Page MenuHomeFreeBSD

git_bdragon.rtk0.net (Brandon Bergren)
Animal

Projects

User does not belong to any projects.

User Details

User Since
Jun 6 2018, 11:31 PM (27 w, 4 d)

Recent Activity

Yesterday

git_bdragon.rtk0.net added a comment to D18565: Fix warning in PowerPC64 ELFv2 build of rtld-elf..

On second thought, it looks like there might be cases where we should actually be checking the relocation. I'll fix this properly with the ifunc implementation I'll be submitting this week.

Sun, Dec 16, 11:27 PM · PowerPC
git_bdragon.rtk0.net created D18575: Add AT_HWCAP / AT_HWCAP2 to elf64_sysvec_v2..
Sun, Dec 16, 5:37 AM · PowerPC

Fri, Dec 14

git_bdragon.rtk0.net created D18565: Fix warning in PowerPC64 ELFv2 build of rtld-elf..
Fri, Dec 14, 11:44 PM · PowerPC
git_bdragon.rtk0.net added a comment to D18547: Fix llvm-tblgen self-bootstrap.

This could plausibly be either that binutils ld.bfd is more picky than lld about undefined references (in code that is ultimately dead, possibly? JSON.o / MicrosoftMangle.o seem like the sort of thing to me that would end up ultimately trimmed out) or that my build went down a path that is untested on amd64.

Fri, Dec 14, 9:17 PM
git_bdragon.rtk0.net updated the diff for D18547: Fix llvm-tblgen self-bootstrap.

Always include xxhash.cpp in libllvm, to fix buildworld on powerpc64 with binutils ld.

Fri, Dec 14, 9:05 PM
git_bdragon.rtk0.net added a comment to D18547: Fix llvm-tblgen self-bootstrap.

I think it's possible that ld.bfd on powerpc64 is more picky than lld when it comes to missing references inside archives, because I am encountering a build failure on the main compiler as well.

Fri, Dec 14, 8:17 PM
git_bdragon.rtk0.net added a comment to D18547: Fix llvm-tblgen self-bootstrap.

make.conf:

OPTIONS_UNSET=DOCS NLS
WRKDIRPREFIX=/usr/obj
Fri, Dec 14, 4:51 PM
git_bdragon.rtk0.net created D18547: Fix llvm-tblgen self-bootstrap.
Fri, Dec 14, 5:36 AM

Sat, Dec 1

git_bdragon.rtk0.net updated the diff for D18364: Fix PowerPC64 ELFv1-specific problem in __elf_phdr_match_addr() leading to crash in threaded programs that unload libraries..

Replaced the comment block with one that jhibbits wrote that's worded a lot better than mine.

Sat, Dec 1, 1:37 AM · PowerPC

Wed, Nov 28

git_bdragon.rtk0.net created D18364: Fix PowerPC64 ELFv1-specific problem in __elf_phdr_match_addr() leading to crash in threaded programs that unload libraries..
Wed, Nov 28, 1:25 AM · PowerPC

Tue, Nov 27

git_bdragon.rtk0.net created D18352: Allow truss to operate on ELFv2 processes..
Tue, Nov 27, 4:11 AM · PowerPC

Mon, Nov 26

git_bdragon.rtk0.net updated the diff for D18330: Fix thread creation in PowerPC64 ELFv2 processes..

Fix ABI violation that was causing crash on thread exit.

Mon, Nov 26, 6:25 AM · PowerPC

Sun, Nov 25

git_bdragon.rtk0.net created D18330: Fix thread creation in PowerPC64 ELFv2 processes..
Sun, Nov 25, 9:19 AM · PowerPC

Nov 2 2018

git_bdragon.rtk0.net created D17813: powerpc64: Fix "show spr" command on ELFv2 kernels.
Nov 2 2018, 4:04 AM · PowerPC

Oct 31 2018

git_bdragon.rtk0.net added a comment to D17117: rtld: set obj->textsize correctly.

Looks good to me too, and I think it's important for anything doing textrelocs on ppc.

Oct 31 2018, 1:24 AM

Oct 29 2018

git_bdragon.rtk0.net added a comment to D17601: ppc64: limited 32-bit DMA address range.

Additionally, the setting of lowaddr to an address below the end of phys memory will force bus_dmamem_alloc to always do contig malloc, which causes additional pressure and possibility of failure. sys/powerpc/powerpc/busdma_machdep.c badly needs some updating to handle stuff like multiple segment allocations and multiple exclusion ranges.

Oct 29 2018, 6:01 PM
git_bdragon.rtk0.net added a comment to D17601: ppc64: limited 32-bit DMA address range.

On POWER9, the controllers operate in 32 bit and 64 bit mode simultaneously. A full fix will require teaching the dma tag handling to handle multiple exclusion ranges, I imagine. Right now it's hardcoded to only allocate memory between 0x0 and lowaddr, and it ignores highaddr entirely.

Oct 29 2018, 5:55 PM
git_bdragon.rtk0.net added a comment to D17601: ppc64: limited 32-bit DMA address range.

Specifically, here's the problem that we're seeing on the Talos II:

Oct 29 2018, 5:52 PM
git_bdragon.rtk0.net added a comment to D17601: ppc64: limited 32-bit DMA address range.

