aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Document some other recent updateslloyd2008-09-111-2/+4
|
* Use getlogin before getpwuid for Windowslloyd2008-09-111-1/+1
|
* Load asm_amd64 only on requestlloyd2008-09-101-1/+1
|
* Add (incomplete) x86-64 mp_mulop variantlloyd2008-09-102-0/+95
|
* Note new 16x16 Comba operationslloyd2008-09-101-0/+1
|
* Increase Karatsuba cutoffs from 12 to 32 wordslloyd2008-09-091-2/+2
|
* Add 16x16->32 word Comba multiply and squarelloyd2008-09-093-176/+719
|
* Hoist load, since compiler may not be able to do so due to aliasinglloyd2008-09-091-4/+7
|
* Undef macros at end of headerslloyd2008-09-092-1/+11
|
* Add comment about bigint_simple_sqr just being an optimization to work ↵lloyd2008-09-091-1/+6
| | | | around lack of restricted pointers
* Use individual variables intead of an array inside the Salsa20 function,lloyd2008-09-091-37/+65
| | | | shows a 35% speedup on my Core2 with G++ vs previous version.
* Add implementation of Salsa20 stream cipherlloyd2008-09-096-0/+503
|
* In get_cipher, do not call cipher->set_iv unless the IV is non-emptylloyd2008-09-091-1/+4
|
* Fix --bench-typelloyd2008-09-091-10/+17
|
* Add a little more detail to RSA example. Reword some intro text.lloyd2008-09-081-29/+39
|
* Add --test to OptionParser listlloyd2008-09-081-1/+1
|
* Last checkin would crash, because the destructor for the rng would notlloyd2008-09-081-60/+55
| | | | | | | | run until after the block ended, which was after set_global_state(0) had run, causing crashes in the destructors that attempted to free memory. Restructure the argument handling, it is significantly easier to understand now. Only create the rng where it is needed.
* Run test_types every time prior to initialization, instead of just duringlloyd2008-09-081-58/+66
| | | | | | | the test suite. Instead of exiting, just carry on; the full test suite will check if the problem is real or if std::numeric_limits is just wonky.
* New (GCC-generated) code for bigint_monty_redc)lloyd2008-09-081-110/+252
|
* Update release noteslloyd2008-09-071-0/+3
|
* Make text easier to understandlloyd2008-09-071-2/+5
|
* Add the username and hostname to build.hlloyd2008-09-072-4/+7
|
* Bump to 1.7.11 (pre)lloyd2008-09-073-2/+4
|
* Reindentlloyd2008-09-071-172/+172
|
* Disable extensive testing during benchmark key generation: throws off profilerlloyd2008-09-071-0/+3
|
* Comment fixlloyd2008-09-071-1/+1
|
* Do not enable by defaultlloyd2008-09-071-1/+1
|
* asm_amd64/lloyd2008-09-074-15/+227
| | | | Check in current state of the x86-64 assembly
* Reduce how often DSA and DH keys are generated during benchmarkslloyd2008-09-071-2/+2
|
* Disable mp_mulop.S until updatedlloyd2008-09-071-3/+3
|
* Disable mp_mulop.S until updatedlloyd2008-09-071-2/+2
|
* Remove bigint_mul_add_words. It was only used now in two callers,lloyd2008-09-073-33/+45
| | | | | | | bigint_simple_mul and bigint_simple_sqr. Examining these functions made it clear inlining would be beneficial, so these two functions have been moved from an anonymous namespace into mp_mulop.cpp (to allow assembly versions).
* Fix call to asmlloyd2008-09-071-1/+1
|
* Rename alg_amd64 to asm_amd64lloyd2008-09-076-8/+117
| | | | | | Namespace exposed SHA-1 function with botan_ prefix. Add start of mp_monty for x86-64 assembly
* Inline BigInt::Rep::operator[], BigInt::sig_words, and BigInt::Rep::sig_wordslloyd2008-09-072-72/+46
|
* Rename alg_ia32 module to asm_ia32, which is more reflective of actuallloyd2008-09-0711-5/+5
| | | | | | | | contents. Rename sha1core.S to sha1_asm.S, and rename the extern "C" function from sha160_core to botan_sha160_asm_ia32 so it is clear what it is and so it is reasonably namespaced.
* Inline BigInt::operator[]lloyd2008-09-072-18/+2
|
* Typo fixlloyd2008-09-071-1/+1
|
* Combine redundant definitions of blocks variablelloyd2008-09-071-7/+6
|
* For non-release, use strict GCC warn flagslloyd2008-09-071-2/+2
|
* Tweak timer outputlloyd2008-09-071-2/+2
|
* Inline BigInt::is_zerolloyd2008-09-072-13/+9
|
* Inline similarly in karatsuba_mullloyd2008-09-071-11/+32
|
* In karatsuba_square, inline the sequencelloyd2008-09-071-3/+25
| | | | | | | | word carry = bigint_add3_nc(workspace+N, z0, N, z1, N); carry += bigint_add2_nc(z + N2, N, workspace + N, N); bigint_add2_nc(z + N + N2, N2, &carry, 1); It turns out quite a bit can be shared among these function calls
* Comment cleanuplloyd2008-09-071-4/+2
|
* Rewrite without gotoslloyd2008-09-071-135/+11
|
* Inline bigint_sub2 into bigint_monty_redclloyd2008-09-071-5/+20
|
* Inline bigint_cmp in bigint_monty_redc (using goto, the horror; I'm basicallylloyd2008-09-071-27/+10
| | | | | | | prototyping and testing the x86-64 assembly version in C) According to most profiles, bigint_monty_redc alone is responsible for 30%-50% of RSA, DSA, and DH benchmarks. So it seems worth tinkering with a bit.
* Move bigint_monty_redc to its own file to make asm implementations easierlloyd2008-09-072-29/+205
|
* Use i instead of j for iterator varlloyd2008-09-071-4/+4
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-13 19:10:13 +0000