Forcing compiler flags on deps

[Raghu Krishnamurthy]

Hello experts,

I'm working on a GN project(https://git.trustedfirmware.org/hafnium/hafnium.git/), mostly a C project,  and i'm trying to find out how i can do the following. I've tried using gn help and looking through the gn referece docs but could not really find the answer.

I have target, that is a template, that eventually compiles into an executable. The target has the 'deps' field which contain only source_set's containing C files. Each source_set has its own public_config that includes extra defines and flags as required by the source set. What i want to be able to do is to use the same source_sets and create two different target executables, that only differ in a few flags. For example, i want target1 to have -DFOO and target2 to not have -DFOO. If i add a config at the target level, it only applies while linking but not to all the source_set's that the target depends on, ie. all the C files in both target1 and target2 dont compile with -DFOO. I have an example with comments:

config("myconfig_a") {

  defines = ["DEFINESFOR_A"]

}

config("myconfig_b") {

  defines = ["DEFINESFOR_B"]

}

config("myconfig_target1") {

  defines = ["FOO"]

}

source_set("src_a") {

  public_configs = [":myconfig_a"]

  sources = ["a.c"]

}

source_set("src_b") {

  public_configs = [":myconfig_a"]

  sources = ["a.c"]

}

template("my_target") {

  executables(target_name) {

   forward_variables_from(invoker,  [cflags, cflags_c, defines,public_configs, sources])

  }

}

my_target("target1") {

 public_configs = [":myconfig_target1"]

# I want c files in src_a and src_b to also include myconfig_target1 defines

# THIS DOES NOT WORK AS I WANT IT TO

 deps= [

    ":src_a",

    ":src_b"

  ]

}

my_target("target2") {

# I want c files in src_a and src_b to NOT include myconfig_target1 defines

 # THIS WORKS AS EXPECTED

 deps= [

    ":src_a",

    ":src_b"

  ]

}

Thanks

Raghu

[Brett Wilson]

The public configs thing applies only to the targets that depend on you.

A target can never modify its dependencies. Otherwise, that creates a situation where that other target has an ambiguous definition based on who is depending on it. In addition to forcing a target to have multiple definitions, this would also create unsolvable build graphs if the same target is referred as a dependency through multiple paths, each with their own flags.

Brett

[raghu....@gmail.com]

Thanks Brett. Understood. In that case what is the GN way to solve the situation below? How are debug vs production builds typically created for example on a given target? How can I build the same source set with different flags?

Do I just create different source sets for the two targets with different configs?

[Brett Wilson]

Most commonly debug and release would be different builds. The developer would then configure their build with the build args. So your BUILDCONFIG.gn would do something like

  declare_args() {

    is_debug = true  # Default to debug build.

  }

And then in your compiler setup code for your build you have something like:

  if (is_debug) {

   cflags = [ ... ]

  } else {

   cflags = [ ... ]

  }

Then the developer would use "gn args" to set or clear the is_debug clag according to what they want. I maintain and out/debug and out/release directory for most of my projects.

If you really need to do this in the same build, you can use toolchains to define a separate debug and release toolchain that both get compiled. But this is more complicated and normally less developer-friendly.

Brett

[raghu....@gmail.com]

Thanks Brett. I think that answers my question.