Introduce linux_filter.h and replace #include <linux/filter.h>

Introduce linux_filter.h and replace #include <linux/filter.h>

Some toolchain (especially PNaCl for Non-SFI build) does not
provide the linux/filter.h system header.
This CL defines all necessary structs and macros for sandbox,
and replace linux/filter.h with the new header.

This CL is preparation to enable seccomp-bpf sandbox on
nacl_helper_nonsfi.

TEST=Ran bots.
BUG=358465
Committed: https://crrev.com/168a20642b1ceb7fb49e78417671aaec26f5477a
Cr-Commit-Position: refs/heads/master@{#324194}

[hidehiko, 2015-04-01 12:08:27]

Patchset #2 (id:20001) has been deleted

[hidehiko, 2015-04-01 12:08:46]

Patchset #1 (id:1) has been deleted

[hidehiko, 2015-04-01 12:10:37]

hidehiko@chromium.org changed reviewers:
+ mdempsky@chromium.org

[hidehiko, 2015-04-01 12:10:38]

Extracted from https://codereview.chromium.org/1029283003/.
Could you take a look?

Thanks,
- hidehiko

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/d...
File sandbox/linux/bpf_dsl/dump_bpf.cc (left):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/d...
sandbox/linux/bpf_dsl/dump_bpf.cc:82: fprintf(stderr, BPF_OP(iter->code) ==
BPF_NEG
FYI: This causes an error with "-Werror,-Wformat-extra-args".

[jln (very slow on Chromium), 2015-04-01 21:54:53]

jln@chromium.org changed reviewers:
+ jln@chromium.org

[jln (very slow on Chromium), 2015-04-01 21:54:53]

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
File sandbox/linux/system_headers/linux_filter.h (right):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
sandbox/linux/system_headers/linux_filter.h:24: #define BPF_CLASS(code) ((code)
& 0x07)
Could you add #ifndef guards?

It's what has been done so far in linux_seccomp.h and other headers.

At some point, I would love to try to come-up with a way to test these headers
(have tests that automaticlaly compare them to what's available on the system
for instance).

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
sandbox/linux/system_headers/linux_filter.h:60: #endif
Don't forget the // SANDBOX_LINUX_SYSTEM_HEADERS_LINUX_FILTER_H_ comment

[mdempsky, 2015-04-01 22:06:10]

lgtm

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
File sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc (right):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:29: #include
"sandbox/linux/system_headers/linux_filter.h"
Why is this necessary?

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/d...
File sandbox/linux/bpf_dsl/dump_bpf.cc (left):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/d...
sandbox/linux/bpf_dsl/dump_bpf.cc:82: fprintf(stderr, BPF_OP(iter->code) ==
BPF_NEG
On 2015/04/01 12:10:38, hidehiko wrote:
> FYI: This causes an error with "-Werror,-Wformat-extra-args".

Yeah, the rewrite is much better, thanks.  (This is just copy/pasted debugging
code from the original sandbox-bpf implementation that I'd never looked at
closely until now.)

[hidehiko, 2015-04-02 09:04:42]

Thank you for review. PTAL.

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
File sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc (right):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:29: #include
"sandbox/linux/system_headers/linux_filter.h"
On 2015/04/01 22:06:10, mdempsky wrote:
> Why is this necessary?

CodeGen::Program used below is actually the typedef vector<sock_filter>, so that
forward declaration does not work. As I removed linux_filter from
linux_seccomp.h, it is not be included indirectly. Instead, I added it here.

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
File sandbox/linux/system_headers/linux_filter.h (right):

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
sandbox/linux/system_headers/linux_filter.h:24: #define BPF_CLASS(code) ((code)
& 0x07)
On 2015/04/01 21:54:53, jln wrote:
> Could you add #ifndef guards?
> 
> It's what has been done so far in linux_seccomp.h and other headers.
> 
> At some point, I would love to try to come-up with a way to test these headers
> (have tests that automaticlaly compare them to what's available on the system
> for instance).

Done.

https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
sandbox/linux/system_headers/linux_filter.h:60: #endif
On 2015/04/01 21:54:53, jln wrote:
> Don't forget the // SANDBOX_LINUX_SYSTEM_HEADERS_LINUX_FILTER_H_ comment

Done.

