From patchwork Wed Nov 19 10:17:19 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alexey Ishchuk X-Patchwork-Id: 5335751 Return-Path: X-Original-To: patchwork-linux-rdma@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork1.web.kernel.org (Postfix) with ESMTP id BCCE59F3ED for ; Wed, 19 Nov 2014 10:18:31 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id B633720176 for ; Wed, 19 Nov 2014 10:18:30 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 824AB20108 for ; Wed, 19 Nov 2014 10:18:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932078AbaKSKSY (ORCPT ); Wed, 19 Nov 2014 05:18:24 -0500 Received: from e06smtp14.uk.ibm.com ([195.75.94.110]:41095 "EHLO e06smtp14.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755366AbaKSKSU (ORCPT ); Wed, 19 Nov 2014 05:18:20 -0500 Received: from /spool/local by e06smtp14.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 19 Nov 2014 10:18:18 -0000 Received: from d06dlp02.portsmouth.uk.ibm.com (9.149.20.14) by e06smtp14.uk.ibm.com (192.168.101.144) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; Wed, 19 Nov 2014 10:18:17 -0000 Received: from b06cxnps3075.portsmouth.uk.ibm.com (d06relay10.portsmouth.uk.ibm.com [9.149.109.195]) by d06dlp02.portsmouth.uk.ibm.com (Postfix) with ESMTP id 220C6219005E; Wed, 19 Nov 2014 10:17:50 +0000 (GMT) Received: from d06av09.portsmouth.uk.ibm.com (d06av09.portsmouth.uk.ibm.com [9.149.37.250]) by b06cxnps3075.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id sAJAIGx719595600; Wed, 19 Nov 2014 10:18:16 GMT Received: from d06av09.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av09.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id sAJAIFAv031965; Wed, 19 Nov 2014 03:18:16 -0700 Received: from tuxmaker.boeblingen.de.ibm.com (tuxmaker.boeblingen.de.ibm.com [9.152.85.9]) by d06av09.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVin) with ESMTP id sAJAHhIa030388; Wed, 19 Nov 2014 03:18:15 -0700 From: Alexey Ishchuk To: linux-rdma@vger.kernel.org Cc: arlin.r.davis@intel.com, gilr@dev.mellanox.co.il, roland@kernel.org, yishaih@mellanox.com, linux-s390@vger.kernel.org, gmuelas@de.ibm.com, utz.bacher@de.ibm.com, schwidefsky@de.ibm.com, blaschka@linux.vnet.ibm.com, Alexey Ishchuk Subject: [PATCH 2/2] libmlx4: add support for the s390x platform Date: Wed, 19 Nov 2014 11:17:19 +0100 Message-Id: <1416392239-10282-3-git-send-email-aishchuk@linux.vnet.ibm.com> X-Mailer: git-send-email 1.8.5.5 In-Reply-To: <1416392239-10282-1-git-send-email-aishchuk@linux.vnet.ibm.com> References: <1416392239-10282-1-git-send-email-aishchuk@linux.vnet.ibm.com> X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 14111910-0017-0000-0000-000001E51585 Sender: linux-rdma-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org X-Spam-Status: No, score=-6.9 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, T_RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Since, s390x platform requires execution of privileged CPU instructions to work with PCI I/O memory, the PCI I/O memory cannot be directly accessed from the userspace programs via the mapped memory areas. The current implementation of the Inifiniband verbs uses mapped memory areas to write data to device UAR and Blueflame page to initiate the I/O operations, these verbs currently cannot be used on the s390x platfrom without modification. This patch contains the changes to the libmlx4 userspace Mellanox device driver library required to provide support for the DAPL API on the s390x platform. The original code that directly used mapped memory areas to access the PCI I/O memory of the Mellanox networking device is replaced with the new system call invocation for writing the data to mapped memory areas. Signed-off-by: Alexey Ishchuk --- Makefile.am | 2 src/doorbell.h | 8 ++- src/mlx4.h | 2 src/mmio.h | 115 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/qp.c | 17 -------- 5 files changed, 125 insertions(+), 19 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html --- a/Makefile.am +++ b/Makefile.am @@ -12,7 +12,7 @@ src_libmlx4_la_LDFLAGS = -avoid-version mlx4confdir = $(sysconfdir)/libibverbs.d mlx4conf_DATA = mlx4.driver -EXTRA_DIST = src/doorbell.h src/mlx4.h src/mlx4-abi.h src/wqe.h \ +EXTRA_DIST = src/doorbell.h src/mlx4.h src/mlx4-abi.h src/wqe.h src/mmio.h \ src/mlx4.map libmlx4.spec.in mlx4.driver dist-hook: libmlx4.spec --- a/src/doorbell.h +++ b/src/doorbell.h @@ -33,6 +33,8 @@ #ifndef DOORBELL_H #define DOORBELL_H +#include "mmio.h" + #if SIZEOF_LONG == 8 #if __BYTE_ORDER == __LITTLE_ENDIAN @@ -45,7 +47,7 @@ static inline void mlx4_write64(uint32_t val[2], struct mlx4_context *ctx, int offset) { - *(volatile uint64_t *) (ctx->uar + offset) = MLX4_PAIR_TO_64(val); + mmio_writeq((unsigned