Windows for Workgroups: Avoiding Hardware Conflicts

Last reviewed: February 8, 1995
Article ID: Q99859
The information in this article applies to:
  • Microsoft Windows for Workgroups versions 3.1, 3.11
  • Microsoft Windows operating system versions 3.1, 3.11

SUMMARY

This article summarizes four possible sources of hardware conflicts: IRQ lines, upper memory address (UMA), base I/O address, and DMA channels. It also briefly describes the function of each and suggests a strategy for diagnosing and/or preventing hardware conflicts.

MORE INFORMATION

When you install peripheral devices, there are four possible sources of hardware conflicts. (Not all devices use all four parameters or settings, although some do. Most use IRQ, base I/O address, and/or UMA.) The key to avoiding hardware conflicts is to ensure that no two devices are configured to use any of the same settings.

A wide variety of symptoms may indicate hardware conflicts. These include system hangs, general protection (GP) faults (and other error messages), corrupted video displays, corrupted printing, limited functionality from one or more installed devices, intermittent performance, and so on.

Diagnostic programs, such as Microsoft Diagnostics (MSD), can sometimes help determine which IRQs and upper memory ranges are in use and which are available. Currently, however, computer card manufacturers do not adhere to any industry standard for reporting the current status of a card when probed by diagnostic software. Therefore, the only accurate way to determine the current settings for a card is to consult the manufacturer's documentation or inspect each card.

IRQs and DMA channels are single numbers (for example 1, 3, or 7). I/O base and upper memory addresses include or cover a range (for example, D000-D800 or 360h-380h). Typically, these settings are chosen through DIP switches or jumpers on the card. Sometimes the cards are software configurable or both hardware and software configurable. For more information, consult your manufacturer's documentation.

IRQ

An interrupt request line (IRQ) is a line trace etched on the circuit board from the expansion slot to the programmable interrupt controller (PIC) chip through which a peripheral device can notify the CPU. Under most circumstances, each device must use its own IRQ line. IBM PC/AT-compatible design architecture provides for a possibility of 16 IRQ lines (0-15), some of which are reserved for standard devices such as the keyboard, floppy disk drives, the system clock, and so forth. Communication (COM) ports 1 and 3 typically share IRQ 4; COM ports 2 and 4 share IRQ 3. Therefore, a potential conflict exists if devices are installed on both COM ports 1 and 3 or on COM ports 2 and 4.

Available IRQ lines (for sound cards, network adapters, scanners, bus mouse, video adapters, tape drives, SCSI interfaces, and so forth) usually include IRQs 5, 9, 10, 11, 12, 13, and 15. 8-bit expansion slots provide access to IRQs through 7. 16-bit slots generally provide access to IRQs through 15.

Not all devices allow the selection of any IRQ number. Consult the device's manufacturer or its documentation for information on which IRQs can be used by a particular device and select an IRQ that does not conflict with currently installed devices; or, change an installed device's IRQ to a different one to make available an IRQ for the new device.

NOTE: You may need to reconfigure application software settings after changing IRQs of previously installed devices.

UMA

A device's upper memory address (UMA) is a range of address space somewhere between A000 and FFFF for use by the device. This corresponds to "using" some portion of memory between 640 kilobytes (K) and 1 megabyte (MB). While physical system RAM cannot be used by the device, the "addresses" are used and are therefore unavailable to the system or other devices.

Conflicts can occur if two hardware devices try to occupy the same address range or if an application, a device driver, or a terminate-and-stay-resident (TSR) program does not detect that this address range is already in use and tries to occupy it.

The entire UMA range is the same range shared by the video adapter (A000-CFFF typically), the system ROM (0FFF-FFFF typically), and any installed cards that use UMAs. Furthermore, loading device drivers and TSR programs high places them in this range as well and can increase the potential for conflicts to occur.

To avoid UMA conflicts, use the following suggestions:

  • Install cards before you optimize your upper memory area. If you have already optimized the UMA and need to add a new card, load everything low, add the new card, and then try to load items high. Optimizing the UMA after installing cards avoids fragmenting the UMA range. This may allow more (or larger) TSR programs and/or device drivers to be loaded high.
  • Carefully plan the use of the entire UMA range when you install several peripherals and/or when you try to maximize conventional memory. Consult the manufacturer or the documentation to determine the memory range required by each device driver, TSR program, and peripheral. Note the options for this range's "starting" address (for example, C000, D000, or D800). Ensure that no two devices (device drivers or TSR programs loaded high or peripherals) are sharing any part of a range within this space.
  • It may be necessary to use the EMM386.EXE "exclude range" switch (X=) in the CONFIG.SYS file or the EMMexclude= parameter in the [386enh] section of the SYSTEM.INI file. This option prevents software from accessing the specified ranges in cases where the software does not recognize that the range is already occupied.

Base I/O Address

Base I/O address (I/O port) is another "address" or path for system- to-peripheral communication. The address table is 00H through FFFFH and should not be confused with UMA mentioned above. A device's I/O port range has a base address (for example, 300 or 320 or 3A0, and so forth). Each port address covers a range (8, 16, or even 32 bytes); therefore, there exists a possibility of overlap and conflict with devices whose starting addresses are close together. Consult the manufacturer's documentation for range size and starting address options. Ensure that each card has exclusive use of the range it occupies.

DMA Channel

Direct memory access (DMA) channel is a means for peripherals to directly access RAM memory without requiring the services of the CPU. Many devices that move large amounts of data (network cards, sound boards, hard disk drive controllers, and so forth) take advantage of this feature to enhance overall system performance and increase data throughput within certain applications.

The DMA channel is designated by a single number (0 through 7). Not all devices use it. Some can use it as an option. Consult your manufacturer's documentation for information on how, when, and why to use this option. Ensure each device has exclusive use of its channel.

NOTE: DMA conflicts are somewhat less likely than UMA and base I/O address conflicts. Fewer devices use it; and, in the case of a sound card for example, the card's DMA channel may be in use only when the application using the card is running.

Managing Complex Systems

To help manage complex systems, make a chart like the one below, which documents the current settings of each device. Keep this chart available and consult and update it when you install more hardware.

   Hardware         IRQ         Base I/O         UMA      DMA
   -----------------------------------------------------------

   Mouse            Com 2 (3)   COM 2(2F8h)      N/A      N/A
   Device 1         5           300              D000     2
   Device 2         x           x                n/a      x
   Device 3         x           x                x        x


KBCategory: kbnetwork kbdisplay kbsound kbfasttip kbsetup kbtshoot
KBSubcategory: wfw wfwg
Additional reference words: tshoot gpf 3.10 3.11


THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLEDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES, EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY.

Last reviewed: February 8, 1995
© 1998 Microsoft Corporation. All rights reserved. Terms of Use.