Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
// Copyright 2017 Google Inc. All rights reserved.
|
|
|
|
//
|
|
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
// you may not use this file except in compliance with the License.
|
|
|
|
// You may obtain a copy of the License at
|
|
|
|
//
|
|
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
//
|
|
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
// See the License for the specific language governing permissions and
|
|
|
|
// limitations under the License.
|
|
|
|
|
|
|
|
package main
|
|
|
|
|
|
|
|
import (
|
|
|
|
"context"
|
2017-07-15 02:29:29 +08:00
|
|
|
"flag"
|
|
|
|
"fmt"
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
"os"
|
|
|
|
"path/filepath"
|
|
|
|
"strconv"
|
|
|
|
"strings"
|
|
|
|
"time"
|
|
|
|
|
|
|
|
"android/soong/ui/build"
|
|
|
|
"android/soong/ui/logger"
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
"android/soong/ui/status"
|
|
|
|
"android/soong/ui/terminal"
|
2016-08-22 06:17:17 +08:00
|
|
|
"android/soong/ui/tracer"
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
)
|
|
|
|
|
|
|
|
func indexList(s string, list []string) int {
|
|
|
|
for i, l := range list {
|
|
|
|
if l == s {
|
|
|
|
return i
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return -1
|
|
|
|
}
|
|
|
|
|
|
|
|
func inList(s string, list []string) bool {
|
|
|
|
return indexList(s, list) != -1
|
|
|
|
}
|
|
|
|
|
|
|
|
func main() {
|
2018-07-17 10:59:10 +08:00
|
|
|
var stdio terminal.StdioInterface
|
|
|
|
stdio = terminal.StdioImpl{}
|
|
|
|
|
|
|
|
// dumpvar uses stdout, everything else should be in stderr
|
|
|
|
if os.Args[1] == "--dumpvar-mode" || os.Args[1] == "--dumpvars-mode" {
|
|
|
|
stdio = terminal.NewCustomStdio(os.Stdin, os.Stderr, os.Stderr)
|
|
|
|
}
|
|
|
|
|
|
|
|
writer := terminal.NewWriter(stdio)
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
defer writer.Finish()
|
|
|
|
|
|
|
|
log := logger.New(writer)
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
defer log.Cleanup()
|
|
|
|
|
2017-07-15 02:29:29 +08:00
|
|
|
if len(os.Args) < 2 || !(inList("--make-mode", os.Args) ||
|
|
|
|
os.Args[1] == "--dumpvars-mode" ||
|
|
|
|
os.Args[1] == "--dumpvar-mode") {
|
|
|
|
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
log.Fatalln("The `soong` native UI is not yet available.")
|
|
|
|
}
|
|
|
|
|
|
|
|
ctx, cancel := context.WithCancel(context.Background())
|
|
|
|
defer cancel()
|
|
|
|
|
2016-08-22 06:17:17 +08:00
|
|
|
trace := tracer.New(log)
|
|
|
|
defer trace.Close()
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
stat := &status.Status{}
|
|
|
|
defer stat.Finish()
|
|
|
|
stat.AddOutput(terminal.NewStatusOutput(writer, os.Getenv("NINJA_STATUS")))
|
|
|
|
stat.AddOutput(trace.StatusTracer())
|
|
|
|
|
2016-08-22 06:17:17 +08:00
|
|
|
build.SetupSignals(log, cancel, func() {
|
|
|
|
trace.Close()
|
|
|
|
log.Cleanup()
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
stat.Finish()
|
2016-08-22 06:17:17 +08:00
|
|
|
})
|
|
|
|
|
2018-07-23 12:18:45 +08:00
|
|
|
buildCtx := build.Context{ContextImpl: &build.ContextImpl{
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
Context: ctx,
|
|
|
|
Logger: log,
|
|
|
|
Tracer: trace,
|
|
|
|
Writer: writer,
|
|
|
|
Status: stat,
|
2016-08-22 06:17:17 +08:00
|
|
|
}}
|
2017-07-15 02:29:29 +08:00
|
|
|
var config build.Config
|
|
|
|
if os.Args[1] == "--dumpvars-mode" || os.Args[1] == "--dumpvar-mode" {
|
|
|
|
config = build.NewConfig(buildCtx)
|
|
|
|
} else {
|
|
|
|
config = build.NewConfig(buildCtx, os.Args[1:]...)
