PDA

View Full Version : Why does reboot take so long?


opus472
04-20-2008, 01:11 AM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...

steve614
04-20-2008, 02:22 AM
Good question.
I would say it has to do with the processor, but I could be wrong.

What is Tivos processor equivelant to? A Pentium I ? :p

bkdtv
04-20-2008, 03:05 AM
TiVo runs their own software on top of Linux.

The high-definition TiVos use a MIPS 300MHz processor comparable in performance to an Intel Celeron 300, but with built-in MPEG decoding with dedicated hardware acceleration for DVR functions.

A number of other consumer electronics devices are also based on Linux and boot much faster because they load an image from flash memory (similar to suspend/resume on a PC). TiVo could do something like that with a future model.

bkdtv
04-20-2008, 03:18 AM
To the above, I would add that TiVo would be wise to consider a move to a flash-based design for its system software. In the past, one had to trade off capacity (and thus software upgrade potential) with flash, but that is no longer the case.

Currently, TiVos start to crash / reboot when a problem develops with the software. A flash-based design would ensure that the software remains stable when the hard disk starts to develop problems.

ZeoTiVo
04-20-2008, 07:41 AM
To the above, I would add that TiVo would be wise to consider a move to a flash-based design for its system software. In the past, one had to trade off capacity (and thus software upgrade potential) with flash, but that is no longer the case.

Currently, TiVos start to crash / reboot when a problem develops with the software. A flash-based design would ensure that the software remains stable when the hard disk starts to develop problems.


I agree with everything you say excpet that you get 2 out of 3 of large, cheap, reliable.
Can TiVo truly get a large enough flash memory hardware that it can rely on while not jacking the price up too much?

The series 4 is at least a year off. maybe by then?

classicsat
04-20-2008, 08:03 AM
If they can give away thumb drives, yes.

If it were me, I'd run a boot OS and have the loopsets and root database on the Flash drive.

Really though, it isn't the drive interface slowing things down, it is a whole bunch of little procedures and modules it is loading taking time, as well as the kernel security check. I noticed when I hacked my 240, that it booted that much faster when I disabled the kernel security check.

chip_r
04-20-2008, 08:26 AM
I'd say the slow boot time is because it's Linux on a low powered processor. If a system is architected properly, it should use almost all of it's bandwidth during normal operation with little to spare. The Tivos are not like PCs where PC makers don't know what applications will be run and the user directly sees performance increases for higher powered CPUs, more memory, etc (i.e. more hardware costs). CE makers have a compelling need to "right size" hardware to their application. Boot time is a busy time for any OS, so it's slow.

Flash would be installed on the mainboard and flash does fail. Yep, you could use removable flash but that's a compatibility headache for Tivo and how would you reload it, your hard disk?

Tivo partitioned it right. All of the higher reliability parts on the mainboard, hard drive and power supply external.

sinanju
04-20-2008, 08:28 AM
Why should it be fast?

In an ideal world, the TiVo never reboots. In the real world (if it's on a UPS), it reboots maybe twice a year. To refer to the OPs comparison, many folk have Windows up and down more than once a day -- no wonder MSFT is interested in a fast boot -- it's a big part of the Windows experience.

sdavis139
04-20-2008, 09:07 AM
Why should it be fast?

In an ideal world, the TiVo never reboots. In the real world (if it's on a UPS), it reboots maybe twice a year. To refer to the OPs comparison, many folk have Windows up and down more than once a day -- no wonder MSFT is interested in a fast boot -- it's a big part of the Windows experience.

True, but it still shouldn't take 5 minutes to reset - especially on a Linux based OS. I have an old desktop with 512Mb running Ubuntu that does better! ;)

Anyway, the fact that TiVo uses Linux does mean that it should stay running longer... much longer than any Windows machine, and without the weekly patches / forced reboot cycles!!

ZeoTiVo
04-20-2008, 09:21 AM
True, but it still shouldn't take 5 minutes to reset - especially on a Linux based OS. I have an old desktop with 512Mb running Ubuntu that does better! ;)

So does your desktop get two tuners set to recieve input - possibly having to decrypt it or decode/encode to mpeg2 and then write both of those streams to the same hard drive it is booting from along with adding in encryption of those writes to the drive?

Oh and also look for scheduled events (recording) that should be happening and perform those if needed.

This is all in addition to the mormal startup items.

I think if so then the desktop just might boot a little slower

ewilts
04-20-2008, 09:48 AM
To the above, I would add that TiVo would be wise to consider a move to a flash-based design for its system software. In the past, one had to trade off capacity (and thus software upgrade potential) with flash, but that is no longer the case.
Spoken by a person who doesn't understand flash-based storage.

There are many different types of flash and believe it or not, some are actually *slower* than a decent hard drive for some types of operations (writes in particular).

On a gigabyte basis, hard drives are considerably cheaper than decent flash drives - yes, you *DO* trade off capacity when going from hard disk to flash. On a IOPS basis, it can go either way depending on the operations you're doing.

Those little thumb drives they give away at conferences aren't all that great and definitely not something I'd want to see in my TiVo.

bizzy
04-20-2008, 11:15 AM
So does your desktop get two tuners set to recieve input - possibly having to decrypt it or decode/encode to mpeg2 and then write both of those streams to the same hard drive it is booting from along with adding in encryption of those writes to the drive?

Oh and also look for scheduled events (recording) that should be happening and perform those if needed.

This is all in addition to the mormal startup items.