The problem is that the change is applying to 64-bit devices too. It really needs to be restricted to PECs that have 32 bit devices attached.

Oct 29 2018, 5:48 PM

Oct 5 2018

git_bdragon.rtk0.net created D17426: Speed up devel/gdb startup on FreeBSD.
Oct 5 2018, 12:06 AM

Oct 2 2018

git_bdragon.rtk0.net added a comment to D17368: Use nda(4) on powerpc64.

No complaints on this end, it's working just fine here.

Oct 2 2018, 5:27 PM

Oct 1 2018

git_bdragon.rtk0.net added a comment to D17368: Use nda(4) on powerpc64.

oh, THAT's why the cam layer wasn't attaching. Will test this on my equipment now.

Oct 1 2018, 4:59 PM

Sep 19 2018

git_bdragon.rtk0.net updated the diff for D17183: Fix random DMA failure on powernv.

Add a check for powernv by running the opal_check().

Sep 19 2018, 8:02 PM · PowerPC

Sep 17 2018

git_bdragon.rtk0.net added a comment to D17183: Fix random DMA failure on powernv.

Oh, right, I forgot to also put in my original symptoms and train of thought, for people who weren't following along with my chasing of this issue:

Sep 17 2018, 8:31 PM · PowerPC
git_bdragon.rtk0.net added a comment to D17183: Fix random DMA failure on powernv.

hmm, actually, phb3 does the same thing for MSI when initting ioda2 as phb4 does for ioda3. Is this bug also happening on P8?

Sep 17 2018, 7:18 PM · PowerPC
git_bdragon.rtk0.net added a comment to D17183: Fix random DMA failure on powernv.

How about /interrupt-controller@0 is compatible with "ibm,opal-xive-vc" as a check then? That seems to me like a good way to target P9 xics emulation specifically, without matching on P8.

Sep 17 2018, 6:44 PM · PowerPC

Sep 16 2018

git_bdragon.rtk0.net added a comment to D17183: Fix random DMA failure on powernv.

Physical map before:

Sep 16 2018, 9:29 PM · PowerPC
git_bdragon.rtk0.net updated the summary of D17183: Fix random DMA failure on powernv.
Sep 16 2018, 9:23 PM · PowerPC
git_bdragon.rtk0.net created D17183: Fix random DMA failure on powernv.
Sep 16 2018, 9:20 PM · PowerPC

Aug 3 2018

git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

Two more make -j64 buildworld runs complete. I don't know why it's suddenly decided to stop crashing for me, but I will focus on the hardware side if it comes back.

Aug 3 2018, 2:32 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

I figured out the correct SCOM address for interrogating what error happened that is causing my controller freeze, but of course now that I have that figured out, it's not happening.

Aug 3 2018, 4:50 AM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

@kbowling Nah, my fs is fine. I'm saying that this patch is an improvement in that when it crashes, it leaves the fs in a saner state than a crash without the patch.

Aug 3 2018, 3:21 AM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

One definite improvement for me with this patch is that fsck seems to be able to recover the root filesystem after a crash without manual intervention, which I was not seeing before.

Aug 3 2018, 1:15 AM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

I was not on the latest firmware. However, updating the firmware didn't seem to help.

Aug 3 2018, 12:24 AM

Aug 2 2018

git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

Crashed in intx mode as well.

Aug 2 2018, 11:10 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

ok, crashed with the local bits I was wondering about as well.

Aug 2 2018, 10:26 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

ok, so unfortunately, I had another crash when testing with only these changes. My original crash where the controller straight up stops working and won't even reset.

Aug 2 2018, 9:30 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

yep, working on it now.

Aug 2 2018, 8:42 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

I recompiled with just this diff, and I was back to crashing during buildworld. Going to recompile with my local changes again so if it crashes again I have my tools to examine the nvme queues more extensively from the kernel debugger.

Aug 2 2018, 7:44 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

I actually made it through a buildworld for once without panicing for once with this patch. My kernel's not clean at the moment, but I will rerun with only this patch in a bit.

Aug 2 2018, 5:14 PM
git_bdragon.rtk0.net added a comment to D16570: nvme(4): Add bus_dmamap_sync() at the end of the request path.

I actually made it through a buildworld for once without panicing for once with this patch. My kernel's not clean at the moment, but I will rerun with only this patch in a bit.

Aug 2 2018, 5:09 PM
git_bdragon.rtk0.net added a comment to D16552: random: Add PowerPC 'darn' instruction entropy source.

Attaches fine on my Talos II. Seems to work fine, as far as I can tell.

Aug 2 2018, 2:31 PM

Jun 24 2018

git_bdragon.rtk0.net created D15986: Expose stopped cpu contexts to ddb on PowerPC.
Jun 24 2018, 2:04 AM · PowerPC

Jun 23 2018

git_bdragon.rtk0.net changed the edit policy for D15973: Fix debug trap on Talos II (and all of ppc64?).
Jun 23 2018, 12:16 AM · PowerPC
git_bdragon.rtk0.net added a watcher for PowerPC: git_bdragon.rtk0.net.
Jun 23 2018, 12:11 AM
git_bdragon.rtk0.net created D15973: Fix debug trap on Talos II (and all of ppc64?).
Jun 23 2018, 12:07 AM · PowerPC