Hi All, Recently came across an issue which is decribed below. Setup: Port Multiplier with 2 HDD's connected to it; Port multiplier is connected to a SATA port of the Host Controller. Test: HDD1 is mounted and using dd command a file of size 512MB is being written on it. Command: time sh -c "dd if=/dev/zero of=/mnt/hd/ddfile bs=1k count=512000 && sync" While this command is running, if the second HDD on PMP is plugged out, the R/W speed of the first HDD gets affected. Please see the results of experiments on 2.6.31 kernel. Plug Out real 0m34.433s user 0m0.932s sys 0m21.352s Plug in real 0m23.726s user 0m0.908s sys 0m21.388s Normal real 0m23.329s user 0m1.132s sys 0m20.968s The results are similiar with multiple Port Multipliers (Si3726, Si4726 and EXAR PMP) and multiple Host controllers (Si3132 and sata_fsl). As the behaviour with multiple Host controllers and Port Multipliers is the same so this seems to be an issue with libata stack. Is there already a patch for this issue? Any pointers or help to fix this will be greatly appreciated. Best Regards, Anju Applications Engineer Freescale Semiconductor -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
