tinymembench | Simple benchmark for memory throughput and latency | Performance Testing library

by ssvb C Version: v0.4 License: Non-SPDX

X-Ray Key Features Code Snippets Community Discussions(2)Vulnerabilities Install Support

kandi X-RAY | tinymembench Summary

tinymembench is a C library typically used in Testing, Performance Testing applications. tinymembench has no bugs, it has no vulnerabilities and it has low support. However tinymembench has a Non-SPDX License. You can download it from GitHub.

Simple benchmark for memory throughput and latency

Support

Quality

Security

License

Reuse

Support

tinymembench has a low active ecosystem.

It has 230 star(s) with 69 fork(s). There are 29 watchers for this library.

It had no major release in the last 12 months.

There are 9 open issues and 5 have been closed. On average issues are closed in 299 days. There are 9 open pull requests and 0 closed requests.

It has a neutral sentiment in the developer community.

The latest version of tinymembench is v0.4

Quality

tinymembench has no bugs reported.

Security

tinymembench has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.

License

tinymembench has a Non-SPDX License.

Non-SPDX licenses can be open source with a non SPDX compliant license, or non open source licenses, and you need to review them closely before use.

Reuse

tinymembench releases are available to install and integrate.

Top functions reviewed by kandi - BETA

kandi's functional review helps you automatically verify the functionalities of the libraries and avoid rework.
Currently covering the most popular Java, JavaScript and Python libraries. See a Sample of tinymembench

Get all kandi verified functions for this library.

tinymembench Key Features

No Key Features are available at this moment for tinymembench.

tinymembench Examples and Code Snippets

No Code Snippets are available at this moment for tinymembench.

Community Discussions

Trending Discussions on tinymembench

clang assembler error: ambiguous operand size for instruction

Enhanced REP MOVSB for memcpy

QUESTION

clang assembler error: ambiguous operand size for instruction

Asked 2019-May-02 at 03:23

I'm trying build tinymembench with clang and am hitting errors on the assembly code. I can fix the easy ones (remove .func/.endfunc's) but the 'ambiguous operand size for instruction' errors on add and sub surpass my minimal assembly skills. I posted an issue on the repo but it's possible it's no longer being maintained.

Using this source file (removes the .func/.endfunc) I get errors like:

$ clang-8 x86-sse2.S /tmp/x86-sse2-dbaa71.s:86:9: error: ambiguous operand size for instruction 'add' add SRC, 64 ^~~ /tmp/x86-sse2-dbaa71.s:87:9: error: ambiguous operand size for instruction 'add' add DST, 64 ^~~ /tmp/x86-sse2-dbaa71.s:88:9: error: ambiguous operand size for instruction 'sub' sub SIZE, 64 ^~~~ ...

I looked at this answer which looks similar, but I wasn't able to translate it into an answer for these instructions.

...

ANSWER

Answered 2019-May-02 at 03:23

I can fix the easy ones (remove .func/.endfunc's)

The .func macro includes a .set SRC, rsi which defines registers according to the appropriate calling convention (x86-64 System V, Windows x64, or 32-bit with stack args).

Removing it leaves just an undefined SRC symbol which of course is treated as a memory operand. (And add mem, imm doesn't have either operand implying an operand-size, so it's ambiguous.)

Your "fix" introduced this bug.

Use clang -no-integrated-as to use the system assembler instead of clang's built-in assembler. As expected, that builds https://github.com/letrout/tinymembench/blob/master/x86-sse2.S just fine on my Linux desktop with clang7.0.1. (And system assembler = GNU Binutils as 2.31.1)

Source https://stackoverflow.com/questions/55945153

QUESTION

Enhanced REP MOVSB for memcpy

Asked 2019-Jan-09 at 08:55

I would like to use enhanced REP MOVSB (ERMSB) to get a high bandwidth for a custom memcpy.

ERMSB was introduced with the Ivy Bridge microarchitecture. See the section "Enhanced REP MOVSB and STOSB operation (ERMSB)" in the Intel optimization manual if you don't know what ERMSB is.

The only way I know to do this directly is with inline assembly. I got the following function from https://groups.google.com/forum/#!topic/gnu.gcc.help/-Bmlm_EG_fE

...

ANSWER

Answered 2017-Apr-11 at 10:57

There are far more efficient ways to move data. These days, the implementation of memcpy will generate architecture specific code from the compiler that is optimized based upon the memory alignment of the data and other factors. This allows better use of non-temporal cache instructions and XMM and other registers in the x86 world.

When you hard-code rep movsb prevents this use of intrinsics.

Therefore, for something like a memcpy, unless you are writing something that will be tied to a very specific piece of hardware and unless you are going to take the time to write a highly optimized memcpy function in assembly (or using C level intrinsics), you are far better off allowing the compiler to figure it out for you.

Source https://stackoverflow.com/questions/43343231

Community Discussions, Code Snippets contain sources that include Stack Exchange Network

Vulnerabilities

No vulnerabilities reported

Install tinymembench

You can download it from GitHub.

Support

For any new features, suggestions and bugs create an issue on GitHub. If you have any questions check and ask questions on community page Stack Overflow .

Find more information at: