From: Li Zhijian <lizhijian@fujitsu.com>
To: Jason Gunthorpe <jgg@ziepe.ca>, Zhu Yanjun <zyjzyj2000@gmail.com>,
"Leon Romanovsky" <leon@kernel.org>, <linux-rdma@vger.kernel.org>
Cc: Xiao Yang <yangx.jy@fujitsu.com>, <y-goto@fujitsu.com>,
Bob Pearson <rpearsonhpe@gmail.com>,
Mark Bloch <mbloch@nvidia.com>,
Aharon Landau <aharonl@nvidia.com>, Tom Talpey <tom@talpey.com>,
<tomasz.gromadzki@intel.com>,
Dan Williams <dan.j.williams@intel.com>,
<linux-kernel@vger.kernel.org>,
Li Zhijian <lizhijian@fujitsu.com>
Subject: [PATCH v4 1/6] RDMA: Allow registering MR with flush access flags
Date: Fri, 5 Aug 2022 07:46:14 +0000 [thread overview]
Message-ID: <1659685579-2-2-git-send-email-lizhijian@fujitsu.com> (raw)
In-Reply-To: <1659685579-2-1-git-send-email-lizhijian@fujitsu.com>
It makes device/HCA support new FLUSH attributes/capabilities, and it
also makes memory region support new FLUSH access flags.
Users can use ibv_reg_mr(3) to register flush access flags. Only the
access flags also supported by device's capabilities can be registered
successfully.
Once registered successfully, it means the MR is flushable. Similarly,
A flushable MR should also have one or both of GLOBAL_VISIBILITY and
PERSISTENT attributes/capabilities like device/HCA.
CC: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
---
V4: set is_pmem more simply
V2: new scheme check is_pmem # Dan
---
include/rdma/ib_verbs.h | 17 ++++++++++++++++-
include/uapi/rdma/ib_user_ioctl_verbs.h | 2 ++
include/uapi/rdma/ib_user_verbs.h | 2 ++
3 files changed, 20 insertions(+), 1 deletion(-)
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 7c2f76f34f6f..aa174cdcdf5a 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -270,6 +270,10 @@ enum ib_device_cap_flags {
/* The device supports padding incoming writes to cacheline. */
IB_DEVICE_PCI_WRITE_END_PADDING =
IB_UVERBS_DEVICE_PCI_WRITE_END_PADDING,
+ /* Placement type attributes */
+ IB_DEVICE_PLT_GLOBAL_VISIBILITY =
+ IB_UVERBS_DEVICE_PLT_GLOBAL_VISIBILITY,
+ IB_DEVICE_PLT_PERSISTENT = IB_UVERBS_DEVICE_PLT_PERSISTENT,
};
enum ib_kernel_cap_flags {
@@ -1458,10 +1462,14 @@ enum ib_access_flags {
IB_ACCESS_ON_DEMAND = IB_UVERBS_ACCESS_ON_DEMAND,
IB_ACCESS_HUGETLB = IB_UVERBS_ACCESS_HUGETLB,
IB_ACCESS_RELAXED_ORDERING = IB_UVERBS_ACCESS_RELAXED_ORDERING,
+ IB_ACCESS_FLUSH_GLOBAL_VISIBILITY = IB_UVERBS_ACCESS_FLUSH_GLOBAL_VISIBILITY,
+ IB_ACCESS_FLUSH_PERSISTENT = IB_UVERBS_ACCESS_FLUSH_PERSISTENT,
+ IB_ACCESS_FLUSHABLE = IB_ACCESS_FLUSH_GLOBAL_VISIBILITY |
+ IB_ACCESS_FLUSH_PERSISTENT,
IB_ACCESS_OPTIONAL = IB_UVERBS_ACCESS_OPTIONAL_RANGE,
IB_ACCESS_SUPPORTED =
- ((IB_ACCESS_HUGETLB << 1) - 1) | IB_ACCESS_OPTIONAL,
+ ((IB_ACCESS_FLUSH_PERSISTENT << 1) - 1) | IB_ACCESS_OPTIONAL,
};
/*
@@ -4310,6 +4318,7 @@ int ib_dealloc_xrcd_user(struct ib_xrcd *xrcd, struct ib_udata *udata);
static inline int ib_check_mr_access(struct ib_device *ib_dev,
unsigned int flags)
{
+ u64 device_cap = ib_dev->attrs.device_cap_flags;
/*
* Local write permission is required if remote write or
* remote atomic permission is also requested.
@@ -4324,6 +4333,12 @@ static inline int ib_check_mr_access(struct ib_device *ib_dev,
if (flags & IB_ACCESS_ON_DEMAND &&
!(ib_dev->attrs.kernel_cap_flags & IBK_ON_DEMAND_PAGING))
return -EINVAL;
+
+ if ((flags & IB_ACCESS_FLUSH_GLOBAL_VISIBILITY &&
+ !(device_cap & IB_DEVICE_PLT_GLOBAL_VISIBILITY)) ||
+ (flags & IB_ACCESS_FLUSH_PERSISTENT &&
+ !(device_cap & IB_DEVICE_PLT_PERSISTENT)))
+ return -EINVAL;
return 0;
}
diff --git a/include/uapi/rdma/ib_user_ioctl_verbs.h b/include/uapi/rdma/ib_user_ioctl_verbs.h
index 7dd56210226f..32d44ca328b9 100644
--- a/include/uapi/rdma/ib_user_ioctl_verbs.h
+++ b/include/uapi/rdma/ib_user_ioctl_verbs.h
@@ -57,6 +57,8 @@ enum ib_uverbs_access_flags {
IB_UVERBS_ACCESS_ZERO_BASED = 1 << 5,
IB_UVERBS_ACCESS_ON_DEMAND = 1 << 6,
IB_UVERBS_ACCESS_HUGETLB = 1 << 7,
+ IB_UVERBS_ACCESS_FLUSH_GLOBAL_VISIBILITY = 1 << 8,
+ IB_UVERBS_ACCESS_FLUSH_PERSISTENT = 1 << 9,
IB_UVERBS_ACCESS_RELAXED_ORDERING = IB_UVERBS_ACCESS_OPTIONAL_FIRST,
IB_UVERBS_ACCESS_OPTIONAL_RANGE =
diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h
index 7dd903d932e5..a58df0ebcb79 100644
--- a/include/uapi/rdma/ib_user_verbs.h
+++ b/include/uapi/rdma/ib_user_verbs.h
@@ -1331,6 +1331,8 @@ enum ib_uverbs_device_cap_flags {
/* Deprecated. Please use IB_UVERBS_RAW_PACKET_CAP_SCATTER_FCS. */
IB_UVERBS_DEVICE_RAW_SCATTER_FCS = 1ULL << 34,
IB_UVERBS_DEVICE_PCI_WRITE_END_PADDING = 1ULL << 36,
+ IB_UVERBS_DEVICE_PLT_GLOBAL_VISIBILITY = 1ULL << 38,
+ IB_UVERBS_DEVICE_PLT_PERSISTENT = 1ULL << 39,
};
enum ib_uverbs_raw_packet_caps {
--
2.31.1
next prev parent reply other threads:[~2022-08-05 7:40 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-05 7:46 [PATCH v4 0/6] RDMA/rxe: Add RDMA FLUSH operation Li Zhijian
2022-08-05 7:46 ` Li Zhijian [this message]
2022-08-05 7:46 ` [PATCH v4 2/6] RDMA/rxe: Allow registering persistent flag for pmem MR only Li Zhijian
2022-08-05 7:46 ` [PATCH v4 3/6] RDMA/rxe: Implement RC RDMA FLUSH service in requester side Li Zhijian
2022-08-05 7:46 ` [PATCH v4 4/6] RDMA/rxe: Implement flush execution in responder side Li Zhijian
2022-08-05 7:55 ` [PATCH v4 5/6] RDMA/rxe: Implement flush completion Li Zhijian
2022-08-05 7:55 ` [PATCH v4 6/6] RDMA/rxe: Enable RDMA FLUSH capability for rxe device Li Zhijian
2022-08-26 13:37 ` [PATCH v4 0/6] RDMA/rxe: Add RDMA FLUSH operation Jason Gunthorpe
2022-08-29 1:36 ` yangx.jy
2022-08-29 1:40 ` yangx.jy
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=1659685579-2-2-git-send-email-lizhijian@fujitsu.com \
--to=lizhijian@fujitsu.com \
--cc=aharonl@nvidia.com \
--cc=dan.j.williams@intel.com \
--cc=jgg@ziepe.ca \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=mbloch@nvidia.com \
--cc=rpearsonhpe@gmail.com \
--cc=tom@talpey.com \
--cc=tomasz.gromadzki@intel.com \
--cc=y-goto@fujitsu.com \
--cc=yangx.jy@fujitsu.com \
--cc=zyjzyj2000@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