I think if so then the desktop just might boot a little slower

None of that is happening during the boot process. But thanks for the misguided attempt at sarcasm.

aaronwt
04-20-2008, 11:20 AM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...


Windows Vista boots up very fast for me on my 3 Vista machines. Faster than my Xp machines.

bkdtv
04-20-2008, 04:47 PM
Spoken by a person who doesn't understand flash-based storage.

There are many different types of flash and believe it or not, some are actually *slower* than a decent hard drive for some types of operations (writes in particular).

On a gigabyte basis, hard drives are considerably cheaper than decent flash drives - yes, you *DO* trade off capacity when going from hard disk to flash. On a IOPS basis, it can go either way depending on the operations you're doing.I think you are a bit confused. Nowhere did I suggest that TiVo completely replace the hard drive with flash memory.

I suggested that TiVo follow the same model used in other modern CE devices -- such as cell phones and upcoming Blu-ray players. This puts the OS software on non-volatile flash as a pre-booted Linux image. This is much faster, because you are loading a 'pre-booted' image into memory rather than booting the OS. Recordings would still be stored on the hard drive.

The Linux OS and TiVo software total less than 400Mb. Total boot time with this implementation should be around 30 seconds.

Those little thumb drives they give away at conferences aren't all that great and definitely not something I'd want to see in my TiVo.Nor did I suggest that. I'm talking about non-volatile flash memory devices of appropriate size (i.e. 512MB-1GB) specifically designed for this purpose. A number of manufacturers produce these flash devices for cell phones, music players, and other consumer electronics devices. With the aquisition of M-Systems, Sandisk is perhaps the best known manufacturer of these disk-on-a-chip (http://www.sandisk.com/OEM/ProductCatalog(1276)-SanDisk_Modular_Flash_Disks.aspx) devices. Here is their marketing video (http://www.youtube.com/watch?v=-NkyRpcwaX8) on YouTube.

These products have been around for awhile, but only in the past year or so did they become available in volume at capacities greater than 256Mb. Sandisk's product is now available in volume at 512MB, 768MB, 1GB, and 2GB sizes.

lrhorer
04-20-2008, 05:52 PM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...
Well, the boot process does indeed take a while, but there's a lot going on there. Your comparison to Vista is apples and oranges. Fudging just a bit, when Vista is done booting, your PC is only more or less at the point where the "Just a few more minutes" screen pops up. A better comparison would be to set your Windows Startup to automatically run MS Access and load a 500M database file, run a web server, and start playing two HD videos. Start your stopwatch when the power is turned on and stop it when all the applications are fully up and running.

Of course, no matter what, the comparison isn't going to be perfectly one for one, and it's quite possible your PC might still beat the TiVo, but as others have pointed out, fast booting is just not a high priority in a DVR.

lessd
04-20-2008, 05:57 PM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...

This is just my opion as an answer to your question

When I had a DVR cable box it would boot in less than 30 seconds except there would be no guide date in the box, it would take an hour before you would have a days worth of guide data and a day or so before you would have all the two weeks of guid data. I think when TiVo re-boots it has to re index all the guide date because it make no assumptions as to how long the TiVo has been off. This takes time but in about 4 minutes you have your TiVo back with ALL the guide data out about two weeks (unless the TiVo was off for more than a day )

Kenji
04-21-2008, 04:58 PM
TiVo runs their own software on top of Linux.
There's probably also some disk-checkng (fsck) time thrown in.

Da Goon
04-21-2008, 05:07 PM
There's probably also some disk-checkng (fsck) time thrown in.

it runs e2fsck on the /var partition on each boot. if /var gets rebuilt, it only takes about 30 additional seconds.

mattack
04-21-2008, 10:09 PM
BTW, my S1s are WAY WAY WAY WAY faster in booting than my S3 or TivoHD.

Also, the S3 & TivoHD do some kind of channel scan upon reboot too, which takes a while.

ciper
04-21-2008, 10:10 PM
My S1 with 700gb and a 512mb cachecard is also faster than my THD.

CrashHD
04-21-2008, 10:47 PM
With Series2 units, half the boot time is tied up in security checks (initrd verifying files/signatures, etc). My units booted in about 6 minutes unmodified, and about 3 minutes after hacking. I'd call it a reasonable assumption to assume this is still the case with Series3 units.

cramer
04-21-2008, 11:44 PM
A) PROM verification
B) kernel + initrd verification
- and the real time waster -
C) filesystem "anti-tamper" verification

Disable all three of those and the thing will boot in seconds... just like a S1.

JYoung
04-21-2008, 11:45 PM
I timed my TiVo boot times a couple of months back:
Series 1 1:05
Series 2 5:02
Series 3 4:12
TiVo HD 4:44


BTW, anyone ever reboot a mainframe before?
I want to say it took about 10 minutes. Maybe 15.

opus472
04-22-2008, 12:33 AM
I timed my TiVo boot times a couple of months back:
Series 1 1:05
Series 2 5:02
Series 3 4:12
TiVo HD 4:44


Seems like about 3 days when you've got the cable service guy there waiting for it...

BobB
04-22-2008, 10:05 AM
Why should it be fast?

In an ideal world, the TiVo never reboots. In the real world (if it's on a UPS), it reboots maybe twice a year.

