Page MenuHomeFreeBSD
Feed Advanced Search

Jul 24 2015

wma_semihalf.com added inline comments to D3031: PCIe support for ThunderX.
Jul 24 2015, 12:31 PM
wma_semihalf.com updated the diff for D3031: PCIe support for ThunderX.

Reduced the diff to internal Root Complex only.

Jul 24 2015, 12:29 PM
wma_semihalf.com retitled D3184: THUNDERX ERRATA: Apply erratum for mrs ICC_IAR1_EL1 from to THUNDERX ERRATA: Apply erratum for mrs ICC_IAR1_EL1 speculative execution.
Jul 24 2015, 12:05 PM

Jul 22 2015

wma_semihalf.com updated the diff for D3131: Add quirk for Thunder ITS GITS_BASER.

Modified revision to sync with the Linux version of this quirk.

Jul 22 2015, 6:26 AM

Jul 21 2015

wma_semihalf.com added inline comments to D3118: Add support for vendor specific function for PCI devid acquisition.
Jul 21 2015, 2:48 PM
wma_semihalf.com accepted D3121: Don't allow malloc() to wait for resource while holding a lock in ITS.

Just tested on two ThunderX revisions (CRB and EVB) and works great. Please commit.

Jul 21 2015, 9:54 AM
wma_semihalf.com updated the diff for D3009: Add MSI-x support to AHCI driver.

Forgot to include new BAR allocation, fixing it now. It should work even if all BARs (mem+msix+pba) are the same.

Jul 21 2015, 6:35 AM
wma_semihalf.com updated the diff for D3009: Add MSI-x support to AHCI driver.
Jul 21 2015, 6:11 AM

Jul 20 2015

wma_semihalf.com retitled D3134: Implement get_cyclecount on ARM64 from to Implement get_cyclecount on ARM64.
Jul 20 2015, 7:21 AM
wma_semihalf.com retitled D3132: Improve generic_timer startup on secondary cores from to Improve generic_timer startup on secondary cores.
Jul 20 2015, 6:43 AM
wma_semihalf.com retitled D3131: Add quirk for Thunder ITS GITS_BASER from to Add quirk for Thunder ITS GITS_BASER.
Jul 20 2015, 5:12 AM

Jul 17 2015

wma_semihalf.com added inline comments to D3117: Improve ARM64 CPU_MATCH.
Jul 17 2015, 6:20 PM
wma_semihalf.com abandoned D3008: Add ThunderX as FreeBSD target.
Jul 17 2015, 2:11 PM
wma_semihalf.com added reviewers for D3117: Improve ARM64 CPU_MATCH: andrew, emaste, zbb.
Jul 17 2015, 1:02 PM
wma_semihalf.com retitled D3118: Add support for vendor specific function for PCI devid acquisition from to Add support for external function for PCI devid acquisition.
Jul 17 2015, 1:01 PM
wma_semihalf.com retitled D3117: Improve ARM64 CPU_MATCH from to Improve ARM64 CPU_MATCH.
Jul 17 2015, 12:55 PM
wma_semihalf.com retitled D3113: Increase DMAP size on ARM64 from to Increase DMAP size on ARM64.
Jul 17 2015, 6:33 AM

Jul 16 2015

wma_semihalf.com updated the diff for D3093: ARM64 TCR register update.
Jul 16 2015, 8:32 AM
wma_semihalf.com updated the diff for D3107: Fix ARM64 stack calculation.
Jul 16 2015, 8:27 AM
wma_semihalf.com retitled D3107: Fix ARM64 stack calculation from to Fix ARM64 stack calculation.
Jul 16 2015, 8:23 AM
wma_semihalf.com updated the diff for D3093: ARM64 TCR register update.
Jul 16 2015, 8:22 AM
wma_semihalf.com updated the diff for D3094: Fix KSTACK_PAGES issue.
Jul 16 2015, 8:07 AM
wma_semihalf.com retitled D3106: Fix possible coherency issues between PEs related to I-cache from to Fix possible coherency issues between PEs related to I-cache.
Jul 16 2015, 5:09 AM

Jul 15 2015

wma_semihalf.com added a comment to D3094: Fix KSTACK_PAGES issue.

Andrew, I would love to change it in other places as well, but the interaction with so many maintainers will slow down this change and I don't have any mips/powerpc hardware to run tests on.
I have a proposal. Today I will prepare another patch with separate review removing KSTACK_PAGES from param.h in other archs and ask on the fbsd list for some help with testing on more exotic hardware, This allow submitting this to arm+arm64 while waiting for approvals from mips/sparc/x86/etc Will that work for you?

