blob: c713bd535a7132e4f1fc8b39a31f0911919770cc [file] [log] [blame]
# Copyright lowRISC contributors.
# Licensed under the Apache License, Version 2.0, see LICENSE for details.
# SPDX-License-Identifier: Apache-2.0
import logging as log
import pprint
import random
import shlex
from pathlib import Path
from typing import List
from JobTime import JobTime
from LauncherFactory import get_launcher
from sim_utils import (get_cov_summary_table, get_job_runtime,
get_simulated_time)
from tabulate import tabulate
from utils import (VERBOSE, clean_odirs, find_and_substitute_wildcards,
rm_path, subst_wildcards)
class Deploy():
"""
Abstraction to create and maintain a runnable job (builds, runs, etc.).
"""
# Indicate the target for each sub-class.
target = "none"
# List of variable names that are to be treated as "list of commands".
# This tells '_construct_cmd' that these vars are lists that need to
# be joined with '&&' instead of a space.
cmds_list_vars = []
# Represents the weight with which a job of this target is scheduled. These
# initial weights set for each of the targets below are roughly inversely
# proportional to their average runtimes. These are subject to change in
# future. Lower the runtime, the higher chance the it gets scheduled. It is
# useful to customize this only in case of targets that may coexist at a
# time.
# TODO: Allow these to be set in the HJson.
weight = 1
def __str__(self):
return (pprint.pformat(self.__dict__)
if log.getLogger().isEnabledFor(VERBOSE) else self.full_name)
def __init__(self, sim_cfg):
assert self.target is not None
# Cross ref the whole cfg object for ease.
self.sim_cfg = sim_cfg
# A list of jobs on which this job depends.
self.dependencies = []
# Indicates whether running this job requires all dependencies to pass.
# If this flag is set to False, any passing dependency will trigger
# this current job to run
self.needs_all_dependencies_passing = True
# Declare attributes that need to be extracted from the HJSon cfg.
self._define_attrs()
# Set class instance attributes.
self._set_attrs()
# Check if all attributes that are needed are set.
self._check_attrs()
# Do variable substitutions.
self._subst_vars()
# List of vars required to be exported to sub-shell, as a dict.
self.exports = self._process_exports()
# Construct the job's command.
self.cmd = self._construct_cmd()
# Launcher instance created later using create_launcher() method.
self.launcher = None
# Job's wall clock time (a.k.a CPU time, or runtime).
self.job_runtime = JobTime()
def _define_attrs(self):
"""Defines the attributes this instance needs to have.
These attributes are extracted from the Mode object / HJson config with
which this instance is created. There are two types of attributes -
one contributes to the generation of the command directly; the other
provides supplementary information pertaining to the job, such as
patterns that determine whether it passed or failed. These are
represented as dicts, whose values indicate in boolean whether the
extraction was successful.
"""
# These attributes are explicitly used to construct the job command.
self.mandatory_cmd_attrs = {}
# These attributes may indirectly contribute to the construction of the
# command (through substitution vars) or other things such as pass /
# fail patterns.
self.mandatory_misc_attrs = {
"build_mode": False,
"dry_run": False,
"exports": False,
"flow_makefile": False,
"name": False,
}
# Function to parse a dict and extract the mandatory cmd and misc attrs.
def _extract_attrs(self, ddict):
"""Extracts the attributes from the supplied dict.
'ddict' is typically either the Mode object or the entire config
object's dict. It is used to retrieve the instance attributes defined
in 'mandatory_cmd_attrs' and 'mandatory_misc_attrs'.
"""
ddict_keys = ddict.keys()
for key in self.mandatory_cmd_attrs.keys():
if self.mandatory_cmd_attrs[key] is False:
if key in ddict_keys:
setattr(self, key, ddict[key])
self.mandatory_cmd_attrs[key] = True
for key in self.mandatory_misc_attrs.keys():
if self.mandatory_misc_attrs[key] is False:
if key in ddict_keys:
setattr(self, key, ddict[key])
self.mandatory_misc_attrs[key] = True
def _set_attrs(self):
"""Sets additional attributes.
Invokes '_extract_attrs()' to read in all the necessary instance
attributes. Based on those, some additional instance attributes may
be derived. Those are set by this method.
