Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Hot-Insertion of I2C Devices

IP.com Disclosure Number: IPCOM000015956D
Original Publication Date: 2002-May-08
Included in the Prior Art Database: 2003-Jun-21
Document File: 1 page(s) / 34K

Publishing Venue

IBM

Abstract

In today's server systems hot plugging of adapter cards, system components, memory, etc. has become a wide spread function. Often I2C devices reside on these hot-plug or hot-added devices to provide VPD or other system monitoring functions. It is not possible to instruct every I2C agent that may reside on the shared bus to stop "talking" before a hot-plug event occurs. If a hot-plug event were to occur while the I2C bus was not IDLE, the I2C bus could possibly become corrupted. When hot-plugging an I2C device, this invention will ensure that the I2C bus is in an IDLE state before connecting/disconnecting the bus thus preventing I2C bus corruption. The invention consists of a FET switch, circuitry to detect an I2C IDLE condition, and circuitry to communicate with the user/system. For a hot-remove event, the invention first detects that the user/system wants to hot-remove an I2C device. The invention will then monitor the I2C bus for an IDLE state. Once IDLE is detected, the invention will disconnect the I2C device from the bus using a FET switch and then signal to the system/user that the device can be removed. The same basic events occur for a hot-add. The invention hold the FET switch in the off position. The user hot-adds the I2C device and the system supplies power to the device. The invention waits for an IDLE cycle to be detected on the I2C bus and closes the FET switch. Completion of the close can then be communicated to the user/system. 1

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 99% of the total text.

Page 1 of 1

Hot-Insertion of I2C Devices

In today's server systems hot plugging of adapter cards, system components, memory, etc. has become a wide spread function. Often I2C devices reside on these hot-plug or hot-added devices to provide VPD or other system monitoring functions. It is not possible to instruct every I2C agent that may reside on the shared bus to stop "talking" before a hot-plug event occurs. If a hot-plug event were to occur while the I2C bus was not IDLE, the I2C bus could possibly become corrupted. When hot-plugging an I2C device, this invention will ensure that the I2C bus is in an IDLE state before connecting/disconnecting the bus thus preventing I2C bus corruption. The invention consists of a FET switch, circuitry to detect an I2C IDLE condition, and circuitry to communicate with the user/system.

For a hot-remove event, the invention first detects that the user/system wants to hot-remove an I2C device. The invention will then monitor the I2C bus for an IDLE state. Once IDLE is detected, the invention will disconnect the I2C device from the bus using a FET switch and then signal to the system/user that the device can be removed. The same basic events occur for a hot-add. The invention hold the FET switch in the off position. The user hot-adds the I2C device and the system supplies power to the device. The invention waits for an IDLE cycle to be detected on the I2C bus and closes the FET switch. Completion of the close can then be communicated to the use...