Introduction

In case you don't know, nspire CX II is a calculator made by Texas Instruments released in 2018. It has an ARM926EJ-s processor running at 396MHz, 64MB of LPDDR memory, and 128MB of SPI nand flash. Though this is probably already an overkill for a calculator, but, can we get even faster on this calculator? Of course, it would really mean anything. I don't need that additional performance, overclocking a calculator is just like overclocking anything else, it is just purely for fun. If I do need more performance, I would go and pick up a laptop. Let's get started.

Background

To overclock any modern processor, there are two key parts: clock synthesizer, memory controller. The first one boosts up the clock frequency, and the second controls the memory timing to make sure the memory runs at higher frequency (if you are unable to decouple the memory frequency with core frequency). The TI nspire CX II uses a custom SoC which I don't have any documents. The operating system on the nspire CX II is also locked down to disallow any unsigned binary code execution. So to overclock it, there are three steps, 1. Get the "root priviledge" on the nspire OS, 2. Understand how clock generation works on nspire, 3. Write a program to overclock it.

nspire OS

Luckily, the first part has been done by the community. There is a tool called Ndless which exploits some vulnerabilities in the nspire OS to allow unsigned code execution. The version released in September 2019 introduced initial support for nspire CX II. Huge thanks to the ndless team who made this all possible.

Clock synthesizer

The nspire SoC has an internal clock synthesizer (PLL). Since there is no datasheet or manual for the SoC, I have to reverse engineer it to find out how it sets the clock. Luckily, there is a community developed emulator for the nspire CX II, also based on reverse engineering efforts obviously. (Note: CX II suppoort was developed by the same guy who implemented CXII supported for ndless, Vogtinato.)

Note: Translated with www.DeepL.com/Translator (free version)

Introduction

Xserve is Apple's long-discontinued line of servers, powered by the Mac OS X Server operating system. After Xserve was discontinued, OS X Server slowly became just an app. In 2019 Apple introduced a new rackmount Mac Pro, which can also be used as a rackmount server. But the price is not something I can justify spending for fun. The other thing is that with ARM Mac releases coming soon, the x86-64 based Mac Pro will be obsolete in few years. It may not make sense to spend a lot of money on a Mac Pro at this point in time. On the other hand, the old Xserve is already very cheap, with similar performance and price to the contemporaneous Dell PowerEdge R610. Unlike the 2009 Mac Pro, which for some reason still costs a lot now. For this reason as a Mac homelab or for desktop use, the last generation Xserve can be considered a good choice. The following are the main specs of the last Xserve.

Xserve 3,1 (Xserve 2009)

• CPU: 2x Intel Xeon E/X/W5500 series
• RAM: 12x DDR3 REG ECC (Up to 192GB)
• HDD: 3x SAS/SATA
• GFX: nVIDIA GeForce GT120 (MXM-A upgradable)

Xserve in 2020

Here are some concerns when using Xserve in home settings:

As a server

• Too few hard drive bays and SATA hard drive capacity limited to 1TB

As a workstation

• The performance of the 5500 series of processors is seriously outdated
• Severe fan noise
• Weak GPU

Common issues

• No longer supports the latest macOS, if you need to use macOS

This article is about fan noise only.

Introduction

ThinkPad X61 is a classical machine, along with T61 they are the last 4:3 ThinkPads made by Lenovo. I bought my X61s used in 2010 (I simply couldn't afford it back in 2007 when it first came out) and keeps using it as my main laptop (I have a much more powerful desktop PC for most of the stuff though). Many of the components have been upgraded since then, 2GB DDR2 is upgraded to 6GB, 160GB spin drive has been replaced by a 512GB SSD, and the TN screen has been replaced with an AFFS (IPS) screen. The thing that never changed is the processor: the 65nm Core 2 Duo L7500.

X61 comes with various CPU options available when buying new:

X61:

• Core 2 Duo T7100, 1.8 GHz, 2 MB L2, FSB 800, 65 nm, 35 W
• Core 2 Duo T7250, 2.0 GHz, 2 MB L2, FSB 800, 65 nm, 35 W
• Core 2 Duo T7300, 2.0 GHz, 4 MB L2, FSB 800, 65 nm, 35 W
• Core 2 Duo T8100, 2.1 GHz, 3 MB L2, FSB 800, 45 nm, 35 W
• Core 2 Duo T8300, 2.4 GHz, 3 MB L2, FSB 800, 45 nm, 35 W
• Core 2 Duo T9300, 2.5 GHz, 6 MB L2, FSB 800, 45 nm, 35 W

X61s:

• Core 2 Duo L7300, 1.4 GHz, 4 MB L2, FSB 800, 65 nm, 17 W
• Core 2 Duo L7500, 1.6 GHz, 4 MB L2, FSB 800, 65 nm, 17 W
• Core 2 Duo L7700, 1.8 GHz, 4 MB L2, FSB 800, 65 nm, 17 W

However, when buying used, especially in 2020, it is really hard to pick one specific processor. Most of the listings online come with a T7100 or T7300 processor, built with a 65 nm process. If the machine is an X61s, then the 65 nm processor is pretty much the only choice.

Unless the processor can be replaced or upgraded.

It is quite common to replace the processor in T61s, just buy a new processor, install that on to the motherboard, done. Or if one wants to use newer FSB 1066 processors or even quad-core processor, BIOS mod and motherboard mod are available to allow the processor to run on the T61 motherboard.

The X61, on the other hand, has the processor soldered down on to the board, making it harder to replace the processor. And it is much harder to buy BGA processors than just normal PGA processors: pretty much no one would take out the BGA processor from their laptop and just sell that, and it is pretty hard to buy new old stock BGA processors. But the good thing is if the CPU can be indeed bought and installed, all the mods available to the T61 are also available to the X61.

This post is a comprehensive note about how to install probably any Core 2 Duo mobile processors onto X61 if the power supply and cooling allow. And my method uses a PGA processor rather than a BGA processor so the processor is still easily purchasable on eBay or other platforms. Quad-core processors run much hotter and require further ACPI mod which is not described in this post.

I understand that these kinds of mod have been done on X61 many times before. But I just want to document comprehensively how I did it and what each step is.

I am also aware of the existence of the X62/X63 motherboard project. This project only focuses on upgrading the original motherboard.

I used P8600 as an example in this post, but other processors should also work.

All information is provided "as is". I will not be responsible for any damage due to the use or misuse of the information provided. Modding BIOS and hardware is risky and likely to permanently brick the machine if not done properly. Do this at your own risk.

简介

以下介绍的裸机开发均基于i.MX6 Platform SDK完成，最后的版本于2013年发布，在使用gcc9编译时会遇到一些小问题，这里不做具体介绍。所有的开发均在Ubuntu 20.04上完成。目标平台为i.MX6D/Q，大部分也适用于i.MX6S/DL，但是对于i.MX6UL/ULL会需要较大改动。

编译与下载

PlatformSDK的Makefile中已经包含了产生目标镜像的代码，直接编译即可产生可以烧写进SD卡或者通过uuu载入的镜像，SPI Flash有待研究。（uuu是代替了MfgTools的跨平台i.MX USB烧录工具，可支持i.MX6\7\8）

可以直接使用如sudo uuu application.bin的命令下载程序至RAM运行。如果需要写成脚本，可以参考如下的写法：

uuu_version 1.2.135

SDP: boot -f path_to_the_application.bin
SDP: done

如果需要使用JTAG下载，则需要单独编写初始化DDR内存的初始化脚本，暂未进行测试。

串口

在EVK上默认调试串口为UART4，如果实际使用的板子并非使用UART4，则需要修改成板子上对应使用的串口。定义位于board/common/hardware_modules.c中。

内存配置

DDR控制器配置位于启动镜像的DCD区域中。如果使用的内存与EVK不同，则通常需要修改DCD配置。这部分数据位于board/mx6dq/board/dcd.c中。可以考虑使用官方的Programming Aid工具和DDR Stress Test工具来产生合适的DDR控制器配置。

不同板子拥有不同的内存容量。通常会区分出一部分不可cache内存用作DMA使用。主内存映射代码位于sdk/core/src/mmu.c中，具体的DMA内存设定定义在apps/common/platform_init.c中。以下为我在1GB RAM板子上使用的内存分割方案，仅供参考：

mmu_map_l1_range(0x10000000, 0x10000000, 0x20000000, kOuterInner_WB_WA, share_attr, kRWAccess); // First 512MB
mmu_map_l1_range(0x30000000, 0x30000000, 0x20000000, kNoncacheable, kShareable, kRWAccess); // Last 512MB

注：本文内容主要在LPC5528上测试，同样适用于55S2x和55S6x，不确定将来推出的其它系列是否也会同样适用。

高速USB

高速USB需要外部晶振才能使用，外部晶振频率可以在12\16\19.2\20\24\30\32MHz之间选择。由于官方开发板使用的是16MHz的晶振，所有例程和库也是使用16MHz的。如果使用了其它频率的晶振，可以通过设置PLL_SIC寄存器修改。官方库中设置的代码位于fsl_clock.c中：

USBPHY->PLL_SIC     = (USBPHY->PLL_SIC & ~USBPHY_PLL_SIC_PLL_DIV_SEL(0x7)) | USBPHY_PLL_SIC_PLL_DIV_SEL(0x06);

其中0x06即为PLL分频器设定值。以下为可选的值：