes, size_t new_num_bytes) {
+void *m_realloc_maybe(void *ptr, size_t old_num_bytes, size_t new_num_bytes, bool allow_move) {
 #else
-void *m_realloc_maybe(void *ptr, size_t new_num_bytes) {
+void *m_realloc_maybe(void *ptr, size_t new_num_bytes, bool allow_move) {
 #endif
-    void *new_ptr = realloc(ptr, new_num_bytes);
+    void *new_ptr = realloc_ext(ptr, new_num_bytes, allow_move);
 #if MICROPY_MEM_STATS
     // At first thought, "Total bytes allocated" should only grow,
     // after all, it's *total*. But consider for example 2K block