| Message ID | 20221104094016.102049-1-asavkov@redhat.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 61fc5e66f755db24d27ba37ce1ee4873def1a074 |
| Delegated to: | BPF |
| Headers | show |
| Series | [bpf-next] selftests/bpf: fix build-id for liburandom_read.so | expand |
On Fri, Nov 4, 2022 at 10:41 AM Artem Savkov <asavkov@redhat.com> wrote: > > lld produces "fast" style build-ids by default, which is inconsistent > with ld's "sha1" style. Explicitly specify build-id style to be "sha1" > when linking liburandom_read.so the same way it is already done for > urandom_read. > > Signed-off-by: Artem Savkov <asavkov@redhat.com> Acked-by: KP Singh <kpsingh@kernel.org> This was done in https://lore.kernel.org/bpf/20200922232140.1994390-1-morbo@google.com
Hi Artem, On 11/4/22 2:29 PM, KP Singh wrote: > On Fri, Nov 4, 2022 at 10:41 AM Artem Savkov <asavkov@redhat.com> wrote: >> >> lld produces "fast" style build-ids by default, which is inconsistent >> with ld's "sha1" style. Explicitly specify build-id style to be "sha1" >> when linking liburandom_read.so the same way it is already done for >> urandom_read. >> >> Signed-off-by: Artem Savkov <asavkov@redhat.com> > > Acked-by: KP Singh <kpsingh@kernel.org> > > This was done in > https://lore.kernel.org/bpf/20200922232140.1994390-1-morbo@google.com When you say "fix", does it actually fix a failing test case or is it more of a cleanup to align liburandom_read build with urandom_read? From glancing at the code, we only check build id for urandom_read. Cheers, Daniel
On Fri, Nov 4, 2022 at 10:38 AM Daniel Borkmann <daniel@iogearbox.net> wrote: > > Hi Artem, > > On 11/4/22 2:29 PM, KP Singh wrote: > > On Fri, Nov 4, 2022 at 10:41 AM Artem Savkov <asavkov@redhat.com> wrote: > >> > >> lld produces "fast" style build-ids by default, which is inconsistent > >> with ld's "sha1" style. Explicitly specify build-id style to be "sha1" > >> when linking liburandom_read.so the same way it is already done for > >> urandom_read. > >> > >> Signed-off-by: Artem Savkov <asavkov@redhat.com> > > > > Acked-by: KP Singh <kpsingh@kernel.org> > > > > This was done in > > https://lore.kernel.org/bpf/20200922232140.1994390-1-morbo@google.com > > When you say "fix", does it actually fix a failing test case or is it more > of a cleanup to align liburandom_read build with urandom_read? From glancing > at the code, we only check build id for urandom_read. I reworded the subject to "selftests/bpf: Use consistent build-id type for liburandom_read.so" and pushed. Thanks! > > Cheers, > Daniel
Hello: This patch was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <andrii@kernel.org>: On Fri, 4 Nov 2022 10:40:16 +0100 you wrote: > lld produces "fast" style build-ids by default, which is inconsistent > with ld's "sha1" style. Explicitly specify build-id style to be "sha1" > when linking liburandom_read.so the same way it is already done for > urandom_read. > > Signed-off-by: Artem Savkov <asavkov@redhat.com> > > [...] Here is the summary with links: - [bpf-next] selftests/bpf: fix build-id for liburandom_read.so https://git.kernel.org/bpf/bpf-next/c/61fc5e66f755 You are awesome, thank you!
On 11/4/22 11:58 PM, Andrii Nakryiko wrote: > On Fri, Nov 4, 2022 at 10:38 AM Daniel Borkmann <daniel@iogearbox.net> wrote: >> >> Hi Artem, >> >> On 11/4/22 2:29 PM, KP Singh wrote: >>> On Fri, Nov 4, 2022 at 10:41 AM Artem Savkov <asavkov@redhat.com> wrote: >>>> >>>> lld produces "fast" style build-ids by default, which is inconsistent >>>> with ld's "sha1" style. Explicitly specify build-id style to be "sha1" >>>> when linking liburandom_read.so the same way it is already done for >>>> urandom_read. >>>> >>>> Signed-off-by: Artem Savkov <asavkov@redhat.com> >>> >>> Acked-by: KP Singh <kpsingh@kernel.org> >>> >>> This was done in >>> https://lore.kernel.org/bpf/20200922232140.1994390-1-morbo@google.com >> >> When you say "fix", does it actually fix a failing test case or is it more >> of a cleanup to align liburandom_read build with urandom_read? From glancing >> at the code, we only check build id for urandom_read. > > I reworded the subject to "selftests/bpf: Use consistent build-id type > for liburandom_read.so" and pushed. Thanks! Ack, sgtm!
On Fri, Nov 04, 2022 at 03:58:43PM -0700, Andrii Nakryiko wrote: > On Fri, Nov 4, 2022 at 10:38 AM Daniel Borkmann <daniel@iogearbox.net> wrote: > > > > Hi Artem, > > > > On 11/4/22 2:29 PM, KP Singh wrote: > > > On Fri, Nov 4, 2022 at 10:41 AM Artem Savkov <asavkov@redhat.com> wrote: > > >> > > >> lld produces "fast" style build-ids by default, which is inconsistent > > >> with ld's "sha1" style. Explicitly specify build-id style to be "sha1" > > >> when linking liburandom_read.so the same way it is already done for > > >> urandom_read. > > >> > > >> Signed-off-by: Artem Savkov <asavkov@redhat.com> > > > > > > Acked-by: KP Singh <kpsingh@kernel.org> > > > > > > This was done in > > > https://lore.kernel.org/bpf/20200922232140.1994390-1-morbo@google.com > > > > When you say "fix", does it actually fix a failing test case or is it more > > of a cleanup to align liburandom_read build with urandom_read? From glancing > > at the code, we only check build id for urandom_read. I called it a "fix" because it broke expectations of external tools, but the reworded version sounds much better. > I reworded the subject to "selftests/bpf: Use consistent build-id type > for liburandom_read.so" and pushed. Thanks! Thank you, Andrii.
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 79edef1dbda4..5a792987df66 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -182,7 +182,8 @@ endif $(OUTPUT)/liburandom_read.so: urandom_read_lib1.c urandom_read_lib2.c $(call msg,LIB,,$@) $(Q)$(CLANG) $(filter-out -static,$(CFLAGS) $(LDFLAGS)) $^ $(LDLIBS) \ - -fuse-ld=$(LLD) -Wl,-znoseparate-code -fPIC -shared -o $@ + -fuse-ld=$(LLD) -Wl,-znoseparate-code -Wl,--build-id=sha1 \ + -fPIC -shared -o $@ $(OUTPUT)/urandom_read: urandom_read.c urandom_read_aux.c $(OUTPUT)/liburandom_read.so $(call msg,BINARY,,$@)
lld produces "fast" style build-ids by default, which is inconsistent with ld's "sha1" style. Explicitly specify build-id style to be "sha1" when linking liburandom_read.so the same way it is already done for urandom_read. Signed-off-by: Artem Savkov <asavkov@redhat.com> --- tools/testing/selftests/bpf/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)