[hidehiko, 2015-04-06 18:56:25]

jln@, friendly ping?

On 2015/04/02 09:04:42, hidehiko wrote:
> Thank you for review. PTAL.
> 
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
> File sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc (right):
> 
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
> sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:29: #include
> "sandbox/linux/system_headers/linux_filter.h"
> On 2015/04/01 22:06:10, mdempsky wrote:
> > Why is this necessary?
> 
> CodeGen::Program used below is actually the typedef vector<sock_filter>, so
that
> forward declaration does not work. As I removed linux_filter from
> linux_seccomp.h, it is not be included indirectly. Instead, I added it here.
> 
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> File sandbox/linux/system_headers/linux_filter.h (right):
> 
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> sandbox/linux/system_headers/linux_filter.h:24: #define BPF_CLASS(code)
((code)
> & 0x07)
> On 2015/04/01 21:54:53, jln wrote:
> > Could you add #ifndef guards?
> > 
> > It's what has been done so far in linux_seccomp.h and other headers.
> > 
> > At some point, I would love to try to come-up with a way to test these
headers
> > (have tests that automaticlaly compare them to what's available on the
system
> > for instance).
> 
> Done.
> 
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> sandbox/linux/system_headers/linux_filter.h:60: #endif
> On 2015/04/01 21:54:53, jln wrote:
> > Don't forget the // SANDBOX_LINUX_SYSTEM_HEADERS_LINUX_FILTER_H_ comment
> 
> Done.

[hidehiko, 2015-04-08 05:36:22]

Chatted offline with mdempsky@, and I'm submitting.
jln@, please let me know, if you have more comments.

Again, thank you very much for your review!
- hidehiko


On 2015/04/06 18:56:25, hidehiko wrote:
> jln@, friendly ping?
> 
> On 2015/04/02 09:04:42, hidehiko wrote:
> > Thank you for review. PTAL.
> > 
> >
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
> > File sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc (right):
> > 
> >
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/bpf_dsl/b...
> > sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:29: #include
> > "sandbox/linux/system_headers/linux_filter.h"
> > On 2015/04/01 22:06:10, mdempsky wrote:
> > > Why is this necessary?
> > 
> > CodeGen::Program used below is actually the typedef vector<sock_filter>, so
> that
> > forward declaration does not work. As I removed linux_filter from
> > linux_seccomp.h, it is not be included indirectly. Instead, I added it here.
> > 
> >
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> > File sandbox/linux/system_headers/linux_filter.h (right):
> > 
> >
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> > sandbox/linux/system_headers/linux_filter.h:24: #define BPF_CLASS(code)
> ((code)
> > & 0x07)
> > On 2015/04/01 21:54:53, jln wrote:
> > > Could you add #ifndef guards?
> > > 
> > > It's what has been done so far in linux_seccomp.h and other headers.
> > > 
> > > At some point, I would love to try to come-up with a way to test these
> headers
> > > (have tests that automaticlaly compare them to what's available on the
> system
> > > for instance).
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/1053603002/diff/40001/sandbox/linux/system_he...
> > sandbox/linux/system_headers/linux_filter.h:60: #endif
> > On 2015/04/01 21:54:53, jln wrote:
> > > Don't forget the // SANDBOX_LINUX_SYSTEM_HEADERS_LINUX_FILTER_H_ comment
> > 
> > Done.

[hidehiko, 2015-04-08 05:36:26]

The CQ bit was checked by hidehiko@chromium.org

[hidehiko, 2015-04-08 05:36:27]

The patchset sent to the CQ was uploaded after l-g-t-m from
mdempsky@chromium.org
Link to the patchset: https://codereview.chromium.org/1053603002/#ps60001
(title: " ")

[commit-bot: I haz the power, 2015-04-08 05:36:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1053603002/60001

[commit-bot: I haz the power, 2015-04-08 07:09:58]

Committed patchset #2 (id:60001)

[commit-bot: I haz the power, 2015-04-08 07:10:53]

Patchset 2 (id:??) landed as
https://crrev.com/168a20642b1ceb7fb49e78417671aaec26f5477a
Cr-Commit-Position: refs/heads/master@{#324194}