Lenovo T400 report

I use third button only for copy/pasting (also) from xterm. Really miss this feature on other laptops..

I do not use hal nor fancy xorg mappings.

//By "trackpad scrolling" I meant trackpad edge scrolling, which is probably configurable, but not really needed IMHO. Moused works for me.
 
Thank you for reply.

I managed to force to work third button's paste function after I put in /boot/loader.conf file that line:

Code:
hw.psm.synaptics_support="1"

It should affect TouchPad but in my case it affected TrackPoint.

In my Slackware 12.2 and previous I used specific xorg.conf configuration to enable scrolling with TrackPoint. In Slackware 13.0 I have to use for these purposes HAL with special FDI file. In FreeBSD 7.2 any of these methods doesn't work. Without vertical scrolling reading Internet sites is a nightmare. Modern Internet sites are usually wider than my 4:3 screen so horizontal scrolling is also useful.
 
Third button pasting works here out of the box. As I said before, I only use moused.

It should affect TouchPad but in my case it affected TrackPoint.

I do not see how's that connected with Touchpad nor Trackpoint.

Without vertical scrolling reading Internet sites is a nightmare.

I use cursor keys, mostly. There is usually scroll bar in browser.
 
From manufacturers point of view keyboard with TrackPoint is one part and TouchPad with five buttons is the other part. IBM/Lenovo calls each such part FRU.

From drivers developer's point of view keyboard, TrackPoint, TouchPad and five buttons are separate devices. Each should be treated separately.

From user's point of view TrackPoint with three related buttons is one device and TouchPad with two related buttons is the other device.

I presented above user's point of view.

I changed some setting related to TouchPad and it affected TrackPoint's button. This is the essence of my statement.
 
Ok, I just never saw those 3 buttons as part of Trackpoint. Thanks for clarification.

addendum:

(Trackpoint does not have any buttons per se.)
 
From manufacturers point of view keyboard with TrackPoint is one part and TouchPad with five buttons is the other part. IBM/Lenovo calls each such part FRU.

Actually, the top three buttons are part of the keyboard, not part of the touchpad.
 
Carpetsmoker said:
Actually, the top three buttons are part of the keyboard, not part of the touchpad.

I changed keyboard in my old T40 a dozen or so months ago. I forgot since that time that these buttons are a part of a keyboard. Since it's one FRU there is one more reason to associate these buttons with TrackPoint. Thank you Carpetsmoker for the correction.
 
michal said:
What about the sound card? Any luck with snd_hda? It's not working on my system. I get only system beeps (very loud) and sound on headphones (very quiet) but nothing on the speakers. Also volume up/down buttons are not working.
I had speaker working without any particular configuration for what I can remember. But they got mute somewhere between some BETA and RC1.
I had no time to investigate further...
 
I'm also tired, so I could miss something out, but I'm under impression michal asked for mute and up/down hot keys configuration.

As far as speakers going mute between BETA and RC1, that's something new to me, I had no such problem.

Code:
hdac0: HDA Driver Revision: 20090624_0136
hdac0: HDA Codec #0: Conexant CX20561 (Hermosa)

Code:
FreeBSD 8.0-RC1 #0 r197737: Sat Oct  3 21:14:16 CEST 2009
 
Works here.

8.1-STABLE #0 r210522

Code:
hint.hdac.0.cad0.nid22.config="as=1 seq=15"
hint.hdac.0.cad0.nid24.config="as=3"
hint.hdac.0.cad0.nid26.config="as=1"
hint.hdac.0.cad0.nid29.config="as=2"
 
Hello morbit,

what did you do to get your T400 working with FreeBSD - especially suspend/resume.
I tried to get these modes running but on resume the notebook is dead.
Thanks for your answer.

Best regards

JueDan
 
Sorry, I do not use suspend/resume at all*, FYI I currently use 9-STABLE amd64 (9.1-PRERELEASE #0) r241721 without any problems, ath works flawlessly, power saving, KMS too (drm2)...

