This is the mail archive of the newlib@sourceware.org mailing list for the newlib 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] Improve performance of sinf/cosf/sincosf

From: "Jon Beniston" <jon at beniston dot com>
To: <newlib at sourceware dot org>, "'Wilco Dijkstra'" <Wilco dot Dijkstra at arm dot com>
Cc: "'nd'" <nd at arm dot com>
Date: Tue, 19 Jun 2018 11:35:56 +0100
Subject: RE: [PATCH] Improve performance of sinf/cosf/sincosf
References: <DB5PR08MB103064AE7954988DAF0F7C7D837B0@DB5PR08MB1030.eurprd08.prod.outlook.com> <20180618122829.GE3036@calimero.vinschen.de> <DB5PR08MB1030042175C118155172DA4B83710@DB5PR08MB1030.eurprd08.prod.outlook.com> <20180619075022.GB11110@calimero.vinschen.de> <002301d407b8$61636db0$242a4910$@beniston.com>

>16-bit - ~200% worse and wrong results

This can be fixed in reduce_fast, by changing: 

-  int n = ((int)r + 0x800000) >> 24;
+  int n = ((int32_t)r + 0x800000) >> 24;

It's then ~400% slower, but gets the right answer.

Cheers,
Jon

Follow-Ups:
- Re: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Wilco Dijkstra
- Re: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Wilco Dijkstra

References:
- [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Wilco Dijkstra
- Re: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Corinna Vinschen
- Re: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Wilco Dijkstra
- Re: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Corinna Vinschen
- RE: [PATCH] Improve performance of sinf/cosf/sincosf
  - From: Jon Beniston

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