Skip to contentSkip to conversation

bpftool: improve error handing for missing .BTF section

PATCH Applied Changbin Du 5 participants 18 messages Dec 17, 2022 – Jan 5, 2023

Subsystems (5)
Kselftest BPF BPF selftests libbpf +1 more subsystems
lore
Share & tools
lore
#0 lore raw

v3 cover bpftool: improve error handing for missing .BTF section

Display error message for missing ".BTF" section and clean up empty vmlinux.h file.

v3:
- fix typo and make error message consistent. (Andrii Nakryiko)
- split out perf change.
v2:
- remove vmlinux specific error info.
- use builtin target .DELETE_ON_ERROR: to delete empty vmlinux.h

Changbin Du (2):
libbpf: show error info about missing ".BTF" section
bpf: makefiles: do not generate empty vmlinux.h

 tools/bpf/bpftool/Makefile           | 3 +++
 tools/lib/bpf/btf.c                  | 1 +
 tools/testing/selftests/bpf/Makefile | 3 +++
 3 files changed, 7 insertions(+)
Signature · 1 line
2.37.2
#1 lore raw

PATCH v3 1/2 libbpf: show error info about missing ".BTF" section

Show the real problem instead of just saying "No such file or directory".

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux
Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

---
 tools/lib/bpf/btf.c | 1 +
 1 file changed, 1 insertion(+)
tools/lib/bpf/btf.c +1 -0
index 71e165b09ed5..dd2badf1a54e 100644
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 	err = 0;
 
 	if (!btf_data) {
+		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENOENT;
 		goto done;
 	}
Signature · 1 line
2.37.2
#2 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Hi Changbin,

Quoted 4 lines · Changbin Du

Show the real problem instead of just saying "No such file or directory".

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

Thanks,
Leo

Quoted 20 lines
---
 tools/lib/bpf/btf.c | 1 +
 1 file changed, 1 insertion(+)
tools/lib/bpf/btf.c +1 -0
index 71e165b09ed5..dd2badf1a54e 100644
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 	err = 0;
 
 	if (!btf_data) {
+		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENOENT;
 		goto done;
 	}
Signature · 1 line
2.37.2
#3 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Quoted 8 lines · Leo Yan

Hi Changbin,

Quoted 4 lines · Changbin Du

Show the real problem instead of just saying "No such file or directory".

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Quoted 9 lines

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

Quoted 23 lines

Thanks,
Leo

Quoted 20 lines
---
 tools/lib/bpf/btf.c | 1 +
 1 file changed, 1 insertion(+)
tools/lib/bpf/btf.c +1 -0
index 71e165b09ed5..dd2badf1a54e 100644
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 	err = 0;
 
 	if (!btf_data) {
+		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENOENT;
 		goto done;
 	}
Signature · 1 line
2.37.2
Signature · 2 lines
Cheers,
Changbin Du
#4 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

On Tue, Dec 20, 2022 at 09:31:14AM +0800, Changbin Du wrote:

[...]

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Quoted 7 lines
Quoted 5 lines

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Okay, I see. Sorry for noise.

Quoted 7 lines
Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

[...]

Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 	err = 0;
 
 	if (!btf_data) {
+		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

Thanks,
Leo

#5 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Quoted 43 lines · Leo Yan

On Tue, Dec 20, 2022 at 09:31:14AM +0800, Changbin Du wrote:

[...]

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Quoted 7 lines
Quoted 5 lines

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Okay, I see. Sorry for noise.

Quoted 7 lines
Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

[...]

Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Anyways, I've applied this patch set to bpf-next. Thanks.

Quoted 3 lines

Thanks,
Leo

#6 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

On Tue, Dec 20, 2022 at 04:13:13PM -0800, Andrii Nakryiko wrote:

[...]

Quoted 11 lines
Quoted 11 lines
Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 7 lines
Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

Or maybe -ENODATA (No data available) is a better choice?

