On Fri, Jan 19, 2018 at 04:56:31PM -0500, Stefan Berger wrote: > On 01/19/2018 01:42 PM, Eduardo Habkost wrote: > > On Fri, Jan 19, 2018 at 12:10:03PM -0500, Stefan Berger wrote: > > > On 01/19/2018 09:11 AM, Marc-André Lureau wrote: > > > > tpm_crb is a device for TPM 2.0 Command Response Buffer (CRB) > > > > Interface as defined in TCG PC Client Platform TPM Profile (PTP) > > > > Specification Family “2.0” Level 00 Revision 01.03 v22. > > > > > > > > The PTP allows device implementation to switch between TIS and CRB > > > > model at run time, but given that CRB is a simpler device to > > > > implement, I chose to implement it as a different device. > > > > > > > > The device doesn't implement other locality than 0 for now (my laptop > > > > TPM doesn't either, so I assume this isn't so bad) > > > > > > > > The command/reply memory region is statically allocated after the CRB > > > > registers address TPM_CRB_ADDR_BASE + sizeof(struct crb_regs) (I > > > > wonder if the BIOS could or should allocate it instead, or what size > > > > to use, again this seems to fit well expectations) > > > I removed this last sentence now. It's at the right location. > > > > > > > The PTP doesn't specify a particular bus to put the device. So I added > > > > it on the system bus directly, so it could hopefully be used easily on > > > > a different platform than x86. Currently, it fails to init on piix, > > > > because error_on_sysbus_device() check. The check may be changed in a > > > > near future, see discussion on the qemu-devel ML. > > > I think this has to be solved. So I remove these last 2 sentences. I'll > > > have > > > to wait until that other patch series from Eduard is merged since it > > > doesn't > > > start yet. > > The series was just merged to master. It's possible to make a > > machine accept the new device using > > machine_class_allow_dynamic_sysbus_dev(), now. > > I saw that. > > > > However, is it really necessary to make it a sysbus device? > > Having bus-less devices was not possible in the past, but it is > > possible today. > > > What I don't like about it is the fact that I would have to use q35 if we > only extend that machine type to allow this sysbus device. What is the > reason that dynamic sysbus devices have to explicitly be allowed? If we > don't need to limit this device to a certain machine type that may be more > user friendly.
Most sysbus devices don't work unless they have additional supporting machine code to wire them at the right addresses. Devices that work without extra machine code (like *-iommu) are the exception. I think the last time I saw an explanation for this was at: https://www.mail-archive.com/qemu-devel@nongnu.org/msg439549.html -- Eduardo