"""
self._extract_attrs(self.sim_cfg.__dict__)
# Enable GUI mode.
self.gui = self.sim_cfg.gui
# Output directory where the artifacts go (used by the launcher).
self.odir = getattr(self, self.target + "_dir")
# Qualified name disambiguates the instance name with other instances
# of the same class (example: 'uart_smoke' reseeded multiple times
# needs to be disambiguated using the index -> '0.uart_smoke'.
self.qual_name = self.name
# Full name disambiguates across multiple cfg being run (example:
# 'aes:default', 'uart:default' builds.
self.full_name = self.sim_cfg.name + ":" + self.qual_name
# Job name is used to group the job by cfg and target. The scratch path
# directory name is assumed to be uniquified, in case there are more
# than one sim_cfgs with the same name.
self.job_name = "{}_{}".format(
Path(self.sim_cfg.scratch_path).name, self.target)
# Input directories (other than self) this job depends on.
self.input_dirs = []
# Directories touched by this job. These directories are marked
# because they are used by dependent jobs as input.
self.output_dirs = [self.odir]
# Pass and fail patterns.
self.pass_patterns = []
self.fail_patterns = []
def _check_attrs(self):
"""Checks if all required class attributes are set.
Invoked in __init__() after all attributes are extracted and set.
"""
for attr in self.mandatory_cmd_attrs.keys():
if self.mandatory_cmd_attrs[attr] is False:
raise AttributeError("Attribute {!r} not found for "
"{!r}.".format(attr, self.name))
for attr in self.mandatory_misc_attrs.keys():
if self.mandatory_misc_attrs[attr] is False:
raise AttributeError("Attribute {!r} not found for "
"{!r}.".format(attr, self.name))
def _subst_vars(self, ignored_subst_vars=[]):
"""Recursively search and replace substitution variables.
First pass: search within self dict. We ignore errors since some
substitions may be available in the second pass. Second pass: search
the entire sim_cfg object."""
self.__dict__ = find_and_substitute_wildcards(self.__dict__,
self.__dict__,
ignored_subst_vars, True)
self.__dict__ = find_and_substitute_wildcards(self.__dict__,
self.sim_cfg.__dict__,
ignored_subst_vars,
False)
def _process_exports(self):
"""Convert 'exports' as a list of dicts in the HJson to a dict.
Exports is a list of key-value pairs that are to be exported to the
subprocess' environment so that the tools can lookup those options.
DVSim limits how the data is presented in the HJson (the value of a
HJson member cannot be an object). This method converts a list of dicts
into a dict variable, which makes it easy to merge the list of exports
with the subprocess' env where the ASIC tool is invoked.
"""
return {k: str(v) for item in self.exports for k, v in item.items()}
def _construct_cmd(self):
"""Construct the command that will eventually be launched."""
cmd = "make -f {} {}".format(self.flow_makefile, self.target)
if self.dry_run is True:
cmd += " -n"
for attr in sorted(self.mandatory_cmd_attrs.keys()):
value = getattr(self, attr)
if type(value) is list:
# Join attributes that are list of commands with '&&' to chain
# them together when executed as a Make target's recipe.
separator = " && " if attr in self.cmds_list_vars else " "
value = separator.join(item.strip() for item in value)
if type(value) is bool:
value = int(value)
if type(value) is str:
value = shlex.quote(value.strip())
cmd += " {}={}".format(attr, value)
return cmd
def is_equivalent_job(self, item):
"""Checks if job that would be dispatched with 'item' is equivalent to
'self'.
Determines if 'item' and 'self' would behave exactly the same way when
deployed. If so, then there is no point in keeping both. The caller can
choose to discard 'item' and pick 'self' instead. To do so, we check
the final resolved 'cmd' & the exports. The 'name' field will be unique
to 'item' and 'self', so we take that out of the comparison.
"""
if type(self) != type(item):
return False
# Check if the cmd field is identical.
item_cmd = item.cmd.replace(item.name, self.name)
if self.cmd != item_cmd:
return False
# Check if exports have identical set of keys.
if self.exports.keys() != item.exports.keys():
return False
# Check if exports have identical values.
for key, val in self.exports.items():
item_val = item.exports[key]
if type(item_val) is str:
item_val = item_val.replace(item.name, self.name)
if val != item_val:
return False
log.log(VERBOSE, "Deploy job \"%s\" is equivalent to \"%s\"",
item.name, self.name)
return True
def pre_launch(self):
"""Callback to perform additional pre-launch activities.
This is invoked by launcher::_pre_launch().
"""
pass
def post_finish(self, status):
"""Callback to perform additional post-finish activities.
This is invoked by launcher::_post_finish().
"""
pass
def get_log_path(self):
"""Returns the log file path."""
return "{}/{}.log".format(self.odir, self.target)
def get_timeout_mins(self):
"""Returns the timeout in minutes."""
return None
def extract_info_from_log(self, log_text: List):
"""Extracts information pertaining to the job from its log.
This method parses the log text after the job has completed, for the
extraction of information pertaining to the job's performance. This
base class method extracts the job's runtime (i.e. the wall clock time)
as reported by the tool. The tool reported runtime is the most accurate
since it is devoid of the delays incurred due to infrastructure and
setup overhead.
The extended classes may override this method to extract other pieces
of information from the log.
`log_text` is the job's log file contents as a list of lines.
"""
try:
time, unit = get_job_runtime(log_text, self.sim_cfg.tool)
self.job_runtime.set(time, unit)
except RuntimeError as e:
log.warning(f"{self.full_name}: {e} Using dvsim-maintained "
"job_runtime instead.")
self.job_runtime.set(self.launcher.job_runtime_secs, "s")
def create_launcher(self):
"""Creates the launcher instance.
Note that the launcher instance for ALL jobs in the same job group must
be created before the Scheduler starts to dispatch one by one.
"""
# Retain the handle to self for lookup & callbacks.
self.launcher = get_launcher(self)
class CompileSim(Deploy):
"""Abstraction for building the simulation executable."""
target = "build"
cmds_list_vars = ["pre_build_cmds", "post_build_cmds"]
weight = 5
def __init__(self, build_mode, sim_cfg):
self.build_mode_obj = build_mode
self.seed = sim_cfg.build_seed
super().__init__(sim_cfg)
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
# tool srcs
"proj_root": False,
# Flist gen
"sv_flist_gen_cmd": False,
"sv_flist_gen_dir": False,
"sv_flist_gen_opts": False,
# Build
"pre_build_cmds": False,
"build_cmd": False,
"build_dir": False,
"build_opts": False,
"post_build_cmds": False,
})
self.mandatory_misc_attrs.update({
"build_fail_patterns": False,
"build_pass_patterns": False,
"build_timeout_mins": False,
"cov_db_dir": False,
})
def _set_attrs(self):
super()._extract_attrs(self.build_mode_obj.__dict__)
super()._set_attrs()
# Dont run the compile job in GUI mode.
self.gui = False
# 'build_mode' is used as a substitution variable in the HJson.
self.build_mode = self.name
self.job_name += f"_{self.build_mode}"
if self.sim_cfg.cov:
self.output_dirs += [self.cov_db_dir]
self.pass_patterns = self.build_pass_patterns
self.fail_patterns = self.build_fail_patterns
if self.sim_cfg.args.build_timeout_mins is not None:
self.build_timeout_mins = self.sim_cfg.args.build_timeout_mins
if self.build_timeout_mins:
log.debug("Timeout for job \"%s\" is %d minutes.", self.name,
self.build_timeout_mins)
def pre_launch(self):
# Delete old coverage database directories before building again. We
# need to do this because the build directory is not 'renewed'.
rm_path(self.cov_db_dir)
def get_timeout_mins(self):
"""Returns the timeout in minutes.
Limit build jobs to 60 minutes if the timeout is not set.
"""
return self.build_timeout_mins if self.build_timeout_mins is not None else 60
class CompileOneShot(Deploy):
"""Abstraction for building the design (used by non-DV flows)."""
target = "build"
def __init__(self, build_mode, sim_cfg):
self.build_mode_obj = build_mode
super().__init__(sim_cfg)
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
# tool srcs
"proj_root": False,
# Flist gen
"sv_flist_gen_cmd": False,
"sv_flist_gen_dir": False,
"sv_flist_gen_opts": False,
# Build
"build_dir": False,
"build_cmd": False,
"build_opts": False,
"build_log": False,
"build_timeout_mins": False,
"post_build_cmds": False,
"pre_build_cmds": False,
# Report processing
"report_cmd": False,
"report_opts": False
})
self.mandatory_misc_attrs.update({
"build_fail_patterns": False,
"build_pass_patterns": False
})
def _set_attrs(self):
super()._extract_attrs(self.build_mode_obj.__dict__)
super()._set_attrs()
# 'build_mode' is used as a substitution variable in the HJson.
self.build_mode = self.name
self.job_name += f"_{self.build_mode}"
self.fail_patterns = self.build_fail_patterns
self.pass_patterns = self.build_pass_patterns
if self.sim_cfg.args.build_timeout_mins is not None:
self.build_timeout_mins = self.sim_cfg.args.build_timeout_mins
if self.build_timeout_mins:
log.debug("Timeout for job \"%s\" is %d minutes.", self.name,
self.build_timeout_mins)
def get_timeout_mins(self):
"""Returns the timeout in minutes.
Limit build jobs to 60 minutes if the timeout is not set.
"""
return self.build_timeout_mins if self.build_timeout_mins is not None else 60
class RunTest(Deploy):
"""Abstraction for running tests. This is one per seed for each test."""
