From: "Michael Chan" <[EMAIL PROTECTED]> Date: Mon, 19 Dec 2005 13:18:47 -0800
> BTW, I have considered always using config. cycles for such registers. > Config. cycles are non-posted and do not require read back. But our hw > engineers told me that it would still be safer to use mem. cycles even > with the risk of an extended posting delay. And recall how we tried to use PCI config space on the read-back after a full chip reset and we still had to end up adding a udelay() there. Even if the udelay() is "long enough" other events (such as a run of "lspci" from userspace) can cause a PCI config space read to occur at just the wrong time. So what we've setup is a precarious situation where arbitrary users can hang devices on your machine if they can elicit a PCI config space access at just the right time, which is trivial with things like lspci. All in all not a good situation. - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
