Re: [PATCH 2/3] IMA:Define a new template field buf

From: Mimi Zohar
Date: Wed Jun 19 2019 - 14:42:23 EST


On Wed, 2019-06-19 at 11:08 -0700, prakhar srivastava wrote:
> <snip>
> > > if (iint->measured_pcrs & (0x1 << pcr))
> > > diff --git a/security/integrity/ima/ima_init.c b/security/integrity/ima/ima_init.c
> > > index 993d0f1915ff..c8591406c0e2 100644
> > > --- a/security/integrity/ima/ima_init.c
> > > +++ b/security/integrity/ima/ima_init.c
> > > @@ -50,7 +50,7 @@ static int __init ima_add_boot_aggregate(void)
> > > struct ima_template_entry *entry;
> > > struct integrity_iint_cache tmp_iint, *iint = &tmp_iint;
> > > struct ima_event_data event_data = {iint, NULL, boot_aggregate_name,
> > > - NULL, 0, NULL};
> > > + NULL, 0, NULL, NULL, 0};
> > > int result = -ENOMEM;
> > > int violation = 0;
> > > struct {
> > >
> >
> > These changes shouldn't be necessary. Please rebase these patches on
> > top of the latest next-queued-testing branch (git remote update). "IMA: support for per
> > policy rule template formats" is still changing.
> >
> > Minor nit. When re-posting the patches please update the patch titles
> > so that there is a space between the subsystem name and the patch
> > title (eg. "ima: define ...").
> >
> I believe the above event_data changes are needed, to store/read the
> buffer length and buffer itself. The only exception will be if needed will be to
> remove ima-buf as a template instead used a template_fmt in the policy
> with KEXEC_CMDLINE from the "IMA: support for per
> policy rule template formats" is still changing.".
> In my view even ima-buf is needed as it simplifies the usage.
>
> Please let me know if I misunderstood your comment.

The tip of next-queued-testing branch is commit 687d57f90461 ("IMA:
support for per policy rule template formats"). ÂThe current code is:

struct ima_event_data event_data = { .iint = iint,
.filename = boot_aggregate_name };

Mimi