long)(ctx->uar + offset), MLX4_PAIR_TO_64(val)); } #else @@ -53,8 +55,8 @@ static inline void mlx4_write64(uint32_t static inline void mlx4_write64(uint32_t val[2], struct mlx4_context *ctx, int offset) { pthread_spin_lock(&ctx->uar_lock); - *(volatile uint32_t *) (ctx->uar + offset) = val[0]; - *(volatile uint32_t *) (ctx->uar + offset + 4) = val[1]; + mmio_writel((unsigned long)(ctx->uar + offset), val[0]); + mmio_writel((unsigned long)(ctx->uar + offset + 4), val[1]); pthread_spin_unlock(&ctx->uar_lock); } --- a/src/mlx4.h +++ b/src/mlx4.h @@ -74,6 +74,8 @@ #define wc_wmb() asm volatile("sfence" ::: "memory") #elif defined(__ia64__) #define wc_wmb() asm volatile("fwb" ::: "memory") +#elif defined(__s390x__) +#define wc_wmb { asm volatile("" : : : "memory") } #else #define wc_wmb() wmb() #endif --- /dev/null +++ b/src/mmio.h @@ -0,0 +1,115 @@ +#ifndef MMIO_H +#define MMIO_H + +#include +#include +#include +#ifdef __s390x__ + +static inline long mmio_writeb(const unsigned long mmio_addr, + const uint8_t val) +{ + return syscall(__NR_s390_pci_mmio_write, mmio_addr, &val, sizeof(val)); +} + +static inline long mmio_writew(const unsigned long mmio_addr, + const uint16_t val) +{ + return syscall(__NR_s390_pci_mmio_write, mmio_addr, &val, sizeof(val)); +} + +static inline long mmio_writel(const unsigned long mmio_addr, + const uint32_t val) +{ + return syscall(__NR_s390_pci_mmio_write, mmio_addr, &val, sizeof(val)); +} + +static inline long mmio_writeq(const unsigned long mmio_addr, + const uint64_t val) +{ + return syscall(__NR_s390_pci_mmio_write, mmio_addr, &val, sizeof(val)); +} + +static inline long mmio_write(const unsigned long mmio_addr, + const void *val, + const size_t length) +{ + return syscall(__NR_s390_pci_mmio_write, mmio_addr, val, length); +} + +static inline long mmio_readb(const unsigned long mmio_addr, uint8_t *val) +{ + return syscall(__NR_s390_pci_mmio_read, mmio_addr, val, sizeof(*val)); +} + +static inline long mmio_readw(const unsigned long mmio_addr, uint16_t *val) +{ + return syscall(__NR_s390_pci_mmio_read, mmio_addr, val, sizeof(*val)); +} + +static inline long mmio_readl(const unsigned long mmio_addr, uint32_t *val) +{ + return syscall(__NR_s390_pci_mmio_read, mmio_addr, val, sizeof(*val)); +} + +static inline long mmio_readq(const unsigned long mmio_addr, uint64_t *val) +{ + return syscall(__NR_s390_pci_mmio_read, mmio_addr, val, sizeof(*val)); +} + +static inline long mmio_read(const unsigned long mmio_addr, + void *val, + const size_t length) +{ + return syscall(__NR_s390_pci_mmio_read, mmio_addr, val, length); +} + +static inline void mlx4_bf_copy(unsigned long *dst, + unsigned long *src, + unsigned bytecnt) +{ + mmio_write((unsigned long)dst, src, bytecnt); +} + +#else + +#define mmio_writeb(addr, value) \ + (*((volatile uint8_t *)addr) = value) +#define mmio_writew(addr, value) \ + (*((volatile uint16_t *)addr) = value) +#define mmio_writel(addr, value) \ + (*((volatile uint32_t *)addr) = value) +#define mmio_writeq(addr, value) \ + (*((volatile uint64_t *)addr) = value) +#define mmio_write(addr, value, length) \ + memcpy(addr, value, length) + +#define mmio_readb(addr, value) \ + (value = *((volatile uint8_t *)addr)) +#define mmio_readw(addr, value) \ + (value = *((volatile uint16_t *)addr)) +#define mmio_readl(addr, value) \ + (value = *((volatile uint32_t *)addr)) +#define mmio_readq(addr, value) \ + (value = *((volatile uint64_t *)addr)) +#define mmio_read(addr, value, length) \ + memcpy(value, addr, length) + +/* + * Avoid using memcpy() to copy to BlueFlame page, since memcpy() + * implementations may use move-string-buffer assembler instructions, + * which do not guarantee order of copying. + */ +static inline void mlx4_bf_copy(unsigned long *dst, + unsigned long *src, + unsigned bytecnt) +{ + while (bytecnt > 0) { + *dst++ = *src++; + *dst++ = *src++; + bytecnt -= 2 * sizeof(long); + } +} +#endif + +#endif --- a/src/qp.c +++ b/src/qp.c @@ -173,20 +173,6 @@ static void set_data_seg(struct mlx4_wqe dseg->byte_count = htonl(sg->length); } -/* - * Avoid using memcpy() to copy to BlueFlame page, since memcpy() - * implementations may use move-string-buffer assembler instructions, - * which do not guarantee order of copying. - */ -static void mlx4_bf_copy(unsigned long *dst, unsigned long *src, unsigned bytecnt) -{ - while (bytecnt > 0) { - *dst++ = *src++; - *dst++ = *src++; - bytecnt -= 2 * sizeof (long); - } -} - int mlx4_post_send(struct ibv_qp *ibqp, struct ibv_send_wr *wr, struct ibv_send_wr **bad_wr) { @@ -434,7 +420,8 @@ out: */ wmb(); - *(uint32_t *) (ctx->uar + MLX4_SEND_DOORBELL) = qp->doorbell_qpn; + mmio_writel((unsigned long)(ctx->uar + MLX4_SEND_DOORBELL), + qp->doorbell_qpn); } if (nreq)