Peripheral Component Interconnect
In computing, Peripheral Component Interconnect or PCI (in Spanish: Interconexión de Componentes Periféricos), is a standard computer bus to connect peripheral devices directly to the motherboard. These devices can be ICs that fit into it (so-called planar devices in the PCI specification) or expansion cards that fit into connectors. It is common on personal computers, where it has displaced ISA as the bus standard, but it is used on other types of computers as well.
Unlike ISA buses, the PCI bus allows dynamic configuration of a peripheral device. At system boot time, the PCI cards and the BIOS interact and negotiate the resources requested by the PCI card. This allows assignment of the IRQs and port addresses by means of a dynamic process different from the ISA bus, where the IRQs have to be manually configured using external jumpers. The latest revisions of ISA and IBM's MCA bus already incorporated technologies that automated the entire card configuration process, but the PCI bus demonstrated greater efficiency in plug and play technology. Apart from this, the PCI bus provides a detailed description of all the PCI devices connected through the PCI configuration space.
The PCI specification covers the physical size of the bus, electrical characteristics, bus timer, and bus protocols. Copies of the specification are marketed by the “PCI Special Interest Group” (PCI Special Interest Group).
Historical overview
PCI 1.0 was released on June 22, 1992, and was only a component-level specification.
PCI 2.1 was released on June 1, 1995.
PCI was immediately put to use by servers, replacing the MCA and EISA buses as an option to the expansion bus. On PC it was slower to replace VESA Local Bus and did not gain sufficient market penetration until after 1994 with the second generation Pentiums. By 1996 VESA was phased out and companies replaced it in even 80486 computers. Apple adopted PCI for the Power Macintosh (replacing NuBus) in mid-1995 and the Performa (replacing LC PDS) in mid-1996.
New PCI versions added features and performance improvements, including a 66MHz 3.3V standard and a 133MHz standard: called PCI-X. Both PCI-X 1.0b and PCI-X 2.0 are backwards compatible. With the introduction of the serial version of PCI Express in 2004, motherboard manufacturers include fewer and fewer PCI slots in favor of the new standard.
Auto configuration
PCI has two separate 32-bit and 64-bit address spaces, corresponding to the memory and port I/O address of the X86 family of processors. Addressing is assigned by the software. A third address space called “PCI Configuration Space” (PCI Configuration Space), which uses a corrected addressing scheme that allows software to determine the amount of memory and input/output address space needed for each device. Each connected device can request up to six areas of memory space or input/output port spaces through its configuration space register.
In the typical system the firmware (or operating system) queries all PCIs at startup (via PCI configuration space) to find out which devices are present and which resources, telling each device what its location is. accommodation. The PCI configuration space also contains a small amount of information for each device which helps the operating system choose its drivers, or at least have a dialog about system configuration.
Devices may have a ROM containing executable code for x86 or PA-RISC processors, an Open Firmware driver, or an EFI driver. These are normally needed for devices used during system startup, before their drivers are loaded by the operating system.
In addition there are PCI Latency Timers, which are mechanisms for PCI Bus-mastering devices to share the PCI bus more fairly. Where 'fair' in this case means that devices will not use such a large portion of the available PCI bus bandwidth that other devices are not able to do their jobs. This does not apply to PCI-Express.
The way this works is because each PCI device can operate in "bus master" which is required to implement a clock, called a latency clock that limits the time each device can occupy the PCI bus. When the counter reaches 0 the device is requested to leave the bus. If no other device is waiting for ownership of the bus it can simply re-get it and transfer more data.
Hardware specifications
The following specifications represent the version of PCI most commonly used in PCs:
- 33,33 MHz clock with synchronous transfers.
- 32-bit or 64-bit bus width.
- Maximum transfer rate of 133 MB per second on the 32-bit bus (33.33 MHz × 32-bit ÷ 8 bits/byte = 133 MB/s).
- Maximum transfer rate of 266 MB/s on the 64-bit bus.
- 32-bit address space (4 GB).
- 32-bit I/O port space (currently obsolete).
- 256 bytes of configuration space.
- 3.3 V or 5 V depending on the device.
- Reflected-wave switching.
Conventional variants of PCI
- Cardbus is a 32-bit PCMCIA format and 33 MHz PCI.
- Compact PCI uses Eurocard size modules connected to a PCI daughter plate.
- PCI 2.2 works at 66 MHz (requires 3.3 volts on signals), with maximum transfer rate of 503 MiB/s (533MB/s).
