Hello Paul,
On Mon, May 07, 2018 at 12:36:43AM -0700, Paul Eggert wrote:
> Assaf Gordon wrote:
[...]
> > It won't matter to a single user on a desktop, but imagine
> > a shared server at a university, when all of a sudden a sysadmin
> > might see a very very high %SYS load - with almost no way to
Hello,
On Tue, May 08, 2018 at 02:20:07AM +0200, Bruno Haible wrote:
> The following approaches come to mind:
[...]
> * Use the kernel-provided meta-info about the algorithms to decide whether
> to use the kernel API.
>
> In detail: Read /proc/crypto at run time. It consists of records
Hello Bruno and all,
On Tue, May 08, 2018 at 01:10:52AM +0200, Bruno Haible wrote:
> Let me try to summarize.
Thanks for writing this - informative and helpful.
> * We need to consider each of the algorithms md5, sha1 sha256 separately,
> because each algorithm has a different performanc
Matteo Croce wrote:
> Obviously it works only if you always use the same algo and with a
> single thread
Why only in a single thread? Can't you do
socket(AF_ALG, SOCK_SEQPACKET, 0) = 3
bind(3, {sa_family=AF_ALG, sa_data="hash\0... sha1\0..."}, 88) = 0
accept(3, NULL, NULL) = 4
On 05/07/2018 05:20 PM, Bruno Haible wrote:
* Use the kernel-provided meta-info about the algorithms to decide whether
to use the kernel API.
...
Drawback: Does not work if /proc is not mounted.
I would favour the third approach.
I agree. With the Linux kernel these days, mountin
What can we do to get the speedups but avoid the slowdowns, in the two
hairy cases
- the afalg_buffer case,
- the afalg_stream case with non-regular files?
The following approaches come to mind:
* A "tuning" framework like the one from GMP. This is a set of benchmark
programs that the d
Hi all,
Thanks for your benchmarking help and explanations.
Let me try to summarize.
* We need to consider each of the algorithms md5, sha1 sha256 separately,
because each algorithm has a different performance characteristic [1].
This is due to the following factors:
- Some non-Inte
John W. Eaton wrote:
> I also have the vague recollection of RMS telling people to write
>
>Return a list containing a and b.
>
> instead of
>
>The function foo returns ...
>
> or even
>
>Returns a list ...
He apparently doesn't do this any more. Recently RMS reviewed the
document
On 05/06/2018 10:33 AM, Jim Meyering wrote:
Here's the relevant section: (from
https://www.gnu.org/prep/standards/html_node/GNU-Manuals.html)
Whenever possible, please stick to the active voice, avoiding the
passive, and use the present tense, not the future teste. For
instance, write “The funct
Hi,
the test-poll test seems to have a very high failure rate on a
multi-core FreeBSD guest:
$ for i in $(seq 1 1); do
> ./tests/test-poll >/dev/null 2>&1
> echo $?
> done >test-poll.rc
$ grep -c '.*' test-poll.rc
1
$ grep -c '^0$' test-poll.rc
5
$ grep -c '^1$' test
On Mon, May 7, 2018 at 11:55 AM, Matteo Croce wrote:
> On Mon, May 7, 2018 at 4:07 AM, Paul Eggert wrote:
>> Bruno Haible wrote:
>>>
>>> Oops, I goofed with "git diff". Here's the correct patch to test.
>>
>>
>> I tried those bench-md5 benchmarks on two platforms, with somewhat more
>> disappoint
On Mon, May 7, 2018 at 4:07 AM, Paul Eggert wrote:
> Bruno Haible wrote:
>>
>> Oops, I goofed with "git diff". Here's the correct patch to test.
>
>
> I tried those bench-md5 benchmarks on two platforms, with somewhat more
> disappointing results.
>
> I observed a real-time slowdown ranging from 1
Assaf Gordon wrote:
I think that for the same reason that --with-openssl defaults to 'no'
and not to 'auto' applies here.
This is a good point; it's easier to understand and explain if the two options
have similar defaults.
I'm not sure there is significant improvement of using af_alg
comp
13 matches
Mail list logo