Thanks,
Leo

Quoted 12 lines

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Anyways, I've applied this patch set to bpf-next. Thanks.

#7 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Quoted 20 lines · Leo Yan

On Tue, Dec 20, 2022 at 04:13:13PM -0800, Andrii Nakryiko wrote:

[...]

Quoted 15 lines
Quoted 14 lines
Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

Quoted 5 lines
Quoted 3 lines

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

Or maybe -ENODATA (No data available) is a better choice?

-ENODATA sounds good to me, yep.

Quoted 16 lines

Thanks,
Leo

Quoted 12 lines

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Anyways, I've applied this patch set to bpf-next. Thanks.

#8 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

In reply to Leo Yan
Quoted 19 lines · Leo Yan

On Tue, Dec 20, 2022 at 04:13:13PM -0800, Andrii Nakryiko wrote:

[...]

Quoted 15 lines
Quoted 14 lines
Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

Quoted 9 lines
Quoted 3 lines

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

Or maybe -ENODATA (No data available) is a better choice?

Thanks,
Leo

Yan, will you have a patch for this suggestion?

[snip]

Signature · 2 lines
Cheers,
Changbin Du
#9 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Quoted 24 lines · Changbin Du
Quoted 23 lines · Leo Yan

On Tue, Dec 20, 2022 at 04:13:13PM -0800, Andrii Nakryiko wrote:

[...]

Quoted 18 lines
Quoted 11 lines
Quoted 6 lines
Quoted 6 lines

Review static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
(quoted diff) +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

Quoted 6 lines
Quoted 5 lines

Or maybe -ENODATA (No data available) is a better choice?

Thanks,
Leo

Yan, will you have a patch for this suggestion?

You are welcome to send a patch, otherwise, I can cook one.

Thanks,
Leo

#10 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

In reply to Andrii Nakryiko

2022-12-20 16:13 UTC-0800 ~ Andrii Nakryiko <[email protected]>

Quoted 59 lines · Leo Yan
Quoted 43 lines · Changbin Du

[...]

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Quoted 7 lines
Quoted 5 lines

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Okay, I see. Sorry for noise.

Quoted 7 lines
Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

[...]

Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quoted 3 lines

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Apologies for the delay.
What you're proposing is to replace all messages currently looking like this:

	$ bpftool prog
	Error: can't get next program: Operation not permitted

by:

	$ bpftool prog
	Error: can't get next program: -EPERM

Do I understand correctly?

I think the strerror() messages are helpful in some occasions (they _are_ more human-friendly to many users), but it's also true that they're not always precise. With bpftool, "Invalid argument" is a classic when the program doesn't load, and may lead to confusion with the args passed to bpftool on the command line. Then there are the other corner cases like the one discussed in this thread. So, why not.

If we do change, yeah I'd rather have as much of this handling in libbpf itself, and then adjust bpftool to handle the remaining cases, for consistency.

Quentin

#11 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

Quoted 72 lines · Quentin Monnet

2022-12-20 16:13 UTC-0800 ~ Andrii Nakryiko <[email protected]>

Quoted 59 lines · Leo Yan
Quoted 43 lines · Changbin Du

[...]

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Quoted 7 lines
Quoted 5 lines

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Okay, I see. Sorry for noise.

Quoted 7 lines
Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

[...]

Quoted 6 lines
Quoted 6 lines

tools/lib/bpf/btf.c static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
tools/lib/bpf/btf.c +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quoted 3 lines

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Apologies for the delay.
What you're proposing is to replace all messages currently looking like this:

        $ bpftool prog
        Error: can't get next program: Operation not permitted
Quoted 6 lines

by:

        $ bpftool prog
        Error: can't get next program: -EPERM

Do I understand correctly?

yep, that's what I had in mind

Quoted 7 lines

I think the strerror() messages are helpful in some occasions (they _are_ more human-friendly to many users), but it's also true that they're not always precise. With bpftool, "Invalid argument" is a classic when the program doesn't load, and may lead to confusion with the args passed to bpftool on the command line. Then there are the other corner cases like the one discussed in this thread. So, why not.

maybe the right approach would be to have both symbolic error name and its human-readable representation, so for example above

Error: can't get next program: [-EPERM] Operation not permitted

or something like that? And if error value is unknown, just keep it as
integer: "[-5555]" ?

