Topic: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

I'm no expert on ini and if files, but I can figure out some things. But this noe has me stumped.

In DriverPack_MassStorage_wnt5_x86-32.ini I see:

[C]
ms_count=3
ms_1_deviceName="Compaq HP ProLiant SCSI"
ms_1_tag="symc8xx"
ms_1_sysFile="CPQ32FS2.sys"
ms_1_hwids="PCI\VEN_1000&DEV_0001,PCI\VEN_1000&DEV_0003,PCI\VEN_1000&DEV_000a,PCI\VEN_1000&DEV_000b,PCI\VEN_1000&DEV_000c,PCI\VEN_1000&DEV_000f,PCI\VEN_1000&DEV_0012"
ms_1_isBusExtender=false
ms_1_exc_skipIfOS="w2k"
ms_1_exc_replaceIfOS="wxp,w2k3"
ms_2_deviceName="LSI Logic C8100 PCI SCSI Host Adapter"
ms_2_tag="symc810"
ms_2_sysFile="CPQ32FS2.sys"
ms_2_hwids=""
ms_2_isBusExtender=false
ms_2_exc_skipIfOS="w2k"
ms_2_exc_replaceIfOS="wxp,w2k3"
ms_3_deviceName="LSI Logic C896 PCI SCSI Host Adapter"
ms_3_tag="sym_hi"
ms_3_sysFile="CPQ32FS2.sys"
ms_3_hwids=""
ms_3_isBusExtender=false
ms_3_exc_skipIfOS="w2k"
ms_3_exc_replaceIfOS="wxp,w2k3"

What does ms_2_hwids="" mean? Use the hwids listed previously? Or what?

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Well, technically I think it means "all" or "any".  dpms 901 had the hwids filled, but they were empty starting with 1006.  I don't know why.
Is it causing a problem for you?

Read BEFORE you post.  HWID tool   DriverPacks Tutorial   DONATE!
http://driverpacks.net/userbar/admin-1.png
Not all heroes wear capes, some wear Kevlar!

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

http://forum.driverpacks.net/viewtopic.php?id=4305

-----snip-----  from testing team forum ... me last year  ----- snip -----

Mass 10.01.24 nightly Jeff aka OverFlow (Jan 24 2010)
Link Retired

updates appplied to JakeLD's 909r6 (1 September 2009)

updated
Replacement drivers that were missing HWIDs
(If a driver is a replacement it MUST replace all HWIDs supported by the previous driver .

dpti2o  [AD4]
- verified


adpu160m [AD5]
- restored hwid
PCI\VEN_9005&DEV_0081


symc8xx [C] ms_1
- need special case for a single driver that replaces two or more native drivers...
- Fixed, simply assign no HWIDs to one or more (ms_X) section(s)
- combined old ms_2 and ms_3 into ms_1
symc810 [C] ms_2
- see above. replaced.
sym_hi  [C] ms_3
- ms_2 is also replaced
-- see section for example!

------- snip ----------  ------------ snip ----------


the replacement driver with NO HWID needs removed becuase it is included with another driver... by adding a driver with no HWID BASE will instead of replacing a native driver, will remove it. This was neccessary because ONE of our DriverPacks drivers replaces THREE windows included (native) drivers. Again... That section replaces three native drivers with one replacement. IE The three windows included drivers symc8xx, symc810, sym_hi were all replaced by CPQ32FS2 and the first section contains all the HWIDS for all three drivers (because they are all supported by the one driver now)

Don't try to figure it out ... let us do that tongue.  I'm not surprized it stumped you... I had a really fun time coming up with a workaround / fix for that.

Thanks for paying attention!

Jeff

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Thank you. I'm working with a program someone else wrote, which incorporates the MassStorage driverpack into a WinPE build by copying the appropriate files and modifying txtsetup.sif. It didn't understand the blank hwid and inserted an invalid entry into txtsetup.sif so the result wouldn't boot. Now I need to figure out how to remove the drivers.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Which program?

PE 1 or PE 2?

Restore the backed up copy of the txtsetup file... the program did make a backup of the files it modified didn't it? or at least you did right? If not you will have to start over... The undo would take a lot more time than starting over, no doubt about that.

Always make a backup at each stage of a build big_smile.

suggestion:

slipstream DriverPacks into a source, use that txtsetup file and overwrite the corrupt unit. big_smile or cut and paste the good sections into your corrupt sections of txtsetup.

The best way to do this is to use a DriverPacks slipstreamed source for your PE build...

Can you tell us more about what you're doing and why... (I just use BartPE / UBCD4Win with my DriverPacks Mass storage plugin, why not do that?)

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

BartPE is stone age! smile (I used it for years but moved on long ago). And UBCD4Win isn't sufficiently customizable for my needs. I do outsourced IT and in-home computer repair, so I see a wide range of hardware. The ultimate goal is to be able to boot from CD/DVD/Flash drive and see the local disks.

I'm doing both PE1 and PE2, but the current issue is with PE1. I'm using WinBuilder, which has a pretty powerful scripting language. The current issue is a script that inserts the drivers at build time into a txtsetup.sif that's originally copied from the source. I suppose it could be changed to slipstream into the source, but I'd want to have a reliable method of detecting what has been slipstreamed in.

Right now I'm just skipping the driver with the blank hwid, which may be acceptable since it's seldom critical that the most recent driver be available, what's really required is a driver that works.

I don't quite see a good algorithm for processing that section. It appears that I have to read, parse, and store all the information in that section before I can process the information. I might have chosen to put the non-blank hwid string at the end rather than at the beginning so a program would see the blank hwid first and trigger special processing. But what's done is done.

Last edited by JonF (2011-01-21 04:58:24)

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

you can't skip the the driver with the blank HWID... that will cause MEGA issues...
IE duplicated HWIDs = Crash city
You must REMOVE ALL references from txtsetup for the drivers with blank HWIDs that is the point of haveing them in the mass ini.

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

I don't understand the problem ...

Suppose I deleted the [C] section completely before processing. Would that cause problems?

That's effectively what I'm doing. Haven't seen a crash yet.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

the problem is that the DriverPacks driver replaces three drivers that windows includes...

each of the three drivers has a unique set of HWIDs... all of the HWIDs are included in a newer driver... so all three drivers must be eliminated from txtsetup so that the newer one can be added. The script you described that you are running will leave two of the drivers behind so all of the HWIDs that are included in those two drivers will be duplicated. If you run into one of the affected controllers it will find the HWIDs in txtsetup twice... That will be disaster...

Yes if you remove the entire section then the primary problem will not occur. However you will be running very old drivers for that hardware... those drivers are for what were at the time high end controllers. If you try to run outdated drivers that MAY also lead to issues. The best thing to do is to have the script properly deal with "a single updated driver that repalces multiple included drivers" Which is what i did in DriverPacks BASE. big_smile

Jeff


PS i wrote a very detailed description of the problem above describing it three different ways and citing a specific example.

BartPE may be outdated but it seems like it it miles ahead in mass storage support. that is to say it will not suffer from this issue if you use DriverPacks plugins (currently suporting Mass LAN WLAN and Graphics)

PE2 yeah you got me there ... Give me a minute tongue
But with Dism it is almost a waste of my time to mess with it

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Well, it's looking as if the best thing to do is to start from scratch and build the script based on your DPs_Base. Once it's done it will be miles ahead of the BartPE plugin smile.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Cool... let me know if you have any questions!

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Here's a couple of questions ...

Not all the HWIDs that appear in section [C] appear in CPQ32FS2.inf. How are we sure that this one new driver replaces all new hwids?

Should I just mimic the process of setting up and then evaluating the BartPE plugin. Or would one of the pointing it to the I385 directory tha's being built be better? When this script runs, all files that get copied from the source have already been copied.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

JonF wrote:

Here's a couple of questions ...

Not all the HWIDs that appear in section [C] appear in CPQ32FS2.inf. How are we sure that this one new driver replaces all new hwids?

Because we have extensively tested it. Please remember this driver is used for text mode only and is replaced during PE stage with the final PnP driver big_smile. (Text mode drivers are used during the PE stage itself.)


JonF wrote:

Should I just mimic the process of setting up and then evaluating the BartPE plugin. Or would one of the pointing it to the I385 directory tha's being built be better? When this script runs, all files that get copied from the source have already been copied.

Yes BartPE is based on and is very similar to WinPE. That would be the very best place to start. However It is always a good idea to compare it to a Slipstreamed I386 source. We know that our slipstreamed source is 99.9% perfect. if ther are differences (please report them) then the slipstreamed version is more likely to contain the best example.

Answer: Both... lol big_smile.

Jeff

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Hi OverFlow,

thanks for all clear replies (specially to post 3)

just to verify:
so CPQ32FS2.SYS file copied with names SYMC8XX.SYS , SYMC810.SYS , SYM_HI.SYS to overcome 1 driver instead of 3 native driver issue, instead same file with 3 names wink. Can you confirm this is intentional ?
(also re-tested with untouched pebuilder311a , DPs_BASE_1006 , DP_MassStorage_wnt5_x86-32_1101)




some comments about things on this topic:
yes, pebuilder is best reference, since it is stable , trustable, and great work of Bart Lagerweij which shows his accurate far sight I admire a lot.
Heh, comparing with winbuilder, winbuilder have oppositely famous with its unstable and untrustability, but sadly for today it is what many have at hand.
(I hope one day a good programmer provide a better pe-build-scripting tool, hope never ends wink)
anyway,
we, nt5x fans, want more smile
I am making pe1x64 builds since ~2009-10 along with multi language support (chinese etc.) already integrating some custom x64 drivers but sadly we do not have a mass storage package for today.
Is there any hope (take this as a honest request) to have DP_MassStorage_wnt5_x64..... one day, which I believe increase popularity of both pe1x64 and xpx64 smile.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

if the mass ini contains a blank HWID that driver is REMOVED

IE

[C]
ms_count=3
ms_1_deviceName="Compaq HP ProLiant SCSI"
ms_1_tag="symc8xx"
ms_1_sysFile="CPQ32FS2.sys"
ms_1_hwids="PCI\VEN_1000&DEV_0001,PCI\VEN_1000&DEV_0003,PCI\VEN_1000&DEV_000a,PCI\VEN_1000&DEV_000b,PCI\VEN_1000&DEV_000c,PCI\VEN_1000&DEV_000f,PCI\VEN_1000&DEV_0012"
ms_1_isBusExtender=false
ms_1_exc_skipIfOS="w2k"
ms_1_exc_replaceIfOS="wxp,w2k3"
ms_2_deviceName="LSI Logic C8100 PCI SCSI Host Adapter"
ms_2_tag="symc810"
ms_2_sysFile="CPQ32FS2.sys"
ms_2_hwids=""
ms_2_isBusExtender=false
ms_2_exc_skipIfOS="w2k"
ms_2_exc_replaceIfOS="wxp,w2k3"
ms_3_deviceName="LSI Logic C896 PCI SCSI Host Adapter"
ms_3_tag="sym_hi"
ms_3_sysFile="CPQ32FS2.sys"
ms_3_hwids=""
ms_3_isBusExtender=false
ms_3_exc_skipIfOS="w2k"
ms_3_exc_replaceIfOS="wxp,w2k3"

"CPQ32FS2.sys" is renamed to  symc8xx (over writing it)
symc810 and sym_hi are removed completely

HOWEVER the HWIDS that symc810 and sym_hi previously supported are now supported by the new CPQ32FS2.sys driver that is now renamed to symc8xx.sy_

so all the HWIDS from all three drivers are now supported by symc8xx (which is actually CPQ32FS2)

I don't mean to be an @$$ but there is no possable way that our team will ever support XP64. It was, and always will be, one if the biggest POS that MS ever came out with ranking right up there with windows ME and Vista...

Sorry No hope at all. (you asked so nicely i really am sorry)
http://forum.driverpacks.net/viewtopic.php?id=2651
Asked and answered a thousand times hmm

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

OverFlow wrote:

symc810 and sym_hi are removed completely

Thanks, that was what I was expecting too cool
Than only reporting: dps_base - bartpe plugin do not delete these files and entries in txsetup.sif (tested with all default pebuilder etc.), instead it overwrites the way I explained on previous post.
ps: I do not know pebuilder syntax well....

OverFlow wrote:

Sorry No hope at all. (you asked so nicely i really am sorry)
http://forum.driverpacks.net/viewtopic.php?id=2651
Asked and answered a thousand times hmm

Sorry for taking time and thanks for the reply.
I guess in the end, one day, we would have to make a small package to support pe1x64 :>, probably following great DriverPacks arrangements smile.

edit:
further, I tested integrating to winxpsp3enu cd, DPs_BASE_1006 - DP_MassStorage_wnt5_x86-32_1101.7z
this one worked better than bartpe plugin wink and deleted all entries from txtsetup.sif (leaving only symc8xx entries as it was explained before)
still,
SYMC810.SY_ ve SYM_HI.SY_ exists in final result (inside each file there is CPQ32FSFS2.SYS file) , so not complete removal wink
also SYMC8XX.SY_ contains CPQ32FSFS2.SYS but I guess SYMC8XX.SY_ should be containing SYMC8XX.SYS (actually CPQ32FSFS2.SYS)

from both results my conclusion is:
DriverPacks mass storage package arrangements finished wink but dpsbase application not updated following these changes which cause confusions, minor things we all living through time to time, I hope next dpsbase utility will clear all possible confusions.

Last edited by Lancelot (2011-02-01 22:04:32)

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

those files are left behind and unchanged (SYMC810.SY_ SYM_HI.SY_) in case DriverPacks is uninstalled...

So yes it is a complete removal... you may notice that all of the old drivers are retained for purposes of uninstallation.
What you see as unfinished is not the case at all, they are left intentionally and deliberately. big_smile This may be confusing if you don't think it through completely LOL wink. I hope i have cleared your possable confusion (no update to BASE is required, it already is processing the files 100% correctly. BASE will throw an error if the files do not get overwritten during uninstall.

On the other hand
Yes the BARTPE section needs an update that i have not had time to implement yet. No bug report has surfaced as a result (yet).

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

OverFlow wrote:

No bug report has surfaced as a result (yet).

I do not think any bug report for bartpe or anyone asking why new unused files on windows cd since both will work on the field nicely.
I understand (agree , fine.....) your decision not updating dps_base, not urgent at all. big_smile

OverFlow wrote:

I hope i have cleared your possible confusion

YES,
Your explanations %500 cleared all confusions and shines our way tongue,
Thanks a loooot for Strong Support. cool

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Hi,

one more very mini confirmation question about this matter wink

following

OverFlow wrote:

If a driver is a replacement it MUST replace all HWIDs supported by the previous driver .

in xpsp3 txtsetup.sif there are

PCI\VEN_1000&DEV_0002 = "symc8xx"
PCI\VEN_1000&DEV_0004 = "symc8xx"
PCI\VEN_1000&DEV_0005 = "symc8xx"
PCI\VEN_1000&DEV_0006 = "symc8xx"
PCI\VEN_1000&DEV_000D = "symc8xx"

entries,

in dps_base + dps1101 integrated wincd-xpsp3 txtsetup.sif, none of above exists.
in dps_base + dps1101 integrated pebuilder-xpsp3 txtsetup.sif, all above exists.


OverFlow wrote:

Don't try to figure it out ... let us do that tongue

I only just want a confirmation if it is intentional smile so i can follow the shining path on pe1 cool.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

If those HWIDS are missing from

ms_count=3
ms_1_deviceName="Compaq HP ProLiant SCSI"
ms_1_tag="symc8xx"
ms_1_sysFile="CPQ32FS2.sys"
ms_1_hwids="PCI\VEN_1000&DEV_0001,PCI\VEN_1000&DEV_0003,PCI\VEN_1000&DEV_000a, PCI\VEN_1000&DEV_000b,PCI\VEN_1000&DEV_000c,PCI\VEN_1000&DEV_000f,PCI\VEN_1000&DEV_0012"
ms_1_isBusExtender=false
ms_1_exc_skipIfOS="w2k"
ms_1_exc_replaceIfOS="wxp,w2k3"

Then it is a mistake / BUG ... (MR_smartEpants! HELP!!!)

I had that right when i fixed the bug... They must have gotten dropped (again) diring a subsequent update of DriverPack MassStorage.

GOOD EYE my man!!!! We'll get that fixed!

Normaly we say please "help us to help you"  BUT...
Now i think i have to say "Helping you has helped us!" Thank you.

Jeff tongue

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

OOPS!
I'll fix that right away.  Thanks for the keen eye!

Read BEFORE you post.  HWID tool   DriverPacks Tutorial   DONATE!
http://driverpacks.net/userbar/admin-1.png
Not all heroes wear capes, some wear Kevlar!

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Thanks OverFlow , Thanks mr_smartepants

also keep in mind to check missing (not replaced)
PCI\VEN_1000&DEV_0012&SUBSYS_001b0e11 = "sym_hi"
PCI\VEN_1000&DEV_000b&SUBSYS_60040e11 = "sym_hi"
entries too wink
(maybe they are not required anymore, but better a 2nd check made by you guys to be sure ....  , well I am pretty sure you will make a clean check....)

OverFlow wrote:

GOOD EYE my man!!!! We'll get that fixed!

All Credits to Olegpov, who catched missing hwids with his well known sharpe eagle eyes.
And thanks to JonF and Saydin77,
We had some days of confusions since JonF figured out things 18-Jan , I am very happy now everything related will be back to normal.


Overall,
THANKS TO GREAT DP TEAM.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

That's how things get done around here...
It's all about the teamwork!
As individuals this could never be done... As a community / team we contine to succeed.

Thanks to you guys for being part of our team

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

I deeply know and fully agree OverFlow,

"Team Work" is the key to many things, I am always happy writing to dp since everytime I write so far, things solved with good topic posts , self-proving how nice you maintain Good Team Work.

we are a small community here too having full team spirit, feeling the same with you.

just one mini thing I noticed to underline since things will be fixed on next release,
on all tests, CPQ32FS2.SYS compressed inside SYMC8XX.SY_  (I mean inside SYMC8XX.SY_ file there is CPQ32FS2.SYS file with CPQ32FS2.SYS name)
With not being an expert like you guys, I do not know if name mismatch cause trouble.... only noticed and reporting here wink.

Thanks again.

Re: [-] Meaning of a blank hwid in DriverPack_MassStorage_wnt5_x86-32.ini?

Good question... Here's a good answer.

The compressed files have to be renamed so they can coexist in the i386 folder and when they are cached to RAM. They are only extracted when needed. Many drivers will fail if they are renamed. (Actually renamed) so the files INSIDE the cab files keep thier original names. This keeps most or all of them from crashing when used. The possability exists that a maichine MIGHT have two controlers that require two different drivers that both contained a same named file... As far as i know this has never happened. It could... But the odds are millions to one.

Recap It does not matter waht we name the "wrapper" / archive / *.sy_ File... BUT most of the time renameing the contents of the archive / original file will cause failure. Safest thing to do is to NOT rename the original file in the archive.  (i belive this is a some kind of a result of the driver calling itself)

HAGD!

Jeff

DP BartPE Tutorial   DP_BASE Tutorial   HWID's Tool     Read BEFORE you post    UserBars!
http://driverpacks.net/userbar/admin-1.png
The DriverPacks, the DP_Base program, and Support Forum are FREE!.