From patchwork Thu Feb 22 16:18:48 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: James Smart X-Patchwork-Id: 10235881 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id C00C660209 for ; Thu, 22 Feb 2018 16:19:30 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id B99E928555 for ; Thu, 22 Feb 2018 16:19:30 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id AE49B28C28; Thu, 22 Feb 2018 16:19:30 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.0 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, FREEMAIL_FROM, RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id C58B428555 for ; Thu, 22 Feb 2018 16:19:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933249AbeBVQT0 (ORCPT ); Thu, 22 Feb 2018 11:19:26 -0500 Received: from mail-qt0-f194.google.com ([209.85.216.194]:36184 "EHLO mail-qt0-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933234AbeBVQTP (ORCPT ); Thu, 22 Feb 2018 11:19:15 -0500 Received: by mail-qt0-f194.google.com with SMTP id c7so6951098qtn.3 for ; Thu, 22 Feb 2018 08:19:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=XyG/Lq1Cl7+2peJwdjyyDt8NpffgUhLqM3E2LWG95q4=; b=LnYFHw21aAw8vtBOAXDOEwgRctkQvR4XQp95pjwICd71OVUozmxzgv7gGsRKcpklKW Ol5+a1FE5M1DL6h1RV9BYtn3njhz4oMRJ4LgtcVy6QpzhJxzyhHsxlbenaADWxkx39D9 dKnqMsT294Jg3DEAeUzzEc5CiFCRYzgSXcWh5QLZprDg6c8b6xyMkL0oZyGVX1MBtv9+ dOF5kRDlSqgPTEzRF0ocihaMeu6sOgQqKHA6Yf66imRgUHYGbzYQnMmntcPQcPAZpqpx mIhf60ECgDn02Ved4xqFjfds8QEce8FT5uJuOBb6k25+CpPUPg8T4BEvz5IOzxGbiQ9g LKAA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=XyG/Lq1Cl7+2peJwdjyyDt8NpffgUhLqM3E2LWG95q4=; b=TRE3svU/Z/s5S5X+X0+BX7PzBaElQdh3I38XKfQsL3NnUE/Zot0fBYPnWkwFsfrA23 Fpba8aIkLO6rzCxjVE744EzB8vud8z7USdPauLCrlkpa+cBI41anjGF3B9AzmPKejbQQ lTf7+OcMzEqhHOqR8LQ8jHldOQRk5lfYS2ZH2QdfJ3jkUNwLfWAGB8XCFRDvttk/kDDl VtXOM9G2jhsaXgP+Ttiv0IAchC7NGFN8hEEUVtlfGyiKaMadeZMe1xhXtRjYZtx9bddn MUgn4mfMqMwgc6Bgn4otRGjAorA8z6p36m424zepD/91djVqbmzXEDVY36RpMoWM3O8N 1DvQ== X-Gm-Message-State: APf1xPBmV00f4kOVgql213Yq9c7/22v5DxpBKlg6YKoQ6oOI+lFRdUuJ +8ssGq3K7Uew9ZIAUzlMm2vHqw== X-Google-Smtp-Source: AH8x225Jw9LfuWZ2Q4eqbkY0Fo6fgbWHCzRDcXnFzV91TYb+InAZaTIhC+mPwPIj4GKtZEW31zOjkw== X-Received: by 10.200.40.129 with SMTP id i1mr11339496qti.176.1519316354511; Thu, 22 Feb 2018 08:19:14 -0800 (PST) Received: from pallmd1.broadcom.com ([192.19.223.250]) by smtp.gmail.com with ESMTPSA id n29sm347007qtf.18.2018.02.22.08.19.13 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Thu, 22 Feb 2018 08:19:14 -0800 (PST) From: James Smart To: linux-scsi@vger.kernel.org Cc: James Smart , Dick Kennedy , James Smart Subject: [PATCH v5 09/13] lpfc: Add embedded data pointers for enhanced performance Date: Thu, 22 Feb 2018 08:18:48 -0800 Message-Id: <20180222161852.15270-10-jsmart2021@gmail.com> X-Mailer: git-send-email 2.13.1 In-Reply-To: <20180222161852.15270-1-jsmart2021@gmail.com> References: <20180222161852.15270-1-jsmart2021@gmail.com> Sender: linux-scsi-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-scsi@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP The current driver isn't taking advantage of a performance hint whereby the initial data buffer descriptor can be placed in the WQE as well as the SGL. Add the logic to detect support for the feature and to use it when supported. Signed-off-by: Dick Kennedy Signed-off-by: James Smart Reviewed-by: Johannes Thumshirn --- v3: remove unneeded parens combined 6422 msg on 1 line --- drivers/scsi/lpfc/lpfc.h | 2 ++ drivers/scsi/lpfc/lpfc_hw4.h | 3 +++ drivers/scsi/lpfc/lpfc_init.c | 20 ++++++++++++++++++++ drivers/scsi/lpfc/lpfc_nvme.c | 18 ++++++++++++++++++ drivers/scsi/lpfc/lpfc_nvmet.c | 24 ++++++++++++++++++++++++ drivers/scsi/lpfc/lpfc_scsi.c | 8 ++++++-- drivers/scsi/lpfc/lpfc_sli.c | 25 +++++++++++++++++++++---- 7 files changed, 94 insertions(+), 6 deletions(-) diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h index 7aad4a717f13..9136a59b1c5b 100644 --- a/drivers/scsi/lpfc/lpfc.h +++ b/drivers/scsi/lpfc/lpfc.h @@ -840,6 +840,8 @@ struct lpfc_hba { #define LPFC_ENABLE_FCP 1 #define LPFC_ENABLE_NVME 2 #define LPFC_ENABLE_BOTH 3 + uint32_t nvme_embed_pbde; + uint32_t fcp_embed_pbde; uint32_t io_channel_irqs; /* number of irqs for io channels */ struct nvmet_fc_target_port *targetport; lpfc_vpd_t vpd; /* vital product data */ diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h index be8227dfa086..ed5e870c58c3 100644 --- a/drivers/scsi/lpfc/lpfc_hw4.h +++ b/drivers/scsi/lpfc/lpfc_hw4.h @@ -4226,6 +4226,9 @@ struct wqe_common { #define wqe_irsp_SHIFT 4 #define wqe_irsp_MASK 0x00000001 #define wqe_irsp_WORD word11 +#define wqe_pbde_SHIFT 5 +#define wqe_pbde_MASK 0x00000001 +#define wqe_pbde_WORD word11 #define wqe_sup_SHIFT 6 #define wqe_sup_MASK 0x00000001 #define wqe_sup_WORD word11 diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index af92c3c681f7..ecb42cae71f2 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -10608,6 +10608,19 @@ lpfc_get_sli4_parameters(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq) phba->cfg_enable_fc4_type = LPFC_ENABLE_FCP; } + /* Only embed PBDE for if_type 6 */ + if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) == + LPFC_SLI_INTF_IF_TYPE_6) { + phba->fcp_embed_pbde = 1; + phba->nvme_embed_pbde = 1; + } + + /* PBDE support requires xib be set */ + if (!bf_get(cfg_xib, mbx_sli4_parameters)) { + phba->fcp_embed_pbde = 0; + phba->nvme_embed_pbde = 0; + } + /* * To support Suppress Response feature we must satisfy 3 conditions. * lpfc_suppress_rsp module parameter must be set (default). @@ -10639,6 +10652,13 @@ lpfc_get_sli4_parameters(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq) else phba->fcp_embed_io = 0; + lpfc_printf_log(phba, KERN_INFO, LOG_INIT | LOG_NVME, + "6422 XIB %d: FCP %d %d NVME %d %d %d\n", + bf_get(cfg_xib, mbx_sli4_parameters), + phba->fcp_embed_pbde, phba->fcp_embed_io, + phba->nvme_support, phba->nvme_embed_pbde, + phba->cfg_suppress_rsp); + if ((bf_get(cfg_cqpsize, mbx_sli4_parameters) & LPFC_CQ_16K_PAGE_SZ) && (bf_get(cfg_wqpsize, mbx_sli4_parameters) & LPFC_WQ_16K_PAGE_SZ) && (sli4_params->wqsize & LPFC_WQ_SZ128_SUPPORT)) diff --git a/drivers/scsi/lpfc/lpfc_nvme.c b/drivers/scsi/lpfc/lpfc_nvme.c index 3a103d0895a2..5a1a6e24a27f 100644 --- a/drivers/scsi/lpfc/lpfc_nvme.c +++ b/drivers/scsi/lpfc/lpfc_nvme.c @@ -1170,6 +1170,7 @@ lpfc_nvme_prep_io_dma(struct lpfc_vport *vport, struct sli4_sge *sgl = lpfc_ncmd->nvme_sgl; struct scatterlist *data_sg; struct sli4_sge *first_data_sgl; + struct ulp_bde64 *bde; dma_addr_t physaddr; uint32_t num_bde = 0; uint32_t dma_len; @@ -1237,7 +1238,24 @@ lpfc_nvme_prep_io_dma(struct lpfc_vport *vport, data_sg = sg_next(data_sg); sgl++; } + if (phba->nvme_embed_pbde) { + /* Use PBDE support for first SGL only, offset == 0 */ + /* Words 13-15 */ + bde = (struct ulp_bde64 *) + &wqe->words[13]; + bde->addrLow = first_data_sgl->addr_lo; + bde->addrHigh = first_data_sgl->addr_hi; + bde->tus.f.bdeSize = + le32_to_cpu(first_data_sgl->sge_len); + bde->tus.f.bdeFlags = BUFF_TYPE_BDE_64; + bde->tus.w = cpu_to_le32(bde->tus.w); + bf_set(wqe_pbde, &wqe->generic.wqe_com, 1); + } else + bf_set(wqe_pbde, &wqe->generic.wqe_com, 0); + } else { + bf_set(wqe_pbde, &wqe->generic.wqe_com, 0); + /* For this clause to be valid, the payload_length * and sg_cnt must zero. */ diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpfc/lpfc_nvmet.c index a332a6638b1b..a4a32d7ec248 100644 --- a/drivers/scsi/lpfc/lpfc_nvmet.c +++ b/drivers/scsi/lpfc/lpfc_nvmet.c @@ -2150,9 +2150,11 @@ lpfc_nvmet_prep_fcp_wqe(struct lpfc_hba *phba, struct lpfc_iocbq *nvmewqe; struct scatterlist *sgel; union lpfc_wqe128 *wqe; + struct ulp_bde64 *bde; uint32_t *txrdy; dma_addr_t physaddr; int i, cnt; + int do_pbde; int xc = 1; if (!lpfc_is_link_up(phba)) { @@ -2243,6 +2245,7 @@ lpfc_nvmet_prep_fcp_wqe(struct lpfc_hba *phba, /* Word 7 */ bf_set(wqe_pu, &wqe->fcp_tsend.wqe_com, 1); bf_set(wqe_cmnd, &wqe->fcp_tsend.wqe_com, CMD_FCP_TSEND64_WQE); + do_pbde = 0; /* Word 8 */ wqe->fcp_tsend.wqe_com.abort_tag = nvmewqe->iotag; @@ -2355,6 +2358,10 @@ lpfc_nvmet_prep_fcp_wqe(struct lpfc_hba *phba, bf_set(wqe_ar, &wqe->fcp_treceive.wqe_com, 0); bf_set(wqe_cmnd, &wqe->fcp_treceive.wqe_com, CMD_FCP_TRECEIVE64_WQE); + if (phba->nvme_embed_pbde) + do_pbde = 1; + else + do_pbde = 0; /* Word 8 */ wqe->fcp_treceive.wqe_com.abort_tag = nvmewqe->iotag; @@ -2438,6 +2445,7 @@ lpfc_nvmet_prep_fcp_wqe(struct lpfc_hba *phba, bf_set(wqe_pu, &wqe->fcp_trsp.wqe_com, 0); bf_set(wqe_ag, &wqe->fcp_trsp.wqe_com, 1); bf_set(wqe_cmnd, &wqe->fcp_trsp.wqe_com, CMD_FCP_TRSP64_WQE); + do_pbde = 0; /* Word 8 */ wqe->fcp_trsp.wqe_com.abort_tag = nvmewqe->iotag; @@ -2508,9 +2516,25 @@ lpfc_nvmet_prep_fcp_wqe(struct lpfc_hba *phba, bf_set(lpfc_sli4_sge_last, sgl, 1); sgl->word2 = cpu_to_le32(sgl->word2); sgl->sge_len = cpu_to_le32(cnt); + if (do_pbde && i == 0) { + bde = (struct ulp_bde64 *)&wqe->words[13]; + memset(bde, 0, sizeof(struct ulp_bde64)); + /* Words 13-15 (PBDE)*/ + bde->addrLow = sgl->addr_lo; + bde->addrHigh = sgl->addr_hi; + bde->tus.f.bdeSize = + le32_to_cpu(sgl->sge_len); + bde->tus.f.bdeFlags = BUFF_TYPE_BDE_64; + bde->tus.w = cpu_to_le32(bde->tus.w); + } sgl++; ctxp->offset += cnt; } + + if (do_pbde) + bf_set(wqe_pbde, &wqe->generic.wqe_com, 1); + else + bf_set(wqe_pbde, &wqe->generic.wqe_com, 0); ctxp->state = LPFC_NVMET_STE_DATA; ctxp->entry_cnt++; return nvmewqe; diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c index c595046a521b..fb81e8a8fb1c 100644 --- a/drivers/scsi/lpfc/lpfc_scsi.c +++ b/drivers/scsi/lpfc/lpfc_scsi.c @@ -3304,8 +3304,12 @@ lpfc_scsi_prep_dma_buf_s4(struct lpfc_hba *phba, struct lpfc_scsi_buf *lpfc_cmd) dma_offset += dma_len; sgl++; } - /* setup the performance hint (first data BDE) if enabled */ - if (phba->sli3_options & LPFC_SLI4_PERFH_ENABLED) { + /* + * Setup the first Payload BDE. For FCoE we just key off + * Performance Hints, for FC we utilize fcp_embed_pbde. + */ + if ((phba->sli3_options & LPFC_SLI4_PERFH_ENABLED) || + phba->fcp_embed_pbde) { bde = (struct ulp_bde64 *) &(iocb_cmd->unsli3.sli3Words[5]); bde->addrLow = first_data_sgl->addr_lo; diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 8fc11baa88a9..416ba8f8e295 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -6958,10 +6958,15 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba) "0378 No support for fcpi mode.\n"); ftr_rsp++; } - if (bf_get(lpfc_mbx_rq_ftr_rsp_perfh, &mqe->un.req_ftrs)) - phba->sli3_options |= LPFC_SLI4_PERFH_ENABLED; - else - phba->sli3_options &= ~LPFC_SLI4_PERFH_ENABLED; + + /* Performance Hints are ONLY for FCoE */ + if (phba->hba_flag & HBA_FCOE_MODE) { + if (bf_get(lpfc_mbx_rq_ftr_rsp_perfh, &mqe->un.req_ftrs)) + phba->sli3_options |= LPFC_SLI4_PERFH_ENABLED; + else + phba->sli3_options &= ~LPFC_SLI4_PERFH_ENABLED; + } + /* * If the port cannot support the host's requested features * then turn off the global config parameters to disable the @@ -9063,6 +9068,12 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq, } /* Note, word 10 is already initialized to 0 */ + /* Don't set PBDE for Perf hints, just fcp_embed_pbde */ + if (phba->fcp_embed_pbde) + bf_set(wqe_pbde, &wqe->fcp_iwrite.wqe_com, 1); + else + bf_set(wqe_pbde, &wqe->fcp_iwrite.wqe_com, 0); + if (phba->fcp_embed_io) { struct lpfc_scsi_buf *lpfc_cmd; struct sli4_sge *sgl; @@ -9122,6 +9133,12 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq, } /* Note, word 10 is already initialized to 0 */ + /* Don't set PBDE for Perf hints, just fcp_embed_pbde */ + if (phba->fcp_embed_pbde) + bf_set(wqe_pbde, &wqe->fcp_iread.wqe_com, 1); + else + bf_set(wqe_pbde, &wqe->fcp_iread.wqe_com, 0); + if (phba->fcp_embed_io) { struct lpfc_scsi_buf *lpfc_cmd; struct sli4_sge *sgl;