Intel® 945G/945GZ/945GC/ 945P/945PL Express Chipset Family ...
Intel® 945G/945GZ/945GC/ 945P/945PL Express Chipset Family ... Intel® 945G/945GZ/945GC/ 945P/945PL Express Chipset Family ...
System Address Map 222 Intel ® 82945G/82945GZ/82945GC GMCH and 82945P/82945PL MCH Datasheet
10 Functional Description This chapter describes the (G)MCH interfaces and major functional units. 10.1 Host Interface Functional Description The (G)MCH supports the Pentium ® 4 processor subset of the Enhanced Mode Scaleable Bus. The cache line size is 64 bytes. Source synchronous transfer is used for the address and data signals. The address signals are double pumped and a new address can be generated every other bus clock. At 200/267 MHz bus clock, the address signals run at 400/533MT/s. The data is quad pumped and an entire 64 B cache line can be transferred in two bus clocks. At 200/267 MHz bus clock, the data signals run at 800/1066MT/s for a maximum bandwidth of 10.7 GB/s. 10.1.1 FSB IOQ Depth The Scalable bus supports up to 12 simultaneous outstanding transactions. 10.1.2 FSB OOQ Depth The (G)MCH supports only one outstanding deferred transaction on the FSB. 10.1.3 FSB GTL+ Termination The (G)MCH integrates GTL+ termination resistors on die. Also, approximately 2.8 pf (fast) – 3.3 pf (slow) per pad of on die capacitance will be implemented to provide better FSB electrical performance. 10.1.4 FSB Dynamic Bus Inversion The (G)MCH supports Dynamic Bus Inversion (DBI) when driving and when receiving data from the processor. DBI limits the number of data signals that are driven to a low voltage on each quad pumped data phase. This decreases the worst-case power consumption of the (G)MCH. HDINV[3:0]# indicate if the corresponding 16 bits of data are inverted on the bus for each quad pumped data phase: HDINV[3:0]# Data Bits HDINV0# HD15:0# HDINV1# HD31:16# HDINV2# HD47:32# HDINV3# HD63:48# Intel ® 82945G/82945G/82945GC GMCH and 82945P/82945PL MCH Datasheet 223
- Page 171 and 172: 6.1.15 DMIUEMSK—DMI Uncorrectable
- Page 173 and 174: Integrated Graphics Device (D2:F0)
- Page 175 and 176: Integrated Graphics Device (D2:F0)
- Page 177 and 178: 7.1.4 PCISTS2—PCI Status (D2:F0)
- Page 179 and 180: 7.1.7 CLS—Cache Line Size (D2:F0)
- Page 181 and 182: 7.1.11 IOBAR—I/O Base Address (D2
- Page 183 and 184: Integrated Graphics Device (D2:F0)
- Page 185 and 186: 7.1.20 MINGNT—Minimum Grant (D2:F
- Page 187 and 188: Integrated Graphics Device (D2:F0)
- Page 189 and 190: Integrated Graphics Device (D2:F0)
- Page 191 and 192: Integrated Graphics Device (D2:F1)
- Page 193 and 194: 8.1.3 PCICMD2—PCI Command (D2:F1)
- Page 195 and 196: 8.1.5 RID2—Revision Identificatio
- Page 197 and 198: Integrated Graphics Device (D2:F1)
- Page 199 and 200: Integrated Graphics Device (D2:F1)
- Page 201 and 202: Integrated Graphics Device (D2:F1)
- Page 203 and 204: 8.2 Device 2 - PCI I/O Registers In
- Page 205 and 206: 9 System Address Map System Address
- Page 207 and 208: Figure 9-1. System Address Ranges 4
- Page 209 and 210: Compatible SMRAM Address Range (A_0
- Page 211 and 212: 9.2 Main Memory Address Range (1 MB
- Page 213 and 214: 9.3 PCI Memory Address Range (TOLUD
- Page 215 and 216: 9.3.6 PCI Express* Graphics Attach
- Page 217 and 218: 9.4.1 SMM Space Definition System A
- Page 219 and 220: 9.4.5 SMM Space Decode and Transact
- Page 221: 9.4.12 Legacy VGA and I/O Range Dec
- Page 225 and 226: Interleaved Mode Functional Descrip
- Page 227 and 228: 667 MHz (PC 5300) (82945G/82945GC/8
- Page 229 and 230: 10.2.2.2 System Memory Supported Co
- Page 231 and 232: Functional Description Table 10-5.
- Page 233 and 234: 10.3 PCI Express* (Intel ® 82945G/
- Page 235 and 236: Figure 10-2. SDVO Conceptual Block
- Page 237 and 238: Figure 10-3. Concurrent SDVO / PCI
- Page 239 and 240: 10.5 Integrated Graphics Device (In
- Page 241 and 242: 10.5.3 4X Faster Setup Engine Funct
- Page 243 and 244: 10.5.4 Texture Engine Functional De
- Page 245 and 246: 10.5.4.8 Pixel Shader Functional De
- Page 247 and 248: Functional Description to determine
- Page 249 and 250: 10.5.6 2D Engine Functional Descrip
- Page 251 and 252: 10.5.8.1 Cursor Plane Functional De
- Page 253 and 254: Functional Description device is in
- Page 255 and 256: 10.6.2 Digital Display Interface Fu
- Page 257 and 258: 10.6.2.1.5 Control Bus Functional D
- Page 259 and 260: Figure 10-7 illustrates the various
- Page 261 and 262: 11 Electrical Characteristics Elect
- Page 263 and 264: 11.1.1 Power Characteristics Table
- Page 265 and 266: Signal Group Electrical Characteris
- Page 267 and 268: 11.3 DC Characteristics Table 11-5.
- Page 269 and 270: Symbol Signal Group Electrical Char
- Page 271 and 272: Ballout and Package Information 12
10 Functional Description<br />
This chapter describes the (G)MCH interfaces and major functional units.<br />
10.1 Host Interface<br />
Functional Description<br />
The (G)MCH supports the Pentium ® 4 processor subset of the Enhanced Mode Scaleable Bus.<br />
The cache line size is 64 bytes. Source synchronous transfer is used for the address and data<br />
signals. The address signals are double pumped and a new address can be generated every other<br />
bus clock. At 200/267 MHz bus clock, the address signals run at 400/533MT/s. The data is quad<br />
pumped and an entire 64 B cache line can be transferred in two bus clocks. At 200/267 MHz bus<br />
clock, the data signals run at 800/1066MT/s for a maximum bandwidth of 10.7 GB/s.<br />
10.1.1 FSB IOQ Depth<br />
The Scalable bus supports up to 12 simultaneous outstanding transactions.<br />
10.1.2 FSB OOQ Depth<br />
The (G)MCH supports only one outstanding deferred transaction on the FSB.<br />
10.1.3 FSB GTL+ Termination<br />
The (G)MCH integrates GTL+ termination resistors on die. Also, approximately 2.8 pf (fast) –<br />
3.3 pf (slow) per pad of on die capacitance will be implemented to provide better FSB electrical<br />
performance.<br />
10.1.4 FSB Dynamic Bus Inversion<br />
The (G)MCH supports Dynamic Bus Inversion (DBI) when driving and when receiving data from<br />
the processor. DBI limits the number of data signals that are driven to a low voltage on each quad<br />
pumped data phase. This decreases the worst-case power consumption of the (G)MCH.<br />
HDINV[3:0]# indicate if the corresponding 16 bits of data are inverted on the bus for each quad<br />
pumped data phase:<br />
HDINV[3:0]# Data Bits<br />
HDINV0# HD15:0#<br />
HDINV1# HD31:16#<br />
HDINV2# HD47:32#<br />
HDINV3# HD63:48#<br />
Intel ® 82<strong>945G</strong>/82<strong>945G</strong>/82<strong>945G</strong>C GMCH and 82<strong>945P</strong>/82<strong>945P</strong>L MCH Datasheet 223