Liu Xuezhao
Liu Xuezhao
@daos-stack/daos-gatekeeper could it be landed? it need to be backport to 2.2 and 2.0 I think. Thx.
> seems like this requires to add at least a unit test to verify that it's working properly? OK, let me refresh to add a UT.
> Yup, this is good point, i can try, but i guess this is not super helpful for fio cases, since now holding write locks are few cases. I am...
> I indeed have a local patch to do this, i could push this. that's great, thanks.
I ever considered it long time ago. The vector rma should be supported by OFI, but OFI providers with limit setting for the local iov_count and remote rma_iov_count (fi_info::tx_attr::iov_limit/::rma_iov_limit), and...
Hi, For the core affinity, our basic considerations are: 1) Upper layer creates different affinity contexts when init, for example one affinity context for one NUMA node. 2) One affinity...
For affinity of lnet, you can find some slides by googling it. There is a wiki page about it as well: https://wiki.hpdd.intel.com/display/PUB/MDS+SMP+Node+Affinity+High+Level+Design It basically falls into the similar method as...
> I didn't see the new umem API for generating redo log, it'll be added in a follow-on PR? seems can reuse umem_tx_add_range/umem_tx_add_ptr APIs? may need to add a flag...
"@liuxuezhao do you think this patch could have caused it? Or possibly it is a rare issue not yet seen in the testing?" I think that assertion in vos agg...
the original patch was to handle the RF concern right? (if you put multiple shards in the same target during rebuild, after rebuild lost the tgt it will break RF...