Jul 15 2015, 11:41 AM
wma_semihalf.com abandoned D3034: Add BUS_UNMASK_INTR method.

Thanks! That was really helpful. However I'm still confused about the system I have. I didn't powered up the full configuration yet, but I'm afraid it has two root pics then (one per each 48-cpu partition). Nevertheless, I'm abandoning this review for a while and stick to hacking generic_timer. When I test it on actual hw I will get back to this patch. However I'm still thinking the generic mask/unmask pair might be a useful feature.

Jul 15 2015, 11:27 AM
wma_semihalf.com added inline comments to D3093: ARM64 TCR register update.
Jul 15 2015, 10:47 AM
wma_semihalf.com added inline comments to D3094: Fix KSTACK_PAGES issue.
Jul 15 2015, 9:26 AM
wma_semihalf.com updated the diff for D3093: ARM64 TCR register update.

Agreed.

Jul 15 2015, 9:15 AM
wma_semihalf.com retitled D3095: Add identify_cpu to ARMv8 init_secondary routine from to Add identify_cpu to ARMv8 init_secondary routine.
Jul 15 2015, 7:48 AM
wma_semihalf.com updated D3094: Fix KSTACK_PAGES issue.
Jul 15 2015, 7:36 AM
wma_semihalf.com retitled D3094: Fix KSTACK_PAGES issue from to Fix ARMv8 KSTACK_PAGES issue.
Jul 15 2015, 7:36 AM
wma_semihalf.com added a comment to D3009: Add MSI-x support to AHCI driver.

Please give any feedback. We'd like to submit it next week if no other objections are reported.

Jul 15 2015, 6:55 AM
wma_semihalf.com updated subscribers of D3009: Add MSI-x support to AHCI driver.
Jul 15 2015, 6:47 AM
wma_semihalf.com retitled D3093: ARM64 TCR register update from to ARMv8 locore.S cleanup and TCR register update.
Jul 15 2015, 6:45 AM
wma_semihalf.com added a comment to D3034: Add BUS_UNMASK_INTR method.

I guess treating PPIs are separate IRQs number will cause a huge mess. I'm starting to port fbsd on 96 core armv8 platform and the idea of 1536 vectors wasted for PPIs is outrageous.

Jul 15 2015, 5:26 AM

Jul 14 2015

wma_semihalf.com updated the diff for D3034: Add BUS_UNMASK_INTR method.

Moved arm_unmask_irq here.

Jul 14 2015, 9:54 AM

Jul 13 2015

wma_semihalf.com retitled D3066: Fix armv8 secondary PIC initialization order from to Fix armv8 secondary PIC initialization order.
Jul 13 2015, 10:03 AM

Jul 12 2015

wma_semihalf.com added a comment to D3009: Add MSI-x support to AHCI driver.

Any comments?

Jul 12 2015, 4:13 PM
wma_semihalf.com added a comment to D3028: Implement stubs for ACPI PCI routines.

Any other issues except ARM64TOFO? If not, please accept this revision, I'd like to submit it soon.

Jul 12 2015, 4:13 PM
wma_semihalf.com added a comment to D3034: Add BUS_UNMASK_INTR method.

Any comments?

Jul 12 2015, 4:11 PM
wma_semihalf.com added inline comments to D3029: Fix intr_machdep.c for armv8.
Jul 12 2015, 4:11 PM

Jul 9 2015

wma_semihalf.com retitled D3034: Add BUS_UNMASK_INTR method from to Add BUS_UNMASK_INTR method.
Jul 9 2015, 11:21 AM
wma_semihalf.com updated the diff for D3028: Implement stubs for ACPI PCI routines.
Jul 9 2015, 10:34 AM
wma_semihalf.com retitled D3031: PCIe support for ThunderX from to PCIe support for ThunderX.
Jul 9 2015, 7:59 AM
wma_semihalf.com retitled D3030: Rework armv8 CPU identification from to Rework armv8 CPU identification.
Jul 9 2015, 7:11 AM
wma_semihalf.com retitled D3029: Fix intr_machdep.c for armv8 from to Fix intr_machdep.c for armv8.
Jul 9 2015, 5:33 AM
wma_semihalf.com retitled D3028: Implement stubs for ACPI PCI routines from to Implement stubs for ACPI PCI routines.
Jul 9 2015, 5:15 AM

Jul 8 2015

wma_semihalf.com added a reviewer for D3009: Add MSI-x support to AHCI driver: jhb.
Jul 8 2015, 6:15 PM
wma_semihalf.com accepted D3024: Add support for SMP on arm64..

I did notice it's cleaned up (and it's got the annoying DO_AST printf changed to ktr). However, the places where I expect merge conflicts are the same. We had to change stack calculation, sequence in init_secondary, fix the KSTACK_PAGES variable (it does not correspond to the one set in kernconf here)... these are the one from the top of my head.
I'm good with this patch, please commit.