# Initial seed values when running tests (if available).
target = "run"
seeds = []
fixed_seed = None
cmds_list_vars = ["pre_run_cmds", "post_run_cmds"]
def __init__(self, index, test, build_job, sim_cfg):
self.test_obj = test
self.index = index
self.build_seed = sim_cfg.build_seed
self.seed = RunTest.get_seed()
self.simulated_time = JobTime()
super().__init__(sim_cfg)
if build_job is not None:
self.dependencies.append(build_job)
# We did something wrong if build_mode is not the same as the build_job
# arg's name.
assert self.build_mode == build_job.name
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
# tool srcs
"proj_root": False,
"uvm_test": False,
"uvm_test_seq": False,
"sw_images": False,
"sw_build_device": False,
"sw_build_cmd": False,
"sw_build_opts": False,
"run_dir": False,
"pre_run_cmds": False,
"run_cmd": False,
"run_opts": False,
"post_run_cmds": False,
"build_seed": True, # Already set in the constructor.
"seed": True, # Already set in the constructor.
})
self.mandatory_misc_attrs.update({
"cov_db_dir": False,
"cov_db_test_dir": False,
"run_dir_name": False,
"run_fail_patterns": False,
"run_pass_patterns": False,
"run_timeout_mins": False,
"run_timeout_multiplier": False,
})
def _set_attrs(self):
super()._extract_attrs(self.test_obj.__dict__)
super()._set_attrs()
# 'test' is used as a substitution variable in the HJson.
self.test = self.name
self.build_mode = self.test_obj.build_mode.name
self.qual_name = self.run_dir_name + "." + str(self.seed)
self.full_name = self.sim_cfg.name + ":" + self.qual_name
self.job_name += f"_{self.build_mode}"
if self.sim_cfg.cov:
self.output_dirs += [self.cov_db_dir]
# In GUI mode, the log file is not updated; hence, nothing to check.
if not self.gui:
self.pass_patterns = self.run_pass_patterns
self.fail_patterns = self.run_fail_patterns
if self.sim_cfg.args.run_timeout_mins is not None:
self.run_timeout_mins = self.sim_cfg.args.run_timeout_mins
if self.sim_cfg.args.run_timeout_multiplier is not None:
self.run_timeout_multiplier = (
self.sim_cfg.args.run_timeout_multiplier)
if self.run_timeout_mins and self.run_timeout_multiplier:
self.run_timeout_mins = int(self.run_timeout_mins *
self.run_timeout_multiplier)
if self.run_timeout_multiplier:
log.debug("Timeout multiplier for job \"%s\" is %f.",
self.full_name, self.run_timeout_multiplier)
if self.run_timeout_mins:
log.debug("Timeout for job \"%s\" is %d minutes.", self.full_name,
self.run_timeout_mins)
def pre_launch(self):
self.launcher.renew_odir = True
def post_finish(self, status):
if status != 'P':
# Delete the coverage data if available.
rm_path(self.cov_db_test_dir)
@staticmethod
def get_seed():
# If --seeds option is passed, then those custom seeds are consumed
# first. If --fixed-seed <val> is also passed, the subsequent tests
# (once the custom seeds are consumed) will be run with the fixed seed.
if not RunTest.seeds:
if RunTest.fixed_seed:
return RunTest.fixed_seed
for i in range(1000):
seed = random.getrandbits(32)
RunTest.seeds.append(seed)
return RunTest.seeds.pop(0)
def get_timeout_mins(self):
"""Returns the timeout in minutes.
Limit run jobs to 60 minutes if the timeout is not set.
"""
return self.run_timeout_mins if self.run_timeout_mins is not None else 60
def extract_info_from_log(self, log_text: List):
"""Extracts the time the design was simulated for, from the log."""
super().extract_info_from_log(log_text)
try:
time, unit = get_simulated_time(log_text, self.sim_cfg.tool)
self.simulated_time.set(time, unit)
except RuntimeError as e:
log.debug(f"{self.full_name}: {e}")
class CovUnr(Deploy):
"""Abstraction for coverage UNR flow."""
target = "cov_unr"
def __init__(self, sim_cfg):
super().__init__(sim_cfg)
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
# tool srcs
"proj_root": False,
# Need to generate filelist based on build mode
"sv_flist_gen_cmd": False,
"sv_flist_gen_dir": False,
"sv_flist_gen_opts": False,
"build_dir": False,
"cov_unr_build_cmd": False,
"cov_unr_build_opts": False,
"cov_unr_run_cmd": False,
"cov_unr_run_opts": False
})
self.mandatory_misc_attrs.update({
"cov_unr_dir": False,
"cov_merge_db_dir": False,
"build_fail_patterns": False
})
def _set_attrs(self):
super()._set_attrs()
self.qual_name = self.target
self.full_name = self.sim_cfg.name + ":" + self.qual_name
self.input_dirs += [self.cov_merge_db_dir]
# Reuse the build_fail_patterns set in the HJson.
self.fail_patterns = self.build_fail_patterns
class CovMerge(Deploy):
"""Abstraction for merging coverage databases."""
