Re: Possibly out of sync/broken openssl code

2020-02-16 Thread nisarg joshi
So the type MD32_REG_T is defined as either long or int in md32_common.h, and it is being used in md4_dgst.c and rmd_dgst.c for carrying out digest calculations. These calculations have shift operations and UBSan is reporting overflows for these operations because MD32_REG_T is int or long. I check

Re: Possibly out of sync/broken openssl code

2020-02-16 Thread Christos Zoulas
Thanks! Next time I import, I will try to minimize the diffs. I have been upstreaming a lot of our patches to different things lately and it requires patience and baby-sitting. christos > On Feb 16, 2020, at 7:10 AM, nisarg joshi wrote: > > signature.asc Description: Message signed with O

Re: Possibly out of sync/broken openssl code

2020-02-15 Thread Christos Zoulas
Yes, because "unsigned long" on _LP64 is not 32bits. What is the undefined behavior you are trying to fix? christos > On Feb 15, 2020, at 7:03 PM, nisarg joshi wrote: > > Hi community, > > > > While trying to fix UBSan Undefined behavior reports for md4_dgst.c and > rmd_dgst.c files which

Possibly out of sync/broken openssl code

2020-02-15 Thread nisarg joshi
Hi community, While trying to fix UBSan Undefined behavior reports for md4_dgst.c and rmd_dgst.c files which reported issues for the overflow of signed ints etc, I came across a possible out of sync tree of openssl crypto implementations. The UBSan report required usage of unsigned int(or long) f