diff options
author | Paul Selkirk <paul@psgd.org> | 2014-09-03 17:57:37 -0400 |
---|---|---|
committer | Paul Selkirk <paul@psgd.org> | 2014-09-03 17:57:37 -0400 |
commit | 6ff743f2abb05b07bd3c5585f59b14c459c69bfa (patch) | |
tree | cb5c887b2fbfcf7a0fc844451cffd3c04470018d /common/rtl/ipcore/clkmgr_dcm.ncf | |
parent | f06583ef515bae8f24d696e0df75a425bc47c2bb (diff) |
hash_tester should read the correct number of response bytes
Both versions of hash_tester optimistically assumed a READ_OK or
WRITE_OK response, and tried to read the expected number of bytes.
But ERROR and UNKNOWN responses are shorter, which could lead to an
overrun of available data from i2c. With commit a768072, this is no
longer fatal (i2c will zero-pad the response), but it's good form to do
the right thing when possible.
Diffstat (limited to 'common/rtl/ipcore/clkmgr_dcm.ncf')
0 files changed, 0 insertions, 0 deletions