From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4E0B4C433EF for ; Thu, 7 Apr 2022 00:49:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232287AbiDGAvF (ORCPT ); Wed, 6 Apr 2022 20:51:05 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59064 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229775AbiDGAvE (ORCPT ); Wed, 6 Apr 2022 20:51:04 -0400 Received: from mail-vk1-xa2a.google.com (mail-vk1-xa2a.google.com [IPv6:2607:f8b0:4864:20::a2a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 68D011DE6C5 for ; Wed, 6 Apr 2022 17:49:04 -0700 (PDT) Received: by mail-vk1-xa2a.google.com with SMTP id q10so331142vkh.0 for ; Wed, 06 Apr 2022 17:49:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc:content-transfer-encoding; bh=UemPf44OnvW/XK7+t7s199pddCRJfCIPX9/cujiGb14=; b=FpCwV+2DWPPES5IQDGwH4IFL3cTr7sX8oRf+sEW7VK9sjlk0LcLA10id7z+6SY4+3F qIAgVmiuyxv9inZC1Wt52nh6RKSy1F3Z5j49mLDYUWr5a1Pcbsug/jt/mVrXqujFF30Y jDjBdUmgeXZs44/1J+H+MtNUPZBKJ4LPiilmCf48+t2SFnY0Hkg5fu+T6CJ3yB3dBgd7 FAQary1y3hixVouUjfRG2loM/su5lU+u/Z7L/MMTsWlCs55eOjtJ4NScHCeagYSY/JJI eIuyu4VKksiCU4a5qC2BRe4wEh8Z1feDIz1ii3t27xzLnijPOFsXASVoliuFvPbF1APS gdNg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc:content-transfer-encoding; bh=UemPf44OnvW/XK7+t7s199pddCRJfCIPX9/cujiGb14=; b=7vzYLtsgCSyFeFeL1ohWcwrOcYgl47WkbkB7VRJtyp4FV67zF+/LbvIOQE1a1plmwF 4LqDnyx7ONIB3dMfHtmxlcAz6pcoLY0GF4qNGTNfLwTw8Sa0k2BHMdlMFFrURplk25Hp qNQdXNfJC2zyynZtwRDejP3NqN6t4lW6opVHBPQAa3MtFaA9xv/jirJVqo4T/HJA3HAV up4+u3EI9oQ0ED6OztRwzpwtlr64cKWhyUvt82M3aQKtJDqiT96qS6u3Mo8YYG19CS3E mzVq3DXq8GTyxinxIWXxJ5MAz2IJJxoNdUoFK0NzQhrGlbHYcNtsNjH8Mf5lHahVblm3 cclQ== X-Gm-Message-State: AOAM532NyA3/23yNCwjw+B0G21vY3zrNO2+wMtja1LBUj4ggh73G3WkK pJOJL0WXim0cyUI7Vj12Tf12eXbqVH2QJoASM/c= X-Google-Smtp-Source: ABdhPJwdmSBmcA8moYQSCh3emrK23zT4ogUrdS6Y4esBwFaODu9IfKtkHIKFcYx8e/SnkR77dvkk17XmMob+tyGSo/M= X-Received: by 2002:a1f:a982:0:b0:33e:2d80:4493 with SMTP id s124-20020a1fa982000000b0033e2d804493mr4261318vke.37.1649292543103; Wed, 06 Apr 2022 17:49:03 -0700 (PDT) MIME-Version: 1.0 References: <20220404045549.76547-1-hyc.lee@gmail.com> In-Reply-To: From: Hyunchul Lee Date: Thu, 7 Apr 2022 09:48:51 +0900 Message-ID: Subject: Re: [PATCH v2] ksmbd: smbd: handle multiple Buffer Decriptors To: Namjae Jeon Cc: linux-cifs , Steve French , Sergey Senozhatsky Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-cifs@vger.kernel.org 2022=EB=85=84 4=EC=9B=94 6=EC=9D=BC (=EC=88=98) =EC=98=A4=ED=9B=84 10:46, N= amjae Jeon =EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=B1: > > 2022-04-04 13:55 GMT+09:00, Hyunchul Lee : > > Make ksmbd handle multiple buffer descriptors > > when reading and writing files using SMB direct: > > > > - Change the prototype of transport's operations > > to accept a pointer and length of descriptors. > > - Post the work requests of rdma_rw_ctx for > > RDMA r/w in smb_direct_rdma_xmit(), and > > the work request for the READ/WRITE response > > with a remote invaliation in smb_direct_writev(). > > - SMB2_READ/WRITE request needs the number of > > rw credits, (the pages the request wants to > > transfer / the maximum pages which can be > > registered with one MR) to read and write > > a file. > > - Allocate enough RDMA resources for the maximum > > number of rw credits allowed by ksmbd. > This patch seems to be a patch that combines 4 or more patches into > one. Is there a reason you made a patch like this? Can't you split it > into multiple pieces for review? > Okay, I will split this into 3 patches, which are changing the function prototypes, controlling RDMA resources, and Reading/writing with multiple buffer descriptors. > > > > Signed-off-by: Hyunchul Lee > > --- > > changes from v1: > > - use le16_to_cpu() instead of le32_to_cpu() to retrieve > > req->ReadChannelInfoOffset(reported by kernel test bot). > > > > > > fs/ksmbd/connection.c | 32 ++-- > > fs/ksmbd/connection.h | 32 ++-- > > fs/ksmbd/ksmbd_work.h | 4 +- > > fs/ksmbd/smb2pdu.c | 77 ++++----- > > fs/ksmbd/transport_rdma.c | 344 ++++++++++++++++++++++---------------- > > fs/ksmbd/transport_tcp.c | 5 +- > > 6 files changed, 278 insertions(+), 216 deletions(-) > > > > diff --git a/fs/ksmbd/connection.c b/fs/ksmbd/connection.c > > index 208d2cff7bd3..6f036ea9f43b 100644 > > --- a/fs/ksmbd/connection.c > > +++ b/fs/ksmbd/connection.c > > @@ -191,10 +191,10 @@ int ksmbd_conn_write(struct ksmbd_work *work) > > } > > > > ksmbd_conn_lock(conn); > > - sent =3D conn->transport->ops->writev(conn->transport, &iov[0], > > - iov_idx, len, > > - work->need_invalidate_rkey, > > - work->remote_key); > > + sent =3D