From lkcl at lkcl.net Tue Dec 1 16:05:31 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 1 Dec 2015 16:05:31 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/ it's by no means perfect, but i have at least data on-screen. i suspect EM interference is causing the corruption, so another revision of PCB1 will need the LVDS IC layout to be reworked. this is however huge progress. the only 2 parts of PCB1 that need to be checked and debugged are the MicroSD card slot and the 1W speaker/headphone amplifier circuits. the CM108AH USB Audio IC has already been confirmed as operational on the USB bus, as has the GL850G and the 3 USB ports. l. From paul at boddie.org.uk Tue Dec 1 17:25:39 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Tue, 1 Dec 2015 18:25:39 +0100 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: References: Message-ID: <201512011825.39856.paul@boddie.org.uk> On Tuesday 1. December 2015 17.05.31 Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/community_ideas/laptop_15in/news/ > > it's by no means perfect, but i have at least data on-screen. i > suspect EM interference is causing the corruption, so another revision > of PCB1 will need the LVDS IC layout to be reworked. this is however > huge progress. the only 2 parts of PCB1 that need to be checked and > debugged are the MicroSD card slot and the 1W speaker/headphone > amplifier circuits. the CM108AH USB Audio IC has already been > confirmed as operational on the USB bus, as has the GL850G and the 3 > USB ports. Nice work on getting this far, despite the "gothic" font. ;-) Now if only the established players would see sense and support this effort instead of dumping excess stock SoCs into newsagents on the covers of magazines and causing scuffles in the aisles. :-/ OK, well, maybe not that established player, but at least the likes of Adafruit and Sparkfun... Paul From lkcl at lkcl.net Tue Dec 1 17:38:26 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 1 Dec 2015 17:38:26 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: <201512011825.39856.paul@boddie.org.uk> References: <201512011825.39856.paul@boddie.org.uk> Message-ID: On Tue, Dec 1, 2015 at 5:25 PM, Paul Boddie wrote: > On Tuesday 1. December 2015 17.05.31 Luke Kenneth Casson Leighton wrote: >> http://rhombus-tech.net/community_ideas/laptop_15in/news/ >> >> it's by no means perfect, but i have at least data on-screen. i >> suspect EM interference is causing the corruption, so another revision >> of PCB1 will need the LVDS IC layout to be reworked. this is however >> huge progress. the only 2 parts of PCB1 that need to be checked and >> debugged are the MicroSD card slot and the 1W speaker/headphone >> amplifier circuits. the CM108AH USB Audio IC has already been >> confirmed as operational on the USB bus, as has the GL850G and the 3 >> USB ports. > > Nice work on getting this far, despite the "gothic" font. ;-) it gets really weird when doing echo "hello" > /dev/tty1 - the on-screen display is rather... garbled. i'll try a 3rd CPU Card later, and/or perhaps set up dual-output (via HDMI) to check that it's not the framebuffer that's corrupted. > Now if only the established players would see sense and support this effort > instead of dumping excess stock SoCs into newsagents on the covers of > magazines and causing scuffles in the aisles. :-/ yehhh you saw the report a few days ago by Make magazine, which showed that the cost of that $5 board is considerably higher: "Solving this lack of connectivity makes the cost of owning a Pi Zero increase greatly. It’s not just that you need a USB Wi-Fi dongle, you’d need a powered USB hub to have a Wi-Fi dongle, keyboard, and mouse (it only has one USB slot for peripherals; the other is for power)." http://makezine.com/2015/11/28/chip-vs-pi-zero/ further down, it shows that the only thing you need to operate the C.H.I.P is a $1 USB cable, to provide power. the alternative board - which is "only" $5 - requires a whopping *$19* of additional components in order to make it actually useful. l. From lkcl at lkcl.net Tue Dec 1 18:43:11 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 1 Dec 2015 18:43:11 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: References: Message-ID: On Tue, Dec 1, 2015 at 4:05 PM, Luke Kenneth Casson Leighton wrote: > ... the only 2 parts of PCB1 that need to be checked and > debugged are the MicroSD card slot confirmed working. the GPIO pin of the T-FLASH slot had not been soldered, despite having 260 Centigrade temperatures applied from an IR heat lamp placed 2cm above it for around 4 minutes. soldering iron took care of that, and the micro-sd card slot on PCB1 is operational. that means that the laptop can boot directly from external micro-sd, now, when using an A20 CPU Card. reason: space is so tight on the 6-layer 43x78mm EOMA68-A20 PCB that it was impossible to route SDC0 over to the *opposite* side of the PCB whilst at the same time crossing SDC3 in the reverse direction, as the A20 and the DDR3 RAM take up the centre of the board, leaving very little space for routing. there are only 3 signal layers (TOP, BOTTOM, layer3). just the TDA2822 amplifier circuits and the microphone to verify as operational, and to track down why the LCD is wobbly, and PCB1 is done. l. From nico at nicorikken.eu Tue Dec 1 21:31:17 2015 From: nico at nicorikken.eu (Nico Rikken) Date: Tue, 01 Dec 2015 22:31:17 +0100 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: References: Message-ID: <1449005477.29833.21.camel@nicorikken.eu> Great to see further progress. Keep it up! At the Dutch T-Dose conference had a couple of visitors at the FSFE-booth mentioning how great it would be if finally a totally free laptop would become available. Have referred them to this project because this seems the only sensible approach to reach this ultimate end goal, and because it shows just how difficult it is to achieve this goal in a voluntary effort. Getting a fully free laptop is certainly easier set then done. Furthermore I couldn't help but notice that one of the most restricted laptops in the world is being used to create the most free laptop in the world. Regards, Nico (NL) On di, 2015-12-01 at 16:05 +0000, Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/community_ideas/laptop_15in/news/ > > it's by no means perfect, but i have at least data on-screen. i > suspect EM interference is causing the corruption, so another revision > of PCB1 will need the LVDS IC layout to be reworked. this is however > huge progress. the only 2 parts of PCB1 that need to be checked and > debugged are the MicroSD card slot and the 1W speaker/headphone > amplifier circuits. the CM108AH USB Audio IC has already been > confirmed as operational on the USB bus, as has the GL850G and the 3 > USB ports. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 213 bytes Desc: This is a digitally signed message part URL: From lkcl at lkcl.net Tue Dec 1 22:10:13 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 1 Dec 2015 22:10:13 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: <1449005477.29833.21.camel@nicorikken.eu> References: <1449005477.29833.21.camel@nicorikken.eu> Message-ID: On Tue, Dec 1, 2015 at 9:31 PM, Nico Rikken wrote: > Great to see further progress. Keep it up! :) > At the Dutch T-Dose conference had a couple of visitors at the > FSFE-booth mentioning how great it would be if finally a totally free > laptop would become available. Have referred them to this project > because this seems the only sensible approach to reach this ultimate end > goal, and because it shows just how difficult it is to achieve this goal > in a voluntary effort. in a word... yes. i had to make some... interesting design decisions shall we say - going for example with a custom 3D-printed case instead of attempting to work with china-based factories [who will not respond to enquiries for anything less than 10k orders, because they know the cost of injection-mold casts]. and that took 6 *MONTHS* to complete the [15] parts needed, which is just a staggering amount of time - waay more than i was expecting. keeping it simple - going with USB2 not USB3. cutting out all hard drives and using USB or MicroSD storage. finding a low-power 15.6in screen (sub 5 watt). keeping the power consumption to under 15 watts so that a single-cell battery charger IC could be used instead of needing multiple cells (which wouldn't fit into the casework design anyway)... a whole boat-load of details all of which basically steer things in ways that you might not have considered at the outset... > Getting a fully free laptop is certainly easier > set then done. yeah. there's quite a few efforts popped up in the past year, including http://www.powerpc-notebook.org. they're going to run into the same challenges, and i hope that they have the good sense to read the reports of