|
|
|
|
}
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
|
2016-08-22 06:17:17 +08:00
|
|
|
build.SetupOutDir(buildCtx, config)
|
2017-02-05 09:30:44 +08:00
|
|
|
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
logsDir := config.OutDir()
|
2017-02-05 09:30:44 +08:00
|
|
|
if config.Dist() {
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
logsDir = filepath.Join(config.DistDir(), "logs")
|
2017-02-05 09:30:44 +08:00
|
|
|
}
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
os.MkdirAll(logsDir, 0777)
|
|
|
|
log.SetOutput(filepath.Join(logsDir, "soong.log"))
|
|
|
|
trace.SetOutput(filepath.Join(logsDir, "build.trace"))
|
|
|
|
stat.AddOutput(status.NewVerboseLog(log, filepath.Join(logsDir, "verbose.log")))
|
|
|
|
stat.AddOutput(status.NewErrorLog(log, filepath.Join(logsDir, "error.log")))
|
|
|
|
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
if start, ok := os.LookupEnv("TRACE_BEGIN_SOONG"); ok {
|
|
|
|
if !strings.HasSuffix(start, "N") {
|
|
|
|
if start_time, err := strconv.ParseUint(start, 10, 64); err == nil {
|
|
|
|
log.Verbosef("Took %dms to start up.",
|
|
|
|
time.Since(time.Unix(0, int64(start_time))).Nanoseconds()/time.Millisecond.Nanoseconds())
|
2016-08-22 06:17:17 +08:00
|
|
|
buildCtx.CompleteTrace("startup", start_time, uint64(time.Now().UnixNano()))
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
}
|
|
|
|
}
|
2017-07-14 05:27:31 +08:00
|
|
|
|
|
|
|
if executable, err := os.Executable(); err == nil {
|
|
|
|
trace.ImportMicrofactoryLog(filepath.Join(filepath.Dir(executable), "."+filepath.Base(executable)+".trace"))
|
|
|
|
}
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
}
|
|
|
|
|
2017-08-05 03:30:12 +08:00
|
|
|
f := build.NewSourceFinder(buildCtx, config)
|
|
|
|
defer f.Shutdown()
|
|
|
|
build.FindSources(buildCtx, config, f)
|
|
|
|
|
2017-07-15 02:29:29 +08:00
|
|
|
if os.Args[1] == "--dumpvar-mode" {
|
|
|
|
dumpVar(buildCtx, config, os.Args[2:])
|
|
|
|
} else if os.Args[1] == "--dumpvars-mode" {
|
|
|
|
dumpVars(buildCtx, config, os.Args[2:])
|
|
|
|
} else {
|
Add a unified status reporting UI
This adds a new status package that merges the running of "actions"
(ninja calls them edges) of multiple tools into one view of the current
state, and gives that to a number of different outputs.
For inputs:
Kati's output parser has been rewritten (and moved) to map onto the
StartAction/FinishAction API. A byproduct of this is that the build
servers should be able to extract errors from Kati better, since they
look like the errors that Ninja used to write.
Ninja is no longer directly connected to the terminal, but its output is
read via the protobuf frontend API, so it's just another tool whose
output becomes merged together.
multiproduct_kati loses its custom status routines, and uses the common
one instead.
For outputs:
The primary output is the ui/terminal.Status type, which along with
ui/terminal.Writer now controls everything about the terminal output.
Today, this doesn't really change any behaviors, but having all terminal
output going through here allows a more complicated (multi-line / full
window) status display in the future.
The tracer acts as an output of the status package, tracing all the
action start / finish events. This replaces reading the .ninja_log file,
so it now properly handles multiple output files from a single action.
A new rotated log file (out/error.log, or out/dist/logs/error.log) just
contains a description of all of the errors that happened during the
current build.
Another new compressed and rotated log file (out/verbose.log.gz, or
out/dist/logs/verbose.log.gz) contains the full verbose (showcommands)
log of every execution run by the build. Since this is now written on
every build, the showcommands argument is now ignored -- if you want to
get the commands run, look at the log file after the build.
