Computurd explained to me that when since we are setting up with adaptive read ahead your system needs to learn how you are using the computer and will adapt to your usage over time. performance will increase.
Printable View
Computurd explained to me that when since we are setting up with adaptive read ahead your system needs to learn how you are using the computer and will adapt to your usage over time. performance will increase.
i was not referring to the adaptive read ahead...the 9260 has an advanced algorithm that controls how it uses its cache...it basically tailors it to how you use yo9urr system, so there is a "learning cycle" that it uses, as a matter of fact there is even options for that in the MSM. also, if youhaveread any reviews about the 9260 they always mention that the algorithms give them a pain in the ass when they try to bench the card.
hit me on google > I was way off!
These benchmarking does not look right..
my setting so far:
Raid 0 4 Supertalent Ultra 32 GB
Adaptive Read Ahead
Write back
64kb block
IO Cache
Cache enable
Need help if possible...
Notice having IO cached enable, decrease my boot time by 5 second. With Direct IO, it still give me the same bar graph (scraping at 64k) and beyond. Should I increase the block to 512KB?
going to be testing the 9260-8i new FW very shortly...had some RLS to take care of...:lsfight:
The new firmware of the 9260-8i has a great improvement!
i dont think you have seen anything yet from these controllers (lsi). the game is about to change.
it will become very one-sided:D
Comp, what do you mean by this statement? Is there a new raid chip out or is a new one in development? Or do you mean these controllers have massive amounts of improvement through new firmware releases?Quote:
Originally Posted by Computurd
i came across with this PDFs from areca that as an IT makes me wonder for some things. This controller performs only with cache?? isnt a controller that Enterprises can work without All in cache?? there is no performance in Critical Data??? because in raid50 r in raid60 some companies not only the LSI r way way faster!!
Is this controller only for benchmark enthousiast?? Is this controller only for RAM-Cache Benchmarking??
And if is that.. Why should i Pay 600euros for 4 gb of cache???
I think the idea is you use battery backup and your ok?
Also - i can confirm - the 1231 will not boot w/o the memory in the slot - I have tried.
Can anyone summarize the differences in the performance and features of the 9211-8i, 9260-8i and the 1231ML-2G, when used with SSDs such as the X25-M?
AceNZ - all three should work very well with the x25-m.
General assessment - I have seen some huge iometer runs with the 9211 and 8x x25-m - huge throughput and with small file reads the highest anywhere - maybe with the exception of the ioxtreme.
In applications where cache can be used - both the 9260 and the 1231 will do well, usually the 1231 will do a little better as at has a larger cache (512MB vs 2 or 4GB)
The 1231 is now old and has a ceiling of 800-900MB/s - so 4 x25-m would max out the 1231 is seq reads.
We have seen at least one vendor advertise the new Areca 1880 to be available in late May, many folks are waiting to see what that looks like.
Rumor is that the 1880 will be well over $1000, so it would have to be pretty amazing to be worth considering. Also, although I've heard good things about Areca's UI, it seems like their firmware updates have been very infrequent compared to LSI.
How does the 9260 perform when cache is disabled? Can it approach the same level of IOPS as the 9211? The 9211's throughput is attractive on the surface, but its limited configurability strikes me as a potentially serious handicap in real-world scenarios.
I have suggested in another thread that we should collectively query LSI wether a firmware update to enable configuring stripe size (and other things people may want wich is avalible on 9260) is possible. If it is, i would think LSI would get better rep for releasing such a firmware, even if it's only as a beta for us enthusiasts.
When LSI says that the 9211 has a fixed 64K "stripe" size, they must really mean strip size, not stripe size, right?
If so, it seems like that might be part of the reason why random IOPS are good, while I/O for "medium-sized" files isn't as good as the 9260 -- since random I/O would tend to use all drives at once (given a large enough QD), while file I/O would tend to focus I/O more heavily on a smaller number of drives (64K strip x 8 drives = 512K stripe).
I wonder how the performance of the 9211 compares to a 64K strip size on the 9260 (with and without caching enabled)?
man oh man we need to revisit this with fastpath on the LSI :)
I can possibly dig up some 9211 benches and see if I have a direct comparison with 9260 at 64k. Give me a couple days. (8x 256gb crucial c300)Quote:
Originally Posted by AceNZ
mbreslin for the love of christ you need fastpath! it is just amazing, the ultra low latencies! i have never seen 4k scale so beautifully, even ich10r is put to shame. i'd love to do a pcmv run with that and your uber array :)
Your numbers look good sir. I'm sort of interested myself, definitely not 160$ interested.
