Martin Roth has posted comments on this change. ( https://review.coreboot.org/28060 )
Change subject: intel/common/block: Fix issues found by klockwork ......................................................................
Patch Set 1:
(1 comment)
Martin, What tool are you using for this code scan - we're not getting these "soc issues" flagged, using Klocwork Static Code Analysis tool? We are getting different "warnings" in chip.c and gspi.c files.
These are generated using clang's scan-build tool: https://clang-analyzer.llvm.org/scan-build.html
We also run coverity's stataic analysis on a bi-weekly basis. That output is here: https://scan.coverity.com/projects/coreboot?tab=overview That finds still different issues in soc/intel.
https://review.coreboot.org/#/c/28060/1/src/soc/intel/common/block/cpu/mp_in... File src/soc/intel/common/block/cpu/mp_init.c:
https://review.coreboot.org/#/c/28060/1/src/soc/intel/common/block/cpu/mp_in... PS1, Line 137: assert
Links are downstream buses attached to the device as: […]
It will only avoid the null pointer being dereferenced *IF* fatal asserts are turned on. Typically, as I said, asserts in coreboot are non-fatal, so the NULL pointer *WILL* still be dereferenced.
Think of assert in the coreboot codebase as a printk that gets output if the assert fails. Since they are non-fatal, the code still runs beyond this.
https://review.coreboot.org/cgit/coreboot.git/tree/src/Kconfig#n691
https://review.coreboot.org/cgit/coreboot.git/tree/src/include/assert.h#n34
-
Martin Roth (Code Review)