Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add ARM AES asm implementation from Cryptogams #683

Closed
noloader opened this issue Jul 11, 2018 · 2 comments
Closed

Add ARM AES asm implementation from Cryptogams #683

noloader opened this issue Jul 11, 2018 · 2 comments
Labels
Enhancement

Comments

@noloader
Copy link
Collaborator

noloader commented Jul 11, 2018 •
edited
Loading

Add ARM AES asm implementation from Cryptogams.

Cryptogams is Andy Polyakov's project used to create high speed crypto algorithms and share them with other developers. Cryptogams has a dual license. First is the OpenSSL license because Andy contributes to OpenSSL. Second is a BSD license for those who want a more permissive license.

The integration instructions are documented at Cryptogams AES on the OpenSSL wiki.
noloader added a commit that referenced this issue Jul 11, 2018
@noloader
Add ARM AES asm implementation from Cryptogams (GH #683)
3ff7d7f
noloader referenced this issue Jul 11, 2018
@noloader
Disable Cryptogams AES under Clang
1e77a72 
What a surprise... Clang pretends to be GCC with __GNUC__ but fails to consume the source file
@noloader noloader mentioned this issue Jul 11, 2018
Closed
noloader added a commit that referenced this issue Jul 12, 2018
@noloader
Disable same buffer for in and out on ARM A-32 (GH #683)
e580ed5
@noloader
Copy link
Collaborator Author

Cleared at Commit 3ff7d7f0286a.

noloader added a commit that referenced this issue Oct 14, 2018
@noloader
Enable same buffer for in and out on ARM A-32 (GH #683)
49d852e
noloader referenced this issue Apr 13, 2020
@noloader
Fix AES/CFB mode failure on ARMv7
cac2828
noloader added a commit that referenced this issue Mar 17, 2021
@noloader
Avoid memcpy in AdditiveCipherTemplate<S>::ProcessData (GH #683, GH #…
ca123d1 
…1010, PR #1019)

We found we can avoid the memcpy in the previous workaround by using a volatile pointer. The pointer appears to tame the optimizer so the compiler does not short-circuit some calls when outString == inString.
noloader added a commit that referenced this issue Mar 17, 2021
@noloader
Fix CTR mode when using FileSource (GH #683, GH #1010)
71a812e 
We think this is another instance problem that surfaced under GH #683 when inString==outString. It violates aliasing rules and the compiler begins removing code.

The ultimate workaround was to add a member variable m_tempOutString as scratch space when inString==outString. We did not loose much in the way of perforamce for some reason. It looks like AES/CTR lost about 0.03-0.05 cpb.

When combined with the updated xorbuf from GH #1020, the net result was a speedup of 0.1-0.6 cpb. In fact, some ciphers like RC6, gained almost 5 cpb.
noloader added a commit that referenced this issue Mar 17, 2021
@noloader
Fix CTR mode when using FileSource (GH #683, GH #1010)
bbc45dd 
We think this is another instance problem that surfaced under GH #683 when inString==outString. It violates aliasing rules and the compiler begins removing code.

The ultimate workaround was to add a member variable m_tempOutString as scratch space when inString==outString. We did not loose much in the way of perforamce for some reason. It looks like AES/CTR lost about 0.03-0.05 cpb.

When combined with the updated xorbuf from GH #1020, the net result was a speedup of 0.1-0.6 cpb. In fact, some ciphers like RC6, gained almost 5 cpb.
EAddario pushed a commit to EAddario/cryptopp that referenced this issue Apr 10, 2021
@noloader @EAddario
Avoid memcpy in AdditiveCipherTemplate<S>::ProcessData (GH weidai11#683…
eda94a2 
…, GH weidai11#1010, PR weidai11#1019)

We found we can avoid the memcpy in the previous workaround by using a volatile pointer. The pointer appears to tame the optimizer so the compiler does not short-circuit some calls when outString == inString.
EAddario pushed a commit to EAddario/cryptopp that referenced this issue Apr 10, 2021
@noloader @EAddario
Fix CTR mode when using FileSource (GH weidai11#683, GH weidai11#1010)
5a37d60 
We think this is another instance problem that surfaced under GH weidai11#683 when inString==outString. It violates aliasing rules and the compiler begins removing code.

The ultimate workaround was to add a member variable m_tempOutString as scratch space when inString==outString. We did not loose much in the way of perforamce for some reason. It looks like AES/CTR lost about 0.03-0.05 cpb.

When combined with the updated xorbuf from GH weidai11#1020, the net result was a speedup of 0.1-0.6 cpb. In fact, some ciphers like RC6, gained almost 5 cpb.
EAddario pushed a commit to EAddario/cryptopp that referenced this issue Apr 10, 2021
@noloader @EAddario
Fix CTR mode when using FileSource (GH weidai11#683, GH weidai11#1010)
07eef06 
We think this is another instance problem that surfaced under GH weidai11#683 when inString==outString. It violates aliasing rules and the compiler begins removing code.

The ultimate workaround was to add a member variable m_tempOutString as scratch space when inString==outString. We did not loose much in the way of perforamce for some reason. It looks like AES/CTR lost about 0.03-0.05 cpb.

When combined with the updated xorbuf from GH weidai11#1020, the net result was a speedup of 0.1-0.6 cpb. In fact, some ciphers like RC6, gained almost 5 cpb.
noloader added a commit that referenced this issue Feb 8, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
21221ac 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 8, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
39731b8 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 8, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
b41a598 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 8, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
1fc08f1 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 8, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
5ea06dc 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 11, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
75be2f2 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 11, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
dff0489 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
df014bb 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
0910749 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
db33299 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
223f967 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
eeda35a 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 13, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
7b7dd17 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
0fccd24 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
325a385 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
157b101 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
4ff80c3 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
e4dd7e0 
Also see GH #683, GH #1010, GH #1088, GH #1103
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
df1812c 
Also see GH #683, GH #1010, GH #1088, GH #1103
@noloader noloader mentioned this issue Feb 14, 2022
Merged
noloader added a commit that referenced this issue Feb 14, 2022
@noloader
Try fix ProcessData in CFB_CipherTemplate and AdditiveCipherTemplate
b54a41d 
This commit attempts to restore performance while taming the optimizer.

Also see GH #683, GH #1010, GH #1088, GH #1103.
@noloader
Copy link
Collaborator Author

I merged the changes into master last night. You should test master now.

@noloader noloader mentioned this issue Sep 29, 2023
Closed
noloader referenced this issue Sep 30, 2023
@noloader
Remove volatile cast gyrations in strciphr.cpp (GH #1231)
2e23f64 
It turns out we went down a rabbit hole when we added the volatile cast gyrations in an attempt to change the compiler behavior. We are seeing the same failures from AES, Rabbit, HIGHT, HC-128 and HC-256 with and without the gyrations.
We were able to work out the problems with Rabbit, HIGHT, HC-128 and HC-256. See GH #1231 and GH #1234.
We are also not able to successfully cut-in Cryptogams AES on ARMv7, so it is now disabled. See GH #1236.
Since the volatile casts were not a solution, we are backing it out along with associated comments.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@noloader