Skip to contentSkip to conversation

bpf: Add d_path helper - preparation changes

PATCH Superseded Jiri Olsa 3 participants 22 messages Jul 3, 2020 – Jul 8, 2020

Subsystems (12)

v5 of 12 Superseded Latest is v12 →

lore
Share & tools
lore

Jiri Olsa

author reviewer
#0 lore raw

v5 cover bpf: Add d_path helper - preparation changes

hi,
this patchset does preparation work for adding d_path helper, which still needs more work, but the initial set of patches is ready and useful to have.

This patchset adds:
- support to generate BTF ID lists that are resolved during kernel linking and usable within kernel code with following macros:

BTF_ID_LIST(bpf_skb_output_btf_ids)
BTF_ID(struct, sk_buff)

and access it in kernel code via:
extern u32 bpf_skb_output_btf_ids[];

- resolve_btfids tool that scans elf object for .BTF_ids section and resolves its symbols with BTF ID values
- resolving of bpf_ctx_convert struct and several other objects with BTF_ID_LIST

v5 changes:
- added acks to patches I did not change in this version
- split the original patchset into 2 parts and sending the first one where the support to define BTF list is added, the BTF set support will be posted later on [Andrii]
- used u32 instead of int in btf_ids.h [Andrii]
- changed the btf_ids.h define guard [Andrii]
- added resolve_btfids_clean target [Andrii]
- moved resolve_btfids test into prog_tests suite [Andrii]
- fixed BTF type iteration index in test [Andrii]
- removed btf_id checks in bpf_ctx_convert resolve code [Yonghong]
- removed WARN_ON_ONCE from btf_resolve_helper_id [Yonghong]
- added BTF_ID_UNUSED macro [Yonghong]
- fixed out of the tree build failure [0day bot]

Also available at:
https://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
bpf/d_path

thanks,
jirka

---
Jiri Olsa (9):
bpf: Add resolve_btfids tool to resolve BTF IDs in ELF object
bpf: Compile resolve_btfids tool at kernel compilation start
bpf: Add BTF_ID_LIST/BTF_ID/BTF_ID_UNUSED macros
bpf: Resolve BTF IDs in vmlinux image
bpf: Remove btf_id helpers resolving
bpf: Use BTF_ID to resolve bpf_ctx_convert struct
bpf: Add info about .BTF.ids section to btf.rst
tools headers: Adopt verbatim copy of btf_ids.h from kernel sources selftests/bpf: Add test for resolve_btfids

 Documentation/bpf/btf.rst                               |  36 +++++
 Makefile                                                |  25 +++-
 include/asm-generic/vmlinux.lds.h                       |   4 +
 include/linux/btf_ids.h                                 |  87 ++++++++++++
 kernel/bpf/btf.c                                        | 103 ++------------
 kernel/trace/bpf_trace.c                                |   9 +-
 net/core/filter.c                                       |   9 +-
 scripts/link-vmlinux.sh                                 |   6 +
 tools/Makefile                                          |   3 +
 tools/bpf/Makefile                                      |   9 +-
 tools/bpf/resolve_btfids/Build                          |  26 ++++
 tools/bpf/resolve_btfids/Makefile                       |  77 +++++++++++
 tools/bpf/resolve_btfids/main.c                         | 716 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 tools/include/linux/btf_ids.h                           |  87 ++++++++++++
 tools/include/linux/compiler.h                          |   4 +
 tools/testing/selftests/bpf/Makefile                    |  22 ++-
 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c | 170 +++++++++++++++++++++++
 17 files changed, 1289 insertions(+), 104 deletions(-)
 create mode 100644 include/linux/btf_ids.h
 create mode 100644 tools/bpf/resolve_btfids/Build
 create mode 100644 tools/bpf/resolve_btfids/Makefile
 create mode 100644 tools/bpf/resolve_btfids/main.c
 create mode 100644 tools/include/linux/btf_ids.h
 create mode 100644 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c
author reviewer
#1 lore raw

PATCH v5 1/9 bpf: Add resolve_btfids tool to resolve BTF IDs in ELF object

The resolve_btfids tool scans elf object for .BTF_ids section and resolves its symbols with BTF ID values.

It will be used to during linking time to resolve arrays of BTF ID values used in verifier, so these IDs do not need to be resolved in runtime.

The expected layout of .BTF_ids section is described in btfid.c header. Related kernel changes are coming in following changes.

Build issue reported by 0-DAY CI Kernel Test Service.

---
 tools/bpf/resolve_btfids/Build    |  26 ++
 tools/bpf/resolve_btfids/Makefile |  77 ++++
 tools/bpf/resolve_btfids/main.c   | 716 ++++++++++++++++++++++++++++++
 3 files changed, 819 insertions(+)
 create mode 100644 tools/bpf/resolve_btfids/Build
 create mode 100644 tools/bpf/resolve_btfids/Makefile
 create mode 100644 tools/bpf/resolve_btfids/main.c
tools/bpf/resolve_btfids/Build new file +26 -0
new file mode 100644
index 000000000000..c7318cc55341
@@ -0,0 +1,26 @@
+resolve_btfids-y += main.o
+resolve_btfids-y += rbtree.o
+resolve_btfids-y += zalloc.o
+resolve_btfids-y += string.o
+resolve_btfids-y += ctype.o
+resolve_btfids-y += str_error_r.o
+
+$(OUTPUT)rbtree.o: ../../lib/rbtree.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)zalloc.o: ../../lib/zalloc.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)string.o: ../../lib/string.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)ctype.o: ../../lib/ctype.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)str_error_r.o: ../../lib/str_error_r.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
tools/bpf/resolve_btfids/Makefile new file +77 -0

77 lines not on this page · raw message

tools/bpf/resolve_btfids/main.c new file +716 -0

716 lines not on this page · raw message

Signature · 1 line
2.25.4
#2 lore raw

Re: [PATCH v5 bpf-next 1/9] bpf: Add resolve_btfids tool to resolve BTF IDs in ELF object

Quoted 45 lines · Jiri Olsa

The resolve_btfids tool scans elf object for .BTF_ids section and resolves its symbols with BTF ID values.

It will be used to during linking time to resolve arrays of BTF ID values used in verifier, so these IDs do not need to be resolved in runtime.

The expected layout of .BTF_ids section is described in btfid.c header. Related kernel changes are coming in following changes.

Build issue reported by 0-DAY CI Kernel Test Service.

---
 tools/bpf/resolve_btfids/Build    |  26 ++
 tools/bpf/resolve_btfids/Makefile |  77 ++++
 tools/bpf/resolve_btfids/main.c   | 716 ++++++++++++++++++++++++++++++
 3 files changed, 819 insertions(+)
 create mode 100644 tools/bpf/resolve_btfids/Build
 create mode 100644 tools/bpf/resolve_btfids/Makefile
 create mode 100644 tools/bpf/resolve_btfids/main.c
tools/bpf/resolve_btfids/Build new file +16 -0
new file mode 100644
index 000000000000..c7318cc55341
@@ -0,0 +1,26 @@
+resolve_btfids-y += main.o
+resolve_btfids-y += rbtree.o
+resolve_btfids-y += zalloc.o
+resolve_btfids-y += string.o
+resolve_btfids-y += ctype.o
+resolve_btfids-y += str_error_r.o
+
+$(OUTPUT)rbtree.o: ../../lib/rbtree.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)zalloc.o: ../../lib/zalloc.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)string.o: ../../lib/string.c FORCE

tools/bpf/resolve_btfids/Build line 26

Quoted 10 lines
tools/bpf/resolve_btfids/Build new file +10 -0
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)ctype.o: ../../lib/ctype.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)str_error_r.o: ../../lib/str_error_r.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)

Is Build also a Makefile? If that's the case, why not:

$(output)%.o: ../../lib/%.c FORCE

    $(call rule_mkdir)
    $(call if_changed_dep,cc_o_c)

?

Quoted 15 lines
tools/bpf/resolve_btfids/Makefile new file +9 -0
new file mode 100644
index 000000000000..948378ca73d4
@@ -0,0 +1,77 @@
+# SPDX-License-Identifier: GPL-2.0-only
+include ../../scripts/Makefile.include
+
+ifeq ($(srctree),)
+srctree := $(patsubst %/,%,$(dir $(CURDIR)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+endif
+

tools/bpf/resolve_btfids/Makefile line 19

Quoted 10 lines
tools/bpf/resolve_btfids/Makefile new file +10 -0
+ifeq ($(V),1)
+  Q =
+  msg =
+else
+  Q = @
+  msg = @printf '  %-8s %s%s\n' "$(1)" "$(notdir $(2))" "$(if $(3), $(3))";
+  MAKEFLAGS=--no-print-directory
+endif
+
+OUTPUT ?= $(srctree)/tools/bpf/resolve_btfids/

Ok, so this builds nicely for in-tree build, but when I did out-of-tree build (I use KBUILD_OUTPUT, haven't checked specifying O=whatever), I get:

LD vmlinux
BTFIDS vmlinux
/data/users/andriin/linux/scripts/link-vmlinux.sh: line 342:
/data/users/andriin/linux/tools/bpf/resolve_btfids/resolve_btfids: No such file or directory

I suspect because you are assuming OUTPUT to be in srctree? You probably need to adjust for out-of-tree mode.

