aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/src/crypto/curve25519.c (unfollow)
Commit message (Collapse)AuthorFilesLines
2016-07-07go test: put nonce at correct locationJason A. Donenfeld1-1/+1
2016-07-07go test: make more idiomaticJonathan Rudenberg1-40/+65
- gofmt - Give config struct one line per field - Use camel case - Check errors - Log invariants with detail - Use consistent pronouns
2016-07-06tools: use pkg-config in MakefileJason A. Donenfeld1-1/+2
2016-07-05device: move unlikely check to if clauseJason A. Donenfeld1-2/+2
2016-07-05contrib: organize example scripts and add synergyJason A. Donenfeld8-0/+43
2016-07-03receive: protect against impossible conditionsJason A. Donenfeld1-0/+4
It should never be the case that skb->head + skb->transport_header - skb->data is greater than 2^16, but in case the kernel network stack borks this at some point in the future, we don't want this to slyly introduce a vulnerability into WireGuard. Further, really smart compilers might be able to make deductions about data_offset, and optimize accordingly.
2016-07-03tools: always fallback to /dev/urandomJason A. Donenfeld1-10/+8
2016-07-03tools: improve error reporting and detectionJason A. Donenfeld4-24/+43
2016-07-02tai64n: don't forget to add 2^62, to be in specJason A. Donenfeld1-2/+2
2016-07-01contrib: remove extraneous cruftJason A. Donenfeld10-195/+19
We don't want people packaging these or even using these scripts, which are only useful for limited development circumstances, so get rid of them. More widespread development testing techniques still exist in src/debug.mk and src/netns.sh
2016-07-01wg.8: wording tweaksexperimental-0.0.20160630Jason A. Donenfeld1-5/+7
Suggested-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
2016-07-01receive: error conditions are unlikelyJason A. Donenfeld1-3/+3
2016-06-30Readme: the documentation moved to .ioJason A. Donenfeld1-1/+1
2016-06-30Readme: use https instead of httpDaniel Kahn Gillmor3-3/+3
For the websites referenced that offer https instead of http, use https.
2016-06-30Makefile: Add more verbose dependency errorsJason A. Donenfeld1-0/+32
2016-06-30device init: free wq after padataJason A. Donenfeld1-3/+3
The padata free functions make reference to their parent workqueue, so it's important that we wait to free the workqueue after the padata.
2016-06-29chacha20poly1305: use more standard way of testing FPU featuresJason A. Donenfeld1-7/+2
2016-06-29device: remove updating of trans_startJason A. Donenfeld1-2/+0
Per http://lists.openwall.net/netdev/2016/05/03/87 dev->trans_start has been removed, and updates are now supposed to be handled with netif_trans_update, which now updates the particular txqueue's trans_start instead. However, netdev_start_xmit already updates this member after calling ndo_start_xmit, so the new netif_trans_update function smartly makes the comment that for drivers that don't use LLTX, it's not neccessary to call netif_trans_update. Except we do use LLTX, so it would seem again that we do need to be calling netif_trans_update. However, glancing at drivers like vxlan and other similar virtual tunnels, this doesn't seem to be the case. I suspect the reason is that we both also set IFF_NO_QUEUE, so we aren't even using a txqueue for updating. Thus, this patch removes updating of trans_start all together. I believe this should be okay for older kernels too.
2016-06-29Kconfig patching: do not match on NETFILTERJason A. Donenfeld1-1/+1
2016-06-29Kconfig: more fully select dependenciesJason A. Donenfeld1-0/+3
2016-06-25tests: make fatalJason A. Donenfeld13-20/+29
2016-06-25nonce: switch to RFC6479 to better support packet reorderingJason A. Donenfeld3-61/+95
With packets hitting multiple cores, a 64bit backtrack was too small. This algorithm increases our backtrack to 1984bits.
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.