From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 26157 invoked by alias); 15 Jul 2009 04:34:46 -0000 Received: (qmail 26125 invoked by uid 22791); 15 Jul 2009 04:34:41 -0000 X-SWARE-Spam-Status: No, hits=-1.1 required=5.0 tests=AWL,BAYES_00,J_CHICKENPOX_83,SARE_MSGID_LONG40,SPF_PASS X-Spam-Check-By: sourceware.org Received: from wf-out-1314.google.com (HELO wf-out-1314.google.com) (209.85.200.172) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 15 Jul 2009 04:34:35 +0000 Received: by wf-out-1314.google.com with SMTP id 28so1112278wfc.26 for ; Tue, 14 Jul 2009 21:34:29 -0700 (PDT) MIME-Version: 1.0 Received: by 10.142.12.14 with SMTP id 14mr1931359wfl.85.1247632469699; Tue, 14 Jul 2009 21:34:29 -0700 (PDT) In-Reply-To: References: <4A5C453C.4090907@intefo.ch> Date: Wed, 15 Jul 2009 04:34:00 -0000 Message-ID: From: Mandeep Sandhu To: Grant Edwards Cc: ecos-discuss@sources.redhat.com Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact ecos-discuss-help@ecos.sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: ecos-discuss-owner@ecos.sourceware.org Subject: Re: [ECOS] Re: ecos memory footprint X-SW-Source: 2009-07/txt/msg00119.txt.bz2 On Tue, Jul 14, 2009 at 7:51 PM, Grant Edwards w= rote: > On 2009-07-14, Simon Kallweit wrote: >> Mandeep Sandhu schrieb: >> >>> Anybody knows if ECOS can run in 64KB mem (or lesser) with the >>> following features: >>> >>> * USB Device/slave support >>> * SPI Driver support >>> * Networking support (for running a small webserver) >>> >>> I've not yet narrowed down on my h/w yet...but Microchips >>> PIC32MX460F128H seems promising? But the sad part is that ecos >>> doesn't show this in their list of supported h/w. >>> >>> Have ppl run ecos on microcontrollers w/o external SDRAM (I'm >>> trying to conserve power by not having ext RAM)? >> >> This will be rather tough, but probably possible. You should use the >> lwIP networking stack to keep memory usage low. Also, try to use as few >> threads as possible, as this saves you valuable stack memory. > > I agree that it's probably possible, but IMO you're going to > spend a lot of time trying to make everything fit -- effort > that could be better spent on developing and testing your > application. =A0I'd recommend a serious look at using something > intended for a smaller footprint (XMK, FreeRTOS, uC/OS-II, etc.). Yes, I'm evaluating FreeRTOS as well for my requirements. I've just compiled their sample webserver app for the same Atmel platform. And that doesn't look too= good as well! :( ~/freeRTOS/FreeRTOS/Demo/lwIP_Demo_Rowley_ARM7$ arm-elf-size rtosdemo.elf text data bss dec hex filename 119096 2016 62736 183848 2ce28 rtosdemo.elf ~/freeRTOS/FreeRTOS/Demo/lwIP_Demo_Rowley_ARM7$ arm-elf-objdump -h rtosdemo= .elf rtosdemo.elf: file format elf32-littlearm Sections: Idx Name Size VMA LMA File off Algn 0 startup 00000044 00100000 00100000 00008000 2**2 CONTENTS, ALLOC, LOAD, READONLY, CODE 1 prog 0001d0f4 00100044 00100044 00008044 2**2 CONTENTS, ALLOC, LOAD, READONLY, CODE 2 .data 000007e0 00200000 0011d138 00028000 2**2 CONTENTS, ALLOC, LOAD, DATA 3 .bss 0000f510 002007e0 002007e0 000287e0 2**3 ALLOC 4 .debug_abbrev 00007274 00000000 00000000 000287e0 2**0 CONTENTS, READONLY, DEBUGGING 5 .debug_info 0002cb4d 00000000 00000000 0002fa54 2**0 CONTENTS, READONLY, DEBUGGING 6 .debug_line 00007e5c 00000000 00000000 0005c5a1 2**0 CONTENTS, READONLY, DEBUGGING 7 .debug_frame 00003400 00000000 00000000 00064400 2**2 CONTENTS, READONLY, DEBUGGING 8 .debug_pubnames 0000213a 00000000 00000000 00067800 2**0 CONTENTS, READONLY, DEBUGGING 9 .debug_aranges 00000ba8 00000000 00000000 00069940 2**3 CONTENTS, READONLY, DEBUGGING 10 .debug_str 00002410 00000000 00000000 0006a4e8 2**0 CONTENTS, READONLY, DEBUGGING 11 .comment 000005fa 00000000 00000000 0006c8f8 2**0 CONTENTS, READONLY 12 .debug_ranges 000001d8 00000000 00000000 0006cef2 2**0 CONTENTS, READONLY, DEBUGGING This does not give the exact figure (as I think the config here does not have the SPI driver built-in), but I'm using the results here to just get some rough comparative figure vis-a-vis ecos with roughly the same config. I think i'll have to "up" my RAM requirements to have at least 256KB RAM. Though I'm not sure if i'll be able to meet my price budget constraints. -mandeep > > -- > Grant Edwards =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 grante =A0 =A0 =A0 =A0 = =A0 =A0 Yow! I had pancake makeup > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0at =A0= =A0 =A0 =A0 =A0 =A0 =A0 for brunch! > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 visi.com > > > -- > Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos > and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss > > -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss