Re: [PATCH v2 1/1] MdePkg: Refine casting expression result to bigger size

["Wu, Hao A" <hao.a.wu@intel.com>]

> -----Original Message-----
> From: Gao, Liming
> Sent: Wednesday, January 25, 2017 1:58 PM
> To: Wu, Hao A; Laszlo Ersek
> Cc: edk2-devel@ml01.01.org
> Subject: RE: [edk2] [PATCH v2 1/1] MdePkg: Refine casting expression result to
> bigger size
> 
> Hao:
>   For PCILIB_TO_COMMON_ADDRESS, we can't assume its usage in the
> consumer code. There may be some usage in other projects. So, I suggest to
> provide the safe fix.
> 

Hi Liming,

The definition "PCILIB_TO_COMMON_ADDRESS" is defined in
MdePkg/Library/BaseS3PciLib/S3PciLib.c. It will not be consumed outside.

Best Regards,
Hao Wu

> Thanks
> Liming
> >-----Original Message-----
> >From: edk2-devel [mailto:edk2-devel-bounces@lists.01.org] On Behalf Of Wu,
> >Hao A
> >Sent: Wednesday, January 25, 2017 8:26 AM
> >To: Laszlo Ersek <lersek@redhat.com>
> >Cc: edk2-devel@ml01.01.org
> >Subject: Re: [edk2] [PATCH v2 1/1] MdePkg: Refine casting expression result
> >to bigger size
> >
> >> -----Original Message-----
> >> From: Laszlo Ersek [mailto:lersek@redhat.com]
> >> Sent: Tuesday, January 24, 2017 5:54 PM
> >> To: Wu, Hao A
> >> Cc: edk2-devel@ml01.01.org
> >> Subject: Re: [edk2] [PATCH v2 1/1] MdePkg: Refine casting expression result
> >to
> >> bigger size
> >>
> >> On 01/24/17 08:25, Hao Wu wrote:
> >> > There are cases that the operands of an expression are all with rank less
> >> > than UINT64/INT64 and the result of the expression is explicitly casted to
> >> > UINT64/INT64 to fit the target size.
> >> >
> >> > An example will be:
> >> > UINT32 a,b;
> >> > // a and b can be any unsigned int type with rank less than UINT64, like
> >> > // UINT8, UINT16, etc.
> >> > UINT64 c;
> >> > c = (UINT64) (a + b);
> >> >
> >> > Some static code checkers may warn that the expression result might
> >> > overflow within the rank of "int" (integer promotions) and the result is
> >> > then cast to a bigger size.
> >> >
> >> > The commit refines codes by the following rules:
> >> > 1). When the expression will not overflow within the rank of "int", remove
> >> > the explicit type casts:
> >> > c = a + b;
> >> >
> >> > 2). When the expression is possible to overflow the range of unsigned int/
> >> > int:
> >> > c = (UINT64)a + b;
> >> >
> >> > Contributed-under: TianoCore Contribution Agreement 1.0
> >> > Signed-off-by: Hao Wu <hao.a.wu@intel.com>
> >> > ---
> >> >  MdePkg/Library/BaseLib/String.c                              |  4 ++--
> >> >  MdePkg/Library/BasePeCoffLib/BasePeCoff.c                    | 12 +++++-------
> >> >  MdePkg/Library/BaseS3PciLib/S3PciLib.c                       |  4 ++--
> >> >  MdePkg/Library/SmmMemoryAllocationLib/MemoryAllocationLib.c  |  4
> >++--
> >> >  MdePkg/Library/UefiMemoryAllocationLib/MemoryAllocationLib.c |  4
> >++--
> >> >  5 files changed, 13 insertions(+), 15 deletions(-)
> >> >
> >> > diff --git a/MdePkg/Library/BaseLib/String.c
> >> b/MdePkg/Library/BaseLib/String.c
> >> > index e84bf50..4151e0e 100644
> >> > --- a/MdePkg/Library/BaseLib/String.c
> >> > +++ b/MdePkg/Library/BaseLib/String.c
> >> > @@ -586,7 +586,7 @@ InternalHexCharToUintn (
> >> >      return Char - L'0';
> >> >    }
> >> >
> >> > -  return (UINTN) (10 + InternalCharToUpper (Char) - L'A');
> >> > +  return (10 + InternalCharToUpper (Char) - L'A');
> >> >  }
> >> >
> >> >  /**
> >> > @@ -1211,7 +1211,7 @@ InternalAsciiHexCharToUintn (
> >> >      return Char - '0';
> >> >    }
> >> >
> >> > -  return (UINTN) (10 + InternalBaseLibAsciiToUpper (Char) - 'A');
> >> > +  return (10 + InternalBaseLibAsciiToUpper (Char) - 'A');
> >> >  }
> >> >
> >> >
> >> > diff --git a/MdePkg/Library/BasePeCoffLib/BasePeCoff.c
> >> b/MdePkg/Library/BasePeCoffLib/BasePeCoff.c
> >> > index 33cad23..8d1daba 100644
> >> > --- a/MdePkg/Library/BasePeCoffLib/BasePeCoff.c
> >> > +++ b/MdePkg/Library/BasePeCoffLib/BasePeCoff.c
> >> > @@ -15,7 +15,7 @@
> >> >    PeCoffLoaderGetPeHeader() routine will do basic check for PE/COFF
> >header.
> >> >    PeCoffLoaderGetImageInfo() routine will do basic check for whole
> >PE/COFF
> >> image.
> >> >
> >> > -  Copyright (c) 2006 - 2014, Intel Corporation. All rights reserved.<BR>
> >> > +  Copyright (c) 2006 - 2017, Intel Corporation. All rights reserved.<BR>
> >> >    Portions copyright (c) 2008 - 2009, Apple Inc. All rights reserved.<BR>
> >> >    This program and the accompanying materials
> >> >    are licensed and made available under the terms and conditions of the
> >BSD
> >> License
> >> > @@ -703,12 +703,10 @@ PeCoffLoaderGetImageInfo (
> >> >        //
> >> >        DebugDirectoryEntryFileOffset = 0;
> >> >
> >> > -      SectionHeaderOffset = (UINTN)(
> >> > -                               ImageContext->PeCoffHeaderOffset +
> >> > -                               sizeof (UINT32) +
> >> > -                               sizeof (EFI_IMAGE_FILE_HEADER) +
> >> > -                               Hdr.Pe32->FileHeader.SizeOfOptionalHeader
> >> > -                               );
> >> > +      SectionHeaderOffset = ImageContext->PeCoffHeaderOffset +
> >> > +                            sizeof (UINT32) +
> >> > +                            sizeof (EFI_IMAGE_FILE_HEADER) +
> >> > +                            Hdr.Pe32->FileHeader.SizeOfOptionalHeader;
> >> >
> >> >        for (Index = 0; Index < Hdr.Pe32->FileHeader.NumberOfSections;
> >Index++)
> >> {
> >> >          //
> >> > diff --git a/MdePkg/Library/BaseS3PciLib/S3PciLib.c
> >> b/MdePkg/Library/BaseS3PciLib/S3PciLib.c
> >> > index e29f7fe..27342b0 100644
> >> > --- a/MdePkg/Library/BaseS3PciLib/S3PciLib.c
> >> > +++ b/MdePkg/Library/BaseS3PciLib/S3PciLib.c
> >> > @@ -3,7 +3,7 @@
> >> >    the PCI operations to be replayed during an S3 resume. This library class
> >> >    maps directly on top of the PciLib class.
> >> >
> >> > -  Copyright (c) 2006 - 2012, Intel Corporation. All rights reserved.<BR>
> >> > +  Copyright (c) 2006 - 2017, Intel Corporation. All rights reserved.<BR>
> >> >
> >> >    This program and the accompanying materials
> >> >    are licensed and made available under the terms and conditions
> >> > @@ -25,7 +25,7 @@
> >> >  #include <Library/S3PciLib.h>
> >> >
> >> >  #define PCILIB_TO_COMMON_ADDRESS(Address) \
> >> > -        ((UINT64) ((((UINTN) ((Address>>20) & 0xff)) << 24) + (((UINTN)
> >> ((Address>>15) & 0x1f)) << 16) + (((UINTN) ((Address>>12) & 0x07)) << 8) +
> >> ((UINTN) (Address & 0xfff ))))
> >> > +        ((((UINTN) ((Address>>20) & 0xff)) << 24) + (((UINTN)
> >((Address>>15) &
> >> 0x1f)) << 16) + (((UINTN) ((Address>>12) & 0x07)) << 8) + ((UINTN)
> >(Address &
> >> 0xfff )))
> >> >
> >> >  /**
> >> >    Saves a PCI configuration value to the boot script.
> >>
> >> I think this change is potentially unsafe, without auditing all uses of
> >> PCILIB_TO_COMMON_ADDRESS(). In a 32-bit build, the type of the result
> >> will no longer be UINT64 but UINT32, and that can cause problems in
> >> several contexts. For example:
> >>
> >> - as an operand to the sizeof operator
> >> - when it's being relied upon to cause conversion to UINT64, for example
> >> another (UINT32) operand could be added to it
> >> - when it is passed through a variable argument list
> >>
> >> It might be safe, but there's no way to tell without auditing all the
> >> call sites. So let me see...
> >>
> >> Apparently this macro is only passed to S3BootScriptSavePciCfgWrite() as
> >> second argument, within the same file, and that argument is covered by
> >> the function prototype explicitly, with type UINT64. So the change
> >> should be safe.
> >>
> >
> >Thanks for the checking. I did search the whole edk2 repository for the
> >reference of "PCILIB_TO_COMMON_ADDRESS" and it is only comsumed by
> >the
> >function you mentioned.
> >
> >> (I see the same macro definition and kind of invocation in
> >> "QuarkPlatformPkg/Acpi/DxeSmm/AcpiSmm/AcpiSmmPlatform.c"; I didn't
> >try
> >> to audit that file.)
> >>
> >> The rest looks okay too.
> >>
> >> Reviewed-by: Laszlo Ersek <lersek@redhat.com>
> >>
> >
> >Many thanks for the feedbacks and the effort for reviewing the patch.
> >
> >> (If you go ahead and submit a 30-part series that does this kind of
> >> fixup all over the tree, please don't expect me to review it all -- I'm
> >> okay reviewing OvmfPkg and ArmVirtPkg changes, but I can't take on the
> >> rest. This kind of patch cannot be reviewed without consulting a really
> >> wide context.)
> >>
> >
> >I am thinking if the package level patch contains too many changes, I
> >might break it into multiple module-level patches and include module
> >owners/experts to help reviewing them.
> >
> >Best Regards,
> >Hao Wu
> >
> >> Thanks
> >> Laszlo
> >>
> >>
> >> > diff --git
> >> a/MdePkg/Library/SmmMemoryAllocationLib/MemoryAllocationLib.c
> >> b/MdePkg/Library/SmmMemoryAllocationLib/MemoryAllocationLib.c
> >> > index 937165a..592cced 100644
> >> > --- a/MdePkg/Library/SmmMemoryAllocationLib/MemoryAllocationLib.c
> >> > +++ b/MdePkg/Library/SmmMemoryAllocationLib/MemoryAllocationLib.c
> >> > @@ -12,7 +12,7 @@
> >> >    allocation for the Reserved memory types are not supported and will
> >> always
> >> >    return NULL.
> >> >
> >> > -  Copyright (c) 2006 - 2015, Intel Corporation. All rights reserved.<BR>
> >> > +  Copyright (c) 2006 - 2017, Intel Corporation. All rights reserved.<BR>
> >> >    This program and the accompanying materials
> >> >    are licensed and made available under the terms and conditions of the
> >BSD
> >> License
> >> >    which accompanies this distribution.  The full text of the license may be
> >> found at
> >> > @@ -343,7 +343,7 @@ InternalAllocateAlignedPages (
> >> >        Status = gSmst->SmmFreePages (Memory, UnalignedPages);
> >> >        ASSERT_EFI_ERROR (Status);
> >> >      }
> >> > -    Memory         = (EFI_PHYSICAL_ADDRESS) (AlignedMemory +
> >> EFI_PAGES_TO_SIZE (Pages));
> >> > +    Memory         = AlignedMemory + EFI_PAGES_TO_SIZE (Pages);
> >> >      UnalignedPages = RealPages - Pages - UnalignedPages;
> >> >      if (UnalignedPages > 0) {
> >> >        //
> >> > diff --git
> >a/MdePkg/Library/UefiMemoryAllocationLib/MemoryAllocationLib.c
> >> b/MdePkg/Library/UefiMemoryAllocationLib/MemoryAllocationLib.c
> >> > index 3da5e211..3bd3aef 100644
> >> > --- a/MdePkg/Library/UefiMemoryAllocationLib/MemoryAllocationLib.c
> >> > +++ b/MdePkg/Library/UefiMemoryAllocationLib/MemoryAllocationLib.c
> >> > @@ -2,7 +2,7 @@
> >> >    Support routines for memory allocation routines based
> >> >    on boot services for Dxe phase drivers.
> >> >
> >> > -  Copyright (c) 2006 - 2015, Intel Corporation. All rights reserved.<BR>
> >> > +  Copyright (c) 2006 - 2017, Intel Corporation. All rights reserved.<BR>
> >> >    This program and the accompanying materials
> >> >    are licensed and made available under the terms and conditions of the
> >BSD
> >> License
> >> >    which accompanies this distribution.  The full text of the license may be
> >> found at
> >> > @@ -216,7 +216,7 @@ InternalAllocateAlignedPages (
> >> >        Status = gBS->FreePages (Memory, UnalignedPages);
> >> >        ASSERT_EFI_ERROR (Status);
> >> >      }
> >> > -    Memory         = (EFI_PHYSICAL_ADDRESS) (AlignedMemory +
> >> EFI_PAGES_TO_SIZE (Pages));
> >> > +    Memory         = AlignedMemory + EFI_PAGES_TO_SIZE (Pages);
> >> >      UnalignedPages = RealPages - Pages - UnalignedPages;
> >> >      if (UnalignedPages > 0) {
> >> >        //
> >> >
> >
> >_______________________________________________
> >edk2-devel mailing list
> >edk2-devel@lists.01.org
> >https://lists.01.org/mailman/listinfo/edk2-devel