Laszlo, Thanks for the comments. I agree all of them. Since there's still EBC listed in SUPPORTED_ARCHITECTURES in dsc file, I don't see any reason not to add it. But likewise, I'll leave the final decision to package maintainers.
Regards, Jian > -----Original Message----- > From: Laszlo Ersek <ler...@redhat.com> > Sent: Wednesday, November 13, 2019 4:27 PM > To: Wang, Jian J <jian.j.w...@intel.com>; devel@edk2.groups.io > Cc: Kinney, Michael D <michael.d.kin...@intel.com>; Gao, Liming > <liming....@intel.com>; Ard Biesheuvel <ard.biesheu...@linaro.org>; Ni, Ray > <ray...@intel.com> > Subject: Re: [PATCH v2] MdePkg: add null version of RngLib > > Hi Jian, > > thanks for the update. Comments below: > > On 11/13/19 06:35, Jian J Wang wrote: > >> v2: > >> - Change the name from RngLibNull to BaseRngLibNull according to Laszlo's > >> comments > >> - Move the module from SecurityPkg to MdePkg according to Laszlo's > comments > >> - Update commit message according to Laszlo and Ray's comments > > > > This is null version of RngLib which should be used with modules that > > inherit an (indirect) dependency on the RngLib class, but never actually > > call RngLib APIs for consuming randomness. > > > > > > Ref: https://bugzilla.tianocore.org/show_bug.cgi?id=1871 > > Cc: Michael D Kinney <michael.d.kin...@intel.com> > > Cc: Liming Gao <liming....@intel.com> > > Cc: Laszlo Ersek <ler...@redhat.com> > > Cc: Ard Biesheuvel <ard.biesheu...@linaro.org> > > Cc: Ray Ni <ray...@intel.com> > > Signed-off-by: Jian J Wang <jian.j.w...@intel.com> > > --- > > .../Library/BaseRngLibNull/BaseRngLibNull.c | 95 +++++++++++++++++++ > > .../Library/BaseRngLibNull/BaseRngLibNull.inf | 31 ++++++ > > .../Library/BaseRngLibNull/BaseRngLibNull.uni | 14 +++ > > MdePkg/MdePkg.dsc | 1 + > > 4 files changed, 141 insertions(+) > > create mode 100644 MdePkg/Library/BaseRngLibNull/BaseRngLibNull.c > > create mode 100644 MdePkg/Library/BaseRngLibNull/BaseRngLibNull.inf > > create mode 100644 MdePkg/Library/BaseRngLibNull/BaseRngLibNull.uni > > > > diff --git a/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.c > b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.c > > new file mode 100644 > > index 0000000000..13677abc84 > > --- /dev/null > > +++ b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.c > > @@ -0,0 +1,95 @@ > > +/** @file > > + Null version of Random number generator services. > > + > > +Copyright (c) 2019, Intel Corporation. All rights reserved.<BR> > > +SPDX-License-Identifier: BSD-2-Clause-Patent > > + > > +**/ > > + > > +#include <Library/BaseLib.h> > > (1) I think we can drop BaseLib. > > > +#include <Library/DebugLib.h> > > +#include <Library/RngLib.h> > > + > > +/** > > + Generates a 16-bit random number. > > + > > + if Rand is NULL, then ASSERT(). > > + > > + @param[out] Rand Buffer pointer to store the 16-bit random value. > > + > > + @retval TRUE Random number generated successfully. > > + @retval FALSE Failed to generate the random number. > > + > > +**/ > > +BOOLEAN > > +EFIAPI > > +GetRandomNumber16 ( > > + OUT UINT16 *Rand > > + ) > > +{ > > + ASSERT (FALSE); > > + return FALSE; > > +} > > + > > +/** > > + Generates a 32-bit random number. > > + > > + if Rand is NULL, then ASSERT(). > > + > > + @param[out] Rand Buffer pointer to store the 32-bit random value. > > + > > + @retval TRUE Random number generated successfully. > > + @retval FALSE Failed to generate the random number. > > + > > +**/ > > +BOOLEAN > > +EFIAPI > > +GetRandomNumber32 ( > > + OUT UINT32 *Rand > > + ) > > +{ > > + ASSERT (FALSE); > > + return FALSE; > > +} > > + > > +/** > > + Generates a 64-bit random number. > > + > > + if Rand is NULL, then ASSERT(). > > + > > + @param[out] Rand Buffer pointer to store the 64-bit random value. > > + > > + @retval TRUE Random number generated successfully. > > + @retval FALSE Failed to generate the random number. > > + > > +**/ > > +BOOLEAN > > +EFIAPI > > +GetRandomNumber64 ( > > + OUT UINT64 *Rand > > + ) > > +{ > > + ASSERT (FALSE); > > + return FALSE; > > +} > > + > > +/** > > + Generates a 128-bit random number. > > + > > + if Rand is NULL, then ASSERT(). > > + > > + @param[out] Rand Buffer pointer to store the 128-bit random value. > > + > > + @retval TRUE Random number generated successfully. > > + @retval FALSE Failed to generate the random number. > > + > > +**/ > > +BOOLEAN > > +EFIAPI > > +GetRandomNumber128 ( > > + OUT UINT64 *Rand > > + ) > > +{ > > + ASSERT (FALSE); > > + return FALSE; > > +} > > diff --git a/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.inf > b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.inf > > new file mode 100644 > > index 0000000000..f456df1dae > > --- /dev/null > > +++ b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.inf > > @@ -0,0 +1,31 @@ > > +## @file > > +# Null instance of RNG (Random Number Generator) Library. > > +# > > +# Copyright (c) 2019, Intel Corporation. All rights reserved.<BR> > > +# > > +# SPDX-License-Identifier: BSD-2-Clause-Patent > > +# > > +## > > + > > +[Defines] > > + INF_VERSION = 0x00010005 > > (2) When adding a new library instance, I seem to remember that we > recommend referring to the latest published INF spec version (and using > a decimal version reference). That would be "1.29" at this point. > > https://edk2-docs.gitbooks.io/edk-ii-inf-specification/content/#edk-ii-module- > information-inf-file-specification > > I defer to the MdePkg maintainers on this, I just wanted to raise this. > > > > + BASE_NAME = BaseRngLibNull > > + MODULE_UNI_FILE = BaseRngLibNull.uni > > + FILE_GUID = CD8991F8-2061-4084-8C9E-9C6F352DC58D > > + MODULE_TYPE = BASE > > + VERSION_STRING = 1.0 > > + LIBRARY_CLASS = RngLib > > + > > +# > > +# VALID_ARCHITECTURES = IA32 X64 ARM AARCH64 > > +# > > (3) Again, just another idea that I'd like the MdePkg reviewers to > comment on: > > I don't see why this library instance would not work with EBC. > > However, I can also imagine we should not add EBC unless we at least > build-test the lib instance for EBC. > > What is the general approach here? > > (Note that I would never consider adding EBC in OvmfPkg or ArmVirtPkg > INF files, simply because I don't even know what EBC compilers exist. > But this is MdePkg, and a Null instance, so EBC *could* make sense.) > > > > + > > +[Sources] > > + BaseRngLibNull.c > > + > > +[Packages] > > + MdePkg/MdePkg.dec > > + > > +[LibraryClasses] > > + BaseLib > > (4) Same as (1). > > > + DebugLib > > diff --git a/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.uni > b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.uni > > new file mode 100644 > > index 0000000000..f32be6a617 > > --- /dev/null > > +++ b/MdePkg/Library/BaseRngLibNull/BaseRngLibNull.uni > > @@ -0,0 +1,14 @@ > > +// /** @file > > +// Null Instance of RNG (Random Number Generator) Library. > > +// > > +// Copyright (c) 2019, Intel Corporation. All rights reserved.<BR> > > +// > > +// SPDX-License-Identifier: BSD-2-Clause-Patent > > +// > > +// **/ > > + > > + > > +#string STR_MODULE_ABSTRACT #language en-US "Null Instance of > RNG Library" > > + > > +#string STR_MODULE_DESCRIPTION #language en-US "This library > instance should be used with modules that inherit an (indirect) dependency on > the RngLib class, but never actually call RngLib APIs for consuming > randomness." > > + > > diff --git a/MdePkg/MdePkg.dsc b/MdePkg/MdePkg.dsc > > index c1278e7907..0aeafaaacc 100644 > > --- a/MdePkg/MdePkg.dsc > > +++ b/MdePkg/MdePkg.dsc > > @@ -61,6 +61,7 @@ > > > MdePkg/Library/BaseUefiDecompressLib/BaseUefiTianoCustomDecompressLib. > inf > > MdePkg/Library/BaseSmbusLibNull/BaseSmbusLibNull.inf > > MdePkg/Library/BaseSafeIntLib/BaseSafeIntLib.inf > > + MdePkg/Library/BaseRngLibNull/BaseRngLibNull.inf > > OK, so this actually seems to confirm my question (3). This hunk > modifies the [Components] section in the DSC file, and not the > - [Components.IA32, Components.X64], > - [Components.ARM, Components.AARCH64] > sections. > > Furthermore, there is even a [Components.EBC] section in the DSC file. > > So I think "EBC" is still considered a viable ARCH in MdePkg, and we're > actually enabling this lib instance for the EBC arch. Therefore, EBC > should be listed in VALID_ARCHITECTURES, in the INF File -- as long as > we care to add a VALID_ARCHITECTURES comment at all, that is. > > If you agree to fix (1) through (4), you can immediately add: > > Reviewed-by: Laszlo Ersek <ler...@redhat.com> > > Thanks! > Laszlo > > > > > > MdePkg/Library/DxeCoreEntryPoint/DxeCoreEntryPoint.inf > > MdePkg/Library/DxeCoreHobLib/DxeCoreHobLib.inf > > -=-=-=-=-=-=-=-=-=-=-=- Groups.io Links: You receive all messages sent to this group. View/Reply Online (#50577): https://edk2.groups.io/g/devel/message/50577 Mute This Topic: https://groups.io/mt/55296935/21656 Group Owner: devel+ow...@edk2.groups.io Unsubscribe: https://edk2.groups.io/g/devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
