726 Part VI . Programming in Linux GCC relies on file extensions to determine what kind of source code file it is, that is, in which programming language the source code is written. Table 28-1 lists the most common extensions and how GCC interprets them. Table 28-1 GCC s File Naming Conventions Extension Type .a, .so Compiled library code .c C language source code .C, .cc C++ language source code .i Preprocessed C source code .ii Preprocessed C++ source code .m Objective-C source code .o Compiled object code .S, .s Assembly language source code Compiling Multiple Source Code Files Most nontrivial programs consist of multiple source files, and each source file must be compiled to object code before the final link step. To do so, provide gcc the name of each source code file it has to compile. GCC handles the rest. The gcc invocation might resemble: $ gcc file1.c file2.c file3.c -o progname gcc would create file1.o, file2.o, and file3.o and then link them all together to create progname. As an alternative, you can use gcc s -c option on each file individually, which creates object files from each file. Then in a second step, you link the object files together to create an executable. Thus, the single command just shown becomes: $ gcc -c file1.c $ gcc -c file2.c $ gcc -c file3.c $ gcc file1.o file2.o file3.o -o progname
In case you need quality webspace to host and run your web applications, try our personal web hosting services.

Back in my day, we didn’t have all these fancy web front-ends to e-mail with their pretty icons and mouse click crunchy goodness. We were lucky if mail got to us at all, since when I started it was *you* configuring the mail hops for uucp delivery (amongst other services). mail or mailx were good enough for us despite their bare-bones approach to message management. The Unix mbox format standardized things enough that other beasts came on the scene. elm was a contender for a while, but the leader of the pack was – and still is – pine.

pine is a mail tool maintained by the top-notch souls over at the University of Washington. I have used pine for as long as it’s been available to the public (pre-POP3/IMAP), whether it be on a *nix variant or even a PC/Mac. At my previous employer where Exchange was king, pine served me well, even though it had to be eventually used in combination with fetchmail for various technical reasons.

When I finally switched over my personal mail to Google for hosting (got tired of fighting spam) I wound up using the web interface as the primary means of mail retrieval, simply because of my disdain for POP/POP3. Compared to IMAP (well call it the FedEx/UPS of mail), POP (aptly named after the government operated postal service) is a loathsome creature, fit for mole men or politicians and it just wasn’t worth dealing with the fact that I wasn’t always operating on the central message store. pine was rarely executed and I stopped staying current.

It wasn’t until recent frustrations with GMail (and Google’s awesome decision to partially support IMAP) that I started to look at pine again and was very pleasantly surprised to see that it’s entering a new era with the release of alpine. The alpine project is a ground-up re-org of the pine source code with the addition of a really nifty web interface that is powered by tcl CGI scripts.

I managed to have just enough time today to grab the sources, do a build and get it installed and configured on my OS X workstation. It works beautifully with GMail and runs fine on the Mac – even opening attachments, but you can look forward to a full review after I’ve been using it for a few days. I am very eager to get the web version up and running as well if only to have a fallback in the even GMail “goes away”.

Sometimes you find a nifty piece of hardware that you just can’t let go into disuse. This time around it was a Tualatin Pentium 3-S 1266MHz CPU new-old stock, new-in-box. I got it some time ago to upgrade a PC for family that it turns out just upgraded the whole system instead. Thus it sat around in the box until I ran across a mobo to drop it in. Recently I found a system at my favorite shopping destination () that had a Tualatin capable mobo; the DFI CM33-TL just so happens to max out a the 1.26Ghz P3-S I already had. Even nicer is that it is the Rev C board which with the newer BIOS updates can boot from USB and can do 48-bit ATA addressing. Alas, no AGP slot. So why all the love for an old P3 server chip? The later P3-S could outperform the early P4 chips and use half the wattage! So what do we do when all that we have to boot the system with is a non-Microsoft OS and most BIOS update utilities run in Windows, or use disk creation software the runs in Windows/DOS? Luckily it seems that is is possible to update some mobos without having to resort to using an unwanted OS. DFI has made the CM33-TL able to boot from a floppy or run a program under Windows to flash the BIOS - or enter an update mode that simply reads the flash utility and BIOS file from a floppy. It turns out that it is a good thing they enabled all three. Under a fairly standard Ubuntu Linux install I was able create a floppy the the DFI board could update from by combining the BIOS update features in a way DFI didn’t document.

Several steps that worked for me:
1. Nab the BIOS update of choice for your mobo & revision. Be sure your file is correct - close doesn’t cut it with a BIOS. It’s either an exact match or something won’t work right. In my case I could nab the smaller download intended for a Windows-based update utility.
2. Extract the .zip file containing the utility and BIOS image. Many of the .exe files manufacturers supply are programs meant to run under DOS or a DOS shell to create a disk image. By having the .zip we can get around that.
3. Copy the extracted files to a freshly formatted and tested floppy (basic FAT12/MS-DOS format is fine). Having a good floppy is very key to a successful flash. GIGO is an important point to consider when doing something that can brick a system.
4. Reboot the system and be ready to press the BIOS flash key(s) when prompted. On the CM33-TL you press Alt-F2 just after the RAM test and floppy seek.
5. The BIOS will then enter the flash update mode and read the floppy. If it determines the BIOS image is compatible it will begin to flash it to the BIOS chip.
6. Once it’s done enter the BIOS setup and “Load Safe Defaults”. This will let the BIOS set any settings that might cause the system to fail to boot. Go though the menus and set things as you need.
7. Test boot to be sure it works as before. Test boot again using the new features and marvel at the sudden uses that have opened up.

I had been concerned about having to make a bootable floppy for the update but the BIOS option to enter the update mode does not need a fully bootable floppy to operate.

With a system like this it is possible to operate a NAS system with large drives on a chip that boots from a USB thumb drive, operate on older, cheap RAM and uses little power. Having a system that boots from USB allows you to configure the server to spin down drives that are idle and save more power; an OS on a USB device will not need to spin up the main/RAID drives to write logs, etc. Smart choices of hardware can make a cobbled together server operate more efficiently.