Banana them.
Hmm... You have a $3000+ array, and you are not interrested in $160 extra for possibly a 50-100% max performance boost? Compared to the total cost of your rigg it's pocket change.
Possibly computurd could overnight you his key so you could try it? (if that is possible)
BTW, lower latencies means a speed boost even at low queue depths (and singletasking), so you don't have to be in the 100Ks of IOPS to see a benefit. It boosts the "cpu accelerator effect" of your SSDs.
mbreslin,
Get the Key :)
It looks like its worth the $ unless you've converted to the ICH :)
I just think it's a crappy way to do business.
I see crucial has put the firmware update back on their site so I guess I'll order the silly key and do the firmware flashing+new array all at once.
@GullLars: You've hurt my ego when pricing my array, can you please tell me where I can purchase 8 256gb c300s and a 9260 for anywhere in the ballpark of 3000$? If this was even possible I'd buy another 8 just to have 16 drives ready for one of the 1880s.
edit: Best place to order from in the states? LSI directly or?
lol yoiua re correct mbreslin, i was thinking the same thing when i read that, WAY more LOL.
scsi4me has them and they are awesome so thats where i would recommend getting it :)
I agree with mbreslin 100%, not only is it crappy business practice, they should be sued for selling a product that is intentionally capped and then telling you if you want the full benefits of what the product can achieve you must pay extra. Sell the damn product with the key already included and charge more for the product in the first place, jeez. This business tactic is pure greed...
Ordered. I feel like such a sucker!
Edit: I should add I have never thought it was something they planned from the beginning only that they figured out how to get more out of the card in software after the product shipped. The price point seems high considering most other companies release their improvements via free firmware/drivers/software.
mbreslin,
8x C300s cost about $7500 in Norway + $900 for the 9260-8i :)
Compared to the cost of the drives+controller, $155 for they Key is cheap *if* it unleashes the kind of performance I expect it to.
(I do get why you are annoyed, it's just that most customers are medium-large companies and I don't think they'll get upset about the price)
I never said it was expensive, let me quote myself..
The money is nothing compared to the cost of the array and certainly compared to the cost of the entire machine, I never claimed otherwise I just was holding out on principal.Quote:
Originally Posted by mbreslin
What's worse than them milking more money for hardware already purchased is that I so quickly went from "Banana them." to "Ordered.". These forums have a way of making me click buy buttons!
for a system of your uber awesomeness a key is definitely required :)
it will help answer a question of mine as well...see no matter what settings that i use, my random read is about the same regardless. now this means one of two things, either it is the top of my drives, which i am sure it is, or it is the top of the processing power of the card....either way it is still the fastest out there right now imo. i havent done a whole ton of testing yet with load times, that is next on my agenda. had to get up higher on pcmv first :)
now only 980x and 1231 with cache are above me, but i have a plan for that with...well we will see if it works first :P
cant wait to see your and anvils results!!
oh and fellas on teh quick install guide of the fastpath key the instructions are WRONG....yes hard to believe, but they tell you to put it on a pair of pins that arent even on the card, the illustration for the 9260 is actually a 9210. which never hit production. the only reason i knew where to put it was becasue i had dug a pic of the key up on intel documentation a few weeks back. they had a pic of the key on their lsi rebranded 9260. i will post a pic of correct pins to use.
now all i need is some 25nm ssd goodness from intel...
Yeah I saw your result ct, insanely good for a 920!
I will take another run with fastpath and new crucial fw. The voltage needed to get my 980x to 4.8ghz scares me but hopefully it'll survive again. :)
Steve was right of course that it wouldn't be long before I started tinkering again. If the 1880 really shines I'm willing to fork up the 1200$ or whatever the msrp winds up being.
Another 5970 and the ud9 is very tempting as well.
oh that result is decent but not as good as the one i submitted last night :) jeez it took them three days to get that one on there, so teh really good one should be on tomorrow? should be numero uno for the 920 on teh bot as well, did have number two :)
i will have more though, figured out a little trick, just need some time to tinker myself. back to real work tommorow, but after that some more time on defcon-one and i should be able to really set it off!
need some more time to acquaint myself with this controller, its a much different beast now. there is much more to be gained from this FastPath, just takes time and patience.
unless the areca is spectacular i wont be buying it for awhile. lsi is WAY ahead of the game right now with drivers, etc. the card is really maturing, and it should take some time for areca's new card to mature as well, but who knows maybe outta the gate it will smoke the lsi. we shall see. the one im interested is the one with dual processors on it. that looks killer!
the 9260 seems to take higher pcie clocks much better than it did before fastpath, gonna be working in that direction as well soon, havent really toyed with that. the IOH could take some more volts as well, i know i am toasting it, there are still many tools in the box, just gotta pull them out :) cant wait until you get it as well so we can collaborate.
The 9260 doesn't show up for me at 105+ pcie, it may be something to do with having 5 pcie devices and not enough voltage (I bumped 2 ticks but don't really want to go more than that). I got tired of draining loops really fast and vowed not to do it again for just some benchmark so I never ran pcmv with just a highly oc'd 5870 and the 9260 and no other pcie cards. I'm quite sure I could have scored higher. It would be nice to pass corsair with their 3 more drives than me and ln2. If the 1880 is even 5% better than the 9260+fastpath I'll be getting it and probably another 5970 and the ud9. I'm very drawn to having the best available cpu/graphics/storage. I'm very close now and it's probably a ridiculous venture but what can I say, I'm hooked.
does scsi 4me give you the key instantly and you can dl the fp software or is it a disk you need to wait for? I want instant gratification. paul is the new fw for 92608i only or do you have the 4i as well?
Same for me - 104-105 was all I could do with 9211, 1231 went to 119.Quote:
Originally Posted by mbreslin
it is a physical key that you install trans am. it works for -4i but doublecheck when you buy it. by physical key that means you have to plug it in...some pics for ya here.....
@stevero/mbreslin..yup 9260 always been finicky with pcie clocks. what i was alluding to is the fact that it seems to take higher pcie now.
heres some pics...
key....
http://i517.photobucket.com/albums/u...2-50-12847.jpg
card, key is under the sff ports (where you plug in cables)
http://i517.photobucket.com/albums/u...0-11-09770.jpg
closeup of the card, key sticking out of the side
http://i517.photobucket.com/albums/u...0-11-37746.jpg
http://i517.photobucket.com/albums/u...0-16-13680.jpg
CT, you need to slow down with buying all this hardware man, all the extra radiation your polluting your environment with is causing your hair to fall out.... LOL j/k hehe :rofl:
Sorry man, didn't mean to. I for some reason i was thinking 8x x25-M 160GB. I don't know how i managed to confuse it.Quote:
Originally Posted by mbreslin
C300 256GB is about $700 on newegg, = $5600 for the drives
LSI 9260-8i is about 530 on newegg.
$6130 whitout the key, $6290 with the key, a 2,6% increase :)
It was in jest and hope it came across that way. :)
(and since I was an insane/dumb day 1 adopter I got them all from crucial for msrp (799$)) :(
I actually got them within a few minutes of them being available, only 3 as they sold them 3 at a time so it took me a few days.
Edit: Throw in the 9211 and it's over 7k just for storage. I actually winced while typing the previous sentence.
Edit2: If I would have waited for newegg to get them and then lower the price (they had them at msrp for awhile as well) the 800$ saved would sure have helped the sting when I buy a 1200$+ areca. Hindsight..
I got it that it was a somewhat jokefull comment, but i realized putting the ballpark figure well below half the actual cost does not serve your array justice, so i felt an actual apology was in order ;)
Does anyone know if the heatsink mount is a standard size of some sort? Wondering how difficult it will be to actively cool this thing...
I read on newegg that the 9211 overheated often for someone but I never had any problems while benching it. The 9260 heatsink gets about as hot as ddr3 with no fan but it's never had any issues, it gets a little air that's blowing out of a hot rad but that's about it.
To answer your question it seems close to the size of many small blocks I've seen but I'm not certain. I can't imagine needing it possibly in a server environment but then you're going to have some good san ace's blowing on it and the AC set to 'antarctica'.
I think my 1231 overheats from time to time - I just aim a case fan at it and all is well.
ok cool, so no real problems with heat then? I just dont want to feed my 5870 warm air from below since I only use the stock cooling on that.
Areca sell fans fints on Areca 1231ML. I did buy one for my areca 1231ml.