Quoted 6 lines
tools/bpf/resolve_btfids/Makefile new file +6 -0
+
+LIBBPF_SRC := $(srctree)/tools/lib/bpf/
+SUBCMD_SRC := $(srctree)/tools/lib/subcmd/
+
+BPFOBJ     := $(OUTPUT)/libbpf.a
+SUBCMDOBJ  := $(OUTPUT)/libsubcmd.a

[...]

tools/bpf/resolve_btfids/Makefile

Quoted 2 lines
tools/bpf/resolve_btfids/Makefile new file +2 -0
+
+#define BTF_IDS_SECTION        ".BTF.ids"

You haven't updated a bunch of places (cover letter, this patch commit message, maybe somewhere else) after renaming from .BTF_ids, please keep them in sync. Also, while I'm not too strongly against this name, it does sound like this section is part of generic BTF format (as is .BTF and .BTF.ext), which it is not, because it's so kernel-specific. So I'm mildly against it and pro .BTF_ids.

Quoted 8 lines
tools/bpf/resolve_btfids/Makefile new file +8 -0
+#define BTF_ID         "__BTF_ID__"
+
+#define BTF_STRUCT     "struct"
+#define BTF_UNION      "union"
+#define BTF_TYPEDEF    "typedef"
+#define BTF_FUNC       "func"
+#define BTF_SET                "set"
+

[...]

tools/bpf/resolve_btfids/Makefile

Quoted 3 lines
tools/bpf/resolve_btfids/Makefile new file +3 -0
+}
+
+static struct btf *btf__parse_raw(const char *file)

I thought you were going to add this to libbpf itself? Or you planned to do a follow up patch later?

Quoted 33 lines
tools/bpf/resolve_btfids/Makefile new file +33 -0
+{
+       struct btf *btf;
+       struct stat st;
+       __u8 *buf;
+       FILE *f;
+
+       if (stat(file, &st))
+               return NULL;
+
+       f = fopen(file, "rb");
+       if (!f)
+               return NULL;
+
+       buf = malloc(st.st_size);
+       if (!buf) {
+               btf = ERR_PTR(-ENOMEM);
+               goto exit_close;
+       }
+
+       if ((size_t) st.st_size != fread(buf, 1, st.st_size, f)) {
+               btf = ERR_PTR(-EINVAL);
+               goto exit_free;
+       }
+
+       btf = btf__new(buf, st.st_size);
+
+exit_free:
+       free(buf);
+exit_close:
+       fclose(f);
+       return btf;
+}
+

[...]

#3 lore raw

Re: [PATCH v5 bpf-next 1/9] bpf: Add resolve_btfids tool to resolve BTF IDs in ELF object

On Mon, Jul 06, 2020 at 05:34:52PM -0700, Andrii Nakryiko wrote:

SNIP

Quoted 22 lines
Quoted 9 lines

+$(OUTPUT)rbtree.o: ../../lib/rbtree.c FORCE

tools/bpf/resolve_btfids/Build new file +8 -0
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)zalloc.o: ../../lib/zalloc.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)string.o: ../../lib/string.c FORCE

tools/bpf/resolve_btfids/Build line 26

Quoted 10 lines
tools/bpf/resolve_btfids/Build new file +10 -0
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)ctype.o: ../../lib/ctype.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)str_error_r.o: ../../lib/str_error_r.c FORCE
+       $(call rule_mkdir)
+       $(call if_changed_dep,cc_o_c)

Is Build also a Makefile? If that's the case, why not:

Quoted 3 lines

$(output)%.o: ../../lib/%.c FORCE

    $(call rule_mkdir)
    $(call if_changed_dep,cc_o_c)

hum, it is ... I'll try that

Quoted 33 lines

?

Quoted 15 lines
tools/bpf/resolve_btfids/Makefile new file +9 -0
new file mode 100644
index 000000000000..948378ca73d4
@@ -0,0 +1,77 @@
+# SPDX-License-Identifier: GPL-2.0-only
+include ../../scripts/Makefile.include
+
+ifeq ($(srctree),)
+srctree := $(patsubst %/,%,$(dir $(CURDIR)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+endif
+

tools/bpf/resolve_btfids/Makefile line 19

Quoted 10 lines
tools/bpf/resolve_btfids/Makefile new file +10 -0
+ifeq ($(V),1)
+  Q =
+  msg =
+else
+  Q = @
+  msg = @printf '  %-8s %s%s\n' "$(1)" "$(notdir $(2))" "$(if $(3), $(3))";
+  MAKEFLAGS=--no-print-directory
+endif
+
+OUTPUT ?= $(srctree)/tools/bpf/resolve_btfids/

Ok, so this builds nicely for in-tree build, but when I did out-of-tree build (I use KBUILD_OUTPUT, haven't checked specifying O=whatever), I get:

Quoted 8 lines

LD vmlinux
BTFIDS vmlinux
/data/users/andriin/linux/scripts/link-vmlinux.sh: line 342:
/data/users/andriin/linux/tools/bpf/resolve_btfids/resolve_btfids: No such file or directory

I suspect because you are assuming OUTPUT to be in srctree? You probably need to adjust for out-of-tree mode.

ok, make clean did not clean resolve_btfids, so it was still there when I tried the out of the tree build..

Quoted 10 lines
Quoted 7 lines
tools/bpf/resolve_btfids/Makefile new file +6 -0
+
+LIBBPF_SRC := $(srctree)/tools/lib/bpf/
+SUBCMD_SRC := $(srctree)/tools/lib/subcmd/
+
+BPFOBJ     := $(OUTPUT)/libbpf.a
+SUBCMDOBJ  := $(OUTPUT)/libsubcmd.a

[...]

Quoted 9 lines

tools/bpf/resolve_btfids/Makefile

Quoted 2 lines
tools/bpf/resolve_btfids/Makefile new file +2 -0
+
+#define BTF_IDS_SECTION        ".BTF.ids"

You haven't updated a bunch of places (cover letter, this patch commit message, maybe somewhere else) after renaming from .BTF_ids, please keep them in sync. Also, while I'm not too strongly against this name, it does sound like this section is part of generic BTF format (as is .BTF and .BTF.ext), which it is not, because it's so kernel-specific. So I'm mildly against it and pro .BTF_ids.

.BTF_ids it is.. I'll change all the places

Quoted 12 lines
Quoted 9 lines
tools/bpf/resolve_btfids/Makefile new file +8 -0
+#define BTF_ID         "__BTF_ID__"
+
+#define BTF_STRUCT     "struct"
+#define BTF_UNION      "union"
+#define BTF_TYPEDEF    "typedef"
+#define BTF_FUNC       "func"
+#define BTF_SET                "set"
+

[...]

Quoted 6 lines

tools/bpf/resolve_btfids/Makefile

Quoted 3 lines
tools/bpf/resolve_btfids/Makefile new file +3 -0
+}
+
+static struct btf *btf__parse_raw(const char *file)

I thought you were going to add this to libbpf itself? Or you planned to do a follow up patch later?

yes, I did not want to complicate this patchset more, and send this change after.. also then there'll be one more reason to make it a library function ;-)

thanks,
jirka

author reviewer
#4 lore raw

PATCH v5 2/9 bpf: Compile resolve_btfids tool at kernel compilation start

In reply to Jiri Olsa

The resolve_btfids tool will be used during the vmlinux linking, so it's necessary it's ready for it.

---
 Makefile           | 22 ++++++++++++++++++----
 tools/Makefile     |  3 +++
 tools/bpf/Makefile |  9 ++++++++-
 3 files changed, 29 insertions(+), 5 deletions(-)
Makefile +18 -4
index ae5d8220f431..8db4fd8097e0 100644
@@ -1053,9 +1053,10 @@ export mod_sign_cmd
 
 HOST_LIBELF_LIBS = $(shell pkg-config libelf --libs 2>/dev/null || echo -lelf)
 
+has_libelf = $(call try-run,\
+               echo "int main() {}" | $(HOSTCC) -xc -o /dev/null $(HOST_LIBELF_LIBS) -,1,0)
+
 ifdef CONFIG_STACK_VALIDATION
-  has_libelf := $(call try-run,\
-		echo "int main() {}" | $(HOSTCC) -xc -o /dev/null $(HOST_LIBELF_LIBS) -,1,0)
   ifeq ($(has_libelf),1)
     objtool_target := tools/objtool FORCE
   else
@@ -1064,6 +1065,14 @@ ifdef CONFIG_STACK_VALIDATION
   endif
 endif
 
+ifdef CONFIG_DEBUG_INFO_BTF
+  ifeq ($(has_libelf),1)
+    resolve_btfids_target := tools/bpf/resolve_btfids FORCE
+  else
+    ERROR_RESOLVE_BTFIDS := 1
+  endif
+endif
+
 PHONY += prepare0
 
 export MODORDER := $(extmod-prefix)modules.order
@@ -1175,7 +1184,7 @@ prepare0: archprepare
 	$(Q)$(MAKE) $(build)=.
 
