Re: [lfs-support] build time comparison

2020-03-06 Thread Ken Moffat
On Fri, Mar 06, 2020 at 07:42:39AM -0600, Rob wrote:
> Ken Moffat  wrote:
> But for your SSD I'm guessing that you might not be using fstrim
> (either manually, or the ',discard' option in the fstab), and
> therefore writing a new block means it has sufficiently filled up
> that it needs to shuffle things about.
> 
> 
> I thought that trim stuff was all handled automagically by kernels now.

Unless somebody can find a link for that, I don't think so.  It
isn't very long since fedora were deciding whether or not fstrim
should be automatically enabled in their installer, and my memory
says that they went with the systemd service to run it daily.

ĸen
-- 
#define THE_TRUTH_AND_THATS_CUTTIN_ME_OWN_THROAT true
Al Viro, complaining about a 'TRUE' macro in staging

-- 
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style


Re: [lfs-support] build time comparison

2020-03-06 Thread Bruce Dubbs

On 3/6/20 6:11 AM, Thomas Seeling wrote:

Hallo,


I did some comparing of build times.
Building was done on an older i5-6500 with 16 GB RAM, SATA harddisk and SSD.

9.1 build on SSD:
The SBU unit value is equal to 117 seconds.
Total time required to build the system:79.6  SBU

8.3 build on HD:
The SBU unit value is equal to 107 seconds.
Total time required to build the system:205.7  SBU

9.1 build on HD:
The SBU unit value is equal to 107 seconds.
Total time required to build the system:294.9  SBU


I'm not surprised about the speed difference but I wonder why the SBU is
bigger for the SSD build? I was using the same jhalfs configuration file
for the latest 9.1 build.


A long time ago I did an LFS build on a ram disk and compared the time 
to building on an actual disk.  This was before SSDs were available. 
The difference in build times was about 8.5%.


The build time in jhalfs can vary quite a bit.  For my latest reference 
build at -j1 on an ssd the time was 214 SBU at 104 seconds/SBU.


I sometimes build without tests or just a few tests and have seen times 
as low as 29 SBU at -j10 on the same system.


  -- Bruce
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style


Re: [lfs-support] build time comparison

2020-03-06 Thread Thomas Seeling
Hallo Ken,

>> 9.1 build on SSD:
>> The SBU unit value is equal to 117 seconds.
>> Total time required to build the system: 79.6  SBU

> The SBU in LFS is related to the _host_ system

I build LFS on its older version. If it works I build with that LFS to
the 2nd partition (I'm using a concept similar to Android's A/B system).

It's been a long time since I built LFS on a different host system.

Tschau...Thomas
--
Do you wanna be a legend or a passing footprint on the sands of time?
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style


Re: [lfs-support] build time comparison

2020-03-06 Thread Rob
Ken Moffat  wrote:
But for your SSD I'm guessing that you might not be using fstrim
(either manually, or the ',discard' option in the fstab), and
therefore writing a new block means it has sufficiently filled up
that it needs to shuffle things about.


I thought that trim stuff was all handled automagically by kernels now.
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style