pmdabpf: prototype eBPF Kelper module (sustainable-computing.io)

src/pmdas/bpf/README

@@ -1,37 +1,42 @@
 PCP PMDA to load linux BPF modules
 ==================================
 
-This PMDA is capable of collecting and generating arbitrary metrics from kernel-side code running as pre-compiled
-ELF BPF/eBPF modules.
+This PMDA is capable of collecting and generating arbitrary metrics from
+kernel-side code running as pre-compiled ELF BPF/eBPF modules.
 
 Comparison to other PCP PMDAS
 =============================
 
-- pmdabcc runs as python and is a little easier to develop and maintain, has more advanced configuration files,
-  however it requires significantly more runtime memory. It loads and compiles the BCC code through LLVM which
-  has a heavy footprint to work with the fact that kernel structures might change. By comparison, pmdabpf uses
-  pre-compiled ELF-based BPF CO-RE modules to avoid relocation constraints.
+- pmdabcc runs as python and is a little easier to develop and maintain,
+  however it requires significantly more runtime memory. It loads and
+  compiles the BCC code through LLVM which has a heavy footprint to
+  work with the fact that kernel structures might change. By comparison,
+  pmdabpf uses pre-compiled ELF-based BPF CO-RE modules to avoid
+  relocation constraints.
 
-- pmdabpftrace can load and run arbitrary bpftrace code, so it is very flexible, however it does this by shelling
-  out to bpftrace executable which again requires quite a lot of memory.
+- pmdabpftrace can load and run arbitrary bpftrace code, so it is very
+  flexible, however it does this by shelling out to bpftrace executable
+  which again requires quite a lot of memory.
 
 Deployment
 ==========
 
 The file `bpf.conf` lists modules, one per line, that will be started.
 
-A relatively new version of libbpf is required on the system. libbpf 0.1.0 is unlikely to work, however, libbpf 0.4.0
-has been shown to work.
+A relatively new version of libbpf is required on the system. libbpf
+1.0.0 is known to work.
 
 Development
 ===========
 
 To develop additional modules:
-- Follow the examples provided (runqlat and biolatency) to create a new module (which will output a .so). The entry
-  point in your .so will be a load_module() call that should return a newly allocated `struct module` object.
-- Create your bpf code (this will become a .bpf.o). Use the various `_helpers` headers.
-- Ensure `module.h` has correct unique setup for your cluster and metric ids.
-- Add details to `pmns` and `help` files to ensure they match the `module.h` changes.
+- Follow the examples provided (runqlat and biolatency) to create a new
+  module (which will output a .so). The entry point in your .so will
+  be a load_module() call that should return a newly allocated `struct
+  module` object.
+- Create your bpf code (this will become a .bpf.o). Use the various
+  `_helpers` headers.
+- Add your module name to the global known-modules list in `module.h`.
 
 TODO
 ====

src/pmdas/bpf/bpf.conf

@@ -135,3 +135,12 @@ enabled = false
 # cgroup        - string - unset : Trace process in cgroup path
 [biosnoop.so]
 enabled = false
+
+# This tool traces scheduling activity for use in power management.
+#
+# Configuration options:
+# Name              - type    - default
+#
+# process_count - int    - 20    : number of processes to keep in cache
+[kepler.so]
+enabled = false

src/pmdas/bpf/modules/GNUmakefile

@@ -64,6 +64,15 @@ APPS_BPF = \
 	tcpconnlat.bpf.c \
 	vfsstat.bpf.c \
 
+# non-libbpf header files.
+LOCAL_H = \
+	kepler.bpf.h \
+
+# non-libbpf *.bpf.c files.
+LOCAL_BPF = \
+	kepler.bpf.c \
+
+PMDABPF_MODULES += kepler.so
 
 default_pcp default:	$(PMDABPF_MODULES)
 
@@ -99,7 +108,7 @@ $(APPS_BPF):
 vmlinux.h: $(PMDABPF_VMLINUXH)
 	$(LN_S) -f $< $@
 
-%.bpf.o: %.bpf.c vmlinux.h $(APPS_H) $(APPS_BPF)
+%.bpf.o: %.bpf.c vmlinux.h $(APPS_H) $(APPS_BPF) $(LOCAL_H) $(LOCAL_BPF)
 	$(CLANG) -Wall -g -O2 -target bpf -D__TARGET_ARCH_$(PMDABPF_ARCH) \
 		-I. -c $< -o $@
 	$(LLVM_STRIP) -g $@
@@ -108,7 +117,7 @@ vmlinux.h: $(PMDABPF_VMLINUXH)
 	$(BPFTOOL) gen skeleton $< > $@
 
 %.o: %.c
-%.o: %.c %.skel.h $(HELPERS_H) $(APPS_H)
+%.o: %.c %.skel.h $(HELPERS_H) $(APPS_H) $(LOCAL_H)
 	$(CC) -c $(CFLAGS) $(CPPFLAGS) $< -o $@
 
 %_helpers.o: %_helpers.c