 # All the preparing..
-prepare: prepare0 prepare-objtool
+prepare: prepare0 prepare-objtool prepare-resolve_btfids
 
 # Support for using generic headers in asm-generic
 asm-generic := -f $(srctree)/scripts/Makefile.asm-generic obj
@@ -1188,7 +1197,7 @@ uapi-asm-generic:
 	$(Q)$(MAKE) $(asm-generic)=arch/$(SRCARCH)/include/generated/uapi/asm \
 	generic=include/uapi/asm-generic
 
-PHONY += prepare-objtool
+PHONY += prepare-objtool prepare-resolve_btfids
 prepare-objtool: $(objtool_target)
 ifeq ($(SKIP_STACK_VALIDATION),1)
 ifdef CONFIG_UNWINDER_ORC
@@ -1199,6 +1208,11 @@ else
 endif
 endif
 
+prepare-resolve_btfids: $(resolve_btfids_target)
+ifeq ($(ERROR_RESOLVE_BTFIDS),1)
+	@echo "error: Cannot resolve BTF IDs for CONFIG_DEBUG_INFO_BTF, please install libelf-dev, libelf-devel or elfutils-libelf-devel" >&2
+	@false
+endif
 # Generate some files
 # ---------------------------------------------------------------------------
 
tools/Makefile +3 -0
index bd778812e915..85af6ebbce91 100644
@@ -67,6 +67,9 @@ cpupower: FORCE
 cgroup firewire hv guest bootconfig spi usb virtio vm bpf iio gpio objtool leds wmi pci firmware debugging: FORCE
 	$(call descend,$@)
 
+bpf/%: FORCE
+	$(call descend,$@)
+
 liblockdep: FORCE
 	$(call descend,lib/lockdep)
 
tools/bpf/Makefile +8 -1
index 6df1850f8353..74bc9a105225 100644
@@ -123,5 +123,12 @@ runqslower_install:
 runqslower_clean:
 	$(call descend,runqslower,clean)
 
+resolve_btfids:
+	$(call descend,resolve_btfids)
+
+resolve_btfids_clean:
+	$(call descend,resolve_btfids,clean)
+
 .PHONY: all install clean bpftool bpftool_install bpftool_clean \
-	runqslower runqslower_install runqslower_clean
+	runqslower runqslower_install runqslower_clean \
+	resolve_btfids resolve_btfids_clean
Signature · 1 line
2.25.4
author reviewer
#5 lore raw

PATCH v5 3/9 bpf: Add BTF_ID_LIST/BTF_ID/BTF_ID_UNUSED macros

In reply to Jiri Olsa

Adding support to generate .BTF_ids section that will hold BTF ID lists for verifier.

Adding macros that will help to define lists of BTF ID values placed in .BTF_ids section. They are initially filled with zeros (during compilation) and resolved later during the linking phase by resolve_btfids tool.

Following defines list of one BTF ID value:

BTF_ID_LIST(bpf_skb_output_btf_ids)
BTF_ID(struct, sk_buff)

It also defines following variable to access the list:

  extern u32 bpf_skb_output_btf_ids[];

The BTF_ID_UNUSED macro defines 4 zero bytes. It's used when we want to define 'unused' entry in BTF_ID_LIST, like:

BTF_ID_LIST(bpf_skb_output_btf_ids)
BTF_ID(struct, sk_buff)
BTF_ID_UNUSED
BTF_ID(struct, task_struct)

---
 include/asm-generic/vmlinux.lds.h |  4 ++
 include/linux/btf_ids.h           | 87 +++++++++++++++++++++++++++++++
 2 files changed, 91 insertions(+)
 create mode 100644 include/linux/btf_ids.h
include/asm-generic/vmlinux.lds.h +4 -0
index db600ef218d7..a9b77a4314a8 100644
@@ -641,6 +641,10 @@
 		__start_BTF = .;					\
 		*(.BTF)							\
 		__stop_BTF = .;						\
+	}								\
+	. = ALIGN(4);							\
+	.BTF.ids : AT(ADDR(.BTF.ids) - LOAD_OFFSET) {			\
+		*(.BTF.ids)						\
 	}
 #else
 #define BTF
include/linux/btf_ids.h new file +87 -0

87 lines not on this page · raw message

Signature · 1 line
2.25.4
author reviewer
#6 lore raw

PATCH v5 4/9 bpf: Resolve BTF IDs in vmlinux image

In reply to Jiri Olsa

Using BTF_ID_LIST macro to define lists for several helpers using BTF arguments.

And running resolve_btfids on vmlinux elf object during linking, so the .BTF_ids section gets the IDs resolved.

---
 Makefile                 | 3 ++-
 kernel/trace/bpf_trace.c | 9 +++++++--
 net/core/filter.c        | 9 +++++++--
 scripts/link-vmlinux.sh  | 6 ++++++
 4 files changed, 22 insertions(+), 5 deletions(-)
Makefile +2 -1
index 8db4fd8097e0..def58d4f9ed7 100644
@@ -448,6 +448,7 @@ OBJSIZE		= $(CROSS_COMPILE)size
 STRIP		= $(CROSS_COMPILE)strip
 endif
 PAHOLE		= pahole
+RESOLVE_BTFIDS	= $(srctree)/tools/bpf/resolve_btfids/resolve_btfids
 LEX		= flex
 YACC		= bison
 AWK		= awk
@@ -510,7 +511,7 @@ GCC_PLUGINS_CFLAGS :=
 CLANG_FLAGS :=
 
 export ARCH SRCARCH CONFIG_SHELL BASH HOSTCC KBUILD_HOSTCFLAGS CROSS_COMPILE LD CC
-export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE LEX YACC AWK INSTALLKERNEL
+export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE RESOLVE_BTFIDS LEX YACC AWK INSTALLKERNEL
 export PERL PYTHON PYTHON3 CHECK CHECKFLAGS MAKE UTS_MACHINE HOSTCXX
 export KGZIP KBZIP2 KLZOP LZMA LZ4 XZ
 export KBUILD_HOSTCXXFLAGS KBUILD_HOSTLDFLAGS KBUILD_HOSTLDLIBS LDFLAGS_MODULE
kernel/trace/bpf_trace.c +7 -2
index 1d874d8e4384..e26fbb1dad9c 100644
@@ -13,6 +13,7 @@
 #include <linux/kprobes.h>
 #include <linux/syscalls.h>
 #include <linux/error-injection.h>
+#include <linux/btf_ids.h>
 
 #include <asm/tlb.h>
 
@@ -710,7 +711,9 @@ BPF_CALL_5(bpf_seq_printf, struct seq_file *, m, char *, fmt, u32, fmt_size,
 	return err;
 }
 
-static int bpf_seq_printf_btf_ids[5];
+BTF_ID_LIST(bpf_seq_printf_btf_ids)
+BTF_ID(struct, seq_file)
+
 static const struct bpf_func_proto bpf_seq_printf_proto = {
 	.func		= bpf_seq_printf,
 	.gpl_only	= true,
@@ -728,7 +731,9 @@ BPF_CALL_3(bpf_seq_write, struct seq_file *, m, const void *, data, u32, len)
 	return seq_write(m, data, len) ? -EOVERFLOW : 0;
 }
 
-static int bpf_seq_write_btf_ids[5];
+BTF_ID_LIST(bpf_seq_write_btf_ids)
+BTF_ID(struct, seq_file)
+
 static const struct bpf_func_proto bpf_seq_write_proto = {
 	.func		= bpf_seq_write,
 	.gpl_only	= true,
net/core/filter.c +7 -2
index c5e696e6c315..a859ad681988 100644
@@ -75,6 +75,7 @@
 #include <net/ipv6_stubs.h>
 #include <net/bpf_sk_storage.h>
 #include <net/transp_v6.h>
+#include <linux/btf_ids.h>
 
 /**
  *	sk_filter_trim_cap - run a packet through a socket filter
@@ -3779,7 +3780,9 @@ static const struct bpf_func_proto bpf_skb_event_output_proto = {
 	.arg5_type	= ARG_CONST_SIZE_OR_ZERO,
 };
 
-static int bpf_skb_output_btf_ids[5];
+BTF_ID_LIST(bpf_skb_output_btf_ids)
+BTF_ID(struct, sk_buff)
+
 const struct bpf_func_proto bpf_skb_output_proto = {
 	.func		= bpf_skb_event_output,
 	.gpl_only	= true,
@@ -4173,7 +4176,9 @@ static const struct bpf_func_proto bpf_xdp_event_output_proto = {
 	.arg5_type	= ARG_CONST_SIZE_OR_ZERO,
 };
 
-static int bpf_xdp_output_btf_ids[5];
+BTF_ID_LIST(bpf_xdp_output_btf_ids)
+BTF_ID(struct, xdp_buff)
+
 const struct bpf_func_proto bpf_xdp_output_proto = {
 	.func		= bpf_xdp_event_output,
 	.gpl_only	= true,
scripts/link-vmlinux.sh +6 -0
index 92dd745906f4..e26f02dbedee 100755
@@ -336,6 +336,12 @@ fi
 
 vmlinux_link vmlinux "${kallsymso}" ${btf_vmlinux_bin_o}
 
+# fill in BTF IDs
+if [ -n "${CONFIG_DEBUG_INFO_BTF}" ]; then
+info BTFIDS vmlinux
+${RESOLVE_BTFIDS} vmlinux
+fi
+
 if [ -n "${CONFIG_BUILDTIME_TABLE_SORT}" ]; then
 	info SORTTAB vmlinux
 	if ! sorttable vmlinux; then
Signature · 1 line
2.25.4
#7 lore raw

Re: [PATCH v5 bpf-next 4/9] bpf: Resolve BTF IDs in vmlinux image

Quoted 21 lines · Jiri Olsa

Using BTF_ID_LIST macro to define lists for several helpers using BTF arguments.

And running resolve_btfids on vmlinux elf object during linking, so the .BTF_ids section gets the IDs resolved.

---
 Makefile                 | 3 ++-
 kernel/trace/bpf_trace.c | 9 +++++++--
 net/core/filter.c        | 9 +++++++--
 scripts/link-vmlinux.sh  | 6 ++++++
 4 files changed, 22 insertions(+), 5 deletions(-)
Makefile +0 -0
index 8db4fd8097e0..def58d4f9ed7 100644

Makefile OBJSIZE = $(CROSS_COMPILE)size · line 451

Quoted 4 lines
Makefile +1 -0
@@ -448,6 +448,7 @@ OBJSIZE             = $(CROSS_COMPILE)size
 STRIP          = $(CROSS_COMPILE)strip
 endif
 PAHOLE         = pahole
+RESOLVE_BTFIDS = $(srctree)/tools/bpf/resolve_btfids/resolve_btfids

Oh, this is probably wrong and why out-of-tree build fails. Why don't you follow how this is done for objtool?

Quoted 12 lines
Makefile +1 -1
 LEX            = flex
 YACC           = bison
 AWK            = awk
@@ -510,7 +511,7 @@ GCC_PLUGINS_CFLAGS :=
 CLANG_FLAGS :=
 
 export ARCH SRCARCH CONFIG_SHELL BASH HOSTCC KBUILD_HOSTCFLAGS CROSS_COMPILE LD CC
-export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE LEX YACC AWK INSTALLKERNEL
+export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE RESOLVE_BTFIDS LEX YACC AWK INSTALLKERNEL
 export PERL PYTHON PYTHON3 CHECK CHECKFLAGS MAKE UTS_MACHINE HOSTCXX
 export KGZIP KBZIP2 KLZOP LZMA LZ4 XZ
 export KBUILD_HOSTCXXFLAGS KBUILD_HOSTLDFLAGS KBUILD_HOSTLDLIBS LDFLAGS_MODULE

[...]

#8 lore raw

Re: [PATCH v5 bpf-next 4/9] bpf: Resolve BTF IDs in vmlinux image

Quoted 17 lines · Andrii Nakryiko
Quoted 16 lines · Jiri Olsa

Using BTF_ID_LIST macro to define lists for several helpers using BTF arguments.

And running resolve_btfids on vmlinux elf object during linking, so the .BTF_ids section gets the IDs resolved.

---
 Makefile                 | 3 ++-
 kernel/trace/bpf_trace.c | 9 +++++++--
 net/core/filter.c        | 9 +++++++--
 scripts/link-vmlinux.sh  | 6 ++++++
 4 files changed, 22 insertions(+), 5 deletions(-)
Quoted 12 lines

Makefile OBJSIZE = $(CROSS_COMPILE)size · line 451

Quoted 9 lines
Makefile +1 -0
index 8db4fd8097e0..def58d4f9ed7 100644
@@ -448,6 +448,7 @@ OBJSIZE             = $(CROSS_COMPILE)size
 STRIP          = $(CROSS_COMPILE)strip
 endif
 PAHOLE         = pahole
+RESOLVE_BTFIDS = $(srctree)/tools/bpf/resolve_btfids/resolve_btfids

Oh, this is probably wrong and why out-of-tree build fails. Why don't you follow how this is done for objtool?

right, should be $(objtree) in there

jirka

Quoted 16 lines
Quoted 13 lines
Makefile +1 -1
 LEX            = flex
 YACC           = bison
 AWK            = awk
@@ -510,7 +511,7 @@ GCC_PLUGINS_CFLAGS :=
 CLANG_FLAGS :=
 
 export ARCH SRCARCH CONFIG_SHELL BASH HOSTCC KBUILD_HOSTCFLAGS CROSS_COMPILE LD CC
-export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE LEX YACC AWK INSTALLKERNEL
+export CPP AR NM STRIP OBJCOPY OBJDUMP OBJSIZE READELF PAHOLE RESOLVE_BTFIDS LEX YACC AWK INSTALLKERNEL
 export PERL PYTHON PYTHON3 CHECK CHECKFLAGS MAKE UTS_MACHINE HOSTCXX
 export KGZIP KBZIP2 KLZOP LZMA LZ4 XZ
 export KBUILD_HOSTCXXFLAGS KBUILD_HOSTLDFLAGS KBUILD_HOSTLDLIBS LDFLAGS_MODULE

[...]

author reviewer
#9 lore raw

PATCH v5 5/9 bpf: Remove btf_id helpers resolving

In reply to Jiri Olsa

Now when we moved the helpers btf_id arrays into .BTF_ids section, we can remove the code that resolve those IDs in runtime.

---
 kernel/bpf/btf.c | 89 +++---------------------------------------------
 1 file changed, 5 insertions(+), 84 deletions(-)
kernel/bpf/btf.c +5 -84
index 4c3007f428b1..71140b73ae3c 100644
@@ -4079,96 +4079,17 @@ int btf_struct_access(struct bpf_verifier_log *log,
 	return -EINVAL;
 }
 
-static int __btf_resolve_helper_id(struct bpf_verifier_log *log, void *fn,
-				   int arg)
-{
-	char fnname[KSYM_SYMBOL_LEN + 4] = "btf_";
-	const struct btf_param *args;
-	const struct btf_type *t;
-	const char *tname, *sym;
-	u32 btf_id, i;
-
-	if (IS_ERR(btf_vmlinux)) {
-		bpf_log(log, "btf_vmlinux is malformed\n");
-		return -EINVAL;
-	}
-
-	sym = kallsyms_lookup((long)fn, NULL, NULL, NULL, fnname + 4);
-	if (!sym) {
-		bpf_log(log, "kernel doesn't have kallsyms\n");
-		return -EFAULT;
-	}
-
-	for (i = 1; i <= btf_vmlinux->nr_types; i++) {
-		t = btf_type_by_id(btf_vmlinux, i);
-		if (BTF_INFO_KIND(t->info) != BTF_KIND_TYPEDEF)
-			continue;
-		tname = __btf_name_by_offset(btf_vmlinux, t->name_off);
-		if (!strcmp(tname, fnname))
-			break;
-	}
-	if (i > btf_vmlinux->nr_types) {
-		bpf_log(log, "helper %s type is not found\n", fnname);
-		return -ENOENT;
-	}
-
-	t = btf_type_by_id(btf_vmlinux, t->type);
-	if (!btf_type_is_ptr(t))
-		return -EFAULT;
-	t = btf_type_by_id(btf_vmlinux, t->type);
-	if (!btf_type_is_func_proto(t))
-		return -EFAULT;
-
-	args = (const struct btf_param *)(t + 1);
-	if (arg >= btf_type_vlen(t)) {
-		bpf_log(log, "bpf helper %s doesn't have %d-th argument\n",
-			fnname, arg);
-		return -EINVAL;
-	}
-
-	t = btf_type_by_id(btf_vmlinux, args[arg].type);
-	if (!btf_type_is_ptr(t) || !t->type) {
-		/* anything but the pointer to struct is a helper config bug */
-		bpf_log(log, "ARG_PTR_TO_BTF is misconfigured\n");
-		return -EFAULT;
-	}
-	btf_id = t->type;
-	t = btf_type_by_id(btf_vmlinux, t->type);
-	/* skip modifiers */
-	while (btf_type_is_modifier(t)) {
-		btf_id = t->type;
-		t = btf_type_by_id(btf_vmlinux, t->type);
-	}
-	if (!btf_type_is_struct(t)) {
-		bpf_log(log, "ARG_PTR_TO_BTF is not a struct\n");
-		return -EFAULT;
-	}
-	bpf_log(log, "helper %s arg%d has btf_id %d struct %s\n", fnname + 4,
-		arg, btf_id, __btf_name_by_offset(btf_vmlinux, t->name_off));
-	return btf_id;
-}
-
 int btf_resolve_helper_id(struct bpf_verifier_log *log,
 			  const struct bpf_func_proto *fn, int arg)
 {
-	int *btf_id = &fn->btf_id[arg];
-	int ret;
+	int id;
 
 	if (fn->arg_type[arg] != ARG_PTR_TO_BTF_ID)
 		return -EINVAL;
-
-	ret = READ_ONCE(*btf_id);
-	if (ret)
-		return ret;
-	/* ok to race the search. The result is the same */
-	ret = __btf_resolve_helper_id(log, fn->func, arg);
-	if (!ret) {
-		/* Function argument cannot be type 'void' */
-		bpf_log(log, "BTF resolution bug\n");
-		return -EFAULT;
-	}
-	WRITE_ONCE(*btf_id, ret);
-	return ret;
+	id = fn->btf_id[arg];
+	if (!id || id > btf_vmlinux->nr_types)
+		return -EINVAL;
+	return id;
 }
 
 static int __get_type_size(struct btf *btf, u32 btf_id,
Signature · 1 line
2.25.4
Acked-by on v5 5/9
#10 lore raw

Re: [PATCH v5 bpf-next 5/9] bpf: Remove btf_id helpers resolving

Quoted 9 lines · Jiri Olsa

Now when we moved the helpers btf_id arrays into .BTF_ids section, we can remove the code that resolve those IDs in runtime.

---
 kernel/bpf/btf.c | 89 +++---------------------------------------------
 1 file changed, 5 insertions(+), 84 deletions(-)

LGTM.

Quoted 8 lines
kernel/bpf/btf.c +0 -0
index 4c3007f428b1..71140b73ae3c 100644
@@ -4079,96 +4079,17 @@ int btf_struct_access(struct bpf_verifier_log *log,
        return -EINVAL;
 }
 

[...]

author reviewer
#11 lore raw

PATCH v5 6/9 bpf: Use BTF_ID to resolve bpf_ctx_convert struct

In reply to Jiri Olsa

This way the ID is resolved during compile time,
and we can remove the runtime name search.