Test: m
Test: <built-in tests>
Test: NINJA_ARGS="-t list" m
Test: check the build.trace.gz
Test: check the new log files
Change-Id: If1d8994890d43ef68f65aa10ddd8e6e06dc7013a
2018-05-18 07:37:09 +08:00
|
|
|
if config.IsVerbose() {
|
|
|
|
writer.Print("! The argument `showcommands` is no longer supported.")
|
|
|
|
writer.Print("! Instead, the verbose log is always written to a compressed file in the output dir:")
|
|
|
|
writer.Print("!")
|
|
|
|
writer.Print(fmt.Sprintf("! gzip -cd %s/verbose.log.gz | less -R", logsDir))
|
|
|
|
writer.Print("!")
|
|
|
|
writer.Print("! Older versions are saved in verbose.log.#.gz files")
|
|
|
|
writer.Print("")
|
|
|
|
time.Sleep(5 * time.Second)
|
|
|
|
}
|
|
|
|
|
2017-11-17 09:55:00 +08:00
|
|
|
toBuild := build.BuildAll
|
|
|
|
if config.Checkbuild() {
|
|
|
|
toBuild |= build.RunBuildTests
|
|
|
|
}
|
|
|
|
build.Build(buildCtx, config, toBuild)
|
2017-07-15 02:29:29 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
func dumpVar(ctx build.Context, config build.Config, args []string) {
|
|
|
|
flags := flag.NewFlagSet("dumpvar", flag.ExitOnError)
|
|
|
|
flags.Usage = func() {
|
|
|
|
fmt.Fprintf(os.Stderr, "usage: %s --dumpvar-mode [--abs] <VAR>\n\n", os.Args[0])
|
|
|
|
fmt.Fprintln(os.Stderr, "In dumpvar mode, print the value of the legacy make variable VAR to stdout")
|
|
|
|
fmt.Fprintln(os.Stderr, "")
|
|
|
|
|
|
|
|
fmt.Fprintln(os.Stderr, "'report_config' is a special case that prints the human-readable config banner")
|
|
|
|
fmt.Fprintln(os.Stderr, "from the beginning of the build.")
|
|
|
|
fmt.Fprintln(os.Stderr, "")
|
|
|
|
flags.PrintDefaults()
|
|
|
|
}
|
|
|
|
abs := flags.Bool("abs", false, "Print the absolute path of the value")
|
|
|
|
flags.Parse(args)
|
|
|
|
|
|
|
|
if flags.NArg() != 1 {
|
|
|
|
flags.Usage()
|
|
|
|
os.Exit(1)
|
|
|
|
}
|
|
|
|
|
|
|
|
varName := flags.Arg(0)
|
|
|
|
if varName == "report_config" {
|
|
|
|
varData, err := build.DumpMakeVars(ctx, config, nil, build.BannerVars)
|
|
|
|
if err != nil {
|
|
|
|
ctx.Fatal(err)
|
|
|
|
}
|
|
|
|
|
|
|
|
fmt.Println(build.Banner(varData))
|
|
|
|
} else {
|
|
|
|
varData, err := build.DumpMakeVars(ctx, config, nil, []string{varName})
|
|
|
|
if err != nil {
|
|
|
|
ctx.Fatal(err)
|
|
|
|
}
|
|
|
|
|
|
|
|
if *abs {
|
|
|
|
var res []string
|
|
|
|
for _, path := range strings.Fields(varData[varName]) {
|
|
|
|
if abs, err := filepath.Abs(path); err == nil {
|
|
|
|
res = append(res, abs)
|
|
|
|
} else {
|
|
|
|
ctx.Fatalln("Failed to get absolute path of", path, err)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
fmt.Println(strings.Join(res, " "))
|
|
|
|
} else {
|
|
|
|
fmt.Println(varData[varName])
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
func dumpVars(ctx build.Context, config build.Config, args []string) {
|
|
|
|
flags := flag.NewFlagSet("dumpvars", flag.ExitOnError)
|
|
|
|
flags.Usage = func() {
|
|
|
|
fmt.Fprintf(os.Stderr, "usage: %s --dumpvars-mode [--vars=\"VAR VAR ...\"]\n\n", os.Args[0])
|
|
|
|
fmt.Fprintln(os.Stderr, "In dumpvars mode, dump the values of one or more legacy make variables, in")
|
|
|
|
fmt.Fprintln(os.Stderr, "shell syntax. The resulting output may be sourced directly into a shell to")
|
|
|
|
fmt.Fprintln(os.Stderr, "set corresponding shell variables.")
|
|
|
|
fmt.Fprintln(os.Stderr, "")
|
|
|
|
|
|
|
|
fmt.Fprintln(os.Stderr, "'report_config' is a special case that dumps a variable containing the")
|
|
|
|
fmt.Fprintln(os.Stderr, "human-readable config banner from the beginning of the build.")