Paul thanks for the heads up! what performance increase can we all expect from that little chip. Did you have any old screens you saved with the same lsi config settings/stripe size cpu speed etc? you can show us a before and after results?Quote:
Originally Posted by Computurd
That little chip is the BIGGEST possible profit margin LSI could ever think of. I mean how much to manufacture that tiny thing? A few dollars per unit at most? Amazing stuff these companies think of. I just hope this greedy idea doesn't move on to other hardware items, like hard drives, motherboards or CPU's. What a rip..... sorry for the rant, but Im still amazed they are actually getting away with this. That thing should be $20 max... lol
exactly why I made this youtube presentation for the LSI guys:Quote:
Originally Posted by SkOrPn
enjoy!
http://www.youtube.com/watch?v=8Sb33g13Cr8
couldnt agree more skorpn. highway robbery. i will post a little 4k comparison here in a few trans am, show ya some befores and afters. gonna complete my whole chart, comparing with this :
http://i517.photobucket.com/albums/u...tfirmeware.png
http://i517.photobucket.com/albums/u...firmeware2.png
wtf comp are the results on the left the ones before the key and the right is after? instead of blowing money on these stupid sandforce drives I shoulda spent the money on the key. That is amazing!
I dont care how cheap you guys say LSI is for doing this. they made it available and it is what it is and its still a hell of a lot cheaper than a 1k areca card. I think I'm gonna blow the sandforces off on ebay and revert back to the indinlinx and get the key. I am not seeing a huge boost on the sandforce to the indlinlinx with the lsi except on the 4k but that might be be fixed with the chip. Im gonna go back to the indlilinx drives for a moment and do a proper back to back comparison and then you guys can tell me what I should do.
It isn't cheap, its just wrong, but the LSI guys are smiling all the way to the bank thinking "whats so wrong about it?" eh? For results like that I wouldn't think twice about paying what LSI wants for that little thing, lol... Just wish it could be pre-installed and sold with the card instead, so at least it doesn't look like they have taken advantage of customers who do not have the key.Quote:
Originally Posted by trans am
Love the song trans... :up:
no no fellas....that is all the same stuff BEFORE the key....i am going to make a chart to match that AFTER the key....the left is random and the right is sequential
:ROTF::ROTF: got that damn song stuck in my head trans amQuote:
I did my settings and hit F10
windows started
booting again
I got into desktop
yeah man!
double clicked two thousand and one
pouring ln2 out
before its too late
while watching the goddamn frame rates
lol
05/25/2010 8:08 A.M. OUT FOR DELIVERY
Is there a common iometer config everyone is using? I'd like to do full scaling 1-8 drives with/without fp.
Thanks.
umm not really, just set up 4k runs, both seq and random, or use crystal disk mark for fast comparisons. how familiar with iometer are you? if you need a guide real fast lemme know.
here is some i use...
well i have put my drives through the wringer the last few days LOL
http://service.futuremark.com/compare?pcmv=324602
@computurd: That's the score I said was good the other day, I saw it while browsing results. Hopefully I'll have something new to put up soon. My key is here, going to spend tomorrow benching and maybe try for a new pcmv pb.
well up to tenth on the pcmv which is UBER lol..need some gen 2 !!
here is the main thing i have to say about FASTPATH...on the right is fastpath, on the left is before FP with the latest firmware.
http://i517.photobucket.com/albums/u...omparsions.png
If you find yourself with some free time on your hands, here's huge the post i promised earlier in another thread. It was around 3500 words in norwegian, but i cut it down a bit in translation, so it ended up just above 2000 words here ;)
http://www.xtremesystems.org/forums/...d.php?t=252695
I posted it in a new thread i made for general SSD discussions, rants, curiosities, etc. I'll probably post more there later, and will use that thread to gather up links to usefull or interresting threads, reviews, and general info.
@ Paul - very nice improvement in pcmv - congrats!
@ mbreslin - based on what Paul is seeing with fastpath - i am bracing for a huge score from you! best of luck!
very nice!
where do you actually order the fastpath key ??
SCSI4ME have them for $158:
http://www.scsi4me.com/lsi00247-fast...9260-9280.html
It's where I got mine, fast shipping, about to install now.
before you beat up some pcmv mbreslin if you could just show me some 4k scaling..wondering if it is the top of my drives or the max the iop will do...think its max for my drives...
Computurd, does the Fastpath key work best with Indilinx or Sandforce based drives?
oh i am sure it works best with sandforce. results are coming now on the performnace with other drive types. sandforce drives are better, and i think any drives that are faster will do better.
Ok thanks for the info. Something mbreslin said made me think Indilinx were better with this fastpath key. I was thinking of replacing my Vertex with Agility 2's, but probably just going to wait for next generation as there is always something new just around the corner.
the intel refresh with 25nm and (hopefully, if rumors are true) 6gb/s sata connections are just around the corner. that is what i am waiting for, but that also depends upon jetstream and how it performs. either intel or jetstream.