---
 kernel/bpf/btf.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)
kernel/bpf/btf.c +6 -8
index 71140b73ae3c..a710e3ee1f18 100644
@@ -18,6 +18,7 @@
 #include <linux/sort.h>
 #include <linux/bpf_verifier.h>
 #include <linux/btf.h>
+#include <linux/btf_ids.h>
 #include <linux/skmsg.h>
 #include <linux/perf_event.h>
 #include <net/sock.h>
@@ -3621,12 +3622,15 @@ static int btf_translate_to_vmlinux(struct bpf_verifier_log *log,
 	return kern_ctx_type->type;
 }
 
+BTF_ID_LIST(bpf_ctx_convert_btf_id)
+BTF_ID(struct, bpf_ctx_convert)
+
 struct btf *btf_parse_vmlinux(void)
 {
 	struct btf_verifier_env *env = NULL;
 	struct bpf_verifier_log *log;
 	struct btf *btf = NULL;
-	int err, btf_id;
+	int err;
 
 	env = kzalloc(sizeof(*env), GFP_KERNEL | __GFP_NOWARN);
 	if (!env)
@@ -3659,14 +3663,8 @@ struct btf *btf_parse_vmlinux(void)
 	if (err)
 		goto errout;
 
-	/* find struct bpf_ctx_convert for type checking later */
-	btf_id = btf_find_by_name_kind(btf, "bpf_ctx_convert", BTF_KIND_STRUCT);
-	if (btf_id < 0) {
-		err = btf_id;
-		goto errout;
-	}
 	/* btf_parse_vmlinux() runs under bpf_verifier_lock */
-	bpf_ctx_convert.t = btf_type_by_id(btf, btf_id);
+	bpf_ctx_convert.t = btf_type_by_id(btf, bpf_ctx_convert_btf_id[0]);
 
 	/* find bpf map structs for map_ptr access checking */
 	err = btf_vmlinux_map_ids_init(btf, log);
Signature · 1 line
2.25.4
Acked-by on v5 6/9
#12 lore raw

Re: [PATCH v5 bpf-next 6/9] bpf: Use BTF_ID to resolve bpf_ctx_convert struct

Quoted 6 lines · Jiri Olsa

This way the ID is resolved during compile time,
and we can remove the runtime name search.

---

Quoted 3 lines
 kernel/bpf/btf.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

[...]

author reviewer
#13 lore raw

PATCH v5 7/9 bpf: Add info about .BTF.ids section to btf.rst

In reply to Jiri Olsa

Updating btf.rst doc with info about .BTF.ids section

---
 Documentation/bpf/btf.rst | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)
Documentation/bpf/btf.rst +36 -0
index 4d565d202ce3..17ebd78d2c9e 100644
@@ -691,6 +691,42 @@ kernel API, the ``insn_off`` is the instruction offset in the unit of ``struct
 bpf_insn``. For ELF API, the ``insn_off`` is the byte offset from the
 beginning of section (``btf_ext_info_sec->sec_name_off``).
 
+4.2 .BTF.ids section
+====================
+
+The .BTF.ids section encodes BTF ID values that are used within the kernel.
+
+This section is created during the kernel compilation with the help of
+macros defined in ``include/linux/btf_ids.h`` header file. Kernel code can
+use them to create lists and sets (sorted lists) of BTF ID values.
+
+The ``BTF_ID_LIST`` and ``BTF_ID`` macros define unsorted list of BTF ID values,
+with following syntax::
+
+  BTF_ID_LIST(list)
+  BTF_ID(type1, name1)
+  BTF_ID(type2, name2)
+
+resulting in following layout in .BTF.ids section::
+
+  __BTF_ID__type1__name1__1:
+  .zero 4
+  __BTF_ID__type2__name2__2:
+  .zero 4
+
+The ``int list[];`` variable is defined to access the list.
+
+The ``BTF_ID_UNUSED`` macro defines 4 zero bytes. It's used when we
+want to define unused entry in BTF_ID_LIST, like::
+
+      BTF_ID_LIST(bpf_skb_output_btf_ids)
+      BTF_ID(struct, sk_buff)
+      BTF_ID_UNUSED
+      BTF_ID(struct, task_struct)
+
+All the BTF ID lists and sets are compiled in the .BTF.ids section and
+resolved during the linking phase of kernel build by ``resolve_btfids`` tool.
+
 5. Using BTF
 ************
 
Signature · 1 line
2.25.4
author reviewer
#14 lore raw

PATCH v5 8/9 tools headers: Adopt verbatim copy of btf_ids.h from kernel sources

In reply to Jiri Olsa

It will be needed by bpf selftest for resolve_btfids tool.

Also adding __PASTE macro as btf_ids.h dependency, which is defined in:

include/linux/compiler_types.h

but because tools/include do not have this header, I'm putting the macro into linux/compiler.h header.

---
 tools/include/linux/btf_ids.h  | 87 ++++++++++++++++++++++++++++++++++
 tools/include/linux/compiler.h |  4 ++
 2 files changed, 91 insertions(+)
 create mode 100644 tools/include/linux/btf_ids.h
tools/include/linux/btf_ids.h new file +87 -0
new file mode 100644
index 000000000000..d317150bc9e3
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_BTF_IDS_H
+#define _LINUX_BTF_IDS_H
+
+#include <linux/compiler.h> /* for __PASTE */
+
+/*
+ * Following macros help to define lists of BTF IDs placed
+ * in .BTF_ids section. They are initially filled with zeros
+ * (during compilation) and resolved later during the
+ * linking phase by btfid tool.
+ *
+ * Any change in list layout must be reflected in btfid
+ * tool logic.
+ */
+
+#define BTF_IDS_SECTION ".BTF.ids"
+
+#define ____BTF_ID(symbol)				\
+asm(							\
+".pushsection " BTF_IDS_SECTION ",\"a\";       \n"	\
+".local " #symbol " ;                          \n"	\
+".type  " #symbol ", @object;                  \n"	\
+".size  " #symbol ", 4;                        \n"	\
+#symbol ":                                     \n"	\
+".zero 4                                       \n"	\
+".popsection;                                  \n");
+
+#define __BTF_ID(symbol) \
+	____BTF_ID(symbol)
+
+#define __ID(prefix) \
+	__PASTE(prefix, __COUNTER__)
+
+/*
+ * The BTF_ID defines unique symbol for each ID pointing
+ * to 4 zero bytes.
+ */
+#define BTF_ID(prefix, name) \
+	__BTF_ID(__ID(__BTF_ID__##prefix##__##name##__))
+
+/*
+ * The BTF_ID_LIST macro defines pure (unsorted) list
+ * of BTF IDs, with following layout:
+ *
+ * BTF_ID_LIST(list1)
+ * BTF_ID(type1, name1)
+ * BTF_ID(type2, name2)
+ *
+ * list1:
+ * __BTF_ID__type1__name1__1:
+ * .zero 4
+ * __BTF_ID__type2__name2__2:
+ * .zero 4
+ *
+ */
+#define __BTF_ID_LIST(name)				\
+asm(							\
+".pushsection " BTF_IDS_SECTION ",\"a\";       \n"	\
+".local " #name ";                             \n"	\
+#name ":;                                      \n"	\
+".popsection;                                  \n");	\
+
+#define BTF_ID_LIST(name)				\
+__BTF_ID_LIST(name)					\
+extern u32 name[];
+
+/*
+ * The BTF_ID_UNUSED macro defines 4 zero bytes.
+ * It's used when we want to define 'unused' entry
+ * in BTF_ID_LIST, like:
+ *
+ *   BTF_ID_LIST(bpf_skb_output_btf_ids)
+ *   BTF_ID(struct, sk_buff)
+ *   BTF_ID_UNUSED
+ *   BTF_ID(struct, task_struct)
+ */
+
+#define BTF_ID_UNUSED					\
+asm(							\
+".pushsection " BTF_IDS_SECTION ",\"a\";       \n"	\
+".zero 4                                       \n"	\
+".popsection;                                  \n");
+
+
+#endif
tools/include/linux/compiler.h +4 -0
index 9f9002734e19..6eac24d44e81 100644
@@ -201,4 +201,8 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s
 # define __fallthrough
 #endif
 
+/* Indirect macros required for expanded argument pasting, eg. __LINE__. */
+#define ___PASTE(a, b) a##b
+#define __PASTE(a, b) ___PASTE(a, b)
+
 #endif /* _TOOLS_LINUX_COMPILER_H */
Signature · 1 line
2.25.4
#15 lore raw

Re: [PATCH v5 bpf-next 8/9] tools headers: Adopt verbatim copy of btf_ids.h from kernel sources

Quoted 25 lines · Jiri Olsa

It will be needed by bpf selftest for resolve_btfids tool.

Also adding __PASTE macro as btf_ids.h dependency, which is defined in:

include/linux/compiler_types.h

but because tools/include do not have this header, I'm putting the macro into linux/compiler.h header.

---
 tools/include/linux/btf_ids.h  | 87 ++++++++++++++++++++++++++++++++++
 tools/include/linux/compiler.h |  4 ++
 2 files changed, 91 insertions(+)
 create mode 100644 tools/include/linux/btf_ids.h
tools/include/linux/btf_ids.h new file +0 -0
new file mode 100644
index 000000000000..d317150bc9e3

tools/include/linux/btf_ids.h line 10

Quoted 10 lines
tools/include/linux/btf_ids.h new file +10 -0
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_BTF_IDS_H
+#define _LINUX_BTF_IDS_H
+
+#include <linux/compiler.h> /* for __PASTE */
+
+/*
+ * Following macros help to define lists of BTF IDs placed
+ * in .BTF_ids section. They are initially filled with zeros

One more inconsistency with .BTF_ids vs .BTF.ids (probably same in the original header, which I missed).

tools/include/linux/btf_ids.h line 12

Quoted 2 lines
tools/include/linux/btf_ids.h new file +2 -0
+ * (during compilation) and resolved later during the
+ * linking phase by btfid tool.

typo: resolve_btfids tool

Quoted 5 lines
tools/include/linux/btf_ids.h new file +5 -0
+ *
+ * Any change in list layout must be reflected in btfid
+ * tool logic.
+ */
+

[...]

#16 lore raw

Re: [PATCH v5 bpf-next 8/9] tools headers: Adopt verbatim copy of btf_ids.h from kernel sources

Quoted 31 lines · Andrii Nakryiko
Quoted 30 lines · Jiri Olsa

It will be needed by bpf selftest for resolve_btfids tool.

Also adding __PASTE macro as btf_ids.h dependency, which is defined in:

include/linux/compiler_types.h

but because tools/include do not have this header, I'm putting the macro into linux/compiler.h header.

