Page MenuHomeFreeBSD

Fix SHA256 usage on older CESA versions
ClosedPublic

Authored by mw_semihalf.com on Jan 17 2017, 3:51 PM.
Tags
None
Referenced Files
Unknown Object (File)
Thu, Nov 14, 6:00 PM
Unknown Object (File)
Sep 16 2024, 5:43 PM
Unknown Object (File)
Sep 9 2024, 1:34 AM
Unknown Object (File)
Sep 8 2024, 6:17 PM
Unknown Object (File)
Sep 8 2024, 5:37 AM
Unknown Object (File)
Sep 7 2024, 11:39 AM
Unknown Object (File)
Aug 6 2024, 8:59 AM
Unknown Object (File)
Aug 6 2024, 12:47 AM
Subscribers

Details

Summary

Adding SHA256 support to Marvell crypto driver resulted in regression
for older SoC's, not capable of handling this mode in hardware.

Diff Detail

Lint
Lint Skipped
Unit
Tests Skipped

Event Timeline

mw_semihalf.com retitled this revision from to Fix SHA256 usage on older CESA versions.
mw_semihalf.com updated this object.
mw_semihalf.com edited the test plan for this revision. (Show Details)
mw_semihalf.com added reviewers: ian, zbb, imp, fabient.
mw_semihalf.com set the repository for this revision to rS FreeBSD src repository - subversion.
mw_semihalf.com added a subscriber: ARM.
zbb edited edge metadata.
This revision is now accepted and ready to land.Jan 17 2017, 6:30 PM

imho, this is papering over problem in DTS. If CESA device have more variants, then each one must have different 'compatible' string.
In other words, you cannot use same 'compatible' string for different devices (if the difference cannot be determined at runtime).

BTW, why you can't use 'standard' DTS for Marvell SOCs, even with FBSD specific overlay?

Ok, will add compatible-based revision differentiation instead of the ifdefs.

mw_semihalf.com edited edge metadata.
mw_semihalf.com removed rS FreeBSD src repository - subversion as the repository for this revision.

Updated version contains implemented SOC-ID based hardware differentiation.

This revision now requires review to proceed.Jan 19 2017, 5:52 PM
This revision was automatically updated to reflect the committed changes.