Fresh unixbench (4.1 from ports tree, same as then) results! (clang3.1 system/gcc47 ports), compare with http://forums.freebsd.org/showpost.php?p=13464&postcount=23

Code:
                     INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        116700,0 23125985,7     1981,7
Double-Precision Whetstone                      55,0     3520,9      640,2
Execl Throughput                                43,0     2454,5      570,8
File Copy 1024 bufsize 2000 maxblocks         3960,0    86980,0      219,6
File Copy 256 bufsize 500 maxblocks           1655,0    97202,0      587,3
File Copy 4096 bufsize 8000 maxblocks         5800,0    54872,0       94,6
Pipe Throughput                              12440,0  1272614,1     1023,0
Pipe-based Context Switching                  4000,0    58523,8      146,3
Process Creation                               126,0     7388,8      586,4
Shell Scripts (8 concurrent)                     6,0     1180,7     1967,8
System Call Overhead                         15000,0   869252,7      579,5
                                                                 =========
     FINAL SCORE                                                     527,8

This is the same system (not a single change in hardware), updated from source consistently...

*Upon resume laptop was never 'dead' here, just display was not brought back properly. But I was checking it ages ago. I currently use KMS/drm2 so cannot unload graphic driver to test it properly.

After increasing 4GB->8GB (Kingston HyperX KHX1600C9S3P1K2/8G, I hope they will retain value better than value (heh.) sticks... Plus they run 6-6-6-18 / DDR3-1066, if it ever mattered. Memory chips are made by Hynix.):

Code:
                     INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        116700,0 23462617,5     2010,5
Double-Precision Whetstone                      55,0     3561,0      647,5
Execl Throughput                                43,0     2451,2      570,0
File Copy 1024 bufsize 2000 maxblocks         3960,0    86142,0      217,5
File Copy 256 bufsize 500 maxblocks           1655,0    93024,0      562,1
File Copy 4096 bufsize 8000 maxblocks         5800,0    55081,0       95,0
Pipe Throughput                              12440,0  1256947,4     1010,4
Pipe-based Context Switching                  4000,0    56164,2      140,4
Process Creation                               126,0     7403,3      587,6
Shell Scripts (8 concurrent)                     6,0     1189,0     1981,7
System Call Overhead                         15000,0   869894,5      579,9
                                                                 =========
     FINAL SCORE                                                     524,4

Not much difference... But libreoffice compile now does only saturate about half of memory :)

Huge variability- after 20h uptime:

Code:
                     INDEX VALUES            
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        116700,0 23592407,2     2021,6
Double-Precision Whetstone                      55,0     3549,2      645,3       
Execl Throughput                                43,0     2394,7      556,9      
File Copy 1024 bufsize 2000 maxblocks         3960,0    86828,0      219,3      
File Copy 256 bufsize 500 maxblocks           1655,0    96044,0      580,3
File Copy 4096 bufsize 8000 maxblocks         5800,0    55006,0       94,8
Pipe Throughput                              12440,0  1270230,8     1021,1
Pipe-based Context Switching                  4000,0   121242,2      303,1
Process Creation                               126,0     6852,7      543,9
Shell Scripts (8 concurrent)                     6,0     1143,0     1905,0
System Call Overhead                         15000,0   873253,7      582,2
                                                                 =========
     FINAL SCORE                                                     558,1
 
Same machine, FreeBSD 10.0-STABLE #0 r261091 amd64, a bit disappointing... I think it's connected with new BSD licensed tools.
Code:
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        116700,0 19479405,7     1669,2
Double-Precision Whetstone                      55,0     3417,7      621,4
Execl Throughput                                43,0     1956,3      455,0
File Copy 1024 bufsize 2000 maxblocks         3960,0    84302,0      212,9
File Copy 256 bufsize 500 maxblocks           1655,0    73788,0      445,8
File Copy 4096 bufsize 8000 maxblocks         5800,0    55651,0       96,0
Pipe Throughput                              12440,0  1096226,5      881,2
Pipe-based Context Switching                  4000,0   186495,0      466,2
Process Creation                               126,0     7381,4      585,8
Shell Scripts (8 concurrent)                     6,0      510,6      851,0
System Call Overhead                         15000,0   793704,9      529,1
                                                                 =========
     FINAL SCORE                                                     497,9
 
Thanks for posting this - great timing. I've got a T400 and am considering throwing 10.0 at it. It's currently running Windows 7 x64 with 9.2 in Virtualbox but it's such a waste as I don't actually use Windows that much. Will "invert" the installation making Windows the guest.