---
 tools/include/linux/btf_ids.h  | 87 ++++++++++++++++++++++++++++++++++
 tools/include/linux/compiler.h |  4 ++
 2 files changed, 91 insertions(+)
 create mode 100644 tools/include/linux/btf_ids.h
tools/include/linux/btf_ids.h new file +5 -0
new file mode 100644
index 000000000000..d317150bc9e3
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_BTF_IDS_H
+#define _LINUX_BTF_IDS_H
+
Quoted 8 lines

tools/include/linux/btf_ids.h line 10

Quoted 5 lines
tools/include/linux/btf_ids.h new file +5 -0
+#include <linux/compiler.h> /* for __PASTE */
+
+/*
+ * Following macros help to define lists of BTF IDs placed
+ * in .BTF_ids section. They are initially filled with zeros

One more inconsistency with .BTF_ids vs .BTF.ids (probably same in the original header, which I missed).

ok

Quoted 5 lines

tools/include/linux/btf_ids.h line 12

Quoted 3 lines
tools/include/linux/btf_ids.h new file +2 -0
+ * (during compilation) and resolved later during the
+ * linking phase by btfid tool.

typo: resolve_btfids tool

will fix

thanks,
jirka

Quoted 9 lines
Quoted 6 lines
tools/include/linux/btf_ids.h new file +5 -0
+ *
+ * Any change in list layout must be reflected in btfid
+ * tool logic.
+ */
+

[...]

author reviewer
#17 lore raw

PATCH v5 9/9 selftests/bpf: Add test for resolve_btfids

In reply to Jiri Olsa

Adding resolve_btfids test under test_progs suite.

It's possible to use btf_ids.h header and its logic in user space application, so we can add easy test for it.

The test defines BTF_ID_LIST and checks it gets properly resolved.

For this reason the test_progs binary (and other binaries that use TRUNNER* macros) is processed with resolve_btfids tool, which resolves BTF IDs in .BTF.ids section.

---
 tools/testing/selftests/bpf/Makefile          |  22 ++-
 .../selftests/bpf/prog_tests/resolve_btfids.c | 170 ++++++++++++++++++
 2 files changed, 190 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c
tools/testing/selftests/bpf/Makefile +20 -2
index 1f9c696b3edf..b47a685d12bd 100644
@@ -190,6 +190,16 @@ else
 	cp "$(VMLINUX_H)" $@
 endif
 
+$(SCRATCH_DIR)/resolve_btfids: $(BPFOBJ)				\
+			       $(TOOLSDIR)/bpf/resolve_btfids/main.c	\
+			       $(TOOLSDIR)/lib/rbtree.c			\
+			       $(TOOLSDIR)/lib/zalloc.c			\
+			       $(TOOLSDIR)/lib/string.c			\
+			       $(TOOLSDIR)/lib/ctype.c			\
+			       $(TOOLSDIR)/lib/str_error_r.c
+	$(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/resolve_btfids	\
+	OUTPUT=$(SCRATCH_DIR)/ BPFOBJ=$(BPFOBJ)
+
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
 # such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
@@ -333,7 +343,8 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_BPF_SKELS)				\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 	$$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-	cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+	cd $$(@D) && $$(CC) -I. $$(CFLAGS) $(TRUNNER_EXTRA_CFLAGS)	\
+	-c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		       %.c						\
@@ -355,6 +366,7 @@ $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)			\
 			     | $(TRUNNER_BINARY)-extras
 	$$(call msg,BINARY,,$$@)
 	$$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
+	$(TRUNNER_BINARY_EXTRA_CMD)
 
 endef
 
@@ -365,7 +377,10 @@ TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c	\
 			 network_helpers.c testing_helpers.c		\
 			 flow_dissector_load.h
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read				\
-		       $(wildcard progs/btf_dump_test_case_*.c)
+		       $(wildcard progs/btf_dump_test_case_*.c)		\
+		       $(SCRATCH_DIR)/resolve_btfids
+TRUNNER_EXTRA_CFLAGS := -D"BUILD_STR(s)=\#s" -DVMLINUX_BTF="BUILD_STR($(VMLINUX_BTF))"
+TRUNNER_BINARY_EXTRA_CMD := $(SCRATCH_DIR)/resolve_btfids --btf $(VMLINUX_BTF) test_progs
 TRUNNER_BPF_BUILD_RULE := CLANG_BPF_BUILD_RULE
 TRUNNER_BPF_CFLAGS := $(BPF_CFLAGS) $(CLANG_CFLAGS)
 TRUNNER_BPF_LDFLAGS := -mattr=+alu32
@@ -373,6 +388,7 @@ $(eval $(call DEFINE_TEST_RUNNER,test_progs))
 
 # Define test_progs-no_alu32 test runner.
 TRUNNER_BPF_BUILD_RULE := CLANG_NOALU32_BPF_BUILD_RULE
+TRUNNER_BINARY_EXTRA_CMD := $(SCRATCH_DIR)/resolve_btfids --btf $(VMLINUX_BTF) test_progs-no_alu32
 TRUNNER_BPF_LDFLAGS :=
 $(eval $(call DEFINE_TEST_RUNNER,test_progs,no_alu32))
 
@@ -392,6 +408,8 @@ TRUNNER_EXTRA_FILES :=
 TRUNNER_BPF_BUILD_RULE := $$(error no BPF objects should be built)
 TRUNNER_BPF_CFLAGS :=
 TRUNNER_BPF_LDFLAGS :=
+TRUNNER_EXTRA_CFLAGS :=
+TRUNNER_BINARY_EXTRA_CMD :=
 $(eval $(call DEFINE_TEST_RUNNER,test_maps))
 
 # Define test_verifier test runner.
tools/testing/selftests/bpf/prog_tests/resolve_btfids.c new file +170 -0

170 lines not on this page · raw message

Signature · 1 line
2.25.4
#18 lore raw

Re: [PATCH v5 bpf-next 9/9] selftests/bpf: Add test for resolve_btfids

Quoted 28 lines · Jiri Olsa

Adding resolve_btfids test under test_progs suite.

It's possible to use btf_ids.h header and its logic in user space application, so we can add easy test for it.

The test defines BTF_ID_LIST and checks it gets properly resolved.

For this reason the test_progs binary (and other binaries that use TRUNNER* macros) is processed with resolve_btfids tool, which resolves BTF IDs in .BTF.ids section.

---
 tools/testing/selftests/bpf/Makefile          |  22 ++-
 .../selftests/bpf/prog_tests/resolve_btfids.c | 170 ++++++++++++++++++
 2 files changed, 190 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c
tools/testing/selftests/bpf/Makefile +0 -0
index 1f9c696b3edf..b47a685d12bd 100644
@@ -190,6 +190,16 @@ else
        cp "$(VMLINUX_H)" $@
 endif
 

tools/testing/selftests/bpf/Makefile else · line 202

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+$(SCRATCH_DIR)/resolve_btfids: $(BPFOBJ)                               \
+                              $(TOOLSDIR)/bpf/resolve_btfids/main.c    \
+                              $(TOOLSDIR)/lib/rbtree.c                 \
+                              $(TOOLSDIR)/lib/zalloc.c                 \
+                              $(TOOLSDIR)/lib/string.c                 \
+                              $(TOOLSDIR)/lib/ctype.c                  \
+                              $(TOOLSDIR)/lib/str_error_r.c
+       $(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/resolve_btfids \
+       OUTPUT=$(SCRATCH_DIR)/ BPFOBJ=$(BPFOBJ)
+

please indent OUTPUT, so it doesn't look like it's a separate command

Quoted 14 lines
tools/testing/selftests/bpf/Makefile +2 -1
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
 # such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
@@ -333,7 +343,8 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:                   \
                      $(TRUNNER_BPF_SKELS)                              \
                      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
        $$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-       cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+       cd $$(@D) && $$(CC) -I. $$(CFLAGS) $(TRUNNER_EXTRA_CFLAGS)      \
+       -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:                          \
                       %.c                                              \

tools/testing/selftests/bpf/Makefile $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS) \ · line 369

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +1 -0
@@ -355,6 +366,7 @@ $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)                   \
                             | $(TRUNNER_BINARY)-extras
        $$(call msg,BINARY,,$$@)
        $$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
+       $(TRUNNER_BINARY_EXTRA_CMD)

no need to make this generic, just write out resolve_btfids here explicitly

tools/testing/selftests/bpf/Makefile TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c \ · line 383

Quoted 12 lines
tools/testing/selftests/bpf/Makefile +4 -1
 endef
 
@@ -365,7 +377,10 @@ TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c     \
                         network_helpers.c testing_helpers.c            \
                         flow_dissector_load.h
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read                          \
-                      $(wildcard progs/btf_dump_test_case_*.c)
+                      $(wildcard progs/btf_dump_test_case_*.c)         \
+                      $(SCRATCH_DIR)/resolve_btfids
+TRUNNER_EXTRA_CFLAGS := -D"BUILD_STR(s)=\#s" -DVMLINUX_BTF="BUILD_STR($(VMLINUX_BTF))"
+TRUNNER_BINARY_EXTRA_CMD := $(SCRATCH_DIR)/resolve_btfids --btf $(VMLINUX_BTF) test_progs

I hope we can get rid of this, see suggestion below.

Quoted 5 lines
tools/testing/selftests/bpf/Makefile +0 -0
 TRUNNER_BPF_BUILD_RULE := CLANG_BPF_BUILD_RULE
 TRUNNER_BPF_CFLAGS := $(BPF_CFLAGS) $(CLANG_CFLAGS)
 TRUNNER_BPF_LDFLAGS := -mattr=+alu32
@@ -373,6 +388,7 @@ $(eval $(call DEFINE_TEST_RUNNER,test_progs))
 

[...]

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int duration;
+
+static struct btf *btf__parse_raw(const char *file)

another copy here...

