From patchwork Wed Feb 23 05:09:29 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 12756351 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 AE3EBC4332F for ; Wed, 23 Feb 2022 05:10:49 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S238155AbiBWFLO (ORCPT ); Wed, 23 Feb 2022 00:11:14 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43410 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S238117AbiBWFKu (ORCPT ); Wed, 23 Feb 2022 00:10:50 -0500 Received: from ams.source.kernel.org (ams.source.kernel.org [IPv6:2604:1380:4601:e00::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BEBCC6929B; Tue, 22 Feb 2022 21:10:20 -0800 (PST) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ams.source.kernel.org (Postfix) with ESMTPS id 3D48EB81E81; Wed, 23 Feb 2022 05:10:20 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 72EA9C340F3; Wed, 23 Feb 2022 05:10:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1645593018; bh=0Gg7pQShoekXL9aMv8rYa3c3b7cFK7gk0kHmhFu+mIw=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=HpFMq4mwVita/wOJzifp8lZyPAdJ+tlInVmIM8Cptb5k1lb1s9sqOwhuKT9LkGwg5 uJ/uyQC0TSwp43cke1MjC1+AOaMAEkR0C+y1Ge67XBOUTyMvgF9Se0JdxfbJ8tDnRK eMkq2VLFBZaG99z6sThjE5S7DttPVUf3Ptw9uhDI5rv9Kvo4dyppQR2f3CKpXRHxwB vW4asa37HaMKAfYthsbtFBaCbzhvINitXJcyYLQyzrE64H3AyTmHHBvYkj/jxK5Nia ZON47exBO51gpUeEektnlrHpZLBX7iTedg5PqG/oluVUe/ri5xzBlKJaXZijqEdnTz jsgQGqeWmHUIQ== From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski , Jason Gunthorpe Cc: Leon Romanovsky , linux-rdma@vger.kernel.org, netdev@vger.kernel.org, Saeed Mahameed , Saeed Mahameed Subject: [mlx5-next 14/17] net/mlx5: cmdif, Refactor error handling and reporting of async commands Date: Tue, 22 Feb 2022 21:09:29 -0800 Message-Id: <20220223050932.244668-15-saeed@kernel.org> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220223050932.244668-1-saeed@kernel.org> References: <20220223050932.244668-1-saeed@kernel.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: Saeed Mahameed Same as the new mlx5_cmd_do API, report all information to callers and let them handle the error values and outbox parsing. The user callback status "work->user_callback(status)" is now similar to the error rc code returned from the blocking mlx5_cmd_do() version, and now is defined as follows: -EREMOTEIO : Command executed by FW, outbox.status != MLX5_CMD_STAT_OK. Caller must check FW outbox status. 0 : Command execution successful, outbox.status == MLX5_CMD_STAT_OK. < 0 : Command couldn't execute, FW or driver induced error. Signed-off-by: Saeed Mahameed Signed-off-by: Saeed Mahameed --- drivers/infiniband/hw/mlx5/mr.c | 15 ++++- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 61 +++++++++++-------- include/linux/mlx5/driver.h | 3 +- 3 files changed, 52 insertions(+), 27 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 157d862fb864..06e4b8cea6bd 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -140,6 +140,19 @@ static int destroy_mkey(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) return mlx5_core_destroy_mkey(dev->mdev, mr->mmkey.key); } +static void create_mkey_warn(struct mlx5_ib_dev *dev, int status, void *out) +{ + if (status == -ENXIO) /* core driver is not available */ + return; + + mlx5_ib_warn(dev, "async reg mr failed. status %d\n", status); + if (status != -EREMOTEIO) /* driver specific failure */ + return; + + /* Failed in FW, print cmd out failure details */ + mlx5_cmd_out_err(dev->mdev, MLX5_CMD_OP_CREATE_MKEY, 0, out); +} + static void create_mkey_callback(int status, struct