search for: sess_cmd_map

...ag pre-allocation support of target fabric descriptor memory, utilizing Kent's latest in-flight per-cpu ida bits here: [PATCH v3] lib/idr.c rewrite, percpu ida/tag allocator http://marc.info/?l=linux-kernel&m=137308733526328&w=2 Patch #1 includes target-core setup of se_sess->sess_cmd_map + se_sess->sess_tag_pool resources at session creation time, using fabric independent code in transport_init_session_tags(). Patch #2 is the initial conversion of vhost-scsi fabric code to use per-cpu ida logic for obtaining a new tcm_vhost_cmd descriptor via vhost_scsi_get_tag() during vhost_w...

...ag pre-allocation support of target fabric descriptor memory, utilizing Kent's latest in-flight per-cpu ida bits here: [PATCH v3] lib/idr.c rewrite, percpu ida/tag allocator http://marc.info/?l=linux-kernel&m=137308733526328&w=2 Patch #1 includes target-core setup of se_sess->sess_cmd_map + se_sess->sess_tag_pool resources at session creation time, using fabric independent code in transport_init_session_tags(). Patch #2 is the initial conversion of vhost-scsi fabric code to use per-cpu ida logic for obtaining a new tcm_vhost_cmd descriptor via vhost_scsi_get_tag() during vhost_w...

...tandalone version into target-pending/for-next as a v3.12 item. Jens has also expressed interest in a blk-mq conversion to use these per-cpu-ida primatives, so getting this piece merged for v3.12 would make life easier for both of us. ;) The second patch includes target-core setup of se_sess->sess_cmd_map + se_sess->sess_tag_pool resources at session creation time, using fabric independent code in transport_init_session_tags(). The third patch is the initial conversion of vhost-scsi fabric code to use per-cpu ida logic for obtaining a new tcm_vhost_cmd descriptor via vhost_scsi_get_tag() during...

...tandalone version into target-pending/for-next as a v3.12 item. Jens has also expressed interest in a blk-mq conversion to use these per-cpu-ida primatives, so getting this piece merged for v3.12 would make life easier for both of us. ;) The second patch includes target-core setup of se_sess->sess_cmd_map + se_sess->sess_tag_pool resources at session creation time, using fabric independent code in transport_init_session_tags(). The third patch is the initial conversion of vhost-scsi fabric code to use per-cpu ida logic for obtaining a new tcm_vhost_cmd descriptor via vhost_scsi_get_tag() during...

...t sbp_target_request *req; - int tag; + int tag, cpu; - tag = percpu_ida_alloc(&se_sess->sess_tag_pool, TASK_RUNNING); + tag = sbitmap_queue_get(&se_sess->sess_tag_pool, &cpu); if (tag < 0) return ERR_PTR(-ENOMEM); req = &((struct sbp_target_request *)se_sess->sess_cmd_map)[tag]; memset(req, 0, sizeof(*req)); req->se_cmd.map_tag = tag; + req->se_cmd.map_cpu = cpu; req->se_cmd.tag = next_orb; return req; @@ -1460,7 +1461,8 @@ static void sbp_free_request(struct sbp_target_request *req) kfree(req->pg_tbl); kfree(req->cmd_buf); - percpu_id...

...tag = sbitmap_get(&svq->scsi_tags, 0, false); > if (tag < 0) { > pr_err("Unable to obtain tag for vhost_scsi_cmd\n"); > return ERR_PTR(-ENOMEM); > } After this change, cpu is uninitialized. > > - cmd = &((struct vhost_scsi_cmd *)se_sess->sess_cmd_map)[tag]; > + cmd = &svq->scsi_cmds[tag]; > sg = cmd->tvc_sgl; > prot_sg = cmd->tvc_prot_sgl; > pages = cmd->tvc_upages; > @@ -1065,11 +1069,11 @@ static void vhost_scsi_submission_work(struct work_struct *work) > scsi_command_size(cdb), VHOST_SCSI_MAX_CDB_...

On Tue, 2018-06-12 at 09:15 -0700, Matthew Wilcox wrote: > On Tue, Jun 12, 2018 at 03:22:42PM +0000, Bart Van Assche wrote: > > On Tue, 2018-05-15 at 09:00 -0700, Matthew Wilcox wrote: > > > diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c > > > index 025dc2d3f3de..cdf671c2af61 100644 > > > ---

Removing the percpu_ida code nets over 400 lines of removal. It's not as spectacular as deleting an entire architecture, but it's still a worthy reduction in lines of code. Untested due to lack of hardware and not understanding how to set up a target platform. Changes from v1: - Fixed bugs pointed out by Jens in iscsit_wait_for_tag() - Abstracted out tag freeing as requested by Bart

Removing the percpu_ida code nets over 400 lines of removal. It's not as spectacular as deleting an entire architecture, but it's still a worthy reduction in lines of code. Untested due to lack of hardware and not understanding how to set up a target platform. Changes from v1: - Fixed bugs pointed out by Jens in iscsit_wait_for_tag() - Abstracted out tag freeing as requested by Bart

From: Matthew Wilcox <mawilcox at microsoft.com> This is a pretty rough-and-ready conversion of the target drivers from using percpu_ida to sbitmap. It compiles; I don't have a target setup, so it's completely untested. I haven't tried to do anything particularly clever here, so it's possible that, for example, the wait queue in iscsi_target_util could be more clever, like