From patchwork Fri Sep 27 18:56:42 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: [master] gen-lockedsig-cache: Replace glob lookup with hash to filename lookup X-Patchwork-Submitter: Konrad Scherer X-Patchwork-Id: 174640 Message-Id: <20190927185642.29251-1-konrad.scherer@windriver.com> To: Date: Fri, 27 Sep 2019 14:56:42 -0400 From: Konrad Scherer List-Id: Patches and discussions about the oe-core layer From: Konrad Scherer Using the glob function to map signatures to sstate files is very slow when the sstate is large and accessed over nfs. The lookup now only loads the necessary prefixes and doesn't use glob as all. Unfortunately I don't have access to the systems where the performance isse was noticed and on my test system the glob is fast enough that the performance numbers aren't useful. I could verify that file list returned by the new code is the same. [YOCTO #13539] Signed-off-by: Konrad Scherer --- meta/lib/oe/copy_buildsystem.py | 3 ++- scripts/gen-lockedsig-cache | 44 +++++++++++++++++++++++++++++---- 2 files changed, 41 insertions(+), 6 deletions(-) -- 2.23.0 -- _______________________________________________ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core diff --git a/meta/lib/oe/copy_buildsystem.py b/meta/lib/oe/copy_buildsystem.py index 31a84f5b06..c515683608 100644 --- a/meta/lib/oe/copy_buildsystem.py +++ b/meta/lib/oe/copy_buildsystem.py @@ -259,7 +259,8 @@ def create_locked_sstate_cache(lockedsigs, input_sstate_cache, output_sstate_cac bb.note('Generating sstate-cache...') nativelsbstring = d.getVar('NATIVELSBSTRING') - bb.process.run("gen-lockedsig-cache %s %s %s %s %s" % (lockedsigs, input_sstate_cache, output_sstate_cache, nativelsbstring, filterfile or '')) + stdout, _ = bb.process.run("gen-lockedsig-cache %s %s %s %s %s" % (lockedsigs, input_sstate_cache, output_sstate_cache, nativelsbstring, filterfile or '')) + bb.debug(stdout) if fixedlsbstring and nativelsbstring != fixedlsbstring: nativedir = output_sstate_cache + '/' + nativelsbstring if os.path.isdir(nativedir): diff --git a/scripts/gen-lockedsig-cache b/scripts/gen-lockedsig-cache index e3076e11a5..ae5e09d89f 100755 --- a/scripts/gen-lockedsig-cache +++ b/scripts/gen-lockedsig-cache @@ -5,9 +5,9 @@ import os import sys -import glob import shutil import errno +import time def mkdir(d): try: @@ -16,6 +16,33 @@ def mkdir(d): if e.errno != errno.EEXIST: raise e +# extract the hash from past the last colon to last underscore +def extract_sha(filename): + return filename.split(':')[7].split('_')[0] + +# get all files in a directory, extract hash and make +# a map from hash to list of file with that hash +def map_sha_to_files(dir_, prefix, sha_map): + sstate_prefix_path = dir_ + '/' + prefix + '/' + sstate_files = os.listdir(sstate_prefix_path) + for f in sstate_files: + sha = extract_sha(f) + if sha not in sha_map: + sha_map[sha] = [] + sha_map[sha].append(sstate_prefix_path + f) + +# given a prefix build a map of hash to list of files +def build_sha_cache(prefix): + sha_map = {} + + sstate_dir = sys.argv[2] + map_sha_to_files(sstate_dir, prefix, sha_map) + + native_sstate_dir = sys.argv[2] + sys.argv[4] + map_sha_to_files(native_sstate_dir, prefix, sha_map) + + return sha_map + if len(sys.argv) < 5: print("Incorrect number of arguments specified") print("syntax: gen-lockedsig-cache [filterfile]") @@ -41,12 +68,19 @@ with open(sys.argv[1]) as f: sigs.append(sig) print('Gathering file list') +start_time = time.perf_counter() files = set() +sstate_content_cache = {} for s in sigs: - p = sys.argv[2] + "/" + s[:2] + "/*" + s + "*" - files |= set(glob.glob(p)) - p = sys.argv[2] + "/%s/" % sys.argv[4] + s[:2] + "/*" + s + "*" - files |= set(glob.glob(p)) + prefix = s[:2] + if prefix not in sstate_content_cache: + sstate_content_cache[prefix] = build_sha_cache(prefix) + + for f in sstate_content_cache[prefix][s]: + files.add(f) + +elapsed = time.perf_counter() - start_time +print("Gathering file list took %.1fs" % elapsed) print('Processing files') for f in files: