On 1/17/24 15:12, Gerd Hoffmann wrote: >> This patch is good: >> >> Reviewed-by: Laszlo Ersek <ler...@redhat.com> >> >> but the series shouldn't stop here. In "OvmfPkg/Tcg/Tcg2Config", we're >> left with an INF file (Tcg2ConfigPei.inf) that still references >> "Tpm12Support.h", and the common C source file "Tcg2ConfigPeim.c" still >> calls the one API -- InternalTpm12Detect() -- declared in that header >> file. The only remaining implementation of InternalTpm12Detect() is now >> in "Tpm12SupportNull.c", and all it does is "return EFI_UNSUPPORTED". >> >> Therefore, in a subsequent patch, "Tpm12SupportNull.c" and >> "Tpm12Support.h" should be removed, both from the tree, and from the >> remaining INF file. Furthermore, the InternalTpm12Detect() call in >> "Tcg2ConfigPeim.c", and everything that depends on the success of that >> call, now counts as dead code, and should be removed. >> >> And *that* in turn means that we should also remove >> "gEfiTpmDeviceInstanceTpm12Guid" from the [Guids] section of the >> remaining INF file. > > Yes, I noticed there is more to cleanup, but I wanted have something > quick and proven (== identical to setting TPM1_ENABLE=FALSE) out of > the door to fix the build / CI problems.
Right, I now see that this would eliminate the symbol conflict. Either way, Jiewen has now reviewed Doug's patches for the symbol conflict, so I'm going to merge those (and your v3 VirtNorFlashDxe series, too). Laszlo > > Touching the code for additional cleanups requires a bit more attention > and testing, I'll keep that on my radar though. > > thanks & take care, > Gerd > > > > > > -=-=-=-=-=-=-=-=-=-=-=- Groups.io Links: You receive all messages sent to this group. View/Reply Online (#114005): https://edk2.groups.io/g/devel/message/114005 Mute This Topic: https://groups.io/mt/103764205/21656 Group Owner: devel+ow...@edk2.groups.io Unsubscribe: https://edk2.groups.io/g/devel/leave/9847357/21656/1706620634/xyzzy [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