conn->transport->ops->writev(conn->transport, > > + &iov[0], iov_idx, len, > > + work->need_invalidate_rkey, > > + work->remote_key); > > ksmbd_conn_unlock(conn); > > > > if (sent < 0) { > > @@ -205,31 +205,35 @@ int ksmbd_conn_write(struct ksmbd_work *work) > > return 0; > > } > > > > -int ksmbd_conn_rdma_read(struct ksmbd_conn *conn, void *buf, > > - unsigned int buflen, u32 remote_key, u64 remote_= offset, > > - u32 remote_len) > > +int ksmbd_conn_rdma_read(struct ksmbd_conn *conn, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len) > > { > > int ret =3D -EINVAL; > > > > + ksmbd_conn_lock(conn); > > if (conn->transport->ops->rdma_read) > > ret =3D conn->transport->ops->rdma_read(conn->transport, > > buf, buflen, > > - remote_key, remote_= offset, > > - remote_len); > > + desc, desc_len); > > + ksmbd_conn_unlock(conn); > Is it related to multiple buffer descriptors? No, I will remove these from this patch. > > > return ret; > > } > > > > -int ksmbd_conn_rdma_write(struct ksmbd_conn *conn, void *buf, > > - unsigned int buflen, u32 remote_key, > > - u64 remote_offset, u32 remote_len) > > +int ksmbd_conn_rdma_write(struct ksmbd_conn *conn, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len) > > { > > int ret =3D -EINVAL; > > > > + ksmbd_conn_lock(conn); > > if (conn->transport->ops->rdma_write) > > ret =3D conn->transport->ops->rdma_write(conn->transport, > > buf, buflen, > > - remote_key, remote= _offset, > > - remote_len); > > + desc, desc_len); > > + ksmbd_conn_unlock(conn); > > return ret; > > } > > > > diff --git a/fs/ksmbd/connection.h b/fs/ksmbd/connection.h > > index 7a59aacb5daa..51722d3a8cf6 100644 > > --- a/fs/ksmbd/connection.h > > +++ b/fs/ksmbd/connection.h > > @@ -119,14 +119,18 @@ struct ksmbd_transport_ops { > > void (*disconnect)(struct ksmbd_transport *t); > > void (*shutdown)(struct ksmbd_transport *t); > > int (*read)(struct ksmbd_transport *t, char *buf, unsigned int si= ze); > > - int (*writev)(struct ksmbd_transport *t, struct kvec *iovs, int n= iov, > > - int size, bool need_invalidate_rkey, > > + int (*writev)(struct ksmbd_transport *t, > > + struct kvec *iovs, int niov, int size, > > + bool need_invalidate, > > unsigned int remote_key); > > - int (*rdma_read)(struct ksmbd_transport *t, void *buf, unsigned i= nt len, > > - u32 remote_key, u64 remote_offset, u32 remote_le= n); > > - int (*rdma_write)(struct ksmbd_transport *t, void *buf, > > - unsigned int len, u32 remote_key, u64 remote_of= fset, > > - u32 remote_len); > > + int (*rdma_read)(struct ksmbd_transport *t, > > + void *buf, unsigned int len, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len); > > + int (*rdma_write)(struct ksmbd_transport *t, > > + void *buf, unsigned int len, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len); > > }; > > > > struct ksmbd_transport { > > @@ -148,12 +152,14 @@ struct ksmbd_conn *ksmbd_conn_alloc(void); > > void ksmbd_conn_free(struct ksmbd_conn *conn); > > bool ksmbd_conn_lookup_dialect(struct ksmbd_conn *c); > > int ksmbd_conn_write(struct ksmbd_work *work); > > -int ksmbd_conn_rdma_read(struct ksmbd_conn *conn, void *buf, > > - unsigned int buflen, u32 remote_key, u64 remote_= offset, > > - u32 remote_len); > > -int ksmbd_conn_rdma_write(struct ksmbd_conn *conn, void *buf, > > - unsigned int buflen, u32 remote_key, u64 remote= _offset, > > - u32 remote_len); > > +int ksmbd_conn_rdma_read(struct ksmbd_conn *conn, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len); > > +int ksmbd_conn_rdma_write(struct ksmbd_conn *conn, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len); > > void ksmbd_conn_enqueue_request(struct ksmbd_work *work); > > int ksmbd_conn_try_dequeue_request(struct ksmbd_work *work); > > void ksmbd_conn_init_server_callbacks(struct ksmbd_conn_ops *ops); > > diff --git a/fs/ksmbd/ksmbd_work.h b/fs/ksmbd/ksmbd_work.h > > index 5ece58e40c97..58bfc661000d 100644 > > --- a/fs/ksmbd/ksmbd_work.h > > +++ b/fs/ksmbd/ksmbd_work.h > > @@ -69,9 +69,9 @@ struct ksmbd_work { > > bool encrypted:1; > > /* Is this SYNC or ASYNC ksmbd_work */ > > bool syncronous:1; > > - bool need_invalidate_rkey:1; > > + bool need_invalidate_rkey:1; > > > > - unsigned int remote_key; > > + unsigned int remote_key; > You change only need_invalidate_rkey and remote_key to tab? What > about other variables in this structure? > > /* cancel works */ > > int async_id; > > void **cancel_argv; > > diff --git a/fs/ksmbd/smb2pdu.c b/fs/ksmbd/smb2pdu.c > > index 3bf6c56c654c..8d41e4966905 100644 > > --- a/fs/ksmbd/smb2pdu.c > > +++ b/fs/ksmbd/smb2pdu.c > > @@ -6115,11 +6115,11 @@ static noinline int smb2_read_pipe(struct ksmbd= _work > > *work) > > return err; > > } > > > > -static int smb2_set_remote_key_for_rdma(struct ksmbd_work *work, > > - struct smb2_buffer_desc_v1 *desc, > > - __le32 Channel, > > - __le16 ChannelInfoOffset, > > - __le16 ChannelInfoLength) > > +static int smb2_validate_rdma_buffer_descs(struct ksmbd_work *work, > > + struct smb2_buffer_desc_v1 *de= sc, > > + __le32 Channel, > > + __le16 ChannelInfoOffset, > > + __le16 ChannelInfoLength) > > { > > unsigned int i, ch_count; > > > > @@ -6136,15 +6136,13 @@ static int smb2_set_remote_key_for_rdma(struct > > ksmbd_work *work, > > le32_to_cpu(desc[i].length)); > > } > > } > > - if (ch_count !