From: Vaibhav Jain <vaibhav@linux.ibm.com>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>, mpe@ellerman.id.au
Cc: Oliver O'Halloran <oohall@gmail.com>,
linuxppc-dev@lists.ozlabs.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
Subject: Re: [PATCH v2 2/2] powerpc/nvdimm: use H_SCM_QUERY hcall on H_OVERLAP error
Date: Wed, 04 Sep 2019 12:29:09 +0530 [thread overview]
Message-ID: <87v9u8tuma.fsf@vajain21.in.ibm.com> (raw)
In-Reply-To: <20190903123452.28620-2-aneesh.kumar@linux.ibm.com>
Hi Aneesh,
Thanks for the patch. A minor suggestion below:
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com> writes:
> 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.
>
> 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.
>
> 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
>
> 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.kumar@linux.ibm.com>
> ---
> Changes from V1:
> * Use the first block and last block to query the logical bind memory
> * If we fail to query, ubind and retry the bind.
>
>
> arch/powerpc/platforms/pseries/papr_scm.c | 48 +++++++++++++++++++----
> 1 file changed, 40 insertions(+), 8 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/papr_scm.c b/arch/powerpc/platforms/pseries/papr_scm.c
> index 3bef4d298ac6..61883291defc 100644
> --- a/arch/powerpc/platforms/pseries/papr_scm.c
> +++ b/arch/powerpc/platforms/pseries/papr_scm.c
> @@ -65,10 +65,8 @@ static int drc_pmem_bind(struct papr_scm_priv *p)
> cond_resched();
> } while (rc == H_BUSY);
>
> - if (rc) {
> - dev_err(&p->pdev->dev, "bind err: %lld\n", rc);
> + if (rc)
> return rc;
> - }
>
> p->bound_addr = saved;
> dev_dbg(&p->pdev->dev, "bound drc 0x%x to %pR\n", p->drc_index, &p->res);
> @@ -110,6 +108,42 @@ static void drc_pmem_unbind(struct papr_scm_priv *p)
> return;
> }
>
> +static int drc_pmem_query_n_bind(struct papr_scm_priv *p)
> +{
> + unsigned long start_addr;
> + unsigned long end_addr;
> + 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)
> + goto err_out;
> + start_addr = ret[0];
> +
> + /* Make sure the full region is bound. */
> + rc = plpar_hcall(H_SCM_QUERY_BLOCK_MEM_BINDING, ret,
> + p->drc_index, p->blocks - 1);
> + if (rc)
> + goto err_out;
> + end_addr = ret[0];
> +
> + if ((end_addr - start_addr) != ((p->blocks - 1) * p->block_size))
> + goto err_out;
> +
> + p->bound_addr = start_addr;
> + dev_dbg(&p->pdev->dev, "bound drc 0x%x to %pR\n", p->drc_index, &p->res);
> + return rc;
> +
> +err_out:
> + dev_info(&p->pdev->dev,
> + "Failed to query, trying an unbind followed by bind");
> + drc_pmem_unbind(p);
> + return drc_pmem_bind(p);
> +}
Would have preferred error handling for bind failure to be done at
single location i.e in papr_scm_probe() rather than in
drc_pmem_query_n_bind().
> +
> +
> static int papr_scm_meta_get(struct papr_scm_priv *p,
> struct nd_cmd_get_config_data_hdr *hdr)
> {
> @@ -430,13 +464,11 @@ static int papr_scm_probe(struct platform_device *pdev)
> rc = drc_pmem_bind(p);
>
> /* 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);
> - }
> + if (rc == H_OVERLAP)
> + rc = drc_pmem_query_n_bind(p);
>
> if (rc != H_SUCCESS) {
> + dev_err(&p->pdev->dev, "bind err: %d\n", rc);
> rc = -ENXIO;
> goto err;
> }
> --
> 2.21.0
>
--
Vaibhav Jain <vaibhav@linux.ibm.com>
Linux Technology Center, IBM India Pvt. Ltd.
next prev parent reply other threads:[~2019-09-04 7:01 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-03 12:34 [PATCH v2 1/2] powerpc/nvdimm: Use HCALL error as the return value Aneesh Kumar K.V
2019-09-03 12:34 ` [PATCH v2 2/2] powerpc/nvdimm: use H_SCM_QUERY hcall on H_OVERLAP error Aneesh Kumar K.V
2019-09-04 6:59 ` Vaibhav Jain [this message]
2019-09-24 10:59 ` Aneesh Kumar K.V
2019-09-04 5:39 ` [PATCH v2 1/2] powerpc/nvdimm: Use HCALL error as the return value Vaibhav Jain
2019-09-24 11:02 ` Aneesh Kumar K.V
2019-09-25 11:05 ` Michael Ellerman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87v9u8tuma.fsf@vajain21.in.ibm.com \
--to=vaibhav@linux.ibm.com \
--cc=aneesh.kumar@linux.ibm.com \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=oohall@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).