Does yours have switchable graphics? This has a Radeon and Intel graphics in it. Would rather just turn off the Radeon as it produces noisy yellows and I don't need high performance.
 
I only have Intel GM45 (X4500). Works as good as on 9-STABLE. Thanks for acknowledgement- I've almost thought I'm just leaving notes for myself, it's appreciated. I've once loaded GENERIC kernel- empty cardbus slot (cbb) makes an interrupt storm, but that's all. Regarding Radeon, there is Gallium now https://wiki.freebsd.org/Graphics
 
Thanks for info - will approach this over the weekend. I think you can kill the cardbus support in the BIOS So this might be a starting point. I don't have any CardBus devices installed so obviously depends on that. Really don't want the Radeon even powered up as it kills battery life rapidly. I'm actually beginning to wish I bought the model with only the X4500 in it.
 
I don't compile cbb usually, so that's not a problem for me. I've loaded GENERIC once for testing purposes. You should know, that syscons is not available for a systems where KMS/new X.org is enabled. That means no system console (blank screen once you kill X session or switch to a system console- but system does not hang). There is a newcons project, but it's not MFC'd yet. https://wiki.freebsd.org/Newcons
 
Code:
                     INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        116700,0 18268560,2     1565,4
Double-Precision Whetstone                      55,0     3516,2      639,3
Execl Throughput                                43,0     2015,1      468,6
File Copy 1024 bufsize 2000 maxblocks         3960,0    85683,0      216,4
File Copy 256 bufsize 500 maxblocks           1655,0    75441,0      455,8
File Copy 4096 bufsize 8000 maxblocks         5800,0    55932,0       96,4
Pipe Throughput                              12440,0  1076461,5      865,3
Pipe-based Context Switching                  4000,0   182937,2      457,3
Process Creation                               126,0     7755,6      615,5
Shell Scripts (8 concurrent)                     6,0      532,8      888,0
System Call Overhead                         15000,0   796356,2      530,9
                                                                 =========
     FINAL SCORE                                                     502,2

Full clang 3.4, FreeBSD 10.0-STABLE #0 r263667 amd64. As a side note, vao_demo from libva-intel-driver-1.2.0 certainly works (VAAPI under Intel GM45), but I'm not sure if it buys me anything. Looks pretty anyway (there is a relevant gstreamer plugin in ports collection available, but I don't see a difference in CPU usage anyway).
 
Code:
OS:  FreeBSD 12.2-STABLE #0 r368359 amd64
CC:  FreeBSD clang version 10.0.1
CPU: Intel(R) Core(TM)2 Duo CPU     T9900  @ 3.06GHz
RAM: 8GB DDR3 1066 Kingston HyperX
HDD: Samsung SSD 860 PRO 1TB

unixbench-5.1.3_2

2 CPUs in system; running 1 parallel copy of tests

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   27171381.5   2328.3
Double-Precision Whetstone                       55.0       4389.4    798.1
Execl Throughput                                 43.0       2052.4    477.3
File Copy 1024 bufsize 2000 maxblocks          3960.0     469074.5   1184.5
File Copy 256 bufsize 500 maxblocks            1655.0     126976.5    767.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     919475.4   1585.3
Pipe Throughput                               12440.0     296790.5    238.6
Pipe-based Context Switching                   4000.0     126437.2    316.1
Process Creation                                126.0       8297.8    658.6
Shell Scripts (1 concurrent)                     42.4       5911.9   1394.3
Shell Scripts (8 concurrent)                      6.0       1090.1   1816.8
System Call Overhead                          15000.0     476533.6    317.7
                                                                   ========
System Benchmarks Index Score                                         780.5

