openkylin
/
platform_build
mirror of https://gitee.com/openkylin/platform_build.git
9
0
Fork 1
Code Issues Projects Releases Wiki Activity

Adjust the size limit for splitting large files 

Modify the number of max blocks per transfer to reduce splitting of
large files during patch generation. This will reduce the size of
final incremental OTA updater package. In specific, the size limit
for each file to transfer is set to 1/8 cachesize instead of a fix
number.

Bug: 26472126

Change-Id: Iffb135e378c4c7a91640f2fc4c40e9e93d1fe4bb
(cherry picked from commit bb86e1d9c2)
This commit is contained in:
Tianjie Xu 2016-01-13 16:14:10 -08:00 committed by Tao Bao
parent bbf3687eb8
commit 083f65d66d
1 changed files with 18 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
tools/releasetools/blockimgdiff.py
View File

@ -966,29 +966,36 @@ class BlockImageDiff(object):
too many blocks (greater than MAX_BLOCKS_PER_DIFF_TRANSFER), we split it
into smaller pieces by getting multiple Transfer()s.
The downside is that after splitting, we can no longer use imgdiff but
only bsdiff."""
MAX_BLOCKS_PER_DIFF_TRANSFER = 1024
The downside is that after splitting, we may increase the package size
since the split pieces don't align well. According to our experiments,
1/8 of the cache size as the per-piece limit appears to be optimal.
Compared to the fixed 1024-block limit, it reduces the overall package
size by 30% volantis, and 20% for angler and bullhead."""
# We care about diff transfers only.
if style != "diff" or not split:
Transfer(tgt_name, src_name, tgt_ranges, src_ranges, style, by_id)
return
pieces = 0
cache_size = common.OPTIONS.cache_size
split_threshold = 0.125
max_blocks_per_transfer = int(cache_size * split_threshold /
self.tgt.blocksize)
# Change nothing for small files.
if (tgt_ranges.size() <= MAX_BLOCKS_PER_DIFF_TRANSFER and
src_ranges.size() <= MAX_BLOCKS_PER_DIFF_TRANSFER):
if (tgt_ranges.size() <= max_blocks_per_transfer and
src_ranges.size() <= max_blocks_per_transfer):
Transfer(tgt_name, src_name, tgt_ranges, src_ranges, style, by_id)
return
pieces = 0
while (tgt_ranges.size() > MAX_BLOCKS_PER_DIFF_TRANSFER and
src_ranges.size() > MAX_BLOCKS_PER_DIFF_TRANSFER):
while (tgt_ranges.size() > max_blocks_per_transfer and
src_ranges.size() > max_blocks_per_transfer):
tgt_split_name = "%s-%d" % (tgt_name, pieces)
src_split_name = "%s-%d" % (src_name, pieces)
tgt_first = tgt_ranges.first(MAX_BLOCKS_PER_DIFF_TRANSFER)
src_first = src_ranges.first(MAX_BLOCKS_PER_DIFF_TRANSFER)
tgt_first = tgt_ranges.first(max_blocks_per_transfer)
src_first = src_ranges.first(max_blocks_per_transfer)
Transfer(tgt_split_name, src_split_name, tgt_first, src_first, style,
by_id)