Messages in this thread | | | From | "Kani, Toshimitsu" <> | Subject | Re: [PATCH v2 4/7] ghes_edac: avoid multiple calls to dmi_walk() | Date | Thu, 17 Aug 2017 21:08:40 +0000 |
| |
On Wed, 2017-08-16 at 11:51 -0600, Toshi Kani wrote: > On Wed, 2017-08-16 at 19:40 +0200, Borislav Petkov wrote: > > On Wed, Aug 16, 2017 at 05:28:50PM +0000, Kani, Toshimitsu wrote: : > > > I will test the patch with an SCI when I got a chance. I won't > > > be able to test other notification types or race conditions, > > > though. > > > > Thanks, here's the latest version with the atomic registration too. > > Sure, I will test with this version.
I briefly tested the patch with an SCI. The error reporting part seems to work fine.
There are two issues I noticed:
1. It creates mc0 and mc1. I think this is because you called edac_mc_alloc() with mc_num 1.
2. 'ras-mc-ctl --layout' does not show all DIMMs.
# ras-mc-ctl --layout +------------------------+ | mc0 | mc1 | ---------+------------------------+ memory4: | 0 MB | 16384 MB | memory3: | 0 MB | 0 MB | memory2: | 0 MB | 0 MB | memory1: | 0 MB | 0 MB | memory0: | 0 MB | 8192 MB | ---------+-----------------------+
# ls /sys/bus/mc1/devices csrow0 csrow12 csrow23 dimm0 dimm12 dimm23 mc1 csrow11 csrow19 csrow4 dimm11 dimm19 dimm4
Thanks, -Toshi
| |