Quoted 4 lines

If we do change, yeah I'd rather have as much of this handling in libbpf itself, and then adjust bpftool to handle the remaining cases, for consistency.

we can teach libbpf_strerror_r() to do this and if bpftool is going to use it consistently then it would get the benefit automatically

Quoted 2 lines

Quentin

#12 lore raw

Re: [PATCH v3 1/2] libbpf: show error info about missing ".BTF" section

2023-01-03 15:46 UTC-0800 ~ Andrii Nakryiko <[email protected]>

Quoted 90 lines · Quentin Monnet
Quoted 72 lines

2022-12-20 16:13 UTC-0800 ~ Andrii Nakryiko <[email protected]>

Quoted 59 lines · Leo Yan
Quoted 43 lines · Changbin Du

[...]

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Now will print below info:
libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux

Recently I encountered the same issue, it could be caused by:
either missing to install tool pahole or missing to enable kernel configuration CONFIG_DEBUG_INFO_BTF.

Quoted 7 lines
Quoted 5 lines

Could we give explict info for reasoning failure? Like:

"libbpf: failed to find '.BTF' ELF section in /home/changbin/work/linux/vmlinux, please install pahole and enable CONFIG_DEBUG_INFO_BTF=y for kernel building".

This is vmlinux special information and similar tips are removed from patch V2. libbpf is common for all ELFs.

Okay, I see. Sorry for noise.

Quoted 7 lines
Quoted 6 lines
Quoted 1 line

Error: failed to load BTF from /home/changbin/work/linux/vmlinux: No such file or directory

This log is confusing when we can find vmlinux file but without BTF section. Consider to use a separate patch to detect vmlinux not found case and print out "No such file or directory"?

I think it's already there. If the file doesn't exist, open will fail.

[...]

Quoted 6 lines
Quoted 6 lines

Review static struct btf *btf_parse_elf(const char *path, struct btf *base_btf, · line 994

Quoted 6 lines
(quoted diff) +1 -0
@@ -990,6 +990,7 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
  err = 0;
 
  if (!btf_data) {
+         pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
          err = -ENOENT;

btf_parse_elf() returns -ENOENT when ELF file doesn't contain BTF section, therefore, bpftool dumps error string "No such file or directory". It's confused that actually vmlinux is existed.

Quoted 3 lines

I am wondering if we can use error -LIBBPF_ERRNO__FORMAT (or any better choice?) to replace -ENOENT at here, this can avoid bpftool to outputs "No such file or directory" in this case.

The only really meaningful error code would be -ESRCH, which strerror() will translate to "No such process", which is also completely confusing.

In general, I always found these strerror() messages extremely unhelpful and confusing. I wonder if we should make an effort to actually emit symbolic names of errors instead (literally, "-ENOENT" in this case). This is all tooling for engineers, I find -ENOENT or -ESRCH much more meaningful as an error message, compared to "No such file" seemingly human-readable interpretation.

Quoted 3 lines

Quenting, what do you think about the above proposal for bpftool? We can have some libbpf helper internally and do it in libbpf error messages as well and just reuse the logic in bpftool, perhaps?

Apologies for the delay.
What you're proposing is to replace all messages currently looking like this:

        $ bpftool prog
        Error: can't get next program: Operation not permitted
Quoted 6 lines

by:

        $ bpftool prog
        Error: can't get next program: -EPERM

Do I understand correctly?

yep, that's what I had in mind

Quoted 7 lines

I think the strerror() messages are helpful in some occasions (they _are_ more human-friendly to many users), but it's also true that they're not always precise. With bpftool, "Invalid argument" is a classic when the program doesn't load, and may lead to confusion with the args passed to bpftool on the command line. Then there are the other corner cases like the one discussed in this thread. So, why not.

