Hi Dedhi, Thanks for the information. I'm glad to see the new driver is usable. Actually e1000g-orig just called one more function "e1000_cleanup_led()" during the chip initialization than e1000g-new. I'm working with Intel to investigate the impacts of it to 82541.
When you mentioned the performance is still have room to improve, I also concern about it. I wonder if you ever working with a build that the performance of 82541 is good. I just want to determine if it's a regression or not. Thanks for your help! Miles Xu Dedhi Sujatmiko wrote: > Hi Miles Xu, > > I add another host in the mix, it is alfa, an Ubuntu 8.04 AMD64 with > Realtek 8169 Gigabit Ethernet. Not a perfect candidate for GE test, but > at least making the problem isolated between Solaris and Linux. > > From charlie, the host having 82541 to alfa > > 1. Using original driver from snv b_103 > > r...@charlie # scp sol-10-u6-ga1-x86-dvd.iso de...@alfa-storage:/home/dedhi > de...@alfa-storage's password: > sol-10-u6-ga1-x86-dv 15% |*********** > | 337 MB - stalled -^CKilled by > signal 2. > > Taking too long, many stall.... more than 10 minutes I kill it > > 2. Using e1000-orig from you > > r...@charlie # ls -la /kernel/drv/amd64/e1000g* > -rwxr-xr-x 1 root sys 621168 Jan 11 11:59 > /kernel/drv/amd64/e1000g > -rwxr-xr-x 1 root sys 429544 Nov 15 03:23 > /kernel/drv/amd64/e1000g.dedhi > > o...@charlie # scp sol-10-u6-ga1-x86-dvd.iso de...@alfa-storage:/home/dedhi > de...@alfa-storage's password: > sol-10-u6-ga1-x86-dv 66% |************************************ > | 1464 MB 09:03 ETA^C > > Taking too long, many stall.... more than 10 minutes I kill it > > 3. Using e1000g-new from you > r...@charlie # ls -la /kernel/drv/amd64/e1000g* > -rwxr-xr-x 1 root sys 621184 Jan 11 12:12 > /kernel/drv/amd64/e1000g > -rwxr-xr-x 1 root sys 429544 Nov 15 03:23 > /kernel/drv/amd64/e1000g.dedhi > > r...@charlie # scp sol-10-u6-ga1-x86-dvd.iso de...@alfa-storage:/home/dedhi > de...@alfa-storage's password: > sol-10-u6-ga1-x86-dv 100% > |*****************************************************************************| > > 2201 MB 02:53 > > > now the system is more stable. However, the performance still have many > rooms for improvement. > > For example Fast Ethernet speed is not that many different, it will > clock at ~10 MB/s, where this driver clock in at ~13 MB/sec. > > Since charlie is having rge0 onboard interface, using it is faster than > the Intel 82572 PCI-e, it can clock in at ~19.5 MB/s. > > bash-3.2# scp sol-10-u6-ga1-x86-dvd.iso de...@alfa-storage:/home/dedhi > de...@alfa-storage's password: > sol-10-u6-ga1-x86-dv 100% > |*******************************************************| 2201 MB 01:53 > > regards, > > Dedhi > > Min Miles Xu wrote: > >> Hi Dedhi, >> The issue seems to be related with 82541. Here I made several e1000g >> binaries which should work before snv_105. I don't know your system is >> 32bits or 64 bits. >> Could you help me try them and tell me the result? Thanks! >> >> >> Regards, >> >> Miles Xu >> >> >> Dedhi Sujatmiko wrote: >> >>> Hi Min Xu, >>> >>> Currently I only have 2 hosts connected, one using Intel 82572, >>> another