Price vs Performance.... The Agility 2 is only $180 for the 50GB drives. I want 4 drives to replace my 4 vertex at a cost of $800 or less, lol.... thats my limit...Quote:
Originally Posted by Computurd
I'd wait for 6Gbps drives Skorpn, barefoot drives have pretty decent performance, so $800 to replace 4 of them wouldn't give you the same level of performance boost as when you got them in the first place. (unless you're doing some heavy database or VM work)
I wouldn't replace the Vertexs with barefoot drives, they are barefoot drives, lol, I would replace them with Sandforce drives. But yeah I agree, I think I will try and hold out for something new that wont break the bank. :up:Quote:
Originally Posted by GullLars
I meant that Vertexs as barefoot drives have pretty decent performance, i'm sorry if it came over any other way. The point is, for most uses, the $800 to replace with newer SSDs will give you a performance boost, but nothing like when you first bought them. If you wait for the next gen 6Gbps drives, you might get a tangable performance increase for $800 (that will ofc require a 6Gbps controller).
Yeah I agree, but those agility's look awful nice for their price, lol. Im going to wait until someone releases an affordable sata3 drive, then start planning my upgrade...Quote:
Originally Posted by GullLars
Yes, the smart money is going to be waiting right now LOL waiting for the next big thing!
Anyone try the LSI 9211 new firmware ver. 07.00.00.00?
http://www.lsi.com/storage_home/prod...-8i/index.html
Product: Firmware
Version: 07.00.00.00
Release Date: 05-AUG-10
Release Type: GCA
Defects Fixed (64)
Enhancements Implemented (35)
Defects Fixed (64):
SCGCQ00135004 - SPV: Fault 0x0501 when drive is pulled.
SCGCQ00135037 - SPV: Virtual Expander does not reply to SMP Report Zone Manager Password Request.
SCGCQ00136195 - 1901 Fault due to a Port open fail interrupt without a matching transport error.
SCGCQ00136685 - SPV: Add check to verify if a Virtual Function is allowed to access a Virtual Port.
SCGCQ00138178 - Write Cache enable to be done after we are done with all in memory things
SCGCQ00138417 - IR : Resync can be Pending and in progress at the same time on a Raid 10 Volume
SCGCQ00141209 - Status FIS filelds are zero SATA Passthrough reply for PIO-in cmds
SCGCQ00141555 - Unable to execute diagnostic console commands after fault
SCGCQ00141606 - Change hardcoded fault code to a properly defined code.
SCGCQ00141622 - Calculation of pl memory may be bigger than necessary for some regions.
SCGCQ00141891 - Degraded Physical Disk coming as optimal when a diag reset is issued on a pending resync raid10 volume in Linux
SCGCQ00142962 - (IOP-Only) PCIe 2.0 Hardware Fix for TB A.0 is not conditional
SCGCQ00142963 - (IOP-Only) Interrupt Coalescing is not initialized properly if Multi-MSIX reply is not enabled
SCGCQ00143378 - Flashing a thunder bolt or meteor with DDR enabled FW cause a 1808 fault followed by a recursive 265d fault
SCGCQ00143655 - Read Long command to SATA device does not complete
SCGCQ00144748 - Obtaining proper value of Standby condition timer from Power condition mode page (page 0x1a) for a mode select command.
SCGCQ00145610 - Converting SATA Standby count to proper Standby condition timer value in Power condition mode page (page 0x1a) for mode sense command.
SCGCQ00145728 - Removing dead code for volume inline update
also new FW, drivers and storage manager for 9260-8i.....http://www.lsi.com/channel/products/...-8i/index.html
gah doorules! you dangle lovely firmares in front of me like carrots, yet i cant use them damn evga mobo!
new firmware 9211 do NOT install 2118IT.bin
Thanks for the warning Tilt
BTW, do you know of any board where one can disable pci slot scanning?
(last effort before I RMA the 9260 that hangs)
quote from LSI
If these steps were not followed and the MegaRAID controller hangs at power-on, try the following steps to attempt to recover the MegaRAID controller:
1. Remove the MegaRAID controller from the system
2. Disable the PCI slot scan in system BIOS
3. Insert the MegaRAID controller into the slot where the PCI slot scan was disabled and boot to DOS
4. Use "MegaCli -adpm0flash -f mr2108fw.rom" to reflash this firmware image (Mode 0 flash MUST be done in DOS)
5. Reboot the system to either the OS or DOS
5. Use "MegaCli -adpfwflash -f <older fw file name> -noverchk -a0" to flash the older firmware version
6. Use "MegaCLI -facdefset -a0" to load the factory defaults
7. Reboot the system and the MegaRAID controller should be recovered
@ Anvil - i don't think any of the ggbt boards have this.
Quote:
Originally Posted by Anvil
EVGA's and some new asus like R3E.
u can boot from HDD not from RAID and go to CMD with administration rights
then put the sas2flash and the bin file on the main folder of C:\
type c:\sas2flash -f firmwarename.bin and then hit enter.
well there is a new firmware fo the 9260 that is MAY fix the issue with the EVGA mobos, but im not gonna test it. just got this one back from RMA
So any evgan's tested new firmware for 9260? I cant do that because i used to connect to work? Any word from lsi/evga?
asked the same question of LSI, here is reply:
kinda leaves it open...needs to be tested. I am selling mine so i do not want to mess with it. there is some big nvdata changes, and that was the problem. the changelog denotes a compression/decompression logic addition that may just fix the problem.Quote:
None of our firmware releases have been tested by LSI on EVGA boards to this point. None of the defects listed were specifically referencing EVGA compatibility, though there were some NVDATA changes made.
Kind regards,
Brent VanDyke
Technical Support Engineer
Global Support Services
I've been running a raid 5 setup on a 1231ML with 8 raptors without incident for well over a year. Last week I bought 4 Phoenix Pro 120Gb SSDs, installed and configured them in a raid 0 in Windows, booted from the dvd and installed Windows 7 without a hitch but when I tried to boot back into Windows on my raid 5, it blue screened. Seems like one of my drives dropped out of the raid 5. Through many hours of trial and frustration, I discovered that when the SSDs are plugged in to the controller, whether they are configured or not, they cause a conflict that prevents the raptors from functioning. I installed Windows on a single raptor, then created a raid 0 with the 8 raptors, initialized it but as soon as I attempt to write on it, it drops off and when I reboot, the 1231 hangs until I turn my system off. When I disconnect the SSDs, the problem disappears. Strangest thing I've ever seen.
Already fixed the problem?Quote:
Originally Posted by Tiltevros
I cant find data, how much cache the 9211 have onboard... can anybody help?
16mb
LSI SAS 9211-8i 08.00.00 new firmware release
http://www.lsi.com/storage_home/prod...-8i/index.html
Product: Firmware
Version: 08.00.00.00
Release Date: 02-NOV-10
Release Type: GCA
Defects Fixed (64)
Enhancements Implemented (34)
Defects Fixed (64):
SCGCQ00102232 - PL: Unused phys not disabled on some boards
SCGCQ00150956 - Unwarranted SAS TOPOLOGY CHANGE LIST EVENT with PhyStatus change indication
SCGCQ00151226 - Possible to send a NCQ command during NCQ error handling
SCGCQ00151272 - 6001 format when inquiry is issued while format is active to a SATA device
SCGCQ00151465 - PCI Class Code SubCode value is incorrect in Thunderbolt NVDATA
SCGCQ00125128 - IR: (IR only) Removing a multipath IR disk multiple times results in a fault (0x8109).
SCGCQ00150267 - Unplugging USB end device during IOP USB TEST hangs firmware
SCGCQ00151004 - USB: Enumeration somestime causes fault 0xD04
SCGCQ00153033 - A config page request to erase log entries in the flash fails.
SCGCQ00155949 - Host and Library API names to contain 'i' to denote interfacing API.
SCGCQ00155966 - Phase 8 IT: Fault 265D during driver load and device discovery in parallel
SCGCQ00156138 - Hot Spare not getting pulled in after removing a member drive from IM volume
SCGCQ00156250 - USB - Remove debug memory allocation from USB init
SCGCQ00157653 - PM: Need to disable SAS link power mgmt due to outstanding HW issues
SCGCQ00157667 - USB - CBW, CSW and Status incorrectly has only one instance for two ports
SCGCQ00157920 - Link Reset is happening within one second after getting invalid initial FIS
SCGCQ00157932 - (IOP-Only) Need to add additional hardware EDRAM failure detection
SCGCQ00158934 - Initialization of DDR controller fails when the memory clock is 667MHz
SCGCQ00151857 - (GEMINI) Program Interrrupt seen when in FAULT routine
SCGCQ00155202 - Discovery Error 80h
SCGCQ00152992 - 0x0614 fault when aborting SATA passthrough messages
SCGCQ00099988 - PL: Msync needed to update VP_ID of SCSI IO request before posting the IO to SAS Core.