mlx5_async_work *context) { struct mlx5_ib_mr *mr = @@ -149,7 +162,7 @@ static void create_mkey_callback(int status, struct mlx5_async_work *context) unsigned long flags; if (status) { - mlx5_ib_warn(dev, "async reg mr failed. status %d\n", status); + create_mkey_warn(dev, status, mr->out); kfree(mr); spin_lock_irqsave(&ent->lock, flags); ent->pending--; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index a2f87a686a18..823d5808d5a0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -760,6 +760,18 @@ struct mlx5_ifc_mbox_in_bits { u8 reserved_at_40[0x40]; }; +void mlx5_cmd_out_err(struct mlx5_core_dev *dev, u16 opcode, u16 op_mod, void *out) +{ + u32 syndrome = MLX5_GET(mbox_out, out, syndrome); + u8 status = MLX5_GET(mbox_out, out, status); + + mlx5_core_err_rl(dev, + "%s(0x%x) op_mod(0x%x) failed, status %s(0x%x), syndrome (0x%x), err(%d)\n", + mlx5_command_str(opcode), opcode, op_mod, + cmd_status_str(status), status, syndrome, cmd_status_to_err(status)); +} +EXPORT_SYMBOL(mlx5_cmd_out_err); + static void cmd_status_print(struct mlx5_core_dev *dev, void *in, void *out) { u16 opcode, op_mod; @@ -778,10 +790,7 @@ static void cmd_status_print(struct mlx5_core_dev *dev, void *in, void *out) err = cmd_status_to_err(status); if (!uid && opcode != MLX5_CMD_OP_DESTROY_MKEY) - mlx5_core_err_rl(dev, - "%s(0x%x) op_mod(0x%x) failed, status %s(0x%x), syndrome (0x%x), err(%d)\n", - mlx5_command_str(opcode), opcode, op_mod, - cmd_status_str(status), status, syndrome, err); + mlx5_cmd_out_err(dev, opcode, op_mod, out); else mlx5_core_dbg(dev, "%s(0x%x) op_mod(0x%x) uid(%d) failed, status %s(0x%x), syndrome (0x%x), err(%d)\n", @@ -1686,20 +1695,18 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force callback = ent->callback; context = ent->context; - err = ent->ret; - if (!err && !ent->status) { + err = ent->ret ? : ent->status; + if (err > 0) /* Failed in FW, command didn't execute */ + err = deliv_status_to_err(err); + + if (!err) err = mlx5_copy_from_msg(ent->uout, ent->out, ent->uout_size); - err = mlx5_cmd_check(dev, err, ent->in->first.data, - ent->uout); - } - mlx5_free_cmd_msg(dev, ent->out); free_msg(dev, ent->in); - err = err ? err : ent->status; /* final consumer is done, release ent */ cmd_ent_put(ent); callback(err, context); @@ -1870,6 +1877,18 @@ static int cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out, return err; } +/* preserve -EREMOTEIO for outbox.status != OK, otherwise return err as is */ +static int cmd_status_err(int err, void *out) +{ + if (err) /* -EREMOTEIO is preserved */ + return err == -EREMOTEIO ? -EIO : err; + + if (MLX5_GET(mbox_out, out, status) != MLX5_CMD_STAT_OK) + return -EREMOTEIO; + + return 0; +} + /** * mlx5_cmd_do - Executes a fw command, wait for completion. * Unlike mlx5_cmd_exec, this function will not translate or intercept @@ -1892,13 +1911,7 @@ int mlx5_cmd_do(struct mlx5_core_dev *dev, void *in, int in_size, void *out, int { int err = cmd_exec(dev, in, in_size, out, out_size, NULL, NULL, false); - if (err) /* -EREMOTEIO is preserved */ - return err == -EREMOTEIO ? -EIO : err; - - if (MLX5_GET(mbox_out, out, status) != MLX5_CMD_STAT_OK) - return -EREMOTEIO; - - return 0; + return cmd_status_err(err, out); } EXPORT_SYMBOL(mlx5_cmd_do); @@ -1942,12 +1955,7 @@ int mlx5_cmd_exec_polling(struct mlx5_core_dev *dev, void *in, int in_size, { int err = cmd_exec(dev, in, in_size, out, out_size, NULL, NULL, true); - if (err) /* -EREMOTEIO is preserved */ - return err == -EREMOTEIO ? -EIO : err; - - if (MLX5_GET(mbox_out, out, