User Details
- User Since
- May 10 2014, 2:21 PM (653 w, 4 d)
Thu, Nov 12
FEAT_UAO needs handling on exception entry
Is the FEAT_DIT field missing?
Wed, Nov 11
Mon, Nov 9
Can you rebase past rGfb1fb7df9db5c133e36c09201fbedbddb32e37c5? This should use the new KERNEL_REG macro as it shouldn't be available to userspace.
Fri, Nov 6
Thu, Nov 5
Wed, Nov 4
Mon, Nov 2
Can you include more context. If uploading from the web use git format-patch -U999999, see also https://wiki.freebsd.org/Phabricator
This should be included with the other virtio devices, after virtio_blk
We should remove FDT_BASE, but will need to update the u-boot port to use the correct fdt offset first.
Oct 15 2026
Oct 13 2026
I think we also need to handle IRS_IIDR and IRS_PE_*.
Oct 12 2026
Oct 9 2026
Oct 8 2026
Oct 7 2026
Oct 6 2026
I've created D60407 that should handle both coherent & non-coherent cases
Oct 5 2026
Oct 2 2026
There is a race if two threads are writing at the same time, or if one is reading while another is writing. We could add a flag that gets atomically set when handling the info struct & cleared when done.
I talked with the Arm Linux IOMMU/DMA maintainer about this & we concluded replacing the dsb's with dmb's are correct. I have HW & a local patch that I can test next week.
Oct 1 2026
Sep 29 2026
You can replace the existing dsb calls in with the appropriate dmb from the map == NULL block.
I think the dmb instructions are correct even for the non-coherent case
I think these TLBI_TTL_UNKNOWN values can specify the level, but it would be useful for someone else to check
Sep 25 2026
Do we have somewhere we can document the contract with userspace & when it needs to manage the cache? e.g. removing write & enabling execute on a page will cause the kernel to sync the cache, but having both together means userspace needs to perform a sync operation.
