On Thu, Aug 29, 2019 at 4:34 PM Aneesh Kumar K.V
<aneesh.ku...@linux.ibm.com> wrote:
>
> Right now we force an unbind of SCM memory at drcindex on H_OVERLAP error.
> This really slows down operations like kexec where we get the H_OVERLAP
> error because we don't go through a full hypervisor re init.
Maybe we should be unbinding it on a kexec().
> H_OVERLAP error for a H_SCM_BIND_MEM hcall indicates that SCM memory at
> drc index is already bound. Since we don't specify a logical memory
> address for bind hcall, we can use the H_SCM_QUERY hcall to query
> the already bound logical address.
This is a little sketchy since we might have crashed during the
initial bind. Checking if the last block is bound to where we expect
it to be might be a good idea. If it's not where we expect it to be,
then an unbind->bind cycle is the only sane thing to do.
> Boot time difference with and without patch is:
>
> [ 5.583617] IOMMU table initialized, virtual merging enabled
> [ 5.603041] papr_scm ibm,persistent-memory:ibm,pmemory@44104001: Retrying
> bind after unbinding
> [ 301.514221] papr_scm ibm,persistent-memory:ibm,pmemory@44108001: Retrying
> bind after unbinding
> [ 340.057238] hv-24x7: read 1530 catalog entries, created 537 event attrs (0
> failures), 275 descs
Is the unbind significantly slower than a bind? Or is the region here
just massive?
> after fix
>
> [ 5.101572] IOMMU table initialized, virtual merging enabled
> [ 5.116984] papr_scm ibm,persistent-memory:ibm,pmemory@44104001: Querying
> SCM details
> [ 5.117223] papr_scm ibm,persistent-memory:ibm,pmemory@44108001: Querying
> SCM details
> [ 5.120530] hv-24x7: read 1530 catalog entries, created 537 event attrs (0
> failures), 275 descs
>
> Signed-off-by: Aneesh Kumar K.V <aneesh.ku...@linux.ibm.com>
> ---
> arch/powerpc/platforms/pseries/papr_scm.c | 26 ++++++++++++++++++++---
> 1 file changed, 23 insertions(+), 3 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/papr_scm.c
> b/arch/powerpc/platforms/pseries/papr_scm.c
> index 220e595cb579..4b74cfe7b334 100644
> --- a/arch/powerpc/platforms/pseries/papr_scm.c
> +++ b/arch/powerpc/platforms/pseries/papr_scm.c
> @@ -110,6 +110,27 @@ static void drc_pmem_unbind(struct papr_scm_priv *p)
> return;
> }
>
> +static int drc_pmem_query(struct papr_scm_priv *p)
> +{
> + unsigned long ret[PLPAR_HCALL_BUFSIZE];
> + int64_t rc;
> +
> +
> + rc = plpar_hcall(H_SCM_QUERY_BLOCK_MEM_BINDING, ret,
> + p->drc_index, 0);
> +
> + if (rc) {
> + dev_err(&p->pdev->dev, "Failed to bind SCM");
> + return rc;
> + }
> +
> + p->bound_addr = ret[0];
> + dev_dbg(&p->pdev->dev, "bound drc 0x%x to %pR\n", p->drc_index,
> &p->res);
> +
> + return 0;
> +}
> +
> +
> static int papr_scm_meta_get(struct papr_scm_priv *p,
> struct nd_cmd_get_config_data_hdr *hdr)
> {
> @@ -431,9 +452,8 @@ static int papr_scm_probe(struct platform_device *pdev)
>
> /* If phyp says drc memory still bound then force unbound and retry */
> if (rc == H_OVERLAP) {
> - dev_warn(&pdev->dev, "Retrying bind after unbinding\n");
> - drc_pmem_unbind(p);
> - rc = drc_pmem_bind(p);
> + dev_warn(&pdev->dev, "Querying SCM details\n");
That's a pretty vague message. If we're going to treat leaving the
region bound over kexec() as normal then you might want to bump it
down to pr_info() or so.
> + rc = drc_pmem_query(p);
> }
>
> if (rc != H_SUCCESS) {
> --
> 2.21.0
>
