Re: [Libguestfs] [nbdkit PATCH] data, memory: Optimize .zero > PAGE_SIZE

When sparse_array_zero() is used for a range larger than a page, there's no need to waste time in memset() or is_zero() - we already know the page will be free()d. Signed-off-by: Eric Blake <eblake(a)redhat.com&gt; --- Here's a fun one :) common/sparse/sparse.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/common/sparse/sparse.c b/common/sparse/sparse.c index cb44743c..5e085763 100644 --- a/common/sparse/sparse.c +++ b/common/sparse/sparse.c @@ -343,10 +343,13 @@ sparse_array_zero (struct sparse_array *sa, uint32_t count, uint64_t offset) n = count; if (p) { - memset (p, 0, n); + if (n < PAGE_SIZE) + memset (p, 0, n); + else + assert (p == *l2_page); /* If the whole page is now zero, free it. */ - if (is_zero (*l2_page, PAGE_SIZE)) { + if (n == PAGE_SIZE || is_zero (*l2_page, PAGE_SIZE)) {

if (sa->debug) nbdkit_debug ("%s: freeing zero page at offset %" PRIu64, __func__, offset);