Re: [Qemu-devel] [PULL v1 0/5] Merge tpm 2018/10/29 v1

[Marc-André Lureau]

Hi Peter

On Tue, Oct 30, 2018 at 2:45 PM Peter Maydell <address@hidden> wrote:
>
> On 29 October 2018 at 15:19, Stefan Berger <address@hidden> wrote:
> > From: Stefan Berger <address@hidden>
> >
> > This pull request fixes a couple of TPM support related issues,
> > such as full initialization of a variable to quiet down valgrind,
> > a possible race in the TPM related test cases and marking test
> > cases as skipped if swtpm was not found in PATH.
> >
> >    Stefan
> >
> > The following changes since commit a4d710251fa5aa9ec26de4626f11c78500195d12:
> >
> >   Merge remote-tracking branch 
> > 'remotes/berrange/tags/qcrypto-next-pull-request' into staging (2018-10-24 
> > 22:08:42 +0100)
> >
> > are available in the Git repository at:
> >
> >   git://github.com/stefanberger/qemu-tpm.git tags/pull-tpm-2018-10-29-1
> >
> > for you to fetch changes up to f9da599490730cf0c3016f15225d7f1ee15bec75:
> >
> >   tpm: Zero-init structure to avoid uninitialized variables in valgrind log 
> > (2018-10-27 10:33:18 -0400)
> >
> > ----------------------------------------------------------------
> > Marc-André Lureau (2):
> >       tests/tpm: fix tpm_util_swtpm_has_tpm2()
> >       tests/tpm: mark swtpm test as skipped instead of successful
> >
> > Stefan Berger (3):
> >       docs: tpm: Mention implemented TPM CRB interface emulation and specs
> >       MAINTAINERS: Change my email address to the new domain
> >       tpm: Zero-init structure to avoid uninitialized variables in valgrind 
> > log
>
> Hi; this gives some test failures (all hosts):
>
> TEST: tests/tpm-crb-swtpm-test... (pid=305)
>   /i386/tpm/crb-swtpm/test:                                            FAIL
> GTester: last random seed: R02Sf628ca48da88252a6b95465fef3b855f
>   /i386/tpm/crb-swtpm-migration/test:                                  FAIL
> GTester: last random seed: R02Sf628ca48da88252a6b95465fef3b855f
> FAIL: tests/tpm-crb-swtpm-test
> TEST: tests/tpm-crb-test... (pid=311)
>   /i386/tpm-crb/test:                                                  OK
> PASS: tests/tpm-crb-test
> TEST: tests/tpm-tis-swtpm-test... (pid=324)
>   /i386/tpm/tis-swtpm/test:                                            FAIL
> GTester: last random seed: R02Sca4f303959770d4ca9f8f12b784dba0d
>   /i386/tpm/tis-swtpm-migration/test:                                  FAIL
> GTester: last random seed: R02Sca4f303959770d4ca9f8f12b784dba0d
> FAIL: tests/tpm-tis-swtpm-test
> TEST: tests/tpm-tis-test... (pid=331)
>   /i386/tpm-tis/test_check_localities:                                 OK
>   /i386/tpm-tis/test_check_access_reg:                                 OK
>   /i386/tpm-tis/test_check_access_reg_seize:                           OK
>   /i386/tpm-tis/test_check_access_reg_release:                         OK
>   /i386/tpm-tis/test_check_transmit:                                   OK
> PASS: tests/tpm-tis-test
>

What version of swtpm & libtpms is installed?

thanks

> and a compile failure on OSX:
> /Users/pm215/src/qemu-for-merges/hw/tpm/tpm_emulator.c:161:21: error:
> suggest braces around initialization of subobject
> [-Werror,-Wmissing-braces]
>     ptm_loc loc = { 0 };
>                     ^
>                     {}
>
> (what you want here is just "ptm_loc loc = {};" -- the "{ 0 }" syntax
> is the C standard one but not all compilers can handle it without warnings
> if the first member in the struct happens to be a substruct.)
>
> thanks
> -- PMM
>


-- 
Marc-André Lureau