Quoted 6 lines
tools/testing/selftests/bpf/Makefile +6 -0
+{
+       struct btf *btf;
+       struct stat st;
+       __u8 *buf;
+       FILE *f;
+

[...]

Quoted 6 lines
tools/testing/selftests/bpf/Makefile +6 -0
+
+BTF_ID_LIST(test_list)
+BTF_ID_UNUSED
+BTF_ID(typedef, pid_t)
+BTF_ID(struct,  sk_buff)
+BTF_ID(union,   thread_union)

tools/testing/selftests/bpf/Makefile

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+BTF_ID(func,    memcpy)
+
+struct symbol {
+       const char      *name;
+       int              type;
+       int              id;
+};
+
+struct symbol test_symbols[] = {
+       { "unused",       -1,                0 },

could use BTF_KIND_UNKN here instead of -1

Quoted 6 lines
tools/testing/selftests/bpf/Makefile +6 -0
+       { "pid_t",        BTF_KIND_TYPEDEF, -1 },
+       { "sk_buff",      BTF_KIND_STRUCT,  -1 },
+       { "thread_union", BTF_KIND_UNION,   -1 },
+       { "memcpy",       BTF_KIND_FUNC,    -1 },
+};
+

[...]

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int resolve_symbols(void)
+{
+       const char *path = VMLINUX_BTF;

This build-time parameter passing to find the original VMLINUX_BTF really sucks, IMO.

Why not use the btf_dump tests approach and have our own small "vmlinux BTF", which resolve_btfids would use to resolve these IDs? See how btf_dump_xxx.c files define BTFs that are used in tests. You can do something similar here, and use a well-known BPF object file as a source of BTF, both here in a test and in Makefile for --btf param to resolve_btfids?

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+       struct btf *btf;
+       int type_id;
+       __u32 nr;
+

[...]

#19 lore raw

Re: [PATCH v5 bpf-next 9/9] selftests/bpf: Add test for resolve_btfids

Quoted 29 lines · Andrii Nakryiko
Quoted 28 lines · Jiri Olsa

Adding resolve_btfids test under test_progs suite.

It's possible to use btf_ids.h header and its logic in user space application, so we can add easy test for it.

The test defines BTF_ID_LIST and checks it gets properly resolved.

For this reason the test_progs binary (and other binaries that use TRUNNER* macros) is processed with resolve_btfids tool, which resolves BTF IDs in .BTF.ids section.

---
 tools/testing/selftests/bpf/Makefile          |  22 ++-
 .../selftests/bpf/prog_tests/resolve_btfids.c | 170 ++++++++++++++++++
 2 files changed, 190 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c
tools/testing/selftests/bpf/Makefile +0 -0
index 1f9c696b3edf..b47a685d12bd 100644
@@ -190,6 +190,16 @@ else
        cp "$(VMLINUX_H)" $@
 endif
 
Quoted 12 lines

tools/testing/selftests/bpf/Makefile else · line 202

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+$(SCRATCH_DIR)/resolve_btfids: $(BPFOBJ)                               \
+                              $(TOOLSDIR)/bpf/resolve_btfids/main.c    \
+                              $(TOOLSDIR)/lib/rbtree.c                 \
+                              $(TOOLSDIR)/lib/zalloc.c                 \
+                              $(TOOLSDIR)/lib/string.c                 \
+                              $(TOOLSDIR)/lib/ctype.c                  \
+                              $(TOOLSDIR)/lib/str_error_r.c
+       $(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/resolve_btfids \
+       OUTPUT=$(SCRATCH_DIR)/ BPFOBJ=$(BPFOBJ)
+

please indent OUTPUT, so it doesn't look like it's a separate command

ok

Quoted 12 lines
Quoted 12 lines
tools/testing/selftests/bpf/Makefile +2 -1
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
 # such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
@@ -333,7 +343,8 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:                   \
                      $(TRUNNER_BPF_SKELS)                              \
                      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
        $$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-       cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+       cd $$(@D) && $$(CC) -I. $$(CFLAGS) $(TRUNNER_EXTRA_CFLAGS)      \
+       -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
Quoted 9 lines

tools/testing/selftests/bpf/Makefile $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS) \ · line 369

Quoted 7 lines
tools/testing/selftests/bpf/Makefile +1 -0
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:                          \
                       %.c                                              \
@@ -355,6 +366,7 @@ $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)                   \
                             | $(TRUNNER_BINARY)-extras
        $$(call msg,BINARY,,$$@)
        $$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
+       $(TRUNNER_BINARY_EXTRA_CMD)

no need to make this generic, just write out resolve_btfids here explicitly

currently resolve_btfids fails if there's no .BTF.ids section found, but we can make it silently pass i nthis case and then we can invoke it for all the binaries

Quoted 24 lines

tools/testing/selftests/bpf/Makefile TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c \ · line 383

Quoted 13 lines
tools/testing/selftests/bpf/Makefile +4 -1
 endef
 
@@ -365,7 +377,10 @@ TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c     \
                         network_helpers.c testing_helpers.c            \
                         flow_dissector_load.h
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read                          \
-                      $(wildcard progs/btf_dump_test_case_*.c)
+                      $(wildcard progs/btf_dump_test_case_*.c)         \
+                      $(SCRATCH_DIR)/resolve_btfids
+TRUNNER_EXTRA_CFLAGS := -D"BUILD_STR(s)=\#s" -DVMLINUX_BTF="BUILD_STR($(VMLINUX_BTF))"
+TRUNNER_BINARY_EXTRA_CMD := $(SCRATCH_DIR)/resolve_btfids --btf $(VMLINUX_BTF) test_progs

I hope we can get rid of this, see suggestion below.

Quoted 5 lines
tools/testing/selftests/bpf/Makefile +0 -0
 TRUNNER_BPF_BUILD_RULE := CLANG_BPF_BUILD_RULE
 TRUNNER_BPF_CFLAGS := $(BPF_CFLAGS) $(CLANG_CFLAGS)
 TRUNNER_BPF_LDFLAGS := -mattr=+alu32
@@ -373,6 +388,7 @@ $(eval $(call DEFINE_TEST_RUNNER,test_progs))
 

[...]

Quoted 6 lines

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int duration;
+
+static struct btf *btf__parse_raw(const char *file)

another copy here...

ok

Quoted 20 lines
Quoted 7 lines
tools/testing/selftests/bpf/Makefile +6 -0
+{
+       struct btf *btf;
+       struct stat st;
+       __u8 *buf;
+       FILE *f;
+

[...]

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+
+BTF_ID_LIST(test_list)
+BTF_ID_UNUSED
+BTF_ID(typedef, pid_t)
+BTF_ID(struct,  sk_buff)
+BTF_ID(union,   thread_union)
+BTF_ID(func,    memcpy)
+
+struct symbol {
+       const char      *name;
Quoted 8 lines

tools/testing/selftests/bpf/Makefile

Quoted 6 lines
tools/testing/selftests/bpf/Makefile +6 -0
+       int              type;
+       int              id;
+};
+
+struct symbol test_symbols[] = {
+       { "unused",       -1,                0 },

could use BTF_KIND_UNKN here instead of -1

ok

Quoted 19 lines
Quoted 7 lines
tools/testing/selftests/bpf/Makefile +6 -0
+       { "pid_t",        BTF_KIND_TYPEDEF, -1 },
+       { "sk_buff",      BTF_KIND_STRUCT,  -1 },
+       { "thread_union", BTF_KIND_UNION,   -1 },
+       { "memcpy",       BTF_KIND_FUNC,    -1 },
+};
+

[...]

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int resolve_symbols(void)
+{
+       const char *path = VMLINUX_BTF;

This build-time parameter passing to find the original VMLINUX_BTF really sucks, IMO.

Quoted 6 lines

Why not use the btf_dump tests approach and have our own small "vmlinux BTF", which resolve_btfids would use to resolve these IDs? See how btf_dump_xxx.c files define BTFs that are used in tests. You can do something similar here, and use a well-known BPF object file as a source of BTF, both here in a test and in Makefile for --btf param to resolve_btfids?

well VMLINUX_BTF is there and those types are used are not going away any time soon ;-) but yea, we can do that.. we do this also for bpftrace, it's nicer

jirka

#20 lore raw

Re: [PATCH v5 bpf-next 9/9] selftests/bpf: Add test for resolve_btfids

Quoted 68 lines · Jiri Olsa
Quoted 29 lines · Andrii Nakryiko
Quoted 28 lines · Jiri Olsa

Adding resolve_btfids test under test_progs suite.

It's possible to use btf_ids.h header and its logic in user space application, so we can add easy test for it.

The test defines BTF_ID_LIST and checks it gets properly resolved.

For this reason the test_progs binary (and other binaries that use TRUNNER* macros) is processed with resolve_btfids tool, which resolves BTF IDs in .BTF.ids section.

