Re: [edk2-devel] [PATCH] UefiCpuPkg RegisterCpuFeaturesLib: NumberOfCpus may be uninitialized

["Zeng, Star" <star.zeng@intel.com>]

> -----Original Message-----
> From: Laszlo Ersek <lersek@redhat.com>
> Sent: Thursday, January 21, 2021 9:48 PM
> To: devel@edk2.groups.io; Zeng, Star <star.zeng@intel.com>
> Cc: Dong, Eric <eric.dong@intel.com>; Ni, Ray <ray.ni@intel.com>
> Subject: Re: [edk2-devel] [PATCH] UefiCpuPkg RegisterCpuFeaturesLib:
> NumberOfCpus may be uninitialized
> 
> On 01/21/21 10:39, Zeng, Star wrote:
> > NumberOfCpus local variable in GetAcpiCpuData will be uninitialized
> > when CpuS3DataDxe runs before DxeRegisterCpuFeaturesLib (linked by
> > CpuFeaturesDxe) because there is no code to initialize it at
> > (AcpiCpuData != NULL) execution path.
> >
> > The issue is exposed after cefad282fb31aff3e1a6dcbd368cbbffc3fce900
> > and 38ee7bafa72f58982f99ac6f61eef160f80bad69.
> 
> Oops, sorry!
> 
> > There was negligence in that code review.
> > One further topic may be "Could EDK2 CI be enhanced to catch this kind
> > of uninitialized local variable case?". :)
> 
> Indeed I don't know why none of the toolchains caught this issue in CI.
> :/ GCC is otherwise known for its "variable may be used without
> initialization" warnings.
> 
> >
> > This patch fixes this regression issue.
> >
> > Cc: Eric Dong <eric.dong@intel.com>
> > Cc: Ray Ni <ray.ni@intel.com>
> > Cc: Laszlo Ersek <lersek@redhat.com>
> > Signed-off-by: Star Zeng <star.zeng@intel.com>
> > ---
> >  .../Library/RegisterCpuFeaturesLib/RegisterCpuFeaturesLib.c      | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git
> > a/UefiCpuPkg/Library/RegisterCpuFeaturesLib/RegisterCpuFeaturesLib.c
> > b/UefiCpuPkg/Library/RegisterCpuFeaturesLib/RegisterCpuFeaturesLib.c
> > index 7bb92404027f..60daa5cc87f0 100644
> > ---
> > a/UefiCpuPkg/Library/RegisterCpuFeaturesLib/RegisterCpuFeaturesLib.c
> > +++ b/UefiCpuPkg/Library/RegisterCpuFeaturesLib/RegisterCpuFeaturesLib
> > +++ .c
> > @@ -957,6 +957,7 @@ GetAcpiCpuData (
> >      //
> >      // Allocate buffer for empty RegisterTable and PreSmmInitRegisterTable
> for all CPUs
> >      //
> > +    NumberOfCpus = AcpiCpuData->NumberOfCpus;
> >      TableSize = 2 * NumberOfCpus * sizeof (CPU_REGISTER_TABLE);
> >      RegisterTable  = AllocatePages (EFI_SIZE_TO_PAGES (TableSize));
> >      ASSERT (RegisterTable != NULL);
> >
> 
> Reviewed-by: Laszlo Ersek <lersek@redhat.com>
> 
> (Also, I think Ray's R-b from
> <https://edk2.groups.io/g/devel/message/70615> applies here, so I'm going
> to pick that up.)

Laszlo and Ray, thanks for the R-b.
Please help push it by pull request if no other feedback is received.

Thanks,
Star

> 
> Thanks & sorry again
> Laszlo