If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#41
|
|||
|
|||
Strange memory problem
"????? ????? (Igor Leyko)" wrote in message
... "Walter Clayton" wrote in message ... "????? ????? (Igor Leyko)" wrote in message ... "Walter Clayton" wrote in message ... To repeat the cause: The issue is how 9x manages ring 0 and what drivers, especially non-MS supplied drivers expect to see. It's not unusual that those drivers are overflowing memory counters thus causing a false SOS condition. One of the weaknesses of 9x kernel is that when a driver requests memory the kernel can not fail the memory request. Either the getmem is honored or the system goes down. Drivers are loaded much earlier than you're assuming. ;-) Yes, at this moment they are loaded but not running yet. Just loaded. So no memory requests by getmem can be done. You may want to start he http://www.microsoft.com/technet/arc...t1/wrkc05.mspx --and-- http://www.microsoft.com/technet/arc...t6/wrkc28.mspx Great idea. I started from the similar document in an early 1995. It was a preliminary version of ResKit on beta CD. Unfortunately ResKit has very few information about "insuffisient memory for initialization" problem and can not exlpain why VCACHE cause a crash or blue screen. It takes drivers running in real mode to initiate bootstrap. Part of that process is loading and initializing VMM32.vxd which does have to allocate memory for drivers. There is a lot of stuff happening prior to and "concurrent" with VMM32.vxd being loaded. VMM - Virtual Machine Manager (sorry, I'm speaking about VMM, not about VMM32.vxd which is a set of modules) displays "insufficient memory" if it cannot start initialization. I suppose this is clear, isn't it? -- ????? ????? (Igor Leyko) MS MVP Windows - Shell/User VMM is initialized much latter in the boot sequence and very much dependant on not only real mode, but protected mode drivers as well that are issuing demands on memory. You're still circling around the central issue. It's ring 0 and the drivers, including vcache, that reside there that are miscounting real free space. ;-) To put it another way, if you're looking for 100 free blocks, you only expect a value between 1 and 999 on a query of free resources and the return is 1001 then you're going to throw an error since you're going to truncate to a value of 1; which is significantly smaller than 100. |
#43
|
|||
|
|||
Strange memory problem
Well, now I've clicked SmartDrv.exe in Explorer, & a DOS box opened to say...
"You cannot load SMARTDrive when a shell program (such as MS-DOS shell) or Windows is running. Quit Windows or the shell program, and then load SMARTDrive." ....It will only run in a separately pre-opened DOS box, by typing it's name... C:\SmartDrv /? Installs and configures the SMARTDrive disk-caching utility. SMARTDRV [/X] [[drive[+|-]]...] [/u] [/C | /R] [/F | /N] [/L] [/V | /Q | /S] [InitCacheSize [WinCacheSize]] [/E:ElementSize] [/B:BufferSize] /X Disables write-behind caching for all drives. drive Sets caching options on specific drive(s). The specified drive(s) will have write-caching disabled unless you add +. + Enables write-behind caching for the specified drive. - Disables all caching for the specified drive. /U Do not load CD-ROM caching module. /C Writes all information currently in write-cache to hard disk. /R Clears the cache and restarts SMARTDrive. /F Writes cached data before command prompt returns (default). /N Doesn't write cached data before command prompt returns. /L Prevents SMARTDrive from loading itself into upper memory. /V Displays SMARTDrive status messages when loading. /Q Does not display status information. /S Displays additional information about SMARTDrive's status. InitCacheSize Specifies XMS memory (KB) for the cache. WinCacheSize Specifies XMS memory (KB) for the cache with Windows. /E:ElementSize Specifies how many bytes of information to move at one time. /B:BufferSize Specifies the size of the read-ahead buffer. | I don't know what is smartvsd.vxd, I rhink I never had it on my computers. You've got to have it! It's a genuine Win98 file! However, it is something else altogether. -- Thanks or Good Luck, There may be humor in this post, and, Naturally, you will not sue, should things get worse after this, PCR "????? ????? (Igor Leyko)" wrote in message ... | It is not a separate file. Smartdrv.exe is both DOS and Windows program. The | Windows programs have a small DOS part often called "stub". When a program | is started from DOS, DOS runs only stub which dispays "This program require | Windows" or similar text. Windows recognize a main part of a program and | runs it ignoring a stub. In Smartdrv.exe DOS part is not a stub but a useful | program. Its size is ~90% of file. The last 10 % is a Windows driver which | cannot be started from DOS but can be loaded into Windows. Its internal name | is SD.VXD and you may see it in suitable file viewer. | This VXD really does nothing more then displaing a message "A serious error | occurs..." and allow to retry operation or crash Windows to prevent | additional errors. Windows 3.1 users