---
 tools/testing/selftests/bpf/Makefile          |  22 ++-
 .../selftests/bpf/prog_tests/resolve_btfids.c | 170 ++++++++++++++++++
 2 files changed, 190 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/resolve_btfids.c
tools/testing/selftests/bpf/Makefile +0 -0
index 1f9c696b3edf..b47a685d12bd 100644
@@ -190,6 +190,16 @@ else
        cp "$(VMLINUX_H)" $@
 endif
 
Quoted 12 lines

tools/testing/selftests/bpf/Makefile else · line 202

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+$(SCRATCH_DIR)/resolve_btfids: $(BPFOBJ)                               \
+                              $(TOOLSDIR)/bpf/resolve_btfids/main.c    \
+                              $(TOOLSDIR)/lib/rbtree.c                 \
+                              $(TOOLSDIR)/lib/zalloc.c                 \
+                              $(TOOLSDIR)/lib/string.c                 \
+                              $(TOOLSDIR)/lib/ctype.c                  \
+                              $(TOOLSDIR)/lib/str_error_r.c
+       $(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/resolve_btfids \
+       OUTPUT=$(SCRATCH_DIR)/ BPFOBJ=$(BPFOBJ)
+

please indent OUTPUT, so it doesn't look like it's a separate command

ok

Quoted 12 lines
Quoted 12 lines
tools/testing/selftests/bpf/Makefile +2 -1
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
 # such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
@@ -333,7 +343,8 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:                   \
                      $(TRUNNER_BPF_SKELS)                              \
                      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
        $$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-       cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+       cd $$(@D) && $$(CC) -I. $$(CFLAGS) $(TRUNNER_EXTRA_CFLAGS)      \
+       -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
Quoted 9 lines

tools/testing/selftests/bpf/Makefile $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS) \ · line 369

Quoted 7 lines
tools/testing/selftests/bpf/Makefile +1 -0
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:                          \
                       %.c                                              \
@@ -355,6 +366,7 @@ $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)                   \
                             | $(TRUNNER_BINARY)-extras
        $$(call msg,BINARY,,$$@)
        $$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
+       $(TRUNNER_BINARY_EXTRA_CMD)

no need to make this generic, just write out resolve_btfids here explicitly

Quoted 3 lines

currently resolve_btfids fails if there's no .BTF.ids section found, but we can make it silently pass i nthis case and then we can invoke it for all the binaries

ah, I see. Yeah, either we can add an option to resolve_btfids to not error when .BTF_ids is missing (probably best), or we can check whether the test has .BTF_ids section, and if it does - run resolve_btfids on it. Just ignoring errors always is more error-prone, because we won't know if it's a real problem we are ignoring, or missing .BTF_ids.

Quoted 91 lines
Quoted 25 lines
Quoted 6 lines
tools/testing/selftests/bpf/Makefile +0 -0
 endef
 

tools/testing/selftests/bpf/Makefile TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c \ · line 383

Quoted 8 lines
tools/testing/selftests/bpf/Makefile +4 -1
@@ -365,7 +377,10 @@ TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c     \
                         network_helpers.c testing_helpers.c            \
                         flow_dissector_load.h
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read                          \
-                      $(wildcard progs/btf_dump_test_case_*.c)
+                      $(wildcard progs/btf_dump_test_case_*.c)         \
+                      $(SCRATCH_DIR)/resolve_btfids
+TRUNNER_EXTRA_CFLAGS := -D"BUILD_STR(s)=\#s" -DVMLINUX_BTF="BUILD_STR($(VMLINUX_BTF))"
+TRUNNER_BINARY_EXTRA_CMD := $(SCRATCH_DIR)/resolve_btfids --btf $(VMLINUX_BTF) test_progs

I hope we can get rid of this, see suggestion below.

Quoted 5 lines
tools/testing/selftests/bpf/Makefile +0 -0
 TRUNNER_BPF_BUILD_RULE := CLANG_BPF_BUILD_RULE
 TRUNNER_BPF_CFLAGS := $(BPF_CFLAGS) $(CLANG_CFLAGS)
 TRUNNER_BPF_LDFLAGS := -mattr=+alu32
@@ -373,6 +388,7 @@ $(eval $(call DEFINE_TEST_RUNNER,test_progs))
 

[...]

Quoted 6 lines

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int duration;
+
+static struct btf *btf__parse_raw(const char *file)

another copy here...

ok

Quoted 20 lines
Quoted 7 lines
tools/testing/selftests/bpf/Makefile +6 -0
+{
+       struct btf *btf;
+       struct stat st;
+       __u8 *buf;
+       FILE *f;
+

[...]

Quoted 10 lines
tools/testing/selftests/bpf/Makefile +10 -0
+
+BTF_ID_LIST(test_list)
+BTF_ID_UNUSED
+BTF_ID(typedef, pid_t)
+BTF_ID(struct,  sk_buff)
+BTF_ID(union,   thread_union)
+BTF_ID(func,    memcpy)
+
+struct symbol {
+       const char      *name;
Quoted 8 lines

tools/testing/selftests/bpf/Makefile

Quoted 6 lines
tools/testing/selftests/bpf/Makefile +6 -0
+       int              type;
+       int              id;
+};
+
+struct symbol test_symbols[] = {
+       { "unused",       -1,                0 },

could use BTF_KIND_UNKN here instead of -1

ok

Quoted 19 lines
Quoted 7 lines
tools/testing/selftests/bpf/Makefile +6 -0
+       { "pid_t",        BTF_KIND_TYPEDEF, -1 },
+       { "sk_buff",      BTF_KIND_STRUCT,  -1 },
+       { "thread_union", BTF_KIND_UNION,   -1 },
+       { "memcpy",       BTF_KIND_FUNC,    -1 },
+};
+

[...]

tools/testing/selftests/bpf/Makefile

Quoted 4 lines
tools/testing/selftests/bpf/Makefile +4 -0
+
+static int resolve_symbols(void)
+{
+       const char *path = VMLINUX_BTF;

This build-time parameter passing to find the original VMLINUX_BTF really sucks, IMO.

Quoted 6 lines

Why not use the btf_dump tests approach and have our own small "vmlinux BTF", which resolve_btfids would use to resolve these IDs? See how btf_dump_xxx.c files define BTFs that are used in tests. You can do something similar here, and use a well-known BPF object file as a source of BTF, both here in a test and in Makefile for --btf param to resolve_btfids?

Quoted 3 lines

well VMLINUX_BTF is there and those types are used are not going away any time soon ;-) but yea, we can do that.. we do this also for bpftrace, it's nicer

"VMLINUX_BTF is there" is not really true in a lot of more complicated setups, which is why I'd like to avoid that assumption. E.g., for libbpf Travis CI, we build self-tests in one VM, but run the binary in a different VM. So either vmlinux itself or the path to it might change.

Also, having full control over **small** BTF allows to create various test situations that might be harder to pinpoint in real vmlinux BTF, e.g., same-named entities with different KINDS (typedef vs struct, etc). Then if that fails, debugging this on a small BTF is much-much easier than on a real thing. Real vmlinux BTF is being tested each time you build a kernel and run selftests inside VM either way, so I don't think we lose anything in terms of coverage.

Quoted 3 lines

jirka

#21 lore raw

Re: [PATCH v5 bpf-next 9/9] selftests/bpf: Add test for resolve_btfids

On Tue, Jul 07, 2020 at 10:49:22AM -0700, Andrii Nakryiko wrote:

SNIP

Quoted 25 lines
Quoted 21 lines
Quoted 11 lines
Quoted 11 lines

# Get Clang's default includes on this system, as opposed to those seen by # '-target bpf'. This fixes "missing" files on some architectures/distros,

 # such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
tools/testing/selftests/bpf/Makefile +2 -1
@@ -333,7 +343,8 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:                   \
                      $(TRUNNER_BPF_SKELS)                              \
                      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
        $$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-       cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+       cd $$(@D) && $$(CC) -I. $$(CFLAGS) $(TRUNNER_EXTRA_CFLAGS)      \
+       -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
Quoted 9 lines

tools/testing/selftests/bpf/Makefile $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS) \ · line 369

Quoted 7 lines
tools/testing/selftests/bpf/Makefile +1 -0
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:                          \
                       %.c                                              \
@@ -355,6 +366,7 @@ $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)                   \
                             | $(TRUNNER_BINARY)-extras
        $$(call msg,BINARY,,$$@)
        $$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
+       $(TRUNNER_BINARY_EXTRA_CMD)

no need to make this generic, just write out resolve_btfids here explicitly

Quoted 3 lines

currently resolve_btfids fails if there's no .BTF.ids section found, but we can make it silently pass i nthis case and then we can invoke it for all the binaries

Quoted 6 lines

ah, I see. Yeah, either we can add an option to resolve_btfids to not error when .BTF_ids is missing (probably best), or we can check whether the test has .BTF_ids section, and if it does - run resolve_btfids on it. Just ignoring errors always is more error-prone, because we won't know if it's a real problem we are ignoring, or missing .BTF_ids.

ok, sounds good

Quoted 20 lines
Quoted 15 lines
Quoted 8 lines

tools/testing/selftests/bpf/prog_tests/resolve_btfids.c line 132

Quoted 3 lines
tools/testing/selftests/bpf/prog_tests/resolve_btfids.c new file +3 -0
+static int resolve_symbols(void)
+{
+       const char *path = VMLINUX_BTF;

This build-time parameter passing to find the original VMLINUX_BTF really sucks, IMO.

Quoted 6 lines

Why not use the btf_dump tests approach and have our own small "vmlinux BTF", which resolve_btfids would use to resolve these IDs? See how btf_dump_xxx.c files define BTFs that are used in tests. You can do something similar here, and use a well-known BPF object file as a source of BTF, both here in a test and in Makefile for --btf param to resolve_btfids?

Quoted 3 lines

well VMLINUX_BTF is there and those types are used are not going away any time soon ;-) but yea, we can do that.. we do this also for bpftrace, it's nicer

Quoted 5 lines

"VMLINUX_BTF is there" is not really true in a lot of more complicated setups, which is why I'd like to avoid that assumption. E.g., for libbpf Travis CI, we build self-tests in one VM, but run the binary in a different VM. So either vmlinux itself or the path to it might change.

ok

Quoted 8 lines

Also, having full control over **small** BTF allows to create various test situations that might be harder to pinpoint in real vmlinux BTF, e.g., same-named entities with different KINDS (typedef vs struct, etc). Then if that fails, debugging this on a small BTF is much-much easier than on a real thing. Real vmlinux BTF is being tested each time you build a kernel and run selftests inside VM either way, so I don't think we lose anything in terms of coverage.

agreed, will add that

thanks,
jirka