=3D 1) { > > - ksmbd_debug(RDMA, "RDMA multiple buffer descriptors %d ar= e not supported > > yet\n", > > - ch_count); > > + if (ch_count < 1) > if (!ch_count) I will change it. > > > return -EINVAL; > > - } > > > > - work->need_invalidate_rkey =3D > > - (Channel =3D=3D SMB2_CHANNEL_RDMA_V1_INVALIDATE); > > - work->remote_key =3D le32_to_cpu(desc->token); > > + if (Channel =3D=3D SMB2_CHANNEL_RDMA_V1_INVALIDATE) { > > + work->need_invalidate_rkey =3D true; > > + work->remote_key =3D le32_to_cpu(desc[0].token); > > + } > > return 0; > > } > > > > @@ -6152,14 +6150,12 @@ static ssize_t smb2_read_rdma_channel(struct > > ksmbd_work *work, > > struct smb2_read_req *req, void *da= ta_buf, > > size_t length) > > { > > - struct smb2_buffer_desc_v1 *desc =3D > > - (struct smb2_buffer_desc_v1 *)&req->Buffer[0]; > > int err; > > > > err =3D ksmbd_conn_rdma_write(work->conn, data_buf, length, > > - le32_to_cpu(desc->token), > > - le64_to_cpu(desc->offset), > > - le32_to_cpu(desc->length)); > > + (struct smb2_buffer_desc_v1 *) > > + ((char *)req + le16_to_cpu(req->ReadC= hannelInfoOffset)), > > + le16_to_cpu(req->ReadChannelInfoLengt= h)); > > if (err) > > return err; > > > > @@ -6193,18 +6189,20 @@ int smb2_read(struct ksmbd_work *work) > > > > if (req->Channel =3D=3D SMB2_CHANNEL_RDMA_V1_INVALIDATE || > > req->Channel =3D=3D SMB2_CHANNEL_RDMA_V1) { > > - unsigned int ch_offset =3D le16_to_cpu(req->ReadChannelIn= foOffset); > > + struct smb2_buffer_desc_v1 *descs =3D (struct smb2_buffer= _desc_v1 *) > > + ((char *)req + le16_to_cpu(req->ReadChannelInfoOf= fset)); > > > > - if (ch_offset < offsetof(struct smb2_read_req, Buffer)) { > > + if (le16_to_cpu(req->ReadChannelInfoOffset) < > > + offsetof(struct smb2_read_req, Buffer)) { > > err =3D -EINVAL; > > goto out; > > } > > - err =3D smb2_set_remote_key_for_rdma(work, > > - (struct smb2_buffer_de= sc_v1 *) > > - ((char *)req + ch_offs= et), > > - req->Channel, > > - req->ReadChannelInfoOf= fset, > > - req->ReadChannelInfoLe= ngth); > > + > > + err =3D smb2_validate_rdma_buffer_descs(work, > > + descs, > > + req->Channel, > > + req->ReadChannelInf= oOffset, > > + req->ReadChannelInf= oLength); > > if (err) > > goto out; > > } > > @@ -6252,8 +6250,7 @@ int smb2_read(struct ksmbd_work *work) > > work->aux_payload_buf =3D NULL; > > rsp->hdr.Status =3D STATUS_END_OF_FILE; > > smb2_set_err_rsp(work); > > - ksmbd_fd_put(work, fp); > > - return 0; > > + goto out; > > } > > > > ksmbd_debug(SMB, "nbytes %zu, offset %lld mincount %zu\n", > > @@ -6386,21 +6383,18 @@ static ssize_t smb2_write_rdma_channel(struct > > ksmbd_work *work, > > struct ksmbd_file *fp, > > loff_t offset, size_t length, bool= sync) > > { > > - struct smb2_buffer_desc_v1 *desc; > > char *data_buf; > > int ret; > > ssize_t nbytes; > > > > - desc =3D (struct smb2_buffer_desc_v1 *)&req->Buffer[0]; > > - > > data_buf =3D kvmalloc(length, GFP_KERNEL | __GFP_ZERO); > > if (!data_buf) > > return -ENOMEM; > > > > ret =3D ksmbd_conn_rdma_read(work->conn, data_buf, length, > > - le32_to_cpu(desc->token), > > - le64_to_cpu(desc->offset), > > - le32_to_cpu(desc->length)); > > + (struct smb2_buffer_desc_v1 *) > > + ((char *)req + le16_to_cpu(req->WriteC= hannelInfoOffset)), > > + le16_to_cpu(req->WriteChannelInfoLengt= h)); > > if (ret < 0) { > > kvfree(data_buf); > > return ret; > > @@ -6441,19 +6435,20 @@ int smb2_write(struct ksmbd_work *work) > > > > if (req->Channel =3D=3D SMB2_CHANNEL_RDMA_V1 || > > req->Channel =3D=3D SMB2_CHANNEL_RDMA_V1_INVALIDATE) { > > - unsigned int ch_offset =3D le16_to_cpu(req->WriteChannelI= nfoOffset); > > + struct smb2_buffer_desc_v1 *descs =3D (struct smb2_buffer= _desc_v1 *) > > + ((char *)req + le16_to_cpu(req->WriteChannelInfoO= ffset)); > > > > if (req->Length !=3D 0 || req->DataOffset !