remember such window I suppose. | I have no ideas when this driver can work in Win9x. | | I don't know what is smartvsd.vxd, I rhink I never had it on my computers. | | -- | ????? ????? (Igor Leyko) MS MVP Windows - Shell/User | | | "PCR" wrote in message | ... | I can't find a SmartDRV.vxd anywhere I've looked, including inside | VMM32.VXD. I do find a SmartVSD.vxd, but the lines in my own Bootlog.txt do | say... | | [0017B703] Loading Vxd = C:\Windows\SMARTDRV.EXE | [0017B704] LoadSuccess = C:\Windows\SMARTDRV.EXE | | ..., same as Zhilin's. But it isn't currently loaded (looking in DrWatson's | tabs & DOS's MEM). It does seem that SmartVSD.vxd is loaded & remains | loaded... | | [0017B757] Dynamic load device C:\Windows\system\IOSUBSYS\smartvsd.vxd | [0017B758] Dynamic load success C:\Windows\system\IOSUBSYS\smartvsd.vxd | | | -- | Thanks or Good Luck, | There may be humor in this post, and, | Naturally, you will not sue, | should things get worse after this, | PCR | | "????? ????? (Igor Leyko)" wrote in message | ... | | "Mikhail Zhilin" wrote in message | | ... | | So I suppose it is using (if isn't deleted manually) only at the first | | stage | | of the boot process, as the accelerating engine. This part of my reply | was | | only to attract the attention, that not all is so simple like it | seems... | | | | | | Well, waiting a comments not to me and my knowledge but the subject and my | | article statements I can tell something about SmartDrivet. | | Windows loads not smartdrv.exe but smartdrv.vxd, Windows driver stored in | | smartdrv.exe since its appearing in MS-DOS 6. | | This driver communicates with DOS part of smartdrv. If a serious error | | occures the driver displays error message. It was required in Windows 3.1 | | which did nit have its own disk cashe. It was required in WfW 3.11 because | | it's VCACHE could not cache floppies. | | I believe there is no real need in this driver in Windows 95 and upper but | | it was not removed due to compatibility issues and because developers do | | not like the radical changes in code and algorithms. | | | | -- | | ????? ????? (Igor Leyko) MS MVP Windows - Shell/User | | | | | | | | |
#44
|
|||
|
|||
Strange memory problem
I can't find a SmartDRV.vxd anywhere I've looked, including inside VMM32.VXD. I do find a SmartVSD.vxd, but the lines in my own Bootlog.txt do say... [0017B703] Loading Vxd = C:\Windows\SMARTDRV.EXE [0017B704] LoadSuccess = C:\Windows\SMARTDRV.EXE ------------ My bootlog has the same, except the address allocation is totally different, is there a simple explanation for this? [000AF4AB] Loading Vxd = C:\WINDOWS\SMARTDRV.EXE [000AF4AB] LoadSuccess = C:\WINDOWS\SMARTDRV.EXE These entries are 10th & 11th on the bootlog.txt list. Gekko |
#45
|
|||
|
|||
Strange memory problem
|
#46
|
|||
|
|||
Strange memory problem
"Walter Clayton" wrote in message
... The issue is how 9x manages ring 0 and what drivers, especially non-MS supplied drivers expect to see. It's not unusual that those drivers are overflowing memory counters thus causing a false SOS condition. One of the weaknesses of 9x kernel is that when a driver requests memory the kernel can not fail the memory request. Either the getmem is honored or the system goes down. Drivers are loaded much earlier than you're assuming. ;-) Yes, at this moment they are loaded but not running yet. Just loaded. So no memory requests by getmem can be done. You may want to start he http://www.microsoft.com/technet/arc...t1/wrkc05.mspx --and-- http://www.microsoft.com/technet/arc...t6/wrkc28.mspx Great idea. I started from the similar document in an early 1995. It was a preliminary version of ResKit on beta CD. Unfortunately ResKit has very few information about "insuffisient memory for initialization" problem and can not exlpain why VCACHE cause a crash or blue screen. It takes drivers running in real mode to initiate bootstrap. Part of that process is loading and initializing VMM32.vxd which does have to allocate memory for drivers. There is a lot of stuff happening prior to and "concurrent" with VMM32.vxd being loaded. VMM - Virtual Machine Manager (sorry, I'm speaking about VMM, not about VMM32.vxd which is a set of modules) displays "insufficient memory" if it cannot start initialization. I suppose this is clear, isn't it? VMM is initialized much latter in the boot sequence and very much dependant on not only real mode, but protected mode drivers as well that are issuing demands on memory. Did you tried to detect an exact step when error occur? It is very easy - just look into bootlog.txt. You may see - initialazation do not occur but error appears. You're still circling around the central issue. It's ring 0 and the drivers, including vcache, that reside there that are miscounting real free space. ;-) Can you point to any information source? Yes, of course, it's ring 0 - it it a core initialization process. If it is just a bad driver problem why MS did not fix it in 98SE or Me? -- ????? ????? (Igor Leyko) MS MVP Windows - Shell/User |
#47
|
|||
|
|||
Strange memory problem
My Windows 98 CD does not have smartvsd.vxd.