SCGCQ00149654 - iop usb portstop halts firmware
SCGCQ00149684 - Constant ORR after injecting a wrong-direction data frame
SCGCQ00150990 - Thunderbolt: Implement hardware workaround for bug track 758
SCGCQ00151363 - USB library is not using correct IOP interface functions
SCGCQ00151455 - ManPage 8 Device Type BOOT_ROM and SEEPROM should be unique types
SCGCQ00151755 - PL: Discovery Error MPI2_EVENT_SAS_DISC_DS_MULTI_PORT_DOMAIN
SCGCQ00151758 - Ethernet FWDL: Firmware download from a TFTP Server occasionally fails.
SCGCQ00152176 - Invalid interrupt from memory controller
SCGCQ00152263 - IOP: Uart console command "iop show sbr" only shows 1st copy of SBR correctly
SCGCQ00152526 - Thunderbolt: The SHELL speed is set wrong in SBR
SCGCQ00152549 - (PL-Only) Debug text strings are out of sync with the order of data
SCGCQ00152980 - Meteor with DDR enabled faults with 0002 then 265D at boot up
SCGCQ00153003 - SATL: 620F fault with reassign blocks + invalid parameter list header
SCGCQ00153048 - PL: Port Enable and Config page requests timeout during large topology discovery.
SCGCQ00153107 - Tuning needed for SATA power management
SCGCQ00153117 - SATL: Wrong tag reported when unable to determine tag
SCGCQ00153144 - Manufacturing Page Block create routine should not be part of cfgi library build.
SCGCQ00153253 - Setting the Min and Max value for the resources in Mfgpag9 to ffff leads to fwfault 0x0302
SCGCQ00153284 - CKE signal in Memory subsystem glitches when entering self refresh mode
SCGCQ00153294 - PCIe link width change returns 300a0007 loginfo even when request is honored exactly
SCGCQ00153578 - iop: iopidynmalloc() won't allocate fast - dma-able memory
SCGCQ00153605 - The SEP on a specific enclosure is not being seen.
SCGCQ00154962 - IR Load Balancing
SCGCQ00155425 - Phase 8: IR: Misleading comments in Setting and clearing data Scrub allowed flag
SCGCQ00155602 - Thunderbolt: Bugtrack 770 sys dma to bad host address
SCGCQ00155693 - Firmware might fault with code 0x4300 during SATA device initialization
SCGCQ00155755 - Fault 5803 during discovery of large topology after adjusting Ph8 Alpha FW NVDATA on Liberator B4.
SCGCQ00156721 - Fault 0x265d on port enable.
SCGCQ00125874 - Don't fault if TM sent to an Expander
SCGCQ00155280 - DMA sense data with FW instead of HW to prevent a hang of the DMA transfer.
SCGCQ00146928 - Rescaning of large topology with 90 drives during make and break of wide port not handled properly by driver
SCGCQ00154633 - 265d fault is seen when volume member drives are removed & inserted simultaneously
SCGCQ00156204 - Fault x6815 with Mustang in I/T switching running Target Resets
SCGCQ00157807 - IOP: Fault information sometimes not displayed to UART after a fault
SCGCQ00157838 - USB - FW is not able to communicate with USB end device
SCGCQ00157899 - IR: Possible data corruption with phys devices returning at DMD timer expiration
SCGCQ00158317 - Fault 6051 or Long TM response times for Target Resets
SCGCQ00158423 - Trying to flash IR firmware image on a controller with no NVSRAM faults with 265D fault code
SCGCQ00160490 - Change EEDPTagMode to 0 in default NVDATA
SCGCQ00160605 - SAS target mode transmit DMA engine incorrectly asserts Auto Response Not Sent status
SCGCQ00161827 - Update firmware to check new PCI Device ID for 9205-8e channel board
SCGCQ00161828 - Mustang default NVDATA has Device ID set to 0x80 instead of 0x86 at several places .
Enhancements Implemented (34):
SCGCQ00102813 - IOP: Change Start of Day memory allocation for Raid Accelerator request queues
SCGCQ00118431 - (IOP-Only) Add dynamic lookup of config page buffers at SOD
SCGCQ00128289 - (IOP-Only) Add firmware ManPage9 fixed Min capability
SCGCQ00136668 - Enable sending 3 identify frames
SCGCQ00140397 - MPI2: modify Host Based Discovery Action Request
SCGCQ00141686 - Ph7 IR: Do read IO load balancing for RAID10/RAID1E ASM volumes for better performance
SCGCQ00141834 - MPI2: ability to disable SATA NCQ on a per-device basis
SCGCQ00142578 - MPI2: obsolete MPI2_IOUNITPAGE1_MULTI_PATHING
SCGCQ00142855 - LSI2008 SATA NCQ enablement
SCGCQ00142923 - Allow Config Write Current of SAS IO Unit Page 1 after PortEnable
SCGCQ00142955 - Change how discovery handles spinup hold for sata devices
SCGCQ00143162 - Eliminate PL Config Page restrictions
SCGCQ00145313 - Bit 20 of RAID Volume Page 0 to comply with definition as in MPI 2 spec rev M or later
SCGCQ00146007 - Change DMA engine default threshold values for System and DDR data
SCGCQ00147761 - Customer requested change to DCR requirement placed on IOP by DDR3 library
SCGCQ00147804 - Ph8: Host Based Discovery Request change
SCGCQ00147885 - Method to prevent drivers from resetting firmware during large CLI command
SCGCQ00148227 - Send Execute Device Diagnostic ATA command during SATA Initialization
SCGCQ00148567 - pl: SATL: Support write buffer mode 7.