|
|
|
|
fmt.Fprintln(os.Stderr, "")
|
|
|
|
flags.PrintDefaults()
|
|
|
|
}
|
|
|
|
|
|
|
|
varsStr := flags.String("vars", "", "Space-separated list of variables to dump")
|
|
|
|
absVarsStr := flags.String("abs-vars", "", "Space-separated list of variables to dump (using absolute paths)")
|
|
|
|
|
|
|
|
varPrefix := flags.String("var-prefix", "", "String to prepend to all variable names when dumping")
|
|
|
|
absVarPrefix := flags.String("abs-var-prefix", "", "String to prepent to all absolute path variable names when dumping")
|
|
|
|
|
|
|
|
flags.Parse(args)
|
|
|
|
|
|
|
|
if flags.NArg() != 0 {
|
|
|
|
flags.Usage()
|
|
|
|
os.Exit(1)
|
|
|
|
}
|
|
|
|
|
|
|
|
vars := strings.Fields(*varsStr)
|
|
|
|
absVars := strings.Fields(*absVarsStr)
|
|
|
|
|
|
|
|
allVars := append([]string{}, vars...)
|
|
|
|
allVars = append(allVars, absVars...)
|
|
|
|
|
|
|
|
if i := indexList("report_config", allVars); i != -1 {
|
|
|
|
allVars = append(allVars[:i], allVars[i+1:]...)
|
|
|
|
allVars = append(allVars, build.BannerVars...)
|
|
|
|
}
|
|
|
|
|
|
|
|
if len(allVars) == 0 {
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
varData, err := build.DumpMakeVars(ctx, config, nil, allVars)
|
|
|
|
if err != nil {
|
|
|
|
ctx.Fatal(err)
|
|
|
|
}
|
|
|
|
|
|
|
|
for _, name := range vars {
|
|
|
|
if name == "report_config" {
|
|
|
|
fmt.Printf("%sreport_config='%s'\n", *varPrefix, build.Banner(varData))
|
|
|
|
} else {
|
|
|
|
fmt.Printf("%s%s='%s'\n", *varPrefix, name, varData[name])
|
|
|
|
}
|
|
|
|
}
|
|
|
|
for _, name := range absVars {
|
|
|
|
var res []string
|
|
|
|
for _, path := range strings.Fields(varData[name]) {
|
|
|
|
abs, err := filepath.Abs(path)
|
|
|
|
if err != nil {
|
|
|
|
ctx.Fatalln("Failed to get absolute path of", path, err)
|
|
|
|
}
|
|
|
|
res = append(res, abs)
|
|
|
|
}
|
|
|
|
fmt.Printf("%s%s='%s'\n", *absVarPrefix, name, strings.Join(res, " "))
|
|
|
|
}
|
Add a Go replacement for our top-level Make wrapper
Right now this mostly just copies what Make is doing in
build/core/ninja.mk and build/core/soong.mk. The only major feature it
adds is a rotating log file with some verbose logging.
There is one major functional difference -- you cannot override random
Make variables during the Make phase anymore. The environment variable
is set, and if Make uses ?= or the equivalent, it can still use those
variables. We already made this change for Kati, which also loads all of
the same code and actually does the build, so it has been half-removed
for a while.
The only "UI" this implements is what I'll call "Make Emulation" mode --
it's expected that current command lines will continue working, and
we'll explore alternate user interfaces later.
We're still using Make as a wrapper, but all it does is call into this
single Go program, it won't even load the product configuration. Once
this is default, we can start moving individual users over to using this
directly (still in Make emulation mode), skipping the Make wrapper.
Ideas for the future:
* Generating trace files showing time spent in Make/Kati/Soong/Ninja
(also importing ninja traces into the same stream). I had this working
in a previous version of this patch, but removed it to keep the size
down and focus on the current features.
* More intelligent SIGALRM handling, once we fully remove the Make
wrapper (which hides the SIGALRM)
* Reading the experimental binary output stream from Ninja, so that we
can always save the verbose log even if we're not printing it out to
the console
Test: USE_SOONG_UI=true m -j blueprint_tools
Change-Id: I884327b9a8ae24499eb6c56f6e1ad26df1cfa4e4
2016-08-22 06:17:17 +08:00
|
|
|
}
|