Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
richlowe created a comment on a pull request on richlowe/illumos-gate
This is one of those things where it seems like the long-term fix is probably the separate uts platform for sbsa v. not-sbsa, rather than conditions at runtime, I think.
r1mikey created a comment on a pull request on richlowe/illumos-gate
It might be better to keep the PROM in `unix` and do this all at compile-time. Let's hold off on this one for now.
richlowe pushed 1 commit to arm64-gate richlowe/illumos-gate
- use typed `ddi_prop_lookup...` in cases that might affect aarch64 (fix i386 build) 2829be5
richlowe closed a pull request on richlowe/illumos-gate
use typed DDI interfaces where it might affect AArch64
on fdt-based aarch64, "PROM" properties ultimately come from the FDT. This data is defined (like ieee1275) to be big-endian 32bit integer cells, bytes, or strings. Since we're using the CPU lit...
richlowe pushed 2 commits to arm64-gate richlowe/illumos-gate
richlowe pushed 2 commits to arm64-gate richlowe/illumos-gate
r1mikey opened a pull request on richlowe/illumos-gate
Make PROM usable on both FDT and ACPI systems
Most of the PROM, other than node and prop functions, is the same as what is found on i86pc, and therefore the same between FDT and ACPI systems. Make the PROM itself usable on both, supporting run...
richlowe created a review comment on a pull request on richlowe/illumos-gate
My fear with this (and I also suspect I've been inconsistent, but I tried not to be), is that if someone had foo=0 in a config file to explicitly disable something, that's still boolean true, but f...
richlowe created a review comment on a pull request on richlowe/illumos-gate
It's not, I fixed this but I had to force-push because I'd rebased since I put the review up. Sorry.
richlowe created a comment on a pull request on richlowe/illumos-gate
> This cleans up the current logic nicely, but the current logic doesn't really implement the devicetree spec. > > Do we want to land this as a cleanup and improve the implementation later? W...
richlowe pushed 49 commits to arm64/typed-ddi richlowe/illumos-gate
- 16800 bhyve upstream sync 2024 September Reviewed by: Marco van Wieringen <[email protected]> Revie... 5c4a5fe
- 16806 Some services unexpectedly start in /root Reviewed by: Marco van Wieringen <[email protected]... 82ae164
- 16828 Support AMD DFv4+ UMC decoding Reviewed by: Rich Lowe <[email protected]> Reviewed by: C Fraire <cfraire@me... 05ce395
- Merge branch 'master' of https://github.com/illumos/illumos-gate into arm64-gate # Conflicts: # usr/src/cmd/bhyve/Ma... d5977b3
- 16839 busra comments could expand upon why x86 IRQ 2 is special Reviewed by: Joshua M. Clulow <[email protected]> Revie... 09ded43
- 16841 in.ndpd sometimes loses interface id on restart Reviewed by: Andy Fiddaman <[email protected]> Approved by: ... ba74e13
- 16846 ast: update for gcc 14 Reviewed by: Andy Fiddaman <[email protected]> Reviewed by: Bill Sommerfeld <sommerfe... 1ff6fc1
- 16838 libefi: efi_auto_sense should use sector size Reviewed by: Jason King <[email protected]> Appr... 598dbe5
- 16833 smbsrv: array subscript 12 is above array bounds Reviewed by: Andy Fiddaman <[email protected]> Reviewed by:... 35ae87c
- 16844 pci_prd: array subscript 0 is outside array bounds Reviewed by: Andy Fiddaman <[email protected]> Approved b... e5ab5ea
- 16845 loader: array subscript [0, 536870911] is outside array bounds Reviewed by: Andy Fiddaman <[email protected]... c2eb74c
- 16850 genunix: array subscript [0, 9] is outside array bounds of 'hrtime_t[10]' Reviewed by: Marcel Telka <marcel@tel... 47129be
- 16852 drop gcc 11 from Makefile.master Reviewed by: Bill Sommerfeld <[email protected]> Approved by: Dan McDonal... d378658
- 16550 vmm: writing 1 byte into a region of size 0 Reviewed by: Bill Sommerfeld <[email protected]> Reviewed by: ... 0e34f8d
- 16853 cpuid_pass_basic filling in wrong cpi_extd entry for CPUID_LEAF_EXT_21 Reviewed by: Dan McDonald <[email protected]... a8cd71d
- 16849 const arrays in vmm_gpt.c trigger warnings with newer GCC Reviewed by: Toomas Soome <[email protected]> Reviewed by... a21e12b
- Merge branch 'master' of https://github.com/illumos/illumos-gate into arm64-gate 82efa37
- 15496 dr: array subscript 3 is above array bounds Reviewed by: Bill Sommerfeld <[email protected]> Reviewed by: ... cbb4f59
- 16867 gcc 14 needs to use -gstrict-dwarf afterall Reviewed by: Bill Sommerfeld <[email protected]> Reviewed by: ... 5da8a10
- 16585 ping ignores ipv6 scope id in target address Reviewed by: Andy Fiddaman <[email protected]> Approved by: Dan... 384e273
- and 29 more ...
richlowe pushed 1 commit to arm64-gate richlowe/illumos-gate
- Use loaders to abstract the arch_timer frequency We have a very early FDT dependency in unix for the arch_timer freq... 85eec09
richlowe closed a pull request on richlowe/illumos-gate
Use loaders to abstract the arch_timer frequency
We have a very early FDT dependency in unix for the arch_timer frequency. This dependency exists to deal with the case where the firmware has not programmed the `CNTFRQ_EL0` register, requiring a f...
richlowe pushed 5 commits to arm64-gate richlowe/illumos-gate
- 16873 tools: add lex to build tools Reviewed by: Bill Sommerfeld <[email protected]> Approved by: Dan McDonald <... 2e07277
- 16738 viona should allow configurable packet padding 16801 viona should handle ipv6eh for csum offload Reviewed by: D... ce489a5
- 16884 viona TSO should better handle csum offloads Reviewed by: Dan McDonald <[email protected]> Reviewed by: Toomas Soom... 2317519
- 16892 Update ast feature files Reviewed by: Dan Cross <[email protected]> Reviewed by: Toomas Soome <tsoome@me.... 616c769
- Merge branch 'master' of https://github.com/illumos/illumos-gate into arm64-gate # Conflicts: # exception_lists/pack... b7c1a26
r1mikey created a review comment on a pull request on richlowe/illumos-gate
I'm not sure this is quite right. The spec states that `#interrupt-cells` is used at the root of an interrupt domain, which in turn is described as: _An interrupt domain is the context in which an...
r1mikey created a review comment on a pull request on richlowe/illumos-gate
The logic here seems at odds with `devicetree-specification-v0.4.pdf`. The spec states: _Because the hierarchy of the nodes in the interrupt tree might not match the devicetree, the interrupt-pa...
r1mikey created a review on a pull request on richlowe/illumos-gate
There are quite a few places where it looks like integer properties are being interpreted as bool (i.e. a simple presence check). We probably don't want to change anything structurally or semantica...
r1mikey created a review on a pull request on richlowe/illumos-gate
There are quite a few places where it looks like integer properties are being interpreted as bool (i.e. a simple presence check). We probably don't want to change anything structurally or semantica...
r1mikey created a comment on a pull request on richlowe/illumos-gate
Yep, this is used very early, well before cbe hooks up the timer properly.
richlowe created a review on a pull request on richlowe/illumos-gate
This looks good. I'm assuming this is all way too early to be asking the actual devinfo device tree these questions in arch_timer?
richlowe closed a pull request on richlowe/illumos-gate
Add a virtio nexus for virtio-mmio support
Introduce a virtio nexus driver for use with the virtio MMIO transport. In devicetree systems we are presented with a number of nodes with compatible set to `virtio,mmio` - we have no idea what ...
richlowe created a comment on a pull request on richlowe/illumos-gate
I merged this, after having to fix a bug in arm64-gate to still be able to be build disks (it turns out the mkisofs we're using seems broken a bit). I've been getting the interrupt-controller st...
richlowe pushed 1 commit to arm64-gate richlowe/illumos-gate
- Add a virtio nexus for virtio-mmio support 0b9bcc0
r1mikey opened a pull request on richlowe/illumos-gate
Use loaders to abstract the arch_timer frequency
We have a very early FDT dependency in unix for the arch_timer frequency. This dependency exists to deal with the case where the firmware has not programmed the `CNTFRQ_EL0` register, requiring a f...