This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH] Add sinf and cosf traces

From: Carlos O'Donell <carlos at redhat dot com>
To: Wilco Dijkstra <Wilco dot Dijkstra at arm dot com>, 'GNU C Library' <libc-alpha at sourceware dot org>
Cc: nd <nd at arm dot com>
Date: Fri, 14 Dec 2018 14:44:10 -0500
Subject: Re: [PATCH] Add sinf and cosf traces
References: <DB5PR08MB1030DDA493BCB1B83398F03C83A10@DB5PR08MB1030.eurprd08.prod.outlook.com>

On 12/14/18 11:03 AM, Wilco Dijkstra wrote:
> Add traces for sinf and cosf with easy, hard and mixed cases. They were
> extracted from a trace of 1.4 million calls from a popular benchmark.
> 
> ChangeLog:
> 2018-12-14  Wilco Dijkstra  <wdijkstr@arm.com>
> 
>         * benchtests/cosf-inputs: Add reduced trace.
>         * benchtests/sinf-inputs: Add reduced trace.

One of the goals of the microbenchmark was to tie a benchmark or workload
dataset directly to the customer, user, or application that generated it.
That way if we regressed we could reach out to the stakeholder and discuss
the regression.

Is there any way we can expand the description of the workload you're proposing
so we can recreate it or reach out to the end user of the workload? Even if it's
just "ARM wants this workload to do better. Contact: Wilco Dijkstra <wdijkstr@arm.com>"

Does that make sense?

-- 
Cheers,
Carlos.

Follow-Ups:
- Re: [PATCH] Add sinf and cosf traces
  - From: Wilco Dijkstra

References:
- [PATCH] Add sinf and cosf traces
  - From: Wilco Dijkstra

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]