- PCI 2.3 allows the use of 3.3 volts and universal signaler, but does not support the 5 volts on the cards.
- PCI 3.0 is the official end standard of the bus, with 5 volt support completely removed.
- PCI-X changes the protocol slightly and increases data transfer to 133 MHz (maximum transfer index of 1014 MiB/s).
- PCI-X 2.0 has a ratio of 266 MHz (maximum transfer index of 2035 MiB/s) and also 533 MHz, expands the configuration space to 4096 bytes, adds a 16-bit bus variant and uses 1.5-volt signals.
- Mini PCI is a new PCI 2.2 format to use it internally on laptops.
- PC104/Plus is an industrial bus that uses PCI signals with different connectors.
- Advanced Telecommunications Computing Architecture (ATCA or AdvancedTCA) is the next generation of buses for the telecommunications industry.
- PXI is the extension of the PCI bus for instrumentation and control.
Card dimensions
Full Size Card
The original “full size” PCI card is about 107 mm (4.2 inches) thick and 312 mm (12.283 inches) long. The height includes the card edge connector. However, most modern PCI cards are half length or smaller (see below) and many personal computers cannot fit a full size card.
Backplate card
In addition to these dimensions the size of the backplate is also standardized. The backplate is the metal piece located on the edge that is used to fix it to the chassis and contains the external connectors. The card can be smaller in size, but the backplate must be full size and properly located. Compared to the previous ISA bus, it is located on the opposite side of the board to avoid errors.
Half-length extension card (de facto standard)
This is actually the practical standard today - most modern PCI cards fit within these dimensions:
- Width: 0.6 inches (15.24 mm)
- Depth: 6.9 inches (175,26 mm)
- Height: 4.2 inches (106,68 mm)
Low Profile (Half Height) Card
The PCI organization has defined a standard for low profile cards that is basically suitable for the following ranges:
- Height: 1.42 inches (36.07 mm) to 2.536 inches (64.41 mm)
- Depth: 4,721 inches (119,91 mm) to 6.6 inches (167,64 mm)
The shelf is also lowered in height to a standard 3.118 inches (79.2 mm). The smallest shelf does not fit a standard personal computer. Many manufacturers get around this by supplying both types of shelf (the shelves are typically bolted to the card so changing them is not difficult).
These cards may be known by other names such as thin.
Mini PCI
Mini PCI was added to version 2.2 PCI for use in laptops and uses a 32-bit, 33 MHz bus with powered connections (3.3V only) and bus mastering support. > and DMA. The standard size for Mini PCI cards is approximately 1/4 of their full-size counterparts. Since there is no external access to the card in the same way that there is for desktop PCI cards, there are limitations to the features that p PCI have been developed for, such as Wi-Fi, Fast Ethernet, Bluetooth, modems (often Winmodems), sound cards, cryptographic accelerators, SCSI controllers, IDE/ATA, SATA, combination cards. Regular PCI cards can be used with Mini PCI-equipped hardware and vice versa, using Mini-PCI to PCI and PCI-to-Mini PCI converters. Mini PCI has been superseded by PCI Express Mini Card.
Mini PCI Card Technical Details
Mini PCI cards have a maximum consumption of 2W, which also limits the functionality that can be implemented in this form factor. They require that they also support the PCI CLKRUN# signal, used to start and stop the PCI clock for power control purposes.
There are three card form factors: Type I, Type II, and Type III. The card connector used for each type includes: Type I and II use a 100-pin placement connector, while Type III uses a 124-pin edge connector, e.g. eg the connector for Type I and II differs by this from Type III, where the connector is on the edge of a card, as with an SO-DIMM. The additional 24 pins provide the required supplementary signals to the rear I/O path of the connector system (audio, AC link, LAN, phone line interface). Type II cards have RJ11 and RJ45 connectors mounted. These cards should be located at the edge of the computer or docking station so that the RJ11 and RJ45 ports can be mounted for external access.
Other physical variations
Typical consumer systems specific "N x PCI slots " without specifying the actual dimensions of the available space. In some small form factor systems, this is still not enough for "half-body" enter that slot. Despite this limitation, these systems are still useful because many modern PCI cards are much smaller than half-body cards.
Bank card
Typical PCI cards have one or two key notches, depending on their voltage rating. Cards that require 3.3 volts have a 56.21mm notch on the front of the card (where the external connectors are), while those that require 5 volts have a 104.47mm notch on the front of the card. The calls " Universal cards " they have both key notches and can accept both types of signals.
Contenido relacionado
Charles H. Moore
Convertiplane
Mail server