That's pretty much true of S1 and S2 units, but absolutely not of the HD ones. The HD boxes are so buggy (at least compared to the older ones) that I find I need to force a reboot a couple of times a month to clear up one problem or another (certain recordings won't play, the screen goes black except for the menu, etc). So far (knock wood!) a reboot has always fixed things.

CrashHD
04-22-2008, 12:44 PM
A) PROM verification
B) kernel + initrd verification
- and the real time waster -
C) filesystem "anti-tamper" verification

Disable all three of those and the thing will boot in seconds... just like a S1.
It will boot faster, but not in seconds. I have R10's with prom mods, which I believe disables all three of the above. Boot time is still in the 2 1/2 - 3 minutes range.

Adam1115
04-22-2008, 02:15 PM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...

It doesn't.

My vista reboot takes about 3 minutes to boot. I have to reboot it every week, so in the last 6 months it's taken 3 minutes x 26 weeks = 78 minutes of rebooting.

I've only rebooted my TiVo once in the last six months, it didn't take 78 minutes.

Edited to fix math. ;)

stevevt
04-22-2008, 02:24 PM
...3 minutes x 7 days a week x 26 weeks = 9.1 hours of rebooting.


You reboot your PC every day?

ZXTT95
04-23-2008, 10:11 AM
Some comments:

Flash: I don't see what flash adds, other than cost and another point of failure. An image can be loaded from the hard disk just as easily as from flash. If you're concerned about the integrity of the disk, then you put a CRC on the image. You could even store two images so that if the first failed the CRC check, you could load the second. Since flash is generally better at random access today, loading one large image would probably be faster from disk. The reason why some devices use flash is because they don't have disks.

Re-indexing the guide data: I can see why you'd do an integrity check, but I don't see the need for a re-index. An index is specific to the current state of the data and that hasn't changed when the TiVo is off.

Time to start tuning, playing, etc.: These are all normal functions that the TiVo does very quickly when running. I don't see how they would add much to the startup time.

AbMagFab
04-23-2008, 10:34 AM
To the earlier poster, Linux booted from a USB flash drive is WAY SLOWER than off a hard drive. Like 5-10x slower. Both USB speeds, but more importantly the flash drive speeds are the problem.

SATA drives can transfer at up to 3Gb/s. That's pretty fast. The drive isn't the bottleneck.

classicsat
04-23-2008, 11:27 AM
There are many different types of flash and believe it or not, some are actually *slower* than a decent hard drive for some types of operations (writes in particular).

They all fundamentally work the same, at least on the scale we'd need (1GB range). Just the interface to the outside world would change.

As for write speed, you only need do that once, or not very often at most. All of the writing will be done to the HDD, including software upgrades.


Those little thumb drives they give away at conferences aren't all that great and definitely not something I'd want to see in my TiVo.

The mention of them (at least by myself) is just a point of reference. Because they are cheap, the good stuff can be cheaper, although a premium paid for reliability.

classicsat
04-23-2008, 11:43 AM
I suggested that TiVo follow the same model used in other modern CE devices -- such as cell phones and upcoming Blu-ray players. This puts the OS software on non-volatile flash as a pre-booted Linux image. This is much faster, because you are loading a 'pre-booted' image into memory rather than booting the OS. Recordings would still be stored on the hard drive.

The Linux OS and TiVo software total less than 400Mb. Total boot time with this implementation should be around 30 seconds.

I suggest this:

A minimal boot OS +root database on a DOC, flashed at factory. The boot OS would check for the drive and the main OS and boot image integrity , and check for Kickstart commands. If no OS is on the drive, it would create a drive structure, and download/install the main OS.

The Main OS would boot, and when booted, create a boot image (instant loading ram image). If need be, the boot image can be created, and would be for major changes, such as repeat guided setup, and could be redone with a kickstart code, or from the restart recorder menu.


Nor did I suggest that. I'm talking about non-volatile flash memory devices of appropriate size (i.e. 512MB-1GB)

Some of those el-cheapo MP3 players have minimal boot code embedded on the system chip, and boot from the NAND Flash chip, or HDD.

classicsat
04-23-2008, 11:46 AM
Some comments:

Flash: I don't see what flash adds, other than cost and another point of failure. An image can be loaded from the hard disk just as easily as from flash.

It adds redundancy, in that a corrupted or blank HDD could be loaded from the Flash chip.

ZXTT95
04-23-2008, 12:02 PM
SATA drives can transfer at up to 3Gb/s. That's pretty fast. The drive isn't the bottleneck.

Your bottom line is correct, but that 3 Gb/s number is a little misleading.

3 Gb/s is the SATA/300 interface's raw bit rate. 20% of this is taken up by the 8b10b encoding (every 8 bits sent is encoded as 10 bits), so the actual bit rate before command overhead and latencies is 2.4 Gb/s = 300 MB/s (decimal MB!), hence the name SATA/300. That's still multiple times the fastest 7200 RPM drive's sustained read performance.

My real-world experience is that I get ~60 MB/s over SATA (or back-to-back Gigabit Ethernet for that matter), but ~30 MB/s over USB. That's still much faster than the USB flash drives I have which seem to top out at ~10-20 MB/s (closer to 10 I think). These hard drive numbers are seen when transferring large files, however. With smaller files, resulting in more random access, the hard drives slow down, while flash speed usually stays constant.

ZXTT95
04-23-2008, 12:11 PM
It adds redundancy, in that a corrupted or blank HDD could be loaded from the Flash chip.