-- ????? ????? (Igor Leyko) MS MVP Windows - Shell/User "PCR" wrote in message ... Well, now I've clicked SmartDrv.exe in Explorer, & a DOS box opened to say... "You cannot load SMARTDrive when a shell program (such as MS-DOS shell) or Windows is running. Quit Windows or the shell program, and then load SMARTDrive." ....It will only run in a separately pre-opened DOS box, by typing it's name... C:\SmartDrv /? Installs and configures the SMARTDrive disk-caching utility. SMARTDRV [/X] [[drive[+|-]]...] [/u] [/C | /R] [/F | /N] [/L] [/V | /Q | /S] [InitCacheSize [WinCacheSize]] [/E:ElementSize] [/B:BufferSize] /X Disables write-behind caching for all drives. drive Sets caching options on specific drive(s). The specified drive(s) will have write-caching disabled unless you add +. + Enables write-behind caching for the specified drive. - Disables all caching for the specified drive. /U Do not load CD-ROM caching module. /C Writes all information currently in write-cache to hard disk. /R Clears the cache and restarts SMARTDrive. /F Writes cached data before command prompt returns (default). /N Doesn't write cached data before command prompt returns. /L Prevents SMARTDrive from loading itself into upper memory. /V Displays SMARTDrive status messages when loading. /Q Does not display status information. /S Displays additional information about SMARTDrive's status. InitCacheSize Specifies XMS memory (KB) for the cache. WinCacheSize Specifies XMS memory (KB) for the cache with Windows. /E:ElementSize Specifies how many bytes of information to move at one time. /B:BufferSize Specifies the size of the read-ahead buffer. | I don't know what is smartvsd.vxd, I rhink I never had it on my computers. You've got to have it! It's a genuine Win98 file! However, it is something else altogether. -- Thanks or Good Luck, There may be humor in this post, and, Naturally, you will not sue, should things get worse after this, PCR "????? ????? (Igor Leyko)" wrote in message ... | It is not a separate file. Smartdrv.exe is both DOS and Windows program. The | Windows programs have a small DOS part often called "stub". When a program | is started from DOS, DOS runs only stub which dispays "This program require | Windows" or similar text. Windows recognize a main part of a program and | runs it ignoring a stub. In Smartdrv.exe DOS part is not a stub but a useful | program. Its size is ~90% of file. The last 10 % is a Windows driver which | cannot be started from DOS but can be loaded into Windows. Its internal name | is SD.VXD and you may see it in suitable file viewer. | This VXD really does nothing more then displaing a message "A serious error | occurs..." and allow to retry operation or crash Windows to prevent | additional errors. Windows 3.1 users remember such window I suppose. | I have no ideas when this driver can work in Win9x. | | I don't know what is smartvsd.vxd, I rhink I never had it on my computers. | | -- | ????? ????? (Igor Leyko) MS MVP Windows - Shell/User | | | "PCR" wrote in message | ... | I can't find a SmartDRV.vxd anywhere I've looked, including inside | VMM32.VXD. I do find a SmartVSD.vxd, but the lines in my own Bootlog.txt do | say... | | [0017B703] Loading Vxd = C:\Windows\SMARTDRV.EXE | [0017B704] LoadSuccess = C:\Windows\SMARTDRV.EXE | | ..., same as Zhilin's. But it isn't currently loaded (looking in DrWatson's | tabs & DOS's MEM). It does seem that SmartVSD.vxd is loaded & remains | loaded... | | [0017B757] Dynamic load device C:\Windows\system\IOSUBSYS\smartvsd.vxd | [0017B758] Dynamic load success C:\Windows\system\IOSUBSYS\smartvsd.vxd | | | -- | Thanks or Good Luck, | There may be humor in this post, and, | Naturally, you will not sue, | should things get worse after this, | PCR | | "????? ????? (Igor Leyko)" wrote in message | ... | | "Mikhail Zhilin" wrote in message | | ... | | So I suppose it is using (if isn't deleted manually) only at the first | | stage | | of the boot process, as the accelerating engine. This part of my reply | was | | only to attract the attention, that not all is so simple like it | seems... | | | | | | Well, waiting a comments not to me and my knowledge