Jul 8 2015, 5:42 PM
wma_semihalf.com added a comment to D3024: Add support for SMP on arm64..

That would be great if you could commit this patch "as is". We have few fixes to mp_machdep/locore which are currently rebased against your version. Avoiding merges of post-review fixes will let us save a lot of time...
I will send all our patches once it's pushed. Thanks in advance.

Jul 8 2015, 5:32 PM
wma_semihalf.com added inline comments to D3008: Add ThunderX as FreeBSD target.
Jul 8 2015, 4:06 PM
wma_semihalf.com added a comment to D3008: Add ThunderX as FreeBSD target.

How does MIPS have PCI bus then? Or should that first line be 'option acpi'?

Jul 8 2015, 4:03 PM
wma_semihalf.com added a comment to D3008: Add ThunderX as FreeBSD target.
In D3008#59529, @andrew wrote:

You haven't listed any problems that can't be fixed in GENERIC. It would help if you could tell me the specific problems you are trying to solve and why you couldn't use, for example, quirks to solve them.

Jul 8 2015, 2:44 PM
wma_semihalf.com updated the diff for D3009: Add MSI-x support to AHCI driver.

Added MSI-x table and PBA BAR number resolving in generic way, as per PCIe specification.

Jul 8 2015, 12:44 PM
wma_semihalf.com added a comment to D3008: Add ThunderX as FreeBSD target.

We will not be able to make it generic until the issue with PCI compilation is resolved. Maybe I didn't specified it clearly enough, but in the present FreeBSD state we need to have a separate config files. If you think we can live with fbsd without any IO interfaces working on the real hardware, then we can definitely stick to GENERIC. Keep in mind, that configuring pci in GENERIC does not cope well with an acpi option and makes the compilation fail...
I'd really like to have this change pushed as an enabler for all further ThunderX support. It is only a temporary and will be removed as soon as is no longer necessary.

Jul 8 2015, 11:30 AM
wma_semihalf.com updated the diff for D3012: Add memory barrier to bus_dmamap_sync.

Made it MD.
Regarding mb(), I was expecting a discussion rather to use stronger one, not finer-grained. Here we need to guarantee that mb is valid for every domain (inner+outter shareable) , thus use a full-system one. I was wondering, however, if the dsb() is not more appropriate. Mostly because of scenario what if we're syncing a buffer containing executable code which we jumps to. In that case mb might not be sufficient. I did not find any such place in the kernel, so decided to use less impacting mb instead.

Jul 8 2015, 6:29 AM
wma_semihalf.com updated the diff for D3008: Add ThunderX as FreeBSD target.

Moved some options which will be necessary in the future from THUNDER-88XX to GENERIC. Cleaned up Thunder kenrconf and left only absolutely necessary options.

Jul 8 2015, 5:52 AM

Jul 7 2015

wma_semihalf.com added inline comments to D3008: Add ThunderX as FreeBSD target.
Jul 7 2015, 5:26 PM
wma_semihalf.com added a comment to D3012: Add memory barrier to bus_dmamap_sync.

Sure. However, I'd like to use a negative logic here. I mean, to use mb as a default and disable it for a known list of architectures (afaik only Intel/amd64 don't need a barrier here).

Jul 7 2015, 5:12 PM
wma_semihalf.com added inline comments to D3009: Add MSI-x support to AHCI driver.
Jul 7 2015, 5:00 PM
wma_semihalf.com added inline comments to D3008: Add ThunderX as FreeBSD target.
Jul 7 2015, 4:19 PM
wma_semihalf.com updated the diff for D3009: Add MSI-x support to AHCI driver.

Yes, you're absolutely right. Fixed the logic to avoid allocation for both msix and msi.

Jul 7 2015, 12:46 PM
wma_semihalf.com retitled D3012: Add memory barrier to bus_dmamap_sync from to Add memory barrier to bus_dmamap_sync.
Jul 7 2015, 10:54 AM
wma_semihalf.com added a comment to D3009: Add MSI-x support to AHCI driver.

We are trying to upstream support for ThunderX armv8 system. The producer decided to support only MSI-x for all internal interfaces, including AHCI.

Jul 7 2015, 9:40 AM
wma_semihalf.com retitled D3009: Add MSI-x support to AHCI driver from to Add MSI-x support to AHCI driver.
Jul 7 2015, 7:01 AM
wma_semihalf.com retitled D3008: Add ThunderX as FreeBSD target from to Add ThunderX as FreeBSD target.
Jul 7 2015, 6:10 AM