Granted, although 64 KB should be enough for a boot ROM capable of downloading the full OS from the internet. I could accept that this boot ROM would be in flash so it can be updated, but storing the whole OS in flash seems unnecessary.

mattack
04-23-2008, 09:49 PM
You reboot your PC every day?

I do.. I call it "powering on in the morning" though. Why waste power when I'm never ever going to be connecting to it elsewhere?

(Yes, I feel guilty that my Tivos, and now a router I have plugged in just so my Tivos can talk to each other and I can download shows over wireless, use so much power all the time..)

kb7oeb
04-25-2008, 03:00 AM
It doesn't seem unusual if you think of how long a 300Mhz celeron took to boot

demon
04-26-2008, 10:54 PM
I do.. I call it "powering on in the morning" though. Why waste power when I'm never ever going to be connecting to it elsewhere?

(Yes, I feel guilty that my Tivos, and now a router I have plugged in just so my Tivos can talk to each other and I can download shows over wireless, use so much power all the time..)

Um, you do realize that most modern systems support suspend/sleep mode, so you can just put the machine to sleep with minimal power draw, and save yourself the startup/shutdown time (and all the wasted power involved in restarting everything), right? And the TiVo's power supply takes easily a fifth to a tenth of the typical power draw of a modern desktop PC...

aaronwt
04-26-2008, 11:42 PM
(Yes, I feel guilty that my Tivos, and now a router I have plugged in just so my Tivos can talk to each other and I can download shows over wireless, use so much power all the time..)


Those devices barely use any power. I have several PCs that are on all the time constantly drawing 500watts each. A router barely draws anything and a TiVo doesn't draw much either.
And I assume you have no incandescent bulbs either. That would be wasting more power than the TiVo.

VCD
04-27-2008, 02:12 AM
Any idea why a reboot takes so long on these boxes? Way longer than even Windows Vista...

It does not matter. Reboot, go make a sandwich and come back. All done.

CrashHD
04-27-2008, 09:12 AM
It doesn't seem unusual if you think of how long a 300Mhz celeron took to boot

Using period correct software and hardware with that 300 MHz celeron gives a boot time around 30-40 seconds. NT4 or Win95, 64MB ram, and a 5400rpm 8.4GB hard drive. That's booting a system with an operating system in the range of about 1 gigabyte in size, with a host of different drivers, networking protocols, etc that all have to be loaded.

Now that same hardware with non-period correct(yes, a 300mhz celeron is that old) software, is different. It will struggle to load XP, and surely die under the weight of vista.

Apologies to the OP, as I'm going off topic. My question (purely curiosity). As the Series2 hardware hasn't changed, how has it's software bloated as it went from v3.whatever.it.was.at.S2.introduction to the 9.whatever.it.is.now?

To get back on topic, and answer the original question "Why does reboot take so long?" as compared to Vista, here's what I think:
PC's/OS's are designed with consideration given to boot time. Usually, while a PC is booting, a person is waiting. Tivo is a little different. With the exceptions of installation, power failures, or a malfunction, there is no reason an average joe end user should ever witness a Tivo booting.

lessd
04-27-2008, 11:53 AM
Using period correct software and hardware with that 300 MHz celeron gives a boot time around 30-40 seconds. NT4 or Win95, 64MB ram, and a 5400rpm 8.4GB hard drive. That's booting a system with an operating system in the range of about 1 gigabyte in size, with a host of different drivers, networking protocols, etc that all have to be loaded.

Now that same hardware with non-period correct(yes, a 300mhz celeron is that old) software, is different. It will struggle to load XP, and surely die under the weight of vista.

Apologies to the OP, as I'm going off topic. My question (purely curiosity). As the Series2 hardware hasn't changed, how has it's software bloated as it went from v3.whatever.it.was.at.S2.introduction to the 9.whatever.it.is.now?

To get back on topic, and answer the original question "Why does reboot take so long?" as compared to Vista, here's what I think:
PC's/OS's are designed with consideration given to boot time. Usually, while a PC is booting, a person is waiting. Tivo is a little different. With the exceptions of installation, power failures, or a malfunction, there is no reason an average joe end user should ever witness a Tivo booting.

So TiVo designed a slow boot because most people will not notice it ? There is a reason, it just that we don't know what it is. The Cable Moto box reboots in about 30 seconds, but you have no guide data for about an hour, that the only difference I can see between the two DVRs after a boot (I not talking about operation or interface).

ToddNeedsTiVo
04-27-2008, 12:12 PM
A router barely draws anything and a TiVo doesn't draw much either.

Our power went out the other day for 2.5 hours. I have three small UPS units, same model and age: computer desk upstairs, TiVo HD in living room, and cable modem/router in basement.

I don't know exactly how long the computer or TiVo stayed up because only my wife was home at the beginning of the outage (but apparently under five minutes), but the third UPS kept the router/cable modem going for 2 full hours. The cable service itself didn't go out. She was able to surf on her laptop, which obviously had its own power, quite nicely with a flashlight bouncing off the ceiling. I was impressed with that amount of time for that third UPS. At four years old, I think their batteries are about shot. So, the router/cable modem draw just a tiny trickle.

The most frustrating part was that in five years we've lost power for maybe six hours. This time, though, it cost me the episode of Lost!

vstone
04-27-2008, 12:48 PM
I timed my TiVo boot times a couple of months back:
Series 1 1:05
Series 2 5:02
Series 3 4:12
TiVo HD 4:44


BTW, anyone ever reboot a mainframe before?
I want to say it took about 10 minutes. Maybe 15.We had to reboot a mainframe back in '93. It never came back up. Good thing we were migrating to a LAN!:p

vstone
04-27-2008, 12:58 PM
I would imagine that Tivo puts boot time near the bottom of their priority list.

Howver, Vista can take advantage of some flash memory built into the new hybrid hard drives. The theory is that Vista can boot faster with booting from the flash memory than from the hard drive. So far I think the hybrid drives are only available at 2-1/2" size. I saw something the other day indicating that PC's would transition to 2-1/2" drives (I guess laptops will go to 1-3/8" drives). From the discussion here I'm not sure that a Tivo would benefit much from booting from flash memory, but if it would then maybe the S4 (or at least some variant) will run off of hybrid drives.

aaronwt
04-27-2008, 07:38 PM
Our power went out the other day for 2.5 hours. I have three small UPS units, same model and age: computer desk upstairs, TiVo HD in living room, and cable modem/router in basement.

I don't know exactly how long the computer or TiVo stayed up because only my wife was home at the beginning of the outage (but apparently under five minutes), but the third UPS kept the router/cable modem going for 2 full hours. The cable service itself didn't go out. She was able to surf on her laptop, which obviously had its own power, quite nicely with a flashlight bouncing off the ceiling. I was impressed with that amount of time for that third UPS. At four years old, I think their batteries are about shot. So, the router/cable modem draw just a tiny trickle.

The most frustrating part was that in five years we've lost power for maybe six hours. This time, though, it cost me the episode of Lost!

That's why I have a 1500VA UPS with the extended runtime battery for all my TiVos.(along with other components connected) I get 3 to 4 hours of battery backup which covers those longer power outages.

lrhorer
04-28-2008, 09:07 PM
You reboot your PC every day?
Those running Windows at least that, and often more. The machine at which I am sitting right now must usually be rebooted several times a day, and must be rebooted every time the VPN is either enabled or shut down, which generally happens at least 3 times a day. Otherwise a number of things quit working. The machine on my desk at work gets rebooted at least once or twice a day. I do have one Windows file server which only gets rebooted every 2 or 3 weeks or so.

By comparison, the last time any of my TiVos were rebooted was when they upgraded to 9.3 a couple of weeks ago. Before that, I can't recall the last time either the TiVo HD or the S3 in the theater was rebooted - at least 2 months. The S3 in the Livingroom has an EMP problem which causes it to hang once in a while when the TV is turned on, so it's uptime is usually only a couple of weeks or so, but even it is nowhere as unstable as the Windows workstations.

It's been at least 6 months since either of the Linux servers in my house were rebooted, and one of the HP-UX systems I administer at work hasn't been rebooted since 2005. Another was rebooted 4 months ago.

Da Goon
04-28-2008, 09:30 PM
There is a reason, it just that we don't know what it is.

An unhacked box starts off for nearly 2 minutes checking the hashes of every file in the root filesystem to make sure it hasn't been tampered with. The tivo's logs in /var (partition 9 - particularly the kernel log) will tell you exactly why it takes so long, step by step. Here's a boot log from an unhacked S3 :