Quoted 7 lines

maybe the right approach would be to have both symbolic error name and its human-readable representation, so for example above

Error: can't get next program: [-EPERM] Operation not permitted

or something like that? And if error value is unknown, just keep it as
integer: "[-5555]" ?

That would be great, we'd have both the error name for savvy users and the (more or less accurate) interpretation for others.

Quoted 6 lines
Quoted 3 lines

If we do change, yeah I'd rather have as much of this handling in libbpf itself, and then adjust bpftool to handle the remaining cases, for consistency.

we can teach libbpf_strerror_r() to do this and if bpftool is going to use it consistently then it would get the benefit automatically

Sounds good to me.

#13 lore raw

PATCH v3 2/2 bpf: makefiles: do not generate empty vmlinux.h

In reply to Changbin Du

Remove the empty vmlinux.h if bpftool failed to dump btf info. The empty vmlinux.h can hide real error when reading output of make.

This is done by adding .DELETE_ON_ERROR special target in related makefiles.

---
 tools/bpf/bpftool/Makefile           | 3 +++
 tools/testing/selftests/bpf/Makefile | 3 +++
 2 files changed, 6 insertions(+)
tools/bpf/bpftool/Makefile +3 -0
index 787b857d3fb5..313fd1b09189 100644
@@ -289,3 +289,6 @@ FORCE:
 .PHONY: all FORCE bootstrap clean install-bin install uninstall
 .PHONY: doc doc-clean doc-install doc-uninstall
 .DEFAULT_GOAL := all
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
tools/testing/selftests/bpf/Makefile +3 -0
index c22c43bbee19..205e8c3c346a 100644
@@ -626,3 +626,6 @@ EXTRA_CLEAN := $(TEST_CUSTOM_PROGS) $(SCRATCH_DIR) $(HOST_SCRATCH_DIR)	\
 			       liburandom_read.so)
 
 .PHONY: docs docs-clean
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
Signature · 1 line
2.37.2
#14 lore raw

Re: [PATCH v3 2/2] bpf: makefiles: do not generate empty vmlinux.h

Quoted 6 lines · Changbin Du

Remove the empty vmlinux.h if bpftool failed to dump btf info. The empty vmlinux.h can hide real error when reading output of make.

This is done by adding .DELETE_ON_ERROR special target in related makefiles.

We need to handle the same case for perf building, its makefile linux/tools/perf/Makefile.perf also uses bpftool to generate vmlinux.h, see:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/perf/Makefile.perf#n1067

Please consider to use a separate patch to add the same change in Makefile.perf?

Thanks,
Leo

Quoted 31 lines
---
 tools/bpf/bpftool/Makefile           | 3 +++
 tools/testing/selftests/bpf/Makefile | 3 +++
 2 files changed, 6 insertions(+)
tools/bpf/bpftool/Makefile +3 -0
index 787b857d3fb5..313fd1b09189 100644
@@ -289,3 +289,6 @@ FORCE:
 .PHONY: all FORCE bootstrap clean install-bin install uninstall
 .PHONY: doc doc-clean doc-install doc-uninstall
 .DEFAULT_GOAL := all
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
tools/testing/selftests/bpf/Makefile +3 -0
index c22c43bbee19..205e8c3c346a 100644
@@ -626,3 +626,6 @@ EXTRA_CLEAN := $(TEST_CUSTOM_PROGS) $(SCRATCH_DIR) $(HOST_SCRATCH_DIR)	\
 			       liburandom_read.so)
 
 .PHONY: docs docs-clean
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
Signature · 1 line
2.37.2
#15 lore raw

Re: [PATCH v3 2/2] bpf: makefiles: do not generate empty vmlinux.h

Quoted 8 lines · Leo Yan
Quoted 6 lines · Changbin Du

Remove the empty vmlinux.h if bpftool failed to dump btf info. The empty vmlinux.h can hide real error when reading output of make.

This is done by adding .DELETE_ON_ERROR special target in related makefiles.

Quoted 8 lines

We need to handle the same case for perf building, its makefile linux/tools/perf/Makefile.perf also uses bpftool to generate vmlinux.h, see:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/perf/Makefile.perf#n1067

Please consider to use a separate patch to add the same change in Makefile.perf?

It's alreay there.
https://lore.kernel.org/lkml/[email protected]/T/

Quoted 34 lines

Thanks,
Leo

Quoted 31 lines
---
 tools/bpf/bpftool/Makefile           | 3 +++
 tools/testing/selftests/bpf/Makefile | 3 +++
 2 files changed, 6 insertions(+)
tools/bpf/bpftool/Makefile +3 -0
index 787b857d3fb5..313fd1b09189 100644
@@ -289,3 +289,6 @@ FORCE:
 .PHONY: all FORCE bootstrap clean install-bin install uninstall
 .PHONY: doc doc-clean doc-install doc-uninstall
 .DEFAULT_GOAL := all
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
tools/testing/selftests/bpf/Makefile +3 -0
index c22c43bbee19..205e8c3c346a 100644
@@ -626,3 +626,6 @@ EXTRA_CLEAN := $(TEST_CUSTOM_PROGS) $(SCRATCH_DIR) $(HOST_SCRATCH_DIR)	\
 			       liburandom_read.so)
 
 .PHONY: docs docs-clean
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:
Signature · 1 line
2.37.2
Signature · 2 lines
Cheers,
Changbin Du
Acked-by on v3 2/2
#16 lore raw

Re: [PATCH v3 2/2] bpf: makefiles: do not generate empty vmlinux.h

In reply to Changbin Du

2022-12-18 06:35 UTC+0800 ~ Changbin Du <[email protected]>

Quoted 18 lines

Remove the empty vmlinux.h if bpftool failed to dump btf info. The empty vmlinux.h can hide real error when reading output of make.

This is done by adding .DELETE_ON_ERROR special target in related makefiles.

---
 tools/bpf/bpftool/Makefile           | 3 +++
 tools/testing/selftests/bpf/Makefile | 3 +++
 2 files changed, 6 insertions(+)
tools/bpf/bpftool/Makefile +0 -0
index 787b857d3fb5..313fd1b09189 100644

tools/bpf/bpftool/Makefile FORCE: · line 294

Quoted 6 lines
tools/bpf/bpftool/Makefile +3 -0
@@ -289,3 +289,6 @@ FORCE:
 .PHONY: all FORCE bootstrap clean install-bin install uninstall
 .PHONY: doc doc-clean doc-install doc-uninstall
 .DEFAULT_GOAL := all
+
+# Delete partially updated (corrupted) files on error
+.DELETE_ON_ERROR:

Thanks!

#17 lore raw
In reply to Changbin Du
This series was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:

Hello:

This series was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:

Quoted 11 lines · On Sun, 18 Dec 2022 06:35:07 +0800 you wrote:

Display error message for missing ".BTF" section and clean up empty vmlinux.h file.

v3:
- fix typo and make error message consistent. (Andrii Nakryiko)
- split out perf change.
v2:
- remove vmlinux specific error info.
- use builtin target .DELETE_ON_ERROR: to delete empty vmlinux.h

[...]

Here is the summary with links:
- [v3,1/2] libbpf: show error info about missing ".BTF" section

    https://git.kernel.org/bpf/bpf-next/c/e6b4e1d759d3
  - [v3,2/2] bpf: makefiles: do not generate empty vmlinux.h
    https://git.kernel.org/bpf/bpf-next/c/e7f0d5cdd023

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html