=3D 0 || > > - ch_offset < offsetof(struct smb2_write_req, Buffer)) = { > > + le16_to_cpu(req->WriteChannelInfoOffset) < > > + offsetof(struct smb2_write_req, Buffer)) { > > err =3D -EINVAL; > > goto out; > > } > > - err =3D smb2_set_remote_key_for_rdma(work, > > - (struct smb2_buffer_de= sc_v1 *) > > - ((char *)req + ch_offs= et), > > - req->Channel, > > - req->WriteChannelInfoO= ffset, > > - req->WriteChannelInfoL= ength); > > + err =3D smb2_validate_rdma_buffer_descs(work, > > + descs, > > + req->Channel, > > + req->WriteChannelIn= foOffset, > > + req->WriteChannelIn= foLength); > > if (err) > > goto out; > > } > > diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c > > index e646d79554b8..1eee4be0fe32 100644 > > --- a/fs/ksmbd/transport_rdma.c > > +++ b/fs/ksmbd/transport_rdma.c > > @@ -80,9 +80,9 @@ static int smb_direct_max_fragmented_recv_size =3D 10= 24 * > > 1024; > > /* The maximum single-message size which can be received */ > > static int smb_direct_max_receive_size =3D 8192; > > > > -static int smb_direct_max_read_write_size =3D 524224; > > +static int smb_direct_max_read_write_size =3D 8 * 1024 * 1024; > > > > -static int smb_direct_max_outstanding_rw_ops =3D 8; > > +static int smb_direct_max_outstanding_rw_ops =3D 1; > I wonder why this set is decreased. Because too many RDMA resources have to be allocated for max_read_write_size * max_outstanding_rw_ops. These values mean that a ksmbd connection can handle just one request for 8MB simultaneously, but eight requests for 1MB simultaneously. So nothing has not changed. > > > > static LIST_HEAD(smb_direct_device_list); > > static DEFINE_RWLOCK(smb_direct_device_lock); > > @@ -147,10 +147,12 @@ struct smb_direct_transport { > > atomic_t send_credits; > > spinlock_t lock_new_recv_credits; > > int new_recv_credits; > > - atomic_t rw_avail_ops; > > + int max_rw_credits; > > + int pages_per_rw_credit; > > + atomic_t rw_credits; > > > > wait_queue_head_t wait_send_credits; > > - wait_queue_head_t wait_rw_avail_ops; > > + wait_queue_head_t wait_rw_credits; > > > > mempool_t *sendmsg_mempool; > > struct kmem_cache *sendmsg_cache; > > @@ -159,8 +161,6 @@ struct smb_direct_transport { > > > > wait_queue_head_t wait_send_payload_pending; > > atomic_t send_payload_pending; > > - wait_queue_head_t wait_send_pending; > > - atomic_t send_pending; > > > > struct delayed_work post_recv_credits_work; > > struct work_struct send_immediate_work; > > @@ -208,7 +208,9 @@ struct smb_direct_recvmsg { > > struct smb_direct_rdma_rw_msg { > > struct smb_direct_transport *t; > > struct ib_cqe cqe; > > + int status; > > struct completion *completion; > > + struct list_head list; > > struct rdma_rw_ctx rw_ctx; > > struct sg_table sgt; > > struct scatterlist sg_list[]; > > @@ -377,7 +379,7 @@ static struct smb_direct_transport > > *alloc_transport(struct rdma_cm_id *cm_id) > > t->reassembly_queue_length =3D 0; > > init_waitqueue_head(&t->wait_reassembly_queue); > > init_waitqueue_head(&t->wait_send_credits); > > - init_waitqueue_head(&t->wait_rw_avail_ops); > > + init_waitqueue_head(&t->wait_rw_credits); > > > > spin_lock_init(&t->receive_credit_lock); > > spin_lock_init(&t->recvmsg_queue_lock); > > @@ -388,8 +390,6 @@ static struct smb_direct_transport > > *alloc_transport(struct rdma_cm_id *cm_id) > > > > init_waitqueue_head(&t->wait_send_payload_pending); > > atomic_set(&t->send_payload_pending, 0); > > - init_waitqueue_head(&t->wait_send_pending); > > - atomic_set(&t->send_pending, 0); > > > > spin_lock_init(&t->lock_new_recv_credits); > > > > @@ -419,8 +419,6 @@ static void free_transport(struct smb_direct_transp= ort > > *t) > > ksmbd_debug(RDMA, "wait for all send posted to IB to finish\n"); > > wait_event(t->wait_send_payload_pending, > > atomic_read(&t->send_payload_pending) =3D=3D 0); > > - wait_event(t->wait_send_pending, > > - atomic_read(&t->send_pending) =3D=3D 0); > > > > cancel_work_sync(&t->disconnect_work); > > cancel_delayed_work_sync(&t->post_recv_credits_work); > > @@ -682,10 +680,8 @@ static int smb_direct_read(struct ksmbd_transport = *t, > > char *buf, > > struct smb_direct_transport *st =3D smb_trans_direct_transfort(t)= ; > > > > again: > > - if (st->status !=3D SMB_DIRECT_CS_CONNECTED) { > > - pr_err("disconnected\n"); > > + if (st->status !