the experiences i've encountered along the way, just like i read the experiences of the openpandora, the kosagi laptop and other projects and learned from those. > Furthermore I couldn't help but notice that one of the most restricted > laptops in the world is being used to create the most free laptop in the > world. *sigh* i knowww... it's rather unfortunate that the higher perceived quality of the proprietary software means you end up with a highly profit-rich company that can afford to purchase the best components and make the highest quality hardware products. first thing i did though was blow away the OS, install rEFInd, and boot up debian. i'm not interested - at all - in the proprietary OS, i want a hardware product with an exceptionally high resolution screen, light weight, good build quality, and high specs all round for the money.... and now that i've got one, i won't replace it for 5-8 years. at the time i _did_ try to buy an IBM laptop... their web site failed, at the time, however, preventing and prohibiting me from giving them any money! later on i'll be able to tackle creating high-end hardware - not entirely sure about how to go about that, yet. have to see what SoCs are available in 5-8 years time, as well as what display output standards become commonplace. i can say that i am most certainly *NOT* going to be buying a smartphone. ever. i bought *nine* HTC hand-helds back around 2002 to 2005, i was one of the early reverse-engineers working on removing wince from HTC phones and replacing it with openembedded-built GPE/Familiar based on Angstrom Linux. so i've made the decision: i'm not going to own another smartphone - ever - unless i've made it myself. laptop/workstation however.... yyeah, i have to have something otherwise i can't get anything done at all! l. From joem at martindale-electric.co.uk Wed Dec 2 14:49:25 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 2 Dec 2015 14:49:25 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: References: Message-ID: <1449067836.21913.2.camel@jm-desktop> On Tue, 2015-12-01 at 16:05 +0000, Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/community_ideas/laptop_15in/news/ > > it's by no means perfect, but i have at least data on-screen. i > suspect EM interference is causing the corruption Looks to me like a timing signals error from the CPU with the LCD driver chip trying to do its best to keep sync. Need to check with a scope against the specs of the LCD and what the CPU is generating. From lkcl at lkcl.net Wed Dec 2 22:16:14 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 2 Dec 2015 22:16:14 +0000 Subject: [Arm-netbook] LTC4155 design flaw Message-ID: i've got an enquiry out with Linear, however it looks like there's a serious design flaw with the LTC4155 - not one that's safety-related but one that makes it functionally completely useless. the clue is in looking at page 15 of the datasheet - a diagram which shows the internal operation. in the bottom left corner is a power-selection arrangement, where there are two AND gates. note that they're not symmetrically arranged. the difference between the two is that WALLSNS output is connected to the *input* (inverted) of the USBGT enabler AND gate. what that means is that it is ABSOLUTELY CAST-IRON GUARANTEED - i.e. hard-wired in an inviolate way - that if the 5VDC input is plugged into a wall charger, USB OTG Host Power is ABSOLUTELY 100% WITHOUT FAIL ABSOLUTE guaranteed to be terminated without warning. which: (a) violates the USB-OTG Power Specification (b) is in direct contravention of the statement on page 1 of the LTC4155 datasheet which *clearly* states that the LTC4155 is compliant with USB-OTG Host Power Mode (c) is in clear violation of various laws like "Trade Descriptions Act" of the UK due to the qualifier statement in paragraph 2 which reads "....without needing additional components". all of which is a frickin nuisance, as i now have to go find alternative dual-supply OTG-compliant [boost-mode] battery charger ICs. i've found two already - bq24192i ($7 in low volume and it would require a 2A 5V step-up converter IC as well....], or the MAX77818 which is a 0.4mm pitch BGA - perfect because it even has battery status monitoring built-in, but the 0.4mm pitch BGA _wow_ that's gonna be tough for small china pcb assembly factories to deal with when they're used to QFNs and QFPs. argh! i'll let people know how this goes: i'm expecting to hear from Linear, they *might* confirm that the I2C "REQUEST_OTG" mode does in fact honour the USB OTG Host Power mode properly. if they do, then all is well - otherwise i will need to do a redesign of PCB3. l. From lkcl at lkcl.net Fri Dec 4 12:39:52 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Fri, 4 Dec 2015 12:39:52 +0000 Subject: [Arm-netbook] 15in laptop LCD up on PCB1 In-Reply-To: <1449067836.21913.2.camel@jm-desktop> References: <1449067836.21913.2.camel@jm-desktop> Message-ID: On Wed, Dec 2, 2015 at 2:49 PM, joem wrote: > On Tue, 2015-12-01 at 16:05 +0000, Luke Kenneth Casson Leighton wrote: >> http://rhombus-tech.net/community_ideas/laptop_15in/news/ >> >> it's by no means perfect, but i have at least data on-screen. i >> suspect EM interference is causing the corruption > > Looks to me like a timing signals error from the CPU > with the LCD driver chip trying to do its best to keep sync. it's as if the HSYNC - no matter what parameters are used - is entirely missing. so the LCD takes the very first bits of each line and uses that as "sync". > Need to check with a scope against the specs of the LCD and > what the CPU is generating. yehh my scope's only a max of 75mhz (maplin's digital scope) and i can't ramp things down in order to check, because the 3.4 kernel sunxi lcd code changes the clk_div parameter when the dclk rate is set lower, such that entirely different code is activated. tricky... l. From maillist_arm-netbook at aross.me Fri Dec 4 14:26:24 2015 From: maillist_arm-netbook at aross.me (Alexander Stephen Thomas Ross) Date: Fri, 04 Dec 2015 14:26:24 +0000 Subject: [Arm-netbook] Free Soft & Hardware: ControLeo2 Reflow Oven Message-ID: <5661A290.70607@aross.me> came across this thought it might be on interest maybe. http://www.whizoo.com/ From lkcl at lkcl.net Fri Dec 4 15:31:06 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Fri, 4 Dec 2015 15:31:06 +0000 Subject: [Arm-netbook] Free Soft & Hardware: ControLeo2 Reflow Oven In-Reply-To: <5661A290.70607@aross.me> References: <5661A290.70607@aross.me> Message-ID: On Fri, Dec 4, 2015 at 2:26 PM, Alexander Stephen Thomas Ross wrote: > came across this thought it might be on interest maybe. > http://www.whizoo.com/ not bad. good explanations. thanks alex. From phil at hands.com Sun Dec 6 18:53:51 2015 From: phil at hands.com (Philip Hands) Date: Sun, 06 Dec 2015 19:53:51 +0100 Subject: [Arm-netbook] Paypal sabotage crowdfunding Message-ID: <874mfvs91c.fsf@whist.hands.com> Hi Folks, Since this list has people that are reasonably likely to come up with crowdfunding projects, I thought I'd point at this sorry state of affairs, caused by Paypal: http://neo900.org/news/paypal-trouble-delays-project (the link to http://garethhayes.net/paypal-warning/ is hardly encouraging either) Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] HANDS.COM Ltd. |-| http://www.hands.com/ http://ftp.uk.debian.org/ |(| Hugo-Klemm-Strasse 34, 21075 Hamburg, GERMANY -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 818 bytes Desc: not available URL: From addw at phcomp.co.uk Sun Dec 6 19:55:59 2015 From: addw at phcomp.co.uk (Alain Williams) Date: Sun, 6 Dec 2015 19:55:59 +0000 Subject: [Arm-netbook] Paypal sabotage crowdfunding In-Reply-To: <874mfvs91c.fsf@whist.hands.com> References: <874mfvs91c.fsf@whist.hands.com> Message-ID: <20151206195559.GV6396@phcomp.co.uk> On Sun, Dec 06, 2015 at 07:53:51PM +0100, Phil Hands wrote: > Hi Folks, > > Since this list has people that are reasonably likely to come up with > crowdfunding projects, I thought I'd point at this sorry state of > affairs, caused by Paypal: > > http://neo900.org/news/paypal-trouble-delays-project It advertises an IPv6 address - but does not serve web pages over IPv6 - use IPv4. -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php #include From paul at boddie.org.uk Sun Dec 6 20:29:13 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Sun, 6 Dec 2015 21:29:13 +0100 Subject: [Arm-netbook] Paypal sabotage crowdfunding In-Reply-To: <874mfvs91c.fsf@whist.hands.com> References: <874mfvs91c.fsf@whist.hands.com> Message-ID: <201512062129.13862.paul@boddie.org.uk> On Sunday 6. December 2015 19.53.51 Philip Hands wrote: > Hi Folks, > > Since this list has people that are reasonably likely to come up with > crowdfunding projects, I thought I'd point at this sorry state of > affairs, caused by Paypal: > > http://neo900.org/news/paypal-trouble-delays-project > > (the link to http://garethhayes.net/paypal-warning/ is hardly > encouraging either) I have been somewhat aware of this situation - the Neo900 one, not the Australian one - and I can only say that it is most unfortunate. PayPal has a poor-enough reputation that I refuse to use their service at all: I've seen people struggle with getting an account enabled to be able to pay people fairly urgently, seeing them go round and round in the stupid card- sampling verification loop to no avail; it used to be possible to pay random merchants by card via PayPal, but now they appear to want you to create an account to do so, and even with an account in the bag, a simple transaction becomes something that Kafka wrote up. I've even read reports of jurisdiction- surfing (moving Swiss accounts to Singapore, if I remember correctly), and it wouldn't surprise me if their European residence is in the most leniently- regulated venue that can still offer services to people in the EU. The last time I was confronted with the option to pay by PayPal, I had a nice chat with the merchant (Fritzing Fab) and we agreed on a bank transfer instead, to everybody's satisfaction. Merchants/vendors feel that they need to offer convenient ways of paying for things. Sadly, there are people for whom PayPal has caused no problems who see fit on insisting on using the service for payment. As Werner writes... "However, after many supporters asked for a means to pay by credit card, due to wire transfers being difficult and expensive for them, we looked for a way to accommodate these wishes." Now, it may be the case that some people would otherwise have to use archaic banking systems that don't offer decently-priced transfers to Europe (or within Europe if we're thinking of UK banks). However, convenience may have been king for some people for whom PayPal is easiest but where other options would still have been easy. I would encourage people to reconsider their preference for PayPal even if it causes them slightly more inconvenience and slightly more cost, because the result (as we see here) is neither more convenient nor less costly. Not that anyone really needs to be told this (or maybe thinks that they need to be told this), but anyway. Paul From addw at phcomp.co.uk Sun Dec 6 20:40:45 2015 From: addw at phcomp.co.uk (Alain Williams) Date: Sun, 6 Dec 2015 20:40:45 +0000 Subject: [Arm-netbook] Paypal sabotage crowdfunding In-Reply-To: <201512062129.13862.paul@boddie.org.uk> References: <874mfvs91c.fsf@whist.hands.com> <201512062129.13862.paul@boddie.org.uk> Message-ID: <20151206204045.GB6396@phcomp.co.uk> On Sun, Dec 06, 2015 at 09:29:13PM +0100, Paul Boddie wrote: > PayPal has a poor-enough reputation that I refuse to use their service at all: > I've seen people struggle with getting an account enabled to be able to pay > people fairly urgently, seeing them go round and round in the stupid card- > sampling verification loop to no avail; it used to be possible to pay random > merchants by card via PayPal, but now they appear to want you to create an > account to do so, and even with an account in the bag, a simple transaction I had a big fight with them a few years ago. I bought something (theatre tickets IIRC, no other payment option) via paypal, ticked the box saying that I did NOT want an account; then received an email welcoming me to said account ... long phone calls to Eire and Luxembourg finally got it closed and they then banned me from ever making any other payment via them - although I did make one a year later - and they did not open an account on that occasion. -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php #include From paul at boddie.org.uk Sun Dec 6 21:21:38 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Sun, 6 Dec 2015 22:21:38 +0100 Subject: [Arm-netbook] Paypal sabotage crowdfunding In-Reply-To: <20151206204045.GB6396@phcomp.co.uk> References: <874mfvs91c.fsf@whist.hands.com> <201512062129.13862.paul@boddie.org.uk> <20151206204045.GB6396@phcomp.co.uk> Message-ID: <201512062221.38273.paul@boddie.org.uk> On Sunday 6. December 2015 21.40.45 Alain Williams wrote: > On Sun, Dec 06, 2015 at 09:29:13PM +0100, Paul Boddie wrote: > > PayPal has a poor-enough reputation that I refuse to use their service at > > all: I've seen people struggle with getting an account enabled to be > > able to pay people fairly urgently, seeing them go round and round in > > the stupid card- sampling verification loop to no avail; it used to be > > possible to pay random merchants by card via PayPal, but now they appear > > to want you to create an account to do so, and even with an account in > > the bag, a simple transaction > > I had a big fight with them a few years ago. I bought something (theatre > tickets IIRC, no other payment option) via paypal, ticked the box saying > that I did NOT want an account; then received an email welcoming me to > said account ... long phone calls to Eire and Luxembourg finally got it > closed and they then banned me from ever making any other payment via them > - although I did make one a year later - and they did not open an account > on that occasion. That reminds me of their one-time corporate bedfellows, eBay, and the matter of me closing my account with them after some data spill or other. After having been assured that my account was closed, I would still get "Welcome Paul!" on the site, and still they insisted that the account really was closed. Now it could be possible that they stash salutations in cookies that then activate when people go to the site, although I was quite sure I deleted all my eBay cookies, but I rather suspect that they just keep all the data and pretend that they're not trying to collect every last detail and associate it with my identity (also via a bunch of cookies issued by random tracking and advertising sites operating on their behalf). And I only had an account with them in the first place because a merchant from whom I wanted to buy something seemingly had to do business via eBay and presumably through no other channel. It's all a bit like "friends don't make friends use Facebook" but with buying and selling things and involving real money. Paul From lkcl at lkcl.net Thu Dec 10 04:20:31 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 04:20:31 +0000 Subject: [Arm-netbook] A64 SDK Message-ID: small favour to ask of people (anyone): search "lichee_A64_A5.1_V1.0.tar.gz", this should come up with a link on baidu. endeavour to download this file, then check whether it's corrupted or not. thanks in advance, l. From paul at boddie.org.uk Thu Dec 10 11:05:13 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Thu, 10 Dec 2015 12:05:13 +0100 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: Message-ID: <201512101205.15146.paul@boddie.org.uk> On Thursday 10. December 2015 05.20.31 Luke Kenneth Casson Leighton wrote: > small favour to ask of people (anyone): > search "lichee_A64_A5.1_V1.0.tar.gz", this should come up with a link > on baidu. endeavour to download this file, then check whether it's > corrupted or not. I can't get past the Baidu login page which appears if you click on the download button. Paul From lkcl at lkcl.net Thu Dec 10 13:08:08 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 13:08:08 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: <201512101205.15146.paul@boddie.org.uk> References: <201512101205.15146.paul@boddie.org.uk> Message-ID: On Thu, Dec 10, 2015 at 11:05 AM, Paul Boddie wrote: > On Thursday 10. December 2015 05.20.31 Luke Kenneth Casson Leighton wrote: >> small favour to ask of people (anyone): >> search "lichee_A64_A5.1_V1.0.tar.gz", this should come up with a link >> on baidu. endeavour to download this file, then check whether it's >> corrupted or not. > > I can't get past the Baidu login page which appears if you click on the > download button. ah. from what i recall, that can be solved by creating a qq account. there does exist a baidu downloader for linux.... or something called "Baidu Exporter": http://www.cnx-software.com/2015/01/14/downloading-files-on-baidu-or-via-http-bittorrent-or-metalink-in-linux-with-baiduexporter-aria2-and-yaaw/ i have received a copy of the SDK (via a different route) - the problem is that it took a *month* of retries to extract it out of china, and every single copy has file corruption. the last corrupted version was only corrupted right at the end, meaning that i have alllmost everything needed... ... what i'd like to be able to confirm is, before getting back to allwinner, whether it's *their* file that's corrupted or whether the process of extraction of the file from china is the likely cause. therefore if someone else can obtain the file and verify its integrity, that would be very useful. l. From cnxsoft at cnx-software.com Thu Dec 10 13:52:23 2015 From: cnxsoft at cnx-software.com (Jean-Luc Aufranc) Date: Thu, 10 Dec 2015 20:52:23 +0700 Subject: [Arm-netbook] A64 SDK In-Reply-To: <201512101205.15146.paul@boddie.org.uk> References: <201512101205.15146.paul@boddie.org.uk> Message-ID: <56698397.9080807@cnx-software.com> I got the same issue with the mobile website (as accessed via google on my PC). If I convert the URL to the desktop version, I can start the download: http://pan.baidu.com/share/link?shareid=2502121798&uk=1778654534 On 12/10/2015 06:05 PM, Paul Boddie wrote: > On Thursday 10. December 2015 05.20.31 Luke Kenneth Casson Leighton wrote: >> small favour to ask of people (anyone): >> search "lichee_A64_A5.1_V1.0.tar.gz", this should come up with a link >> on baidu. endeavour to download this file, then check whether it's >> corrupted or not. > I can't get past the Baidu login page which appears if you click on the > download button. > > Paul > > _______________________________________________ > arm-netbook mailing listarm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments toarm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Thu Dec 10 15:16:27 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 15:16:27 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: <56698397.9080807@cnx-software.com> References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: On Thu, Dec 10, 2015 at 1:52 PM, Jean-Luc Aufranc wrote: > I got the same issue with the mobile website (as accessed via google on my > PC). > If I convert the URL to the desktop version, I can start the download: > > http://pan.baidu.com/share/link?shareid=2502121798&uk=1778654534 fantastic. if you are able to get the file in one go and it's complete, please do let us know (md5sum, file size). if it *doesn't* download in one go then you will likely find that a continuation results in corruption, in which case it would be good if you could try that "Baidu Exporter". l. From lkcl at lkcl.net Thu Dec 10 16:37:36 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 16:37:36 +0000 Subject: [Arm-netbook] HX8357D SPI LCD and STM32F072 nucleo dev-board success Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/ managed to *reliably* initialise the adafruit hx8357d LCD from the STM32F072 using 9-bit SPI bit-banging. the yunlea LCD will need to be tested next, with PCB2. l. From gacuest at gmail.com Thu Dec 10 21:15:08 2015 From: gacuest at gmail.com (GaCuest) Date: Thu, 10 Dec 2015 22:15:08 +0100 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: MD5: db1b80ab24d59f3b5266e934f36d6a43 En 10 de diciembre de 2015 en 16:16:49, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Thu, Dec 10, 2015 at 1:52 PM, Jean-Luc Aufranc > wrote: > > I got the same issue with the mobile website (as accessed via google on my > > PC). > > If I convert the URL to the desktop version, I can start the download: > > > > http://pan.baidu.com/share/link?shareid=2502121798&uk=1778654534 > > fantastic. if you are able to get the file in one go and it's > complete, please do let us know (md5sum, file size). > > if it *doesn't* download in one go then you will likely find that a > continuation results in corruption, in which case it would be good if > you could try that "Baidu Exporter". > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Thu Dec 10 22:28:42 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 22:28:42 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: On Thu, Dec 10, 2015 at 9:15 PM, GaCuest wrote: > MD5: db1b80ab24d59f3b5266e934f36d6a43 lkcl at bigmac:~/src/eoma/a64/sdk$ md5sum lichee_A64_A5.1_V1.0.tar.gz dd9b3f36eba4c144e88420c4bb2a82b8 lichee_A64_A5.1_V1.0.tar.gz arrrgh... :) ok - do you have a linux server anywhere, or, manuel, tell you what: could you send me an ssh public key, and would you be happy to rsync the file up to my server? l. From lkcl at lkcl.net Thu Dec 10 23:07:26 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Dec 2015 23:07:26 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: On Thu, Dec 10, 2015 at 9:15 PM, GaCuest wrote: > MD5: db1b80ab24d59f3b5266e934f36d6a43 ok just a thought, before doing that, could you do: tar -tvzf A64_....tar.gz > /tmp/f then report if there are any errors? actually you could probably do: zcat A64_.....tar.gz > /dev/null and it would be just as effective. l. From gacuest at gmail.com Fri Dec 11 09:24:58 2015 From: gacuest at gmail.com (GaCuest) Date: Fri, 11 Dec 2015 10:24:58 +0100 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: I have reupload it: http://george.the-petries.co.uk/shared-write-access/eoma/tmp/lichee_A64_A5.1_V1.0.tar.gz It is the server of Alexander Stephen, and the file is too big, so tomorrow I will delete it. Thanks Alexander! En 10 de diciembre de 2015 en 23:29:06, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Thu, Dec 10, 2015 at 9:15 PM, GaCuest wrote: > > MD5: db1b80ab24d59f3b5266e934f36d6a43 > > lkcl at bigmac:~/src/eoma/a64/sdk$ md5sum lichee_A64_A5.1_V1.0.tar.gz > dd9b3f36eba4c144e88420c4bb2a82b8 lichee_A64_A5.1_V1.0.tar.gz > > arrrgh... :) > > ok - do you have a linux server anywhere, or, manuel, tell you what: > could you send me an ssh public key, and would you be happy to rsync > the file up to my server? > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Fri Dec 11 12:24:38 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Fri, 11 Dec 2015 12:24:38 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: On Fri, Dec 11, 2015 at 9:24 AM, GaCuest wrote: > I have reupload it: http://george.the-petries.co.uk/shared-write-access/eoma/tmp/lichee_A64_A5.1_V1.0.tar.gz on the way - my server's got 100mbit/s ethernet, currently downloading @ 10mbytes/sec, ETA remaining 2mins.. done. tar-command... done. verified as correct. being copied over using rsync to correct the errors. thanks manuel, thanks stephen - you can delete the file now. l. From cnxsoft at cnx-software.com Fri Dec 11 13:49:44 2015 From: cnxsoft at cnx-software.com (Jean-Luc Aufranc) Date: Fri, 11 Dec 2015 20:49:44 +0700 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> Message-ID: <566AD478.3060900@cnx-software.com> Alternatively mirror @ http://www.cnx-software.com/2015/12/10/allwinner-a64-android-5-1-sdk-and-linux-source-code/#comment-517449 On 12/11/2015 07:24 PM, Luke Kenneth Casson Leighton wrote: > On Fri, Dec 11, 2015 at 9:24 AM, GaCuest wrote: >> I have reupload it: http://george.the-petries.co.uk/shared-write-access/eoma/tmp/lichee_A64_A5.1_V1.0.tar.gz > on the way - my server's got 100mbit/s ethernet, currently > downloading @ 10mbytes/sec, ETA remaining 2mins.. done. > tar-command... done. verified as correct. being copied over using > rsync to correct the errors. > > thanks manuel, thanks stephen - you can delete the file now. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Fri Dec 11 14:03:18 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Fri, 11 Dec 2015 14:03:18 +0000 Subject: [Arm-netbook] A64 SDK In-Reply-To: <566AD478.3060900@cnx-software.com> References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> <566AD478.3060900@cnx-software.com> Message-ID: On Fri, Dec 11, 2015 at 1:49 PM, Jean-Luc Aufranc wrote: > Alternatively mirror @ > http://www.cnx-software.com/2015/12/10/allwinner-a64-android-5-1-sdk-and-linux-source-code/#comment-517449 awesome - thanks jean-luc. that will help some new friends i know who would like to work with the A64. i've done a quick preliminary analysis: kernel 3.10 and u-boot source code are there. the u-boot code i have not yet detected any GPL violations: even the NAND source code is there. it's not properly marked up: yet again, those fucking STUPID procedures that tom's ex-manager set up are being run on the code, to destroy and remove all responsibility for copyright ownership, but other than that, even the source code for boot0 and other things such as the SPL initialiser code are all there. the kernel source is a different matter: i've found two clear and blatant GPL violations already (in modules) - one is the usual "libnand" (despite the fact that the u-boot code contains the NAND source code), the other is some "secure" library. now, product-wise, people downloading the GPL-violating A64 SDK may simply create products that *don't use* NAND (or this "secure" library) - booting from SPI, SD or eMMC is perfectly possible instead, and in that way create GPL-compliant products. we already have a clear authoritative written word from an official allwinner representative that any code marked as "all rights reserved" may be modified to state "GPLv2, Copyright Allwinner", although strictly speaking we know that allwinner is supposed to make that mark-up itself... fascinating to be working with products from a company that's fighting itself, left hand not knowing what the right is doing... anyway gotta go, l. From gacuest at gmail.com Fri Dec 11 17:00:34 2015 From: gacuest at gmail.com (GaCuest) Date: Fri, 11 Dec 2015 18:00:34 +0100 Subject: [Arm-netbook] A64 SDK In-Reply-To: References: <201512101205.15146.paul@boddie.org.uk> <56698397.9080807@cnx-software.com> <566AD478.3060900@cnx-software.com> Message-ID: En 11 de diciembre de 2015 en 15:03:36, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Fri, Dec 11, 2015 at 1:49 PM, Jean-Luc Aufranc > wrote: > > Alternatively mirror @ > > http://www.cnx-software.com/2015/12/10/allwinner-a64-android-5-1-sdk-and-linux-source-code/#comment-517449 > > awesome - thanks jean-luc. that will help some new friends i know > who would like to work with the A64. > > i've done a quick preliminary analysis: kernel 3.10 and u-boot source > code are there. the u-boot code i have not yet detected any GPL > violations: even the NAND source code is there. it's not properly > marked up: yet again, those fucking STUPID procedures that tom's > ex-manager set up are being run on the code, to destroy and remove all > responsibility for copyright ownership, but other than that, even the > source code for boot0 and other things such as the SPL initialiser > code are all there. > > the kernel source is a different matter: i've found two clear and > blatant GPL violations already (in modules) - one is the usual > "libnand" (despite the fact that the u-boot code contains the NAND > source code), the other is some "secure" library. > > now, product-wise, people downloading the GPL-violating A64 SDK may > simply create products that *don't use* NAND (or this "secure" > library) - booting from SPI, SD or eMMC is perfectly possible instead, > and in that way create GPL-compliant products. we already have a > clear authoritative written word from an official allwinner > representative that any code marked as "all rights reserved" may be > modified to state "GPLv2, Copyright Allwinner", although strictly > speaking we know that allwinner is supposed to make that mark-up > itself... fascinating to be working with products from a company > that's fighting itself, left hand not knowing what the right is > doing... > > anyway gotta go, >  Well… the other alternative is Snapdragon 410. There is another alternative at present. > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Dec 14 21:45:27 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 14 Dec 2015 21:45:27 +0000 Subject: [Arm-netbook] 15in libre laptop, 1366x768 LCD operational Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/ yippeee! i cleaned up the tracks around the SN75LVDS83b, and also i disabled HDMI (2nd) output entirely in the script.fex. the LCD is now fully operational, and is running at a low-power-saving screen refresh rate of only 37hz. that drops the power consumption of the LCD panel by a huge 40%, as well as allowing the SN75LVDS83b to operate at a lower speed and use less power as well. yaaaa :) From lkcl at lkcl.net Wed Dec 16 13:00:49 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Dec 2015 13:00:49 +0000 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards Message-ID: it's a little early but i wanted to get a poll out of people who might be interested to buy the PCBs (and other parts) as a kit, for both the libre laptop PCBs and the microdesktop units. when i say "early" i mean "there's pretty much no software" - i have preliminary working firmware that i'm using to confirm that various hardware functions work (keyboard, touchscreen, battery monitoring), however as for actual "useable drivers", those do not exist and will need to be written. this is therefore a solicitation for assistance in writing the drivers as much as it is an offer for people to be involved and have extremely-early hardware. the casework for the laptop is available as GPL pyopenscad files: i could potentially print some up for people but would prefer it if you used an online 3d printing service or local hackerspace. the micro-desktop, in direct contrast to the laptop, requires absolutely no drivers whatsoever and works straight away. estimated low-volume (5-10) pricing is going to be somewhere around: * eoma68-jz4775 CPU Card: $300 (cost of 5 PCBs is $1500) * eoma68-a20 CPU Card: same. * laptop main PCB1: $150 * laptop E/C PCB2: $120 * laptop PMIC PCB3: $120 * microdesktop PCB: $150 * laptop battery: $12 * 15.6in LCD: $50 (obtainable on ebay) * Chicony Keyboard: $20 (again, obtainable on ebay) * USB-UART: $5 (obtainable on ebay or adafruit or digikey or anywhere) there are various other sundries that are needed - cables, connectors etc. i can put together kits for people. if we can get as high as 25 units on particular boards, that pricing is likely to drop by something like 25%, at least. if anyone is considering this, please *do* consider prioritising getting the micro-desktop as well, i find it is very useful to have a working setup with the micro-desktop, separate from the laptop PCB, in a totally separate area of my desk. please also bear in mind, i am going through to a 2nd revision of all of the above PCBs, so we will be looking at about 2 months before these boards can go to a (larger) run - i wanted to give people an opportunity to say if they'd like to be involved. l. From paul at boddie.org.uk Wed Dec 16 13:19:35 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Wed, 16 Dec 2015 14:19:35 +0100 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: References: Message-ID: <201512161419.36103.paul@boddie.org.uk> On Wednesday 16. December 2015 14.00.49 Luke Kenneth Casson Leighton wrote: > it's a little early but i wanted to get a poll out of people who might > be interested to buy the PCBs (and other parts) as a kit, for both the > libre laptop PCBs and the microdesktop units. I'm definitely interested in the eoma68-jz4775 card and microdesktop board. The laptop is interesting in its own right, but I don't personally have much of an interest in using laptops these days, so I probably wouldn't want to pursue that part of this straight away. [...] > if anyone is considering this, please *do* consider prioritising > getting the micro-desktop as well, i find it is very useful to have a > working setup with the micro-desktop, separate from the laptop PCB, in > a totally separate area of my desk. > > please also bear in mind, i am going through to a 2nd revision of all > of the above PCBs, so we will be looking at about 2 months before > these boards can go to a (larger) run - i wanted to give people an > opportunity to say if they'd like to be involved. It would be nice to get involved, certainly. Paul From lkcl at lkcl.net Wed Dec 16 15:09:21 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Dec 2015 15:09:21 +0000 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: <201512161419.36103.paul@boddie.org.uk> References: <201512161419.36103.paul@boddie.org.uk> Message-ID: On Wed, Dec 16, 2015 at 1:19 PM, Paul Boddie wrote: > On Wednesday 16. December 2015 14.00.49 Luke Kenneth Casson Leighton wrote: >> it's a little early but i wanted to get a poll out of people who might >> be interested to buy the PCBs (and other parts) as a kit, for both the >> libre laptop PCBs and the microdesktop units. > > I'm definitely interested in the eoma68-jz4775 card and microdesktop board. great. > It would be nice to get involved, certainly. great. so, i'll put you down for the jz4775 and udesktop. would you (or anyone else) possibly be interested in helping with the firmware on the STM32F072? it's basically porting things like the linux i2c touchscreen driver (which is in c) to the stm32f072 with libopencm3 (which is also in c). i've done the hx8357d SPI driver already - got the initialisation and screen-writing done. the main thing is making it all "hang together" - interleaving USB CDCADM with USB HID for mouse *and* keyboard - three separate USB endpoints - and creating some power-management routines, status of which can be monitored over the CDCACM serial interface. and writing an appropriate linux kernel driver to work with that. l. From paul at boddie.org.uk Wed Dec 16 15:18:47 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Wed, 16 Dec 2015 16:18:47 +0100 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: References: <201512161419.36103.paul@boddie.org.uk> Message-ID: <201512161618.47951.paul@boddie.org.uk> On Wednesday 16. December 2015 16.09.21 Luke Kenneth Casson Leighton wrote: > > great. so, i'll put you down for the jz4775 and udesktop. Yes, please do! > would you (or anyone else) possibly be interested in helping with the > firmware on the STM32F072? it's basically porting things like the > linux i2c touchscreen driver (which is in c) to the stm32f072 with > libopencm3 (which is also in c). i've done the hx8357d SPI driver > already - got the initialisation and screen-writing done. the main > thing is making it all "hang together" - interleaving USB CDCADM with > USB HID for mouse *and* keyboard - three separate USB endpoints - and > creating some power-management routines, status of which can be > monitored over the CDCACM serial interface. and writing an > appropriate linux kernel driver to work with that. I don't have that much Linux driver-writing experience, although I've peered into the murky depths of various drivers, including USB-related ones, which was horrible. Still, if you can't find anyone with the expertise who can do this kind of thing in an afternoon, I might be of some assistance. I have looked at USB controllers before now, so it's not completely alien technology to me. Paul From lkcl at lkcl.net Wed Dec 16 15:28:09 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Dec 2015 15:28:09 +0000 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: <201512161618.47951.paul@boddie.org.uk> References: <201512161419.36103.paul@boddie.org.uk> <201512161618.47951.paul@boddie.org.uk> Message-ID: On Wed, Dec 16, 2015 at 3:18 PM, Paul Boddie wrote: > On Wednesday 16. December 2015 16.09.21 Luke Kenneth Casson Leighton wrote: >> >> great. so, i'll put you down for the jz4775 and udesktop. > > Yes, please do! alright, awesome. i've got 5 very-very-first jz4775 cpu cards in assembly at the moment (delayed by the xmas rush), > I don't have that much Linux driver-writing experience, although I've peered > into the murky depths of various drivers, including USB-related ones, which > was horrible. haha > Still, if you can't find anyone with the expertise who can do > this kind of thing in an afternoon, I might be of some assistance. I have > looked at USB controllers before now, so it's not completely alien technology > to me. microcontrollers are actually a very small amount of memory, so if you are writing lots of code, you're doing something badly, badly wrong. it is however a bit of a mind-bender: everything has to be state-driven. it's single-process (no threads, no state-swapping), interrupts and DMA. so everything is done as a huuuuge state-machine... or on polling with sleep loops. l. From paul at boddie.org.uk Wed Dec 16 15:56:51 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Wed, 16 Dec 2015 16:56:51 +0100 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: References: <201512161618.47951.paul@boddie.org.uk> Message-ID: <201512161656.51745.paul@boddie.org.uk> On Wednesday 16. December 2015 16.28.09 Luke Kenneth Casson Leighton wrote: > > microcontrollers are actually a very small amount of memory, so if > you are writing lots of code, you're doing something badly, badly > wrong. it is however a bit of a mind-bender: everything has to be > state-driven. it's single-process (no threads, no state-swapping), > interrupts and DMA. so everything is done as a huuuuge > state-machine... or on polling with sleep loops. Well, the microcontroller stuff I have a bit more experience with, having messed around with a USB controller (MAX3421E) with the Arduino, but then using the NanoNote to do the same thing because it was easier and I was being lazy. ;-) But in other words, not Linux driver stuff. Paul From lkcl at lkcl.net Wed Dec 16 16:30:23 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Dec 2015 16:30:23 +0000 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: <201512161656.51745.paul@boddie.org.uk> References: <201512161618.47951.paul@boddie.org.uk> <201512161656.51745.paul@boddie.org.uk> Message-ID: On Wed, Dec 16, 2015 at 3:56 PM, Paul Boddie wrote: > On Wednesday 16. December 2015 16.28.09 Luke Kenneth Casson Leighton wrote: >> >> microcontrollers are actually a very small amount of memory, so if >> you are writing lots of code, you're doing something badly, badly >> wrong. it is however a bit of a mind-bender: everything has to be >> state-driven. it's single-process (no threads, no state-swapping), >> interrupts and DMA. so everything is done as a huuuuge >> state-machine... or on polling with sleep loops. > > Well, the microcontroller stuff I have a bit more experience with, having > messed around with a USB controller (MAX3421E) with the Arduino, oh great - yeah the STM32F072 is like a grown-up version of the arduino. none of this bullshit installing 160mb of java runtime, a toy-for-an-IDE that does nothing more than wrap sdcc and avr-utils even to the point of modifying simple Makefiles for you. i bought an OSMC back in 2001, it had the same PICs as the arduino, and *nobody* arsed about with 250mb of crap to wrap 5mb worth of libraries and command-line compiler tools! rant over.... :) yeah using libopencm3, it's just Makefiles and gcc arm cross-compilers with no stdlib [gcc-arm-none]. the community surrounding libopencm3 is quite large, so there are actually plenty of examples out there as well as people who can help. l. From phil at hands.com Wed Dec 16 18:02:06 2015 From: phil at hands.com (Philip Hands) Date: Wed, 16 Dec 2015 19:02:06 +0100 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: References: <201512161618.47951.paul@boddie.org.uk> <201512161656.51745.paul@boddie.org.uk> Message-ID: <87si3246ht.fsf@whist.hands.com> Luke Kenneth Casson Leighton writes: > On Wed, Dec 16, 2015 at 3:56 PM, Paul Boddie wrote: >> On Wednesday 16. December 2015 16.28.09 Luke Kenneth Casson Leighton wrote: >>> >>> microcontrollers are actually a very small amount of memory, so if >>> you are writing lots of code, you're doing something badly, badly >>> wrong. it is however a bit of a mind-bender: everything has to be >>> state-driven. it's single-process (no threads, no state-swapping), >>> interrupts and DMA. so everything is done as a huuuuge >>> state-machine... or on polling with sleep loops. >> >> Well, the microcontroller stuff I have a bit more experience with, having >> messed around with a USB controller (MAX3421E) with the Arduino, > > oh great - yeah the STM32F072 is like a grown-up version of the > arduino. none of this bullshit installing 160mb of java runtime, a > toy-for-an-IDE that does nothing more than wrap sdcc and avr-utils > even to the point of modifying simple Makefiles for you. i bought an > OSMC back in 2001, it had the same PICs as the arduino, and *nobody* > arsed about with 250mb of crap to wrap 5mb worth of libraries and > command-line compiler tools! > > rant over.... :) Pretty much exactly why I helped make sure that the java stuff is optional on Debian -- see the 'arduino-core' and 'arduino-mk' packages: https://packages.debian.org/jessie/arduino-mk Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] HANDS.COM Ltd. |-| http://www.hands.com/ http://ftp.uk.debian.org/ |(| Hugo-Klemm-Strasse 34, 21075 Hamburg, GERMANY -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 818 bytes Desc: not available URL: From lkcl at lkcl.net Wed Dec 16 19:19:50 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Dec 2015 19:19:50 +0000 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: <87si3246ht.fsf@whist.hands.com> References: <201512161618.47951.paul@boddie.org.uk> <201512161656.51745.paul@boddie.org.uk> <87si3246ht.fsf@whist.hands.com> Message-ID: On Wed, Dec 16, 2015 at 6:02 PM, Philip Hands wrote: > Luke Kenneth Casson Leighton writes: >> oh great - yeah the STM32F072 is like a grown-up version of the >> arduino. none of this bullshit installing 160mb of java runtime, a >> toy-for-an-IDE that does nothing more than wrap sdcc and avr-utils >> even to the point of modifying simple Makefiles for you. i bought an >> OSMC back in 2001, it had the same PICs as the arduino, and *nobody* >> arsed about with 250mb of crap to wrap 5mb worth of libraries and >> command-line compiler tools! >> >> rant over.... :) > > Pretty much exactly why I helped make sure that the java stuff is > optional on Debian -- see the 'arduino-core' and 'arduino-mk' packages: ah! yes, i remember you mentioning that to me a couple years back, i vaguely remembered, but now you mention it i had installed.... $ dpkg -l | grep arduino ii arduino-core 2:1.0.5+dfsg2-4 ii arduino-mk 1.3.4-1 ... those exact same packages, to do the SPI-testing with an adafruit hx8357d LCD and an arduino uno, a few weeks back. yaay, so the maintainers listened to your advice, split out the core stuff? which is absolutely awesome. so, thank you phil :) btw just for the record (for anyone else reading this in the archives, later), these are the other packages i have installed, to program arduinos *without* the crap 200mb of java-based runtime and IDE: ii avr-libc 1:1.8.0+Atmel3.4.4-1 ii avrdude 6.1-2 ii binutils-avr 2.24+Atmel3.4.4-1 ii gcc-avr 1:4.8.1+Atmel3.4.4-2 this is enough to be able to simply grab Makefiles online (there's lots of examples out there, usually you can search "debian arduino uno makefile" [1]) and e.g. entire adafruit libraries and example programs and get them up-and-running very very quickly and just type "make" and then "make upload". l. [1] http://www.itopen.it/arduino-1-0-development-with-a-makefile/ - if following those instructions replace the first "apt-get install arduino" with phil's recommended packages "apt-get install arduino-core arduino-mk". From phil at hands.com Wed Dec 16 21:11:03 2015 From: phil at hands.com (Philip Hands) Date: Wed, 16 Dec 2015 22:11:03 +0100 Subject: [Arm-netbook] early alpha production run of libre laptop PCBs, microdesktops and CPU Cards In-Reply-To: References: <201512161618.47951.paul@boddie.org.uk> <201512161656.51745.paul@boddie.org.uk> <87si3246ht.fsf@whist.hands.com> Message-ID: <87k2oe3xqw.fsf@whist.hands.com> Luke Kenneth Casson Leighton writes: > On Wed, Dec 16, 2015 at 6:02 PM, Philip Hands wrote: >> Luke Kenneth Casson Leighton writes: >>> oh great - yeah the STM32F072 is like a grown-up version of the >>> arduino. none of this bullshit installing 160mb of java runtime, a >>> toy-for-an-IDE that does nothing more than wrap sdcc and avr-utils >>> even to the point of modifying simple Makefiles for you. i bought an >>> OSMC back in 2001, it had the same PICs as the arduino, and *nobody* >>> arsed about with 250mb of crap to wrap 5mb worth of libraries and >>> command-line compiler tools! >>> >>> rant over.... :) >> >> Pretty much exactly why I helped make sure that the java stuff is >> optional on Debian -- see the 'arduino-core' and 'arduino-mk' packages: > > ah! yes, i remember you mentioning that to me a couple years back, i > vaguely remembered, but now you mention it i had installed.... > > $ dpkg -l | grep arduino > ii arduino-core 2:1.0.5+dfsg2-4 > ii arduino-mk 1.3.4-1 > > ... those exact same packages, to do the SPI-testing with an adafruit > hx8357d LCD and an arduino uno, a few weeks back. yaay, so the > maintainers listened to your advice, split out the core stuff? Not advice -- turning up cold and giving people advice often goes quite badly (as I think you may have demonstrated once or thrice ;-) ) I built the arduino-core package (including the makefile), and getting the author of the best of the several makefile forks to accept some nice patches from other forks of it, so that there's now one obvious upstream makefile to use, while also working out with Scott (who had already done the full 'arduino' package) that we could eliminate the duplication and have the main package depend on mine, and build it all out of the same source. He seems to have since split out the Makefile into the -mk package. > btw just for the record (for anyone else reading this in the archives, > later), these are the other packages i have installed, to program > arduinos *without* the crap 200mb of java-based runtime and IDE: > > ii avr-libc 1:1.8.0+Atmel3.4.4-1 > ii avrdude 6.1-2 > ii binutils-avr 2.24+Atmel3.4.4-1 > ii gcc-avr 1:4.8.1+Atmel3.4.4-2 > > this is enough to be able to simply grab Makefiles online (there's > lots of examples out there, usually you can search "debian arduino uno > makefile" [1]) and e.g. entire adafruit libraries and example programs > and get them up-and-running very very quickly and just type "make" and > then "make upload". > > l. > > [1] http://www.itopen.it/arduino-1-0-development-with-a-makefile/ - if > following those instructions replace the first "apt-get install > arduino" with phil's recommended packages "apt-get install > arduino-core arduino-mk". You only need: apt-get install arduino-mk since that depends on arduino-core, which depends on all the other bits. Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] HANDS.COM Ltd. |-| http://www.hands.com/ http://ftp.uk.debian.org/ |(| Hugo-Klemm-Strasse 34, 21075 Hamburg, GERMANY -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 818 bytes Desc: not available URL: From paul at boddie.org.uk Thu Dec 17 13:06:53 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Thu, 17 Dec 2015 14:06:53 +0100 Subject: [Arm-netbook] LWN: Hands on with the pi-top Message-ID: <201512171406.53778.paul@boddie.org.uk> Hello, I thought that some of you might be interested in this, at least in the context of laptop projects, what other people are doing, and the finished products: http://lwn.net/Articles/667040/ I guess the Raspberry Pi aspect of it means that people are willing to throw their money at something that doesn't really seem to be a particularly good product: tied to the RPi, no speakers or audio outputs (apart from on the RPi board itself), the usual proprietary software dependencies and promotional activity, reportedly dodgy keyboard... Also of interest is that their desktop version recently got funded very quickly indeed. Again, the same RPi dependencies and constraints apply. I guess it shows that there's a market for such stuff, even if (or especially if) you take the RPi cargo culture out of the equation. Paul From lkcl at lkcl.net Thu Dec 17 13:26:16 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Dec 2015 13:26:16 +0000 Subject: [Arm-netbook] LWN: Hands on with the pi-top In-Reply-To: <201512171406.53778.paul@boddie.org.uk> References: <201512171406.53778.paul@boddie.org.uk> Message-ID: On Thu, Dec 17, 2015 at 1:06 PM, Paul Boddie wrote: > Hello, > > I thought that some of you might be interested in this, at least in the > context of laptop projects, what other people are doing, and the finished > products: > > http://lwn.net/Articles/667040/ thanks paul > I guess the Raspberry Pi aspect of it means that people are willing to throw > their money at something that doesn't really seem to be a particularly good > product: tied to the RPi, no speakers or audio outputs (apart from on the RPi > board itself), the usual proprietary software dependencies and promotional > activity, reportedly dodgy keyboard... it sounds very much like it's "first generation". that, and they have gone out early instead of doing a limited run and a slower ramp-up, put it out there to as many people as wanted it, problems-and-all, to get more eyes-on for feedback and potential solutions. that sounds pretty much exactly like one of the core guidelines of successful software libre projects - "release early, release often". let's see if they follow up with the "release often" bit. > Also of interest is that their desktop version recently got funded very > quickly indeed. Again, the same RPi dependencies and constraints apply. I > guess it shows that there's a market for such stuff, very true From lkcl at lkcl.net Thu Dec 24 19:45:59 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 24 Dec 2015 19:45:59 +0000 Subject: [Arm-netbook] chicony keyboard matrix reverse-engineering Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/Chicony_keyboard_matrix_reverse_engineering/ the article about keyboard matrix design, linked from the above, is quite interesting. keyboard design turns out to be fascinatingly complex for something that we take for granted every day. l. From lkcl at lkcl.net Sat Dec 26 04:03:06 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sat, 26 Dec 2015 04:03:06 +0000 Subject: [Arm-netbook] first revision eoma68 jz4775 cpu cards assembled Message-ID: http://rhombus-tech.net/ingenic/jz4775/news/26dec2015_pcb_rev1_populated/ still to be tested but they've actually been assembled. l. From paul at boddie.org.uk Sat Dec 26 17:24:58 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Sat, 26 Dec 2015 17:24:58 +0000 Subject: [Arm-netbook] first revision eoma68 jz4775 cpu cards assembled In-Reply-To: References: Message-ID: <322f8d64b39fd5b937d3a0a7e29bc601@webmail.webfaction.com> On 2015-12-26 04:03, Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/ingenic/jz4775/news/26dec2015_pcb_rev1_populated/ > > still to be tested but they've actually been assembled. Congratulations on getting this far, at least, despite plenty more work to do! Paul From lkcl at lkcl.net Sat Dec 26 17:28:06 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sat, 26 Dec 2015 17:28:06 +0000 Subject: [Arm-netbook] first revision eoma68 jz4775 cpu cards assembled In-Reply-To: <322f8d64b39fd5b937d3a0a7e29bc601@webmail.webfaction.com> References: <322f8d64b39fd5b937d3a0a7e29bc601@webmail.webfaction.com> Message-ID: On Sat, Dec 26, 2015 at 5:24 PM, Paul Boddie wrote: > On 2015-12-26 04:03, Luke Kenneth Casson Leighton wrote: >> >> http://rhombus-tech.net/ingenic/jz4775/news/26dec2015_pcb_rev1_populated/ >> >> still to be tested but they've actually been assembled. > > > Congratulations on getting this far, at least, despite plenty more work to > do! thanks paul. i've reached out to the linux-mips community, i contacted them last year and at least 2 people expressed an interest in helping with board bring-up and kernel development. jz4780 is very similar to jz4775 and there have been several commits which cater for both processors. l. From paul at boddie.org.uk Sat Dec 26 17:38:18 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Sat, 26 Dec 2015 17:38:18 +0000 Subject: [Arm-netbook] first