<strong>PowerPC</strong> <strong>740</strong> <strong>and</strong> <strong>PowerPC</strong> <strong>750</strong> <strong>Microprocessor</strong>CMOS 0.20 µm Copper Technology, PID-8p, PPC<strong>740</strong>L <strong>and</strong> PPC<strong>750</strong>L, dd3.2OverviewThe <strong>750</strong> is targeted for high performance, low power systems <strong>and</strong> supports the following power managementfeatures: doze, nap, sleep, <strong>and</strong> dynamic power management. The <strong>750</strong> consists of a processor core <strong>and</strong> aninternal L2 Tag combined with a dedicated L2 cache interface <strong>and</strong> a 60x bus. The L2 cache is not availablewith the <strong>740</strong>.Figure 1 shows a block diagram of the <strong>750</strong>.Figure 1. <strong>750</strong> Block DiagramCompletionInstruction FetchBranch UnitControl Unit32K ICacheSystemUnitDispatchBHT /BTICFXU1FXU2GPRsRenameBuffersLSUFPRsRenameBuffersFPU32K DCacheL2 TagsL2 Cache60XBIUBIU9/6/2002 Version 2.0 Page 6
<strong>PowerPC</strong> <strong>740</strong> <strong>and</strong> <strong>PowerPC</strong> <strong>750</strong> <strong>Microprocessor</strong>CMOS 0.20 µm Copper Technology, PID-8p, PPC<strong>740</strong>L <strong>and</strong> PPC<strong>750</strong>L, dd3.2FeaturesThis section summarizes features of the implementation of the <strong>PowerPC</strong> <strong>750</strong> architecture. For details, seethe <strong>PowerPC</strong> <strong>740</strong> <strong>and</strong> <strong>PowerPC</strong><strong>750</strong> User’s Manual.• Branch processing unit- Four instructions fetched per clock.- One branch processed per cycle (plus resolving 2 speculations).- Up to 1 speculative stream in execution, 1 additional speculative stream in fetch.- 512-entry branch history table (BHT) for dynamic prediction.- 64-entry, 4-way set associative branch target instruction cache (BTIC) for eliminating branch delayslots.• Dispatch unit- Full hardware detection of dependencies (resolved in the execution units).- Dispatch two instructions to six independent units (system, branch, load/store, fixed-point unit 1,fixed-point unit 2, or floating-point).- Serialization control (predispatch, postdispatch, execution, serialization).• Decode- Register file access.- Forwarding control.- Partial instruction decode.• Load/store unit- One cycle load or store cache access (byte, half word, word, double word).- Effective address generation.- Hits under misses (one outst<strong>and</strong>ing miss).- Single-cycle misaligned access within double word boundary.- Alignment, zero padding, sign extend for integer register file.- Floating-point internal format conversion (alignment, normalization).- Sequencing for load/store multiples <strong>and</strong> string operations.- Store gathering.- Cache <strong>and</strong> TLB instructions.- Big <strong>and</strong> little-endian byte addressing supported.- Misaligned little-endian support in hardware.• Fixed-point units- Fixed-point unit 1 (FXU1); multiply, divide, shift, rotate, arithmetic, logical.- Fixed-point unit 2 (FXU2); shift, rotate, arithmetic, logical.- Single-cycle arithmetic, shift, rotate, logical.- Multiply <strong>and</strong> divide support (multi-cycle).- Early out multiply.• Floating-point unit- Support for IEEE-754 st<strong>and</strong>ard single- <strong>and</strong> double-precision floating-point arithmetic.Page 7 Version 2.0 9/6/2002