=3D SMB_DIRECT_CS_CONNECTED) > > return -ENOTCONN; > > - } > > > > /* > > * No need to hold the reassembly queue lock all the time as we a= re > > @@ -873,13 +869,8 @@ static void send_done(struct ib_cq *cq, struct ib_= wc > > *wc) > > smb_direct_disconnect_rdma_connection(t); > > } > > > > - if (sendmsg->num_sge > 1) { > > - if (atomic_dec_and_test(&t->send_payload_pending)) > > - wake_up(&t->wait_send_payload_pending); > > - } else { > > - if (atomic_dec_and_test(&t->send_pending)) > > - wake_up(&t->wait_send_pending); > > - } > > + if (atomic_dec_and_test(&t->send_payload_pending)) > > + wake_up(&t->wait_send_payload_pending); > > > > /* iterate and free the list of messages in reverse. the list's h= ead > > * is invalid. > > @@ -911,21 +902,12 @@ static int smb_direct_post_send(struct > > smb_direct_transport *t, > > { > > int ret; > > > > - if (wr->num_sge > 1) > > - atomic_inc(&t->send_payload_pending); > > - else > > - atomic_inc(&t->send_pending); > > - > > + atomic_inc(&t->send_payload_pending); > > ret =3D ib_post_send(t->qp, wr, NULL); > > if (ret) { > > pr_err("failed to post send: %d\n", ret); > > - if (wr->num_sge > 1) { > > - if (atomic_dec_and_test(&t->send_payload_pending)= ) > > - wake_up(&t->wait_send_payload_pending); > > - } else { > > - if (atomic_dec_and_test(&t->send_pending)) > > - wake_up(&t->wait_send_pending); > > - } > > + if (atomic_dec_and_test(&t->send_payload_pending)) > > + wake_up(&t->wait_send_payload_pending); > > smb_direct_disconnect_rdma_connection(t); > > } > > return ret; > > @@ -983,18 +965,18 @@ static int smb_direct_flush_send_list(struct > > smb_direct_transport *t, > > } > > > > static int wait_for_credits(struct smb_direct_transport *t, > > - wait_queue_head_t *waitq, atomic_t *credits) > > + wait_queue_head_t *waitq, atomic_t *total_cre= dits, > > + int needed) > > { > > int ret; > > > > do { > > - if (atomic_dec_return(credits) >=3D 0) > > + if (atomic_sub_return(needed, total_credits) >=3D 0) > > return 0; > > - > > - atomic_inc(credits); > > + atomic_add(needed, total_credits); > > ret =3D wait_event_interruptible(*waitq, > > - atomic_read(credits) > 0 |= | > > - t->status !=3D SMB_DIRECT= _CS_CONNECTED); > > + atomic_read(total_credits)= >=3D needed || > > + t->status !=3D SMB_DIRECT_= CS_CONNECTED); > > > > if (t->status !=3D SMB_DIRECT_CS_CONNECTED) > > return -ENOTCONN; > > @@ -1015,7 +997,19 @@ static int wait_for_send_credits(struct > > smb_direct_transport *t, > > return ret; > > } > > > > - return wait_for_credits(t, &t->wait_send_credits, &t->send_credit= s); > > + return wait_for_credits(t, &t->wait_send_credits, &t->send_credit= s, 1); > > +} > > + > > +static int wait_for_rw_credits(struct smb_direct_transport *t, int > > credits) > > +{ > > + return wait_for_credits(t, &t->wait_rw_credits, &t->rw_credits, c= redits); > > +} > > + > > +static int calc_rw_credits(struct smb_direct_transport *t, > > + char *buf, unsigned int len) > > +{ > > + return DIV_ROUND_UP(get_buf_page_count(buf, len), > > + t->pages_per_rw_credit); > > } > > > > static int smb_direct_create_header(struct smb_direct_transport *t, > > @@ -1248,7 +1242,8 @@ static int smb_direct_writev(struct ksmbd_transpo= rt > > *t, > > iov[0].iov_len -=3D 4; > > > > remaining_data_length =3D buflen; > > - ksmbd_debug(RDMA, "Sending smb (RDMA): smb_len=3D%u\n", buflen); > > + ksmbd_debug(RDMA, "Sending smb (RDMA): smb_len=3D%u, inv=3D%d\n", > > + buflen, need_invalidate); > > > > smb_direct_send_ctx_init(st, &send_ctx, need_invalidate, remote_k= ey); > > start =3D i =3D 0; > > @@ -1318,6 +1313,18 @@ static int smb_direct_writev(struct ksmbd_transp= ort > > *t, > > return ret; > > } > > > > +static void smb_direct_free_rdma_rw_msg(struct smb_direct_transport *t= , > > + struct smb_direct_rdma_rw_msg *ms= g, > > + enum dma_data_direction dir) > > +{ > > + if (msg->sgt.orig_nents) { > > + rdma_rw_ctx_destroy(&msg->rw_ctx, t->qp, t->qp->port, > > + msg->sgt.sgl, msg->sgt.nents, dir); > > + sg_free_table_chained(&msg->sgt, SG_CHUNK_SIZE); > > + } > > + kfree(msg); > > +} > > + > > static void read_write_done(struct ib_cq *cq, struct ib_wc *wc, > > enum dma_data_direction dir) > > { > > @@ -1326,19 +1333,14 @@ static void read_write_done(struct ib_cq *cq, s= truct > > ib_wc *wc, > > struct smb_direct_transport *t =3D msg->t; > > > > if (wc->status !=3D IB_WC_SUCCESS) { > > + msg->status =3D -EIO; > > pr_err("read/write error. opcode =3D %d, status =3D %s(%d= )\n", > > wc->opcode, ib_wc_status_msg(wc->status), wc->stat= us); > > - smb_direct_disconnect_rdma_connection(t); > > + if (wc->status !