but the subject and my | | article statements I can tell something about SmartDrivet. | | Windows loads not smartdrv.exe but smartdrv.vxd, Windows driver stored in | | smartdrv.exe since its appearing in MS-DOS 6. | | This driver communicates with DOS part of smartdrv. If a serious error | | occures the driver displays error message. It was required in Windows 3.1 | | which did nit have its own disk cashe. It was required in WfW 3.11 because | | it's VCACHE could not cache floppies. | | I believe there is no real need in this driver in Windows 95 and upper but | | it was not removed due to compatibility issues and because developers do | | not like the radical changes in code and algorithms. | | | | -- | | ????? ????? (Igor Leyko) MS MVP Windows - Shell/User | | | | | | | | |
#48
|
|||
|
|||
Strange memory problem
On Sun, 9 Jul 2006 13:09:01 +0930, "Gekko" wrote:
I can't find a SmartDRV.vxd anywhere I've looked, including inside VMM32.VXD. I do find a SmartVSD.vxd, but the lines in my own Bootlog.txt do say... [0017B703] Loading Vxd = C:\Windows\SMARTDRV.EXE [0017B704] LoadSuccess = C:\Windows\SMARTDRV.EXE ------------ My bootlog has the same, except the address allocation is totally different, is there a simple explanation for this? [000AF4AB] Loading Vxd = C:\WINDOWS\SMARTDRV.EXE [000AF4AB] LoadSuccess = C:\WINDOWS\SMARTDRV.EXE These entries are 10th & 11th on the bootlog.txt list. These numbers are not address allocation. Each number is the time, in hexadecimal form (in milliseconds, if I recall correct), when the event occurred after beginning of the boot process,. These numbers are useful to troubleshoot the boot process, and are using by some diagnostic programs like BootLog Analyzer BLA32. -- Mikhail Zhilin MS MVP (Windows - Shell/User) http://www.aha.ru/~mwz Sorry, no technical support by e-mail. Please reply to the newsgroups only. ====== |
#49
|
|||
|
|||
Strange memory problem
"Andrea" wrote in message
om... I've a client with windows 98 installed. All was working fine since I've installed the antivirus and now a pc with 1GB of ram is unable to open the DOS shell since memory is finished. I really don't know ... Any idea? I don't think that the cause could be the antivirus, because on another client with 256MB of ram, the pc open and close the dos windows withouth any problems. Thanks for any reply This worked for 98SE via msconfig's correction by limiting RAM use in system.ini file. [386Enh] MaxPhysPage=1FFFF Also works in ME on same PC. Same PC has 1GB of RAM. Same PC also has XP. Runs fine on all 1GB of RAM. Same PC works fine when booting from a msdos 7.0 floppy, or 98SE startup floppy diskette. -- Jonny |
#50
|
|||
|
|||
Strange memory problem
"Jonny" wrote in message ... "Andrea" wrote in message om... I've a client with windows 98 installed. All was working fine since I've installed the antivirus and now a pc with 1GB of ram is unable to open the DOS shell since memory is finished. I really don't know ... Any idea? I don't think that the cause could be the antivirus, because on another client with 256MB of ram, the pc open and close the dos windows withouth any problems. Thanks for any reply This worked for 98SE via msconfig's correction by limiting RAM use in system.ini file. [386Enh] MaxPhysPage=1FFFF Also works in ME on same PC. Same PC has 1GB of RAM. Same PC also has XP. Runs fine on all 1GB of RAM. Same PC works fine when booting from a msdos 7.0 floppy, or 98SE startup floppy diskette. Not surprising - it limits the RAM available to Windows to less than 512MB -- Noel Paton (MS-MVP 2002-2006, Windows) Nil Carborundum Illegitemi http://www.crashfixpc.com http://tinyurl.com/6oztj Please read http://dts-l.org/goodpost.htm on how to post messages to NG's |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Here's a poser! | NevBud | General | 25 | January 25th 16 07:06 PM |
on-board memory problem | John | General | 5 | June 1st 05 12:03 AM |
Memory questions | Steve | General | 11 | April 8th 05 08:37 PM |
Strange Sound Problem | James Andrew Wright | General | 0 | August 21st 04 11:54 PM |
RAM | Iqbal | Software & Applications | 22 | July 24th 04 09:04 AM |