2 CPUs in system; running 2 parallel copies of tests

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   53598266.9   4592.8
Double-Precision Whetstone                       55.0       8507.0   1546.7
Execl Throughput                                 43.0       3895.3    905.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     848261.7   2142.1
File Copy 256 bufsize 500 maxblocks            1655.0     230505.7   1392.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    1564690.8   2697.7
Pipe Throughput                               12440.0     698968.0    561.9
Pipe-based Context Switching                   4000.0     227386.9    568.5
Process Creation                                126.0      15404.6   1222.6
Shell Scripts (1 concurrent)                     42.4       8338.5   1966.6
Shell Scripts (8 concurrent)                      6.0       1088.9   1814.8
System Call Overhead                          15000.0     922796.8    615.2
                                                                   ========
System Benchmarks Index Score                                        1371.8

Over the years, Execl Throughput, Pipe Throughput, Pipe-based Context Switching and Shell Scripts (8 concurrent) tests performance has actually dropped, despite hardware upgrades (CPU/HDD/RAM), though raw CPU performance increase is roughly proportional to increase in frequency (+17,5%, +20%).
 
Code:
OS:  FreeBSD 13.1-STABLE #0 stable/13-1149f0ec2 amd64
CC:  FreeBSD clang version 14.0.5
CPU: Intel(R) Core(TM)2 Duo CPU     P9700  @ 2.80GHz
RAM: 8GB DDR3 1066 Kingston HyperX
HDD: Samsung SSD 860 PRO 1TB

unixbench-5.1.3_2

2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       25518433.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4634.4 MWIPS (9.5 s, 7 samples)
Execl Throughput                               1930.2 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        540107.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          148676.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        975256.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              751401.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 201608.7 lps   (10.0 s, 7 samples)
Process Creation                               8343.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6039.5 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1099.2 lpm   (60.0 s, 2 samples)
System Call Overhead                         490093.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   25518433.1   2186.7
Double-Precision Whetstone                       55.0       4634.4    842.6
Execl Throughput                                 43.0       1930.2    448.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     540107.4   1363.9
File Copy 256 bufsize 500 maxblocks            1655.0     148676.4    898.3
File Copy 4096 bufsize 8000 maxblocks          5800.0     975256.1   1681.5
Pipe Throughput                               12440.0     751401.3    604.0
Pipe-based Context Switching                   4000.0     201608.7    504.0
Process Creation                                126.0       8343.7    662.2
Shell Scripts (1 concurrent)                     42.4       6039.5   1424.4
Shell Scripts (8 concurrent)                      6.0       1099.2   1832.1
System Call Overhead                          15000.0     490093.8    326.7
                                                                   ========
System Benchmarks Index Score                                         902.7

2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       50193467.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     8992.3 MWIPS (9.5 s, 7 samples)
Execl Throughput                               3599.4 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1030899.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          279690.6 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1864284.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1456777.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 432365.5 lps   (10.0 s, 7 samples)
Process Creation                              15518.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8514.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1099.2 lpm   (60.0 s, 2 samples)
System Call Overhead                         933636.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   50193467.5   4301.1
Double-Precision Whetstone                       55.0       8992.3   1635.0
Execl Throughput                                 43.0       3599.4    837.1
File Copy 1024 bufsize 2000 maxblocks          3960.0    1030899.4   2603.3
File Copy 256 bufsize 500 maxblocks            1655.0     279690.6   1690.0
File Copy 4096 bufsize 8000 maxblocks          5800.0    1864284.0   3214.3
Pipe Throughput                               12440.0    1456777.1   1171.0
Pipe-based Context Switching                   4000.0     432365.5   1080.9
Process Creation                                126.0      15518.3   1231.6
Shell Scripts (1 concurrent)                     42.4       8514.2   2008.1
Shell Scripts (8 concurrent)                      6.0       1099.2   1832.1
System Call Overhead                          15000.0     933636.5    622.4
                                                                   ========
System Benchmarks Index Score                                        1607.2

Update. It looks it got faster. T9900 was running too hot in practice, not worth it.
 
Back
Top