one using Intel 82541, thus I cannot determine which one is >>> having the problem. However I do have spare NIC for each model. >>> >>> This weekend I will try to test Intel 82572 with Intel 82572, and >>> Intel 82541 with Intel 82541. >>> >>> best regards, >>> >>> Dedhi >>> >>> >>> Min Miles Xu wrote: >>> >>> >>>> Hi Dedhi, >>>> LSO issue was fixed in Build 103. Just for confirmation, you meant >>>> that the issue still persisted when LSO was disabled, right? (It only >>>> affects host golf(Intel 82572). Charlie(Intel 82541) doesn't support >>>> LSO) >>>> First could you identify the issue was on golf or charlie? I got a >>>> separate report complaining 82541 was suffering performance problem >>>> in Build 103. >>>> >>>> Regards, >>>> >>>> Miles Xu >>>> >>>> >>>> Dedhi Sujatmiko wrote: >>>> >>>> >>>>> Ted You wrote: >>>>> >>>>> >>>>>> Hi Dedhi, >>>>>> >>>>>> Could you please let us know the device id of the e1000g NIC on your >>>>>> system? You can get it with the following command: >>>>>> >>>>>> % prtconf -vpPD | grep "pciclass,0200" >>>>>> >>>>>> >>>>>> >>>>> Hi Ted, >>>>> >>>>> at host charlie : >>>>> >>>>> r...@charlie # prtconf -vpPD | grep "pciclass,0200" >>>>> compatible: 'pci8086,1229.8086.1.2' + >>>>> 'pci8086,1229.8086.1' + 'pci8086,1' + 'pci8086,1229.2' + >>>>> 'pci8086,1229' + 'pciclass,020000' + 'pciclass,0200' >>>>> compatible: 'pci8086,107c.8086.1376.5' + >>>>> 'pci8086,107c.8086.1376' + 'pci8086,1376' + 'pci8086,107c.5' + >>>>> 'pci8086,107c' + 'pciclass,020000' + 'pciclass,0200' >>>>> >>>>> It is : http://www.newegg.com/Product/Product.aspx?Item=N82E16833106121 >>>>> Intel PWLA8391GT 10/ 100/ 1000Mbps PCI PRO/1000 GT Desktop Adapter 1 >>>>> x RJ45 - OEM >>>>> >>>>> >>>>> at host golf : >>>>> >>>>> bash-3.2# prtconf -vpPD | grep "pciclass,0200" >>>>> compatible: 'pciex8086,10b9.8086.1083.6' + >>>>> 'pciex8086,10b9.8086.1083' + 'pciex8086,10b9.6' + 'pciex8086,10b9' + >>>>> 'pciexclass,020000' + 'pciexclass,0200' + 'pci8086,10b9.8086.1083.6' >>>>> + 'pci8086,10b9.8086.1083' + 'pci8086,1083' + 'pci8086,10b9.6' + >>>>> 'pci8086,10b9' + 'pciclass,020000' + 'pciclass,0200' >>>>> compatible: 'pciex10ec,8168.1458.e000.2' + >>>>> 'pciex10ec,8168.1458.e000' + 'pciex10ec,8168.2' + 'pciex10ec,8168' + >>>>> 'pciexclass,020000' + 'pciexclass,0200' + 'pci10ec,8168.1458.e000.2' >>>>> + 'pci10ec,8168.1458.e000' + 'pci1458,e000' + 'pci10ec,8168.2' + >>>>> 'pci10ec,8168' + 'pciclass,020000' + 'pciclass,0200' >>>>> compatible: 'pci8086,1229.1' + 'pci8086,1229' + >>>>> 'pciclass,020000' + 'pciclass,0200' >>>>> bash-3.2# >>>>> >>>>> It is : http://www.newegg.com/Product/Product.aspx?Item=N82E16833106033 >>>>> Intel EXPI9301CTBLK 10/ 100/ 1000Mbps PCI-Express Network Adapter 1 >>>>> x RJ45 - Retail >>>>> >>>>> best regards, >>>>> >>>>> Dedhi >>>>> >>>>> >>>> >>>> >>> _______________________________________________ >>> networking-discuss mailing list >>> [email protected] >>> >>> > > _______________________________________________ > networking-discuss mailing list > [email protected] > _______________________________________________ networking-discuss mailing list [email protected]