revision eoma68 jz4775 cpu cards assembled In-Reply-To: References: <322f8d64b39fd5b937d3a0a7e29bc601@webmail.webfaction.com> Message-ID: <10000a546e0cc566da371ee165b0096c@webmail.webfaction.com> On 2015-12-26 17:28, Luke Kenneth Casson Leighton wrote: > > thanks paul. i've reached out to the linux-mips community, i > contacted them last year and at least 2 people expressed an interest > in helping with board bring-up and kernel development. jz4780 is very > similar to jz4775 and there have been several commits which cater for > both processors. Last time I looked, it seemed that the jz4780 had benefited from the work done to support the MIPS Creator CI20, which had probably benefited from the ongoing work done with the jz4770 (in the GCW-Zero and various handheld consoles, as I understand it), so the jz4775 shouldn't bring too many surprises, especially since it comes without the PowerVR nonsense. I guess that even the device tree support in the jz family is fairly complete now, too. Paul From lkcl at lkcl.net Sat Dec 26 18:55:31 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sat, 26 Dec 2015 18:55:31 +0000 Subject: [Arm-netbook] first revision eoma68 jz4775 cpu cards assembled In-Reply-To: <10000a546e0cc566da371ee165b0096c@webmail.webfaction.com> References: <322f8d64b39fd5b937d3a0a7e29bc601@webmail.webfaction.com> <10000a546e0cc566da371ee165b0096c@webmail.webfaction.com> Message-ID: On Sat, Dec 26, 2015 at 5:38 PM, Paul Boddie wrote: > On 2015-12-26 17:28, Luke Kenneth Casson Leighton wrote: >> >> >> thanks paul. i've reached out to the linux-mips community, i >> contacted them last year and at least 2 people expressed an interest >> in helping with board bring-up and kernel development. jz4780 is very >> similar to jz4775 and there have been several commits which cater for >> both processors. > > > Last time I looked, it seemed that the jz4780 had benefited from the work > done to support the MIPS Creator CI20, which had probably benefited from the > ongoing work done with the jz4770 (in the GCW-Zero and various handheld > consoles, as I understand it), so the jz4775 shouldn't bring too many > surprises, especially since it comes without the PowerVR nonsense. I guess > that even the device tree support in the jz family is fairly complete now, > too. good timing. it's still an "unknown" however, so my first instinct is to use "known-good" as a means to confirm the working hardware. that means using the code released by ingenic, which is based (iirc) on linux kernel 3.0.8. *AFTER* the hardware has been confirmed operational, *THEN* the hardware may in turn be used to confirm that the [latest] linux kernel tree works correctly. a very simple very hard lesson learned there: go ONLY from ONE known-good to another. UNDER NO CIRCUMSTANCES attempt to introduce two (or more) highly-complex inter-dependent unknowns. this was why the LCD took almost two maybe as long as three months to get operational, because there were no less than *five* separate and distinct unknowns in the chain, including the linux kernel tree (sunxi-3.4-cb), script.fex file, LVDS cable (which had to be made up by hand), LCD itself, backlight circuit, backlight and lcd GPIO, and the TI LVDS converter IC. and the CPU Card. how many's that so far? 8 or 9 separate unknowns, each inter-dependent? l. From lkcl at lkcl.net Sat Dec 26 22:12:42 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sat, 26 Dec 2015 22:12:42 +0000 Subject: [Arm-netbook] keyboard matrix (and usb-hid stm32f072 firmware) functional Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/PCB2_prototype_functional_keyboard_matrix/ i'll make a short video once the camera's battery is charged up - i have yet to do the keyboard modifiers (shift, ctrl) however everything else is properly functional. with 1st revision prototype firmware available as GPLv2+ source, which has at least proven that the hardware is operational, i can move on to the audio circuits tomorrow, and after that can begin assembly. l. From pbouda at cidles.eu Sun Dec 27 17:17:11 2015 From: pbouda at cidles.eu (Peter Bouda) Date: Sun, 27 Dec 2015 17:17:11 +0000 Subject: [Arm-netbook] keyboard matrix (and usb-hid stm32f072 firmware) functional In-Reply-To: References: Message-ID: <56801D17.9040804@cidles.eu> > http://rhombus-tech.net/community_ideas/laptop_15in/news/PCB2_prototype_functional_keyboard_matrix/ > > i'll make a short video once the camera's battery is charged up - i > have yet to do the keyboard modifiers (shift, ctrl) however everything > else is properly functional. with 1st revision prototype firmware > available as GPLv2+ source, which has at least proven that the > hardware is operational, i can move on to the audio circuits tomorrow, > and after that can begin assembly. > > l. nice! which version of libopencm3 did you use? I just tried the current git repository of the firmware together with libopencm3 master and the paths to the headers seem to be different or some are even missing (libopencm3/usb/misc/string.h, for example). I wanted to see if I can build a gamepad usb-hid example based on your code. Best, Peter From lkcl at lkcl.net Sun Dec 27 17:43:49 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sun, 27 Dec 2015 17:43:49 +0000 Subject: [Arm-netbook] keyboard matrix (and usb-hid stm32f072 firmware) functional In-Reply-To: <56801D17.9040804@cidles.eu> References: <56801D17.9040804@cidles.eu> Message-ID: On Sun, Dec 27, 2015 at 5:17 PM, Peter Bouda wrote: > nice! which version of libopencm3 did you use? I just tried the current git > repository of the firmware together with libopencm3 master and the paths to > the headers seem to be different or some are even missing > (libopencm3/usb/misc/string.h, for example). I wanted to see if I can build > a gamepad usb-hid example based on your code. commit 4abcf1ad255b79085dc7e3d8861e15709848f567 Author: Kuldeep Singh Dhaka Date: Tue Sep 8 11:08:26 2015 +0530 USB rewrite you should be able to find that via a google search on the git commit hash. kuldeep is actively working with the stm32f072 and has made significant enhancements and improvements to libopencm3 which, unfortunately, the core developers do not trust. they are on what can only best be described as a "can't-cope-so-we-must-have-unit-tests" binge, which has hugely stalled development. my understanding is that they feel solely and exclusively responsible for ensuring that the ever-expanding codebase remains "stable", but have not made a push to invite other developers to take responsibility for ensuring that contributions remain current for a particular embedded controller. my advice to you therefore is to use kuldeep's work because he is *actively* working with the stm32f072 right now, and to *not* move from this particular git commit unless it is absolutely, absolutely and i mean absolute 100% essential to do so. if you need I2C i have added to stm32/f0/i2c.h some additional #defines that enable the i2c code to compile and operate. l. From pbouda at cidles.eu Sun Dec 27 18:33:19 2015 From: pbouda at cidles.eu (Peter Bouda) Date: Sun, 27 Dec 2015 18:33:19 +0000 Subject: [Arm-netbook] keyboard matrix (and usb-hid stm32f072 firmware) functional In-Reply-To: References: <56801D17.9040804@cidles.eu> Message-ID: <56802EEF.3030704@cidles.eu> > commit 4abcf1ad255b79085dc7e3d8861e15709848f567 > Author: Kuldeep Singh Dhaka > Date: Tue Sep 8 11:08:26 2015 +0530 > > USB rewrite > > > you should be able to find that via a google search on the git commit > hash. sorry, I cannot find this one, even on Google (?). I am using this one now: https://github.com/kuldeepdhaka/libopencm3/commit/a448f319c863b006042a518db3b54b27deda5512 Is that correct? Best, Peter From lkcl at lkcl.net Sun Dec 27 21:42:06 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sun, 27 Dec 2015 21:42:06 +0000 Subject: [Arm-netbook] keyboard matrix (and usb-hid stm32f072 firmware) functional In-Reply-To: <56802EEF.3030704@cidles.eu> References: <56801D17.9040804@cidles.eu> <56802EEF.3030704@cidles.eu> Message-ID: On Sun, Dec 27, 2015 at 6:33 PM, Peter Bouda wrote: >> commit 4abcf1ad255b79085dc7e3d8861e15709848f567 >> Author: Kuldeep Singh Dhaka >> Date: Tue Sep 8 11:08:26 2015 +0530 >> >> USB rewrite >> >> >> you should be able to find that via a google search on the git commit >> hash. > > > sorry, I cannot find this one, even on Google (?). I am using this one now: > > https://github.com/kuldeepdhaka/libopencm3/commit/a448f319c863b006042a518db3b54b27deda5512 > > Is that correct? parent commit: commit 15a6103f8ed07d4e5b90331e99ec6ed77a6ef1ab Author: Hanspeter Portner Date: Mon Sep 7 09:18:00 2015 +0200 stm32f3: ld: stm32f3x3x[bc] have 32K/40K RAM only. Correct memory sizes in ld/devices.data for: * stm32f3[01]3?c from RAM=48K to RAM=40K. * stm32f303?b from RAM=40K to RAM=32K. Reviewed-by: Karl Palsson close enough.