revision is: 0001810b
FPU revision is: 0003810b
Primary instruction cache 32kb, linesize 32 bytes (2 ways)
Primary data cache 32kb, linesize 32 bytes (2 ways)
Linux version 2.4.20 (build@buildmaster52) (gcc version 3.3.4) #1 Fri Jul 14 03:00:53 PDT 2006
Determined physical RAM map:
memory: 07e27000 @ 001d9000 (usable)
Initial ramdisk at: 0x8018a000 (320381 bytes)
On node 0 totalpages: 32768
zone(0): 32768 pages.
zone(1): 0 pages.
zone(2): 0 pages.
Kernel command line: root=/dev/hda7 dsscon=true console=1,115200
Enable the cache parity protection for MIPS 5KC CPUs.
Monotonic time calibrated: 148.67 counts per usec
Calibrating delay loop... 296.55 BogoMIPS
Contiguous region 1: 12582912 bytes @ address 0x8058a000
Contiguous region 2: 2097152 bytes @ address 0x8118a000
Contiguous region 3: 5402624 bytes @ address 0x8138a000
Contiguous region 8: 67108864 bytes @ address 0x818b1000
Contiguous region of 87191552 bytes total reserved at 0x8058a000.
Memory: 42612k/129180k available (1216k kernel code, 86568k reserved, 65k data, 84k init, 0k highmem)
Dentry cache hash table entries: 16384 (order: 5, 131072 bytes)
Inode cache hash table entries: 8192 (order: 4, 65536 bytes)
Mount-cache hash table entries: 2048 (order: 2, 16384 bytes)
Buffer-cache hash table entries: 8192 (order: 3, 32768 bytes)
Page-cache hash table entries: 32768 (order: 5, 131072 bytes)
Checking for 'wait' instruction... available.
POSIX conformance testing by UNIFIX
BCM7042 MPEG2 Encoder on PCI slot 0xc @ d2000000
BCM7042 MPEG2 Encoder on PCI slot 0xd @ d2400000
BCM7411 Decoder on PCI slot 0xe @ d3000000
cmd 0x2a00002 0x2a00002 BAR0 0xd4000000 BAR1 0xd3000000
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
Broadcom BCM7038IDE: IDE controller on PCI bus 01 dev 00
Broadcom BCM7038IDE: chipset revision 0
Broadcom BCM7038IDE: 100% native mode on irq 49
ide0: BM-DMA at 0x0300-0x0307, BIOS settings: hda:DMA, hdb:DMA
ide1: BM-DMA at 0x0308-0x030f, BIOS settings: hdc:DMA, hdd:DMA
hda: WDC WD2500BS-55RPB1, ATA DISK drive
ide0 at 0x200-0x207,0x242 on irq 49
svwks_ide_dma_speed: hda: mode 0x03, speed 0x45
blk: queue 80180d50, I/O limit 4095Mb (mask 0xffffffff)
hda: 488397168 sectors (250059 MB) w/2048KiB Cache, CHS=30401/255/63, (U)DMA
Partition check:
hda: [mac] hda1 hda2 hda3 hda4 hda5 hda6 hda7 hda8 hda9 hda10 hda11[M] hda12 hda13[M]
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
PPP generic driver version 2.4.2
PPP Deflate Compression module registered
TiVo disk statistics module loaded
Disk cautionlevel is 3
funcount disabled
Kernel Panic Logger registered
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 1024 buckets, 8Kbytes
TCP: Hash tables configured (established 8192 bind 16384)
IP-Config: No network devices available.
ip_conntrack version 2.1 (1024 buckets, 8192 max) - 152 bytes per conntrack
ip_tables: (C) 2000-2002 Netfilter core team
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
RAMDISK: Compressed image found at block 0
Freeing initrd memory: 312k freed
VFS: Mounted root (romfs filesystem) readonly.
Running as /linuxrc - autoscan!
hda: Generic ATA management
Loading signatures file
4037 valid entries loaded
Scan /mnt
Scan /mnt/lost+found
Scan /mnt/bin
Scan /mnt/etc
Scan /mnt/etc/hotplug
Scan /mnt/etc/mempools
Scan /mnt/etc/rc.d
Scan /mnt/etc/rc.d/StageA_PreKickstart
Scan /mnt/etc/rc.d/StageB_PostKickstart
Scan /mnt/etc/rc.d/StageC_MediaInitialization
Scan /mnt/etc/rc.d/StageD_PreMfs
Scan /mnt/etc/rc.d/StageE_PreApplication
Scan /mnt/etc/rc.d/StageF_ApplicationLaunch
Scan /mnt/etc/rc.d/StageG_PostApplication
Scan /mnt/etc/tivoconfig
Scan /mnt/etccombo
Scan /mnt/lib
Scan /mnt/lib/modules
Scan /mnt/opt
Scan /mnt/opt/tivo
Scan /mnt/sbin
Scan /mnt/tvbin
Scan /mnt/tvlib
Scan /mnt/tvlib/firmware
Scan /mnt/tvlib/firmware/cdc
Scan /mnt/tvlib/firmware/prism2
Scan /mnt/tvlib/font
Scan /mnt/tvlib/font/dtvcc
Scan /mnt/tvlib/idl
Scan /mnt/tvlib/itcl3.2
Scan /mnt/tvlib/misc
Scan /mnt/tvlib/modem
Scan /mnt/tvlib/modem/patches
Scan /mnt/tvlib/modem/patches/P2107-V90
Scan /mnt/tvlib/modem/patches/P2107-V90/ram
Scan /mnt/tvlib/modem/patches/P2109-V90
Scan /mnt/tvlib/modem/patches/P2109-V90/ram
Scan /mnt/tvlib/modem/patches/Si2433
Scan /mnt/tvlib/modem/patches/Si2433/C
Scan /mnt/tvlib/modem/patches/Si2433/C/AT
Scan /mnt/tvlib/modem/patches/Si2434
Scan /mnt/tvlib/modem/patches/Si2434/B
Scan /mnt/tvlib/modem/patches/Si2434/B/AT
Scan /mnt/tvlib/modem/patches/Si2456
Scan /mnt/tvlib/modem/patches/Si2456/D
Scan /mnt/tvlib/modem/patches/Si2456/D/AT
Scan /mnt/tvlib/modem/utils
Scan /mnt/tvlib/tcl
Scan /mnt/tvlib/tcl/encoding
Scan /mnt/tvlib/tcl/http
Scan /mnt/tvlib/tcl/http1.0
Scan /mnt/tvlib/tcl/msgcat
Scan /mnt/tvlib/tcl/opt
Scan /mnt/tvlib/tcl/reg
Scan /mnt/tvlib/tcl/tcldom
Scan /mnt/tvlib/tcl/tcllib
Scan /mnt/tvlib/tcl/tcllib/base64
Scan /mnt/tvlib/tcl/tcllib/cmdline
Scan /mnt/tvlib/tcl/tcllib/comm
Scan /mnt/tvlib/tcl/tcllib/counter
Scan /mnt/tvlib/tcl/tcllib/crc
Scan /mnt/tvlib/tcl/tcllib/csv
Scan /mnt/tvlib/tcl/tcllib/des
Scan /mnt/tvlib/tcl/tcllib/exif
Scan /mnt/tvlib/tcl/tcllib/fileutil
Scan /mnt/tvlib/tcl/tcllib/ftp
Scan /mnt/tvlib/tcl/tcllib/inifile
Scan /mnt/tvlib/tcl/tcllib/log
Scan /mnt/tvlib/tcl/tcllib/math
Scan /mnt/tvlib/tcl/tcllib/md4
Scan /mnt/tvlib/tcl/tcllib/md5
Scan /mnt/tvlib/tcl/tcllib/md5crypt
Scan /mnt/tvlib/tcl/tcllib/mime
Scan /mnt/tvlib/tcl/tcllib/multiplexer
Scan /mnt/tvlib/tcl/tcllib/report
Scan /mnt/tvlib/tcl/tcllib/sha1
Scan /mnt/tvlib/tcl/tcllib/struct
Scan /mnt/tvlib/tcl/tcllib/uri
Scan /mnt/tvlib/tcl/tcltest
Scan /mnt/tvlib/tcl/tclxml
Scan /mnt/tvlib/tcl/tv
Scan /mnt/platform
Scan /mnt/platform/etc
Scan /mnt/platform/etc/hotplug
Scan /mnt/platform/etc/mempools
Scan /mnt/platform/lib
Scan /mnt/platform/lib/modules
Scan /mnt/platform/sbin
Scan /mnt/platform/utils
Scan /mnt/platform/utils/DeviceList
Scan /mnt/var
Scan /mnt/proc
Scan /mnt/install
Scan /mnt/dist
Scan /mnt/mnt
Scan /mnt/mnt/cdrom
Scan /mnt/mnt/flash
Scan /mnt/initrd
Scan /mnt/dev
Scan /mnt/res
The filesystem seems to be OK
Scanner main is done
VFS: Mounted root (ext2 filesystem) readonly.
Trying to move old root to /initrd ... okay
Freeing unused kernel memory: 84k freed
Starting rc.sysinit
Running boot Stage A_PreKickstart scripts
Scanning for configuration files
Loading tivoconfig.o
Invoking startup scripts for:
platform 'eiger'
implementation 'Gen05'
implementer 'TiVo'
Releasing /initrd and clearing ramdisk, if they exist
Activating swap partitions
Adding Swap: 131064k swap-space (priority -1)
Loading core system drivers
Loading gpio_Gen05.o
gpio_init: configured as Eiger P1.5/P2
Loading bcm7038tty.o
BCM7038 serial driver loaded, 2 ports starting at /dev/ttyS0
Loading modemtty_Gen05.o
Parallel modem driver loaded, 1 ports starting at /dev/ttyS3
Loading core I2C system drivers
Loading i2c_Gen05.o
Loading i2cflash.o
Loading ircatch_Gen05.o
Loading oled.o
found orange oled module
Found rev 'C' Si2434 modem on /dev/cua3
Checking for Kickstart panic signal
Running boot Stage B_PostKickstart scripts
Cleanup /dev/hda9 pass 1
ext2fs_check_if_mount: No such file or directory while determining whether /dev/hda9 is mounted.
/dev/hda9 was not cleanly unmounted, check forced.
Inode 2052, i_blocks wrong 116 (counted=114). Set i_blocks to counted? yes

Inode 2050, i_blocks wrong 1246 (counted=1240). Set i_blocks to counted? yes

Inode 2056, i_blocks wrong 754 (counted=748). Set i_blocks to counted? yes

Entry 'tcphonehome' in /tmp (4097) has deleted/unused inode 53249.
Clear? yes

Entry 'NetworkApp' in /tmp (4097) has deleted/unused inode 55297.
Clear? yes

Entry 'myworld' in /tmp (4097) has deleted/unused inode 57345.
Clear? yes

Inode 4097 has ref count 11, expecting 8.
Set i_nlinks to count? yes

Fix summary information? yes

/dev/hda9: 78/65536 files (5.1% non-contiguous), 11312/262144 blocks
Cleanup /dev/hda9 pass 2
ext2fs_check_if_mount: No such file or directory while determining whether /dev/hda9 is mounted.
/dev/hda9: clean, 78/65536 files, 11312/262144 blocks
/dev/hda9 is clean after pass 2
Mounting /var
/dev/hda9 on /var type ext2 (rw)
Cleaning up files in /var
Checking space in /var
Mounting initial environment
Starting logging daemons
Checking Panic log
Found Silicon Labs "Si2434" modem, rev C, skipping modem patch
Scanning for phase1 repair scripts
Running boot Stage C_MediaInitialization scripts
Loading input section drivers
Loading tvinput_Gen05.o
VpxOpen[0] chip rev 0x0402
TvOpenMsp[0] boardid = 102489
VpxOpen[1] chip rev 0x0402
TvOpenMsp[1] boardid = 102489
/dev/input loaded
Found Rev C2 of Micronas AVF 49xx.
Loading bcm7042.o
*** 00:00:00.000 encoder: Initializing 7042 device /dev/encoderB
*** 00:00:00.011 encoder: EncoderB device @ base addr d2000000 initialized
*** 00:00:00.018 encoder: Initializing 7042 device /dev/encoderA
*** 00:00:00.031 encoder: EncoderA device @ base addr d2400000 initialized
*** 00:00:00.085 encoder: EncoderA uCode received. Ready to load
*** 00:00:00.091 encoder: Offset: 209072 Length: 209072
*** 00:00:00.192 encoder: Firmware version 2.0813
*** 00:00:00.225 encoder: EncoderB uCode received. Ready to load
*** 00:00:00.230 encoder: Offset: 209072 Length: 209072
*** 00:00:00.332 encoder: Firmware version 2.0813
boardId 0x00102489 : loading fpga for P1.5 or later
Starting programming...
done!
Revision byte: 0x2D
boardId 0x00102489 : loading ati314.o for P2.0 or later
Loading ati314.o
NIM 0 found at address 0x14
Samsung (TDA6651) tuner detected
Downloading 314 demodulator microcode...done!
NIM 1 found at address 0x18
Samsung (TDA6651) tuner detected
Downloading 314 demodulator microcode...done!
Loading pod.o
I see a P1.5 (or later)!
FPGA revision: 2D
Pod module was successfully loaded at C0110060
Loading output section drivers
Loading bcm7038-C2.o
Setting global debug state to debug level 3
brcm contigmem region 8: start 0x018B1000 size 0x04000000
brcm contigmem region 1: start 0x0058A000 size 0x00C00000
bcm7038.o BCHP_Open7038() chip revision = 0x70380024

............... Setting Memory Client Priorities for 200MHz Eiger P1 platform - 2006/01/04
............... 200 MHz (PLL_FREQ_CNTL reg = 0x126)
cfg.ulBufferCnt_SD = 15 cfg.ulBufferCnt_HD = 5 XXX can these be reduced
### BASE_HDMI: No DVI/HDMI Rx Device Available
Initialized tv_bcmencryption, hXpt=0x86712000
Using SGAT, 3 worker slots
Initialized tv_bcmdisplay, hVdc=0x86b66400, s_DisplayHd=c0288b30
Initialized tv_bcmpcm, hXpt=0x86712000
Initialized tv_bcmdecode, hVdc=0x86b66400
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
Loading HDCP keys...
Splash the screen
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
Running boot Stage D_PreMfs scripts
Remote control is TIVO
MFS partition on /dev/hda10
Look for debug board
Loading Eiger podapp
Loading ubuddy.o
Loading TvBus router
Insmoding integrated ethernet port driver
***** HDCP Authentication Initialized
insmod /platform/lib/modules/bcmenet.o mac=XX:XX:XX:XX:XX:XX
Loading bcmenet.o
MAC initialized OK 00:11:D9:05:EC:02
enet : link down
Updating system clock
Time set to: Sun Sep 17 10:56:21 2006
Enabling local route
Setting TCP keepalive parameters
Checking for additional disk
Start fan control
First temperature parameters set:
Terminal temp: 75
Critical temp: 70
Logging temp: 65
Target temp: 48
Lowest fan speed: 7
/tvbin/fancontrol is running in the background.
Starting TvLauncher
Waiting for launcher to start.
Launcher is running.
Scanning for phase2 repair scripts
Running boot Stage E_PreApplication scripts
Checking for database conversions
No upgrade to load
NewSoftware: getting SwSystem name
NewSoftware: failure reading SwSystem master: 0x30001
Scanning for phase3 repair scripts
Running boot Stage F_ApplicationLaunch scripts
Starting Services.
/tvbin/podapp: ATSC/CATV PSIP/POD Manager Version 0.32, compiled on Jul 14 2006
/tvbin/podapp: Using PRODUCTION secrets
/tvbin/podapp: Found the right number of inputs
Microcode version is TiVo!
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
FIRMWARE: Got standby mode update. Standby: NO
Video feed not attached or on during construction
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
About to TmkInit
Logger not initialized! Logging to stdout:
TmkLogger: <133>Sep 17 10:57:31 TmkServer[338]: Enabling port 5353, protocol udp[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1
NetAppCeam.C - HandleHello (line 136): EAM system received Hello from UI session.
Scanning for phase4 repair scripts
Running boot Stage G_PostApplication scripts
Scanning for configuration files
Tuner lock acquired
rc.sysinit is complete
[TV_BCMDISPLAY_SET_VIDEO_OUTPUT_MODE] mode=1

mattack
04-28-2008, 09:39 PM
Um, you do realize that most modern systems support suspend/sleep mode, so you can just put the machine to sleep with minimal power draw, and save yourself the startup/shutdown time (and all the wasted power involved in restarting everything), right?

"minimal" != 0

There is "wasted power involved in restarting". That's illogical.

I just turn my machine on and go get coffee like I'd do anyway (though often it's close to finished booting/logging in by the time I leave my office).

mattack
04-28-2008, 09:40 PM
Those devices barely use any power.

"barely" != 0.

With everyone thinking like this, no wonder we've got global warming.

kb7oeb
04-29-2008, 12:49 AM
Those running Windows at least that, and often more. The machine at which I am sitting right now must usually be rebooted several times a day, and must be rebooted every time the VPN is either enabled or shut down, which generally happens at least 3 times a day. Otherwise a number of things quit working. The machine on my desk at work gets rebooted at least once or twice a day. I do have one Windows file server which only gets rebooted every 2 or 3 weeks or so.

By comparison, the last time any of my TiVos were rebooted was when they upgraded to 9.3 a couple of weeks ago. Before that, I can't recall the last time either the TiVo HD or the S3 in the theater was rebooted - at least 2 months. The S3 in the Livingroom has an EMP problem which causes it to hang once in a while when the TV is turned on, so it's uptime is usually only a couple of weeks or so, but even it is nowhere as unstable as the Windows workstations.

It's been at least 6 months since either of the Linux servers in my house were rebooted, and one of the HP-UX systems I administer at work hasn't been rebooted since 2005. Another was rebooted 4 months ago.

Windows is crap but if you have to reboot that often something is wrong.

opus472
04-29-2008, 12:48 PM
It does not matter. Reboot, go make a sandwich and come back. All done.

Yeah, well, should I also make a sandwich for the Comcast tech who's waiting around for the reboot?