target = "cov_merge"
weight = 10
def __init__(self, run_items, sim_cfg):
# Construct the cov_db_dirs right away from the run_items. This is a
# special variable used in the HJson. The coverage associated with
# the primary build mode needs to be first in the list.
self.cov_db_dirs = []
for run in run_items:
if run.cov_db_dir not in self.cov_db_dirs:
if sim_cfg.primary_build_mode == run.build_mode:
self.cov_db_dirs.insert(0, run.cov_db_dir)
else:
self.cov_db_dirs.append(run.cov_db_dir)
# Early lookup the cov_merge_db_dir, which is a mandatory misc
# attribute anyway. We need it to compute additional cov db dirs.
self.cov_merge_db_dir = subst_wildcards("{cov_merge_db_dir}",
sim_cfg.__dict__)
# Prune previous merged cov directories, keeping past 7 dbs.
prev_cov_db_dirs = clean_odirs(odir=self.cov_merge_db_dir, max_odirs=7)
# If the --cov-merge-previous command line switch is passed, then
# merge coverage with the previous runs.
if sim_cfg.cov_merge_previous:
self.cov_db_dirs += [str(item) for item in prev_cov_db_dirs]
super().__init__(sim_cfg)
self.dependencies += run_items
# Run coverage merge even if one test passes.
self.needs_all_dependencies_passing = False
# Append cov_db_dirs to the list of exports.
self.exports["cov_db_dirs"] = shlex.quote(" ".join(self.cov_db_dirs))
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
"cov_merge_cmd": False,
"cov_merge_opts": False
})
self.mandatory_misc_attrs.update({
"cov_merge_dir": False,
"cov_merge_db_dir": False
})
def _set_attrs(self):
super()._set_attrs()
self.qual_name = self.target
self.full_name = self.sim_cfg.name + ":" + self.qual_name
# For merging coverage db, the precise output dir is set in the HJson.
self.odir = self.cov_merge_db_dir
self.input_dirs += self.cov_db_dirs
self.output_dirs = [self.odir]
class CovReport(Deploy):
"""Abstraction for coverage report generation. """
target = "cov_report"
weight = 10
def __init__(self, merge_job, sim_cfg):
super().__init__(sim_cfg)
self.dependencies.append(merge_job)
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
"cov_report_cmd": False,
"cov_report_opts": False
})
self.mandatory_misc_attrs.update({
"cov_report_dir": False,
"cov_merge_db_dir": False,
"cov_report_txt": False
})
def _set_attrs(self):
super()._set_attrs()
self.qual_name = self.target
self.full_name = self.sim_cfg.name + ":" + self.qual_name
# Keep track of coverage results, once the job is finished.
self.cov_total = ""
self.cov_results = ""
self.cov_results_dict = dict()
def post_finish(self, status):
"""Extract the coverage results summary for the dashboard.
If the extraction fails, an appropriate exception is raised, which must
be caught by the caller to mark the job as a failure.
"""
if self.dry_run or status != 'P':
return
results, self.cov_total = get_cov_summary_table(
self.cov_report_txt, self.sim_cfg.tool)
colalign = (("center", ) * len(results[0]))
self.cov_results = tabulate(results,
headers="firstrow",
tablefmt="pipe",
colalign=colalign)
for tup in zip(*results):
self.cov_results_dict[tup[0]] = tup[1]
class CovAnalyze(Deploy):
"""Abstraction for running the coverage analysis tool."""
target = "cov_analyze"
def __init__(self, sim_cfg):
# Enforce GUI mode for coverage analysis.
sim_cfg.gui = True
super().__init__(sim_cfg)
def _define_attrs(self):
super()._define_attrs()
self.mandatory_cmd_attrs.update({
# tool srcs
"proj_root": False,
"cov_analyze_cmd": False,
"cov_analyze_opts": False
})
self.mandatory_misc_attrs.update({
"cov_analyze_dir": False,
"cov_merge_db_dir": False
})
def _set_attrs(self):
super()._set_attrs()
self.qual_name = self.target
self.full_name = self.sim_cfg.name + ":" + self.qual_name
self.input_dirs += [self.cov_merge_db_dir]