SCGCQ00148569 - pl: SATL: Better handling of invalid signature FIS
SCGCQ00149166 - PL: Add compile option to fault when Discovery detects Discovery Error 0x90 (SMP Timeout)
SCGCQ00150278 - Update Phase 8 Project with new MPI-2 header files
SCGCQ00151472 - Support exiting self-refresh mode in DDR3 library, for battery backed implementations
SCGCQ00151477 - Thunderbolt: Increase max message frame to 1008
SCGCQ00152166 - (NVDUMP only) Add capability to parse XSD files to provide formatted config data
SCGCQ00151361 - Allocate ISTWI device types and GPIO function codes for vendor specific use
SCGCQ00151468 - DDR3 Read and Write Leveling algorithms take too long
SCGCQ00154697 - DDR3 library needs to support different types of Raw Cards
SCGCQ00156735 - Add temperature sensor read capability to uart
SCGCQ00157613 - Thunderbolt SDMegawrapper settings rev 5
SCGCQ00159101 - Add temperature conversion
SCGCQ00151517 - USB: Create a package build
SCGCQ00152660 - Thunderbolt: Updated SDMegaWrapper values
SCGCQ00156191 - (PL-Only) Mask out unused debug code on the IO path when not used
Product: SAS2BIOS
Version: 7.15.00.00
Release Date: 01-NOV-10
Release Type: GCA
Defects Fixed (15)
Enhancements Implemented (7)
Defects Fixed (15):
SCGCQ00148745 - LSI2116 Eval Meteor boards hang going into BIOS CU
SCGCQ00148757 - Add back a partially redundant call to ResetSHEX() because other paths aren't always executed
SCGCQ00151512 - SAS2BIOS - CU: SATA Disk Capacity value always zero for expander attach
SCGCQ00148946 - After expanding second enclosure in the CU, data gets corrupted.
SCGCQ00152052 - SAS2BIOS - Calls to ReadConfigPageRaidPhysdisk0 return invalid data
SCGCQ00157065 - While activating a failed volume the BIOS CU should give some error message if the activation has failed due to any reason.
SCGCQ00158310 - SAS2BIOS CU reports different sizes for hard disks on device properties screen
SCGCQ00155163 - SAS2BIOS CU Failed to showed the first path in BIOS topology
SCGCQ00138021 - WH: Tools report volume size differently
SCGCQ00157562 - BIOSCU hangs while booting/entering CU with 2Falcon IR & 1Meteor IT HBA.
SCGCQ00160446 - One dot (".") remains behind while 'reassigning the blocks' command executed in Device Verify screen.
SCGCQ00153779 - One dot (".") remains behind while erasing the status message of a drive while verifying it.
SCGCQ00154583 - sas2 BIOS CU is not providing more descriptive error information.
SCGCQ00154677 - Invalid topology displayed if vertical scrolling has occurred and an enclosure is unplugged
SCGCQ00154707 - SAS2BIOS returns format success when device is disconnected during format
Enhancements Implemented (7):
SCGCQ00135558 - SAS2BIOS: Add support for non 512-byte sector drives in BIOS CU
SCGCQ00145547 - sas2 BIOS CU should provide more descriptive error information.
SCGCQ00147738 - Handle hardware error in 2208-1 (B0) by alternating all I/O port access and reset SHEX at start of day
SCGCQ00151383 - SAS2BIOS: CU - Add SCSI I/O Command Serial Trace in BIOS CU
SCGCQ00151267 - SAS2BIOS - Add Mustang support to SHEX reset implementation
SCGCQ00148528 - Option ROM Slot and Volume Ordering Enhancement for HP WGBU
SCGCQ00148011 - Handle hardware error in 2208-1 (B0) by alternating all I/O port access in the BIOS CU