=3D IB_WC_WR_FLUSH_ERR) > > + smb_direct_disconnect_rdma_connection(t); > > } > > > > - if (atomic_inc_return(&t->rw_avail_ops) > 0) > > - wake_up(&t->wait_rw_avail_ops); > > - > > - rdma_rw_ctx_destroy(&msg->rw_ctx, t->qp, t->qp->port, > > - msg->sg_list, msg->sgt.nents, dir); > > - sg_free_table_chained(&msg->sgt, SG_CHUNK_SIZE); > > complete(msg->completion); > > - kfree(msg); > > } > > > > static void read_done(struct ib_cq *cq, struct ib_wc *wc) > > @@ -1351,94 +1353,141 @@ static void write_done(struct ib_cq *cq, struc= t > > ib_wc *wc) > > read_write_done(cq, wc, DMA_TO_DEVICE); > > } > > > > -static int smb_direct_rdma_xmit(struct smb_direct_transport *t, void *= buf, > > - int buf_len, u32 remote_key, u64 remote_o= ffset, > > - u32 remote_len, bool is_read) > > +static int smb_direct_rdma_xmit(struct smb_direct_transport *t, > > + void *buf, int buf_len, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len, > > + bool is_read) > > { > > - struct smb_direct_rdma_rw_msg *msg; > > - int ret; > > + struct smb_direct_rdma_rw_msg *msg, *next_msg; > > + int i, ret; > > DECLARE_COMPLETION_ONSTACK(completion); > > - struct ib_send_wr *first_wr =3D NULL; > > + struct ib_send_wr *first_wr; > > + LIST_HEAD(msg_list); > > + char *desc_buf; > > + int credits_needed; > > + unsigned int desc_buf_len; > > + size_t total_length =3D 0; > > + > > + if (t->status !=3D SMB_DIRECT_CS_CONNECTED) > > + return -ENOTCONN; > > + > > + /* calculate needed credits */ > > + credits_needed =3D 0; > > + desc_buf =3D buf; > > + for (i =3D 0; i < desc_len / sizeof(*desc); i++) { > > + desc_buf_len =3D le32_to_cpu(desc[i].length); > > + > > + credits_needed +=3D calc_rw_credits(t, desc_buf, desc_buf= _len); > > + desc_buf +=3D desc_buf_len; > > + total_length +=3D desc_buf_len; > > + if (desc_buf_len =3D=3D 0 || total_length > buf_len || > > + total_length > t->max_rdma_rw_size) > > + return -EINVAL; > > + } > > > > - ret =3D wait_for_credits(t, &t->wait_rw_avail_ops, &t->rw_avail_o= ps); > > + ksmbd_debug(RDMA, "RDMA %s, len %#x, needed credits %#x\n", > > + is_read ? "read" : "write", buf_len, credits_needed); > > + > > + ret =3D wait_for_rw_credits(t, credits_needed); > > if (ret < 0) > > return ret; > > > > - /* TODO: mempool */ > > - msg =3D kmalloc(offsetof(struct smb_direct_rdma_rw_msg, sg_list) = + > > - sizeof(struct scatterlist) * SG_CHUNK_SIZE, GFP_KER= NEL); > > - if (!msg) { > > - atomic_inc(&t->rw_avail_ops); > > - return -ENOMEM; > > - } > > + /* build rdma_rw_ctx for each descriptor */ > > + desc_buf =3D buf; > > + for (i =3D 0; i < desc_len / sizeof(*desc); i++) { > > + msg =3D kzalloc(offsetof(struct smb_direct_rdma_rw_msg, s= g_list) + > > + sizeof(struct scatterlist) * SG_CHUNK_SIZE,= GFP_KERNEL); > > + if (!msg) { > > + ret =3D -ENOMEM; > > + goto out; > > + } > > > > - msg->sgt.sgl =3D &msg->sg_list[0]; > > - ret =3D sg_alloc_table_chained(&msg->sgt, > > - get_buf_page_count(buf, buf_len), > > - msg->sg_list, SG_CHUNK_SIZE); > > - if (ret) { > > - atomic_inc(&t->rw_avail_ops); > > - kfree(msg); > > - return -ENOMEM; > > - } > > + desc_buf_len =3D le32_to_cpu(desc[i].length); > > > > - ret =3D get_sg_list(buf, buf_len, msg->sgt.sgl, msg->sgt.orig_nen= ts); > > - if (ret <=3D 0) { > > - pr_err("failed to get pages\n"); > > - goto err; > > - } > > + msg->t =3D t; > > + msg->cqe.done =3D is_read ? read_done : write_done; > > + msg->completion =3D &completion; > > > > - ret =3D rdma_rw_ctx_init(&msg->rw_ctx, t->qp, t->qp->port, > > - msg->sg_list, get_buf_page_count(buf, buf_= len), > > - 0, remote_offset, remote_key, > > - is_read ? DMA_FROM_DEVICE : DMA_TO_DEVICE)= ; > > - if (ret < 0) { > > - pr_err("failed to init rdma_rw_ctx: %d\n", ret); > > - goto err; > > + msg->sgt.sgl =3D &msg->sg_list[0]; > > + ret =3D sg_alloc_table_chained(&msg->sgt, > > + get_buf_page_count(desc_buf,= desc_buf_len), > > + msg->sg_list, SG_CHUNK_SIZE)= ; > > + if (ret) { > > + kfree(msg); > > + ret =3D -ENOMEM; > > + goto out; > > + } > > + > > + ret =3D get_sg_list(desc_buf, desc_buf_len, > > + msg->sgt.sgl, msg->sgt.orig_nents); > > + if (ret <=3D 0) { > > + sg_free_table_chained(&msg->sgt, SG_CHUNK_SIZE); > > + kfree(msg); > > + goto out; > > + } > > + > > + ret =3D rdma_rw_ctx_init(&msg->rw_ctx, t->qp, t->qp->port= , > > + msg->sgt.sgl, > > + get_buf_page_count(desc_buf, desc_= buf_len), > > + 0, > > + le64_to_cpu(desc[i].offset), > > + le32_to_cpu(desc[i].token), > > + is_read ? DMA_FROM_DEVICE : DMA_TO= _DEVICE); > > + if (ret < 0) { > > + pr_err("failed to init rdma_rw_ctx: %d\n", ret); > > + sg_free_table_chained(&msg->sgt, SG_CHUNK_SIZE); > > + kfree(msg); > > + goto out; > > + } > > + > > + list_add_tail(&msg->list, &msg_list); > > + desc_buf +=3D desc_buf_len; > > } > > > > - msg->t =3D t; > > - msg->cqe.done =3D is_read ? read_done : write_done; > > - msg->completion =3D &completion; > > - first_wr =3D rdma_rw_ctx_wrs(&msg->rw_ctx, t->qp, t->qp->port, > > - &msg->cqe, NULL); > > + /* concatenate work requests of rdma_rw_ctxs */ > > + first_wr =3D NULL; > > + list_for_each_entry_reverse(msg, &msg_list, list) { > > + first_wr =3D rdma_rw_ctx_wrs(&msg->rw_ctx, t->qp, t->qp->= port, > > + &msg->cqe, first_wr); > > + } > > > > ret =3D ib_post_send(t->qp, first_wr, NULL); > > if (ret) { > > - pr_err("failed to post send wr: %d\n", ret); > > - goto err; > > + pr_err("failed to post send wr for RDMA R/W: %d\n", ret); > > + goto out; > > } > > > > + msg =3D list_last_entry(&msg_list, struct smb_direct_rdma_rw_msg,= list); > > wait_for_completion(&completion); > > - return 0; > > - > > -err: > > - atomic_inc(&t->rw_avail_ops); > > - if (first_wr) > > - rdma_rw_ctx_destroy(&msg->rw_ctx, t->qp, t->qp->port, > > - msg->sg_list, msg->sgt.nents, > > - is_read ? DMA_FROM_DEVICE : DMA_TO_DE= VICE); > > - sg_free_table_chained(&msg->sgt, SG_CHUNK_SIZE); > > - kfree(msg); > > + ret =3D msg->status; > > +out: > > + list_for_each_entry_safe(msg, next_msg, &msg_list, list) { > > + list_del(&msg->list); > > + smb_direct_free_rdma_rw_msg(t, msg, > > + is_read ? DMA_FROM_DEVICE : D= MA_TO_DEVICE); > > + } > > + atomic_add(credits_needed, &t->rw_credits); > > + wake_up(&t->wait_rw_credits); > > return ret; > > } > > > > -static int smb_direct_rdma_write(struct ksmbd_transport *t, void *buf, > > - unsigned int buflen, u32 remote_key, > > - u64 remote_offset, u32 remote_len) > > +static int smb_direct_rdma_write(struct ksmbd_transport *t, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len) > > { > > return smb_direct_rdma_xmit(smb_trans_direct_transfort(t), buf, b= uflen, > > - remote_key, remote_offset, > > - remote_len, false); > > + desc, desc_len, false); > > } > > > > -static int smb_direct_rdma_read(struct ksmbd_transport *t, void *buf, > > - unsigned int buflen, u32 remote_key, > > - u64 remote_offset, u32 remote_len) > > +static int smb_direct_rdma_read(struct ksmbd_transport *t, > > + void *buf, unsigned int buflen, > > + struct smb2_buffer_desc_v1 *desc, > > + unsigned int desc_len) > > { > > return smb_direct_rdma_xmit(smb_trans_direct_transfort(t), buf, b= uflen, > > - remote_key, remote_offset, > > - remote_len, true); > > + desc, desc_len, true); > > } > > > > static void smb_direct_disconnect(struct ksmbd_transport *t) > > @@ -1567,8 +1616,8 @@ static int smb_direct_send_negotiate_response(str= uct > > smb_direct_transport *t, > > return ret; > > } > > > > - wait_event(t->wait_send_pending, > > - atomic_read(&t->send_pending) =3D=3D 0); > > + wait_event(t->wait_send_payload_pending, > > + atomic_read(&t->send_payload_pending) =3D=3D 0); > > return 0; > > } > > > > @@ -1638,11 +1687,19 @@ static int smb_direct_prepare_negotiation(struc= t > > smb_direct_transport *t) > > return ret; > > } > > > > +static unsigned int smb_direct_get_max_fr_pages(struct smb_direct_tran= sport > > *t) > > +{ > > + return min_t(unsigned int, > > + t->cm_id->device->attrs.max_fast_reg_page_list_len, > > + 256); > > +} > > + > > static int smb_direct_init_params(struct smb_direct_transport *t, > > struct ib_qp_cap *cap) > > { > > struct ib_device *device =3D t->cm_id->device; > > - int max_send_sges, max_pages, max_rw_wrs, max_send_wrs; > > + int max_send_sges, max_rw_wrs, max_send_wrs; > > + unsigned int max_sge_per_wr, wrs_per_credit; > > > > /* need 2 more sge. because a SMB_DIRECT header will be mapped, > > * and maybe a send buffer could be not page aligned. > > @@ -1654,25 +1711,31 @@ static int smb_direct_init_params(struct > > smb_direct_transport *t, > > return -EINVAL; > > } > > > > - /* > > - * allow smb_direct_max_outstanding_rw_ops of in-flight RDMA > > - * read/writes. HCA guarantees at least max_send_sge of sges for > > - * a RDMA read/write work request, and if memory registration is = used, > > - * we need reg_mr, local_inv wrs for each read/write. > > + /* Calculate the number of work requests for RDMA R/W. > > + * The maximum number of pages which can be registered > > + * with one Memory region can be transferred with one > > + * R/W credit. And at least 4 work requests for each credit > > + * are needed for MR registration, RDMA R/W, local & remote > > + * MR invalidation. > > */ > > t->max_rdma_rw_size =3D smb_direct_max_read_write_size; > > - max_pages =3D DIV_ROUND_UP(t->max_rdma_rw_size, PAGE_SIZE) + 1; > > - max_rw_wrs =3D DIV_ROUND_UP(max_pages, SMB_DIRECT_MAX_SEND_SGES); > > - max_rw_wrs +=3D rdma_rw_mr_factor(device, t->cm_id->port_num, > > - max_pages) * 2; > > - max_rw_wrs *=3D smb_direct_max_outstanding_rw_ops; > > + t->pages_per_rw_credit =3D smb_direct_get_max_fr_pages(t); > > + t->max_rw_credits =3D smb_direct_max_outstanding_rw_ops * > > + DIV_ROUND_UP(t->max_rdma_rw_size, > > + (t->pages_per_rw_credit - 1) * PAGE_SIZE); > > + > > + max_sge_per_wr =3D min_t(unsigned int, device->attrs.max_send_sge= , > > + device->attrs.max_sge_rd); > > + wrs_per_credit =3D max_t(unsigned int, 4, > > + DIV_ROUND_UP(t->pages_per_rw_credit, > > + max_sge_per_wr) + 1); > > + max_rw_wrs =3D t->max_rw_credits * wrs_per_credit; > > > > max_send_wrs =3D smb_direct_send_credit_target + max_rw_wrs; > > if (max_send_wrs > device->attrs.max_cqe || > > max_send_wrs > device->attrs.max_qp_wr) { > > - pr_err("consider lowering send_credit_target =3D %d, or > > max_outstanding_rw_ops =3D %d\n", > > - smb_direct_send_credit_target, > > - smb_direct_max_outstanding_rw_ops); > > + pr_err("consider lowering send_credit_target =3D %d\n", > > + smb_direct_send_credit_target); > > pr_err("Possible CQE overrun, device reporting max_cqe %d= max_qp_wr > > %d\n", > > device->attrs.max_cqe, device->attrs.max_qp_wr); > > return -EINVAL; > > @@ -1707,7 +1770,7 @@ static int smb_direct_init_params(struct > > smb_direct_transport *t, > > > > t->send_credit_target =3D smb_direct_send_credit_target; > > atomic_set(&t->send_credits, 0); > > - atomic_set(&t->rw_avail_ops, smb_direct_max_outstanding_rw_ops); > > + atomic_set(&t->rw_credits, t->max_rw_credits); > > > > t->max_send_size =3D smb_direct_max_send_size; > > t->max_recv_size =3D smb_direct_max_receive_size; > > @@ -1715,12 +1778,10 @@ static int smb_direct_init_params(struct > > smb_direct_transport *t, > > > > cap->max_send_wr =3D max_send_wrs; > > cap->max_recv_wr =3D t->recv_credit_max; > > - cap->max_send_sge =3D SMB_DIRECT_MAX_SEND_SGES; > > + cap->max_send_sge =3D max_sge_per_wr; > > cap->max_recv_sge =3D SMB_DIRECT_MAX_RECV_SGES; > Is there no need to set this value to a value supported by the device? > e.g. device->attrs.max_read_sge Because we don't need more than one(SMB_DIRECT_MAX_REV_SGES), I think it isn't necessary. Thank you for your comments! > > cap->max_inline_data =3D 0; > > - cap->max_rdma_ctxs =3D > > - rdma_rw_mr_factor(device, t->cm_id->port_num, max_pages) = * > > - smb_direct_max_outstanding_rw_ops; > > + cap->max_rdma_ctxs =3D t->max_rw_credits; > > return 0; > > } > > > > @@ -1813,7 +1874,8 @@ static int smb_direct_create_qpair(struct > > smb_direct_transport *t, > > } > > > > t->send_cq =3D ib_alloc_cq(t->cm_id->device, t, > > - t->send_credit_target, 0, IB_POLL_WORKQU= EUE); > > + smb_direct_send_credit_target + cap->max= _rdma_ctxs, > > + 0, IB_POLL_WORKQUEUE); > > if (IS_ERR(t->send_cq)) { > > pr_err("Can't create RDMA send CQ\n"); > > ret =3D PTR_ERR(t->send_cq); > > @@ -1822,8 +1884,7 @@ static int smb_direct_create_qpair(struct > > smb_direct_transport *t, > > } > > > > t->recv_cq =3D ib_alloc_cq(t->cm_id->device, t, > > - cap->max_send_wr + cap->max_rdma_ctxs, > > - 0, IB_POLL_WORKQUEUE); > > + t->recv_credit_max, 0, IB_POLL_WORKQUEUE= ); > > if (IS_ERR(t->recv_cq)) { > > pr_err("Can't create RDMA recv CQ\n"); > > ret =3D PTR_ERR(t->recv_cq); > > @@ -1852,17 +1913,12 @@ static int smb_direct_create_qpair(struct > > smb_direct_transport *t, > > > > pages_per_rw =3D DIV_ROUND_UP(t->max_rdma_rw_size, PAGE_SIZE) + 1= ; > > if (pages_per_rw > t->cm_id->device->attrs.max_sgl_rd) { > > - int pages_per_mr, mr_count; > > - > > - pages_per_mr =3D min_t(int, pages_per_rw, > > - t->cm_id->device->attrs.max_fast_reg= _page_list_len); > > - mr_count =3D DIV_ROUND_UP(pages_per_rw, pages_per_mr) * > > - atomic_read(&t->rw_avail_ops); > > - ret =3D ib_mr_pool_init(t->qp, &t->qp->rdma_mrs, mr_count= , > > - IB_MR_TYPE_MEM_REG, pages_per_mr, 0= ); > > + ret =3D ib_mr_pool_init(t->qp, &t->qp->rdma_mrs, > > + t->max_rw_credits, IB_MR_TYPE_MEM_R= EG, > > + t->pages_per_rw_credit, 0); > > if (ret) { > > pr_err("failed to init mr pool count %d pages %d\= n", > > - mr_count, pages_per_mr); > > + t->max_rw_credits, t->pages_per_rw_credit)= ; > > goto err; > > } > > } > > diff --git a/fs/ksmbd/transport_tcp.c b/fs/ksmbd/transport_tcp.c > > index 8fef9de787d3..4892b0d66a25 100644 > > --- a/fs/ksmbd/transport_tcp.c > > +++ b/fs/ksmbd/transport_tcp.c > > @@ -352,8 +352,9 @@ static int ksmbd_tcp_read(struct ksmbd_transport *t= , > > char *buf, unsigned int to_ > > return ksmbd_tcp_readv(TCP_TRANS(t), &iov, 1, to_read); > > } > > > > -static int ksmbd_tcp_writev(struct ksmbd_transport *t, struct kvec *io= v, > > - int nvecs, int size, bool need_invalidate, > > +static int ksmbd_tcp_writev(struct ksmbd_transport *t, > > + struct kvec *iov, int nvecs, int size, > > + bool need_invalidate, > > unsigned int remote_key) > > > > { > > > > base-commit: 3123109284176b1532874591f7c81f3837bbdc17 > > -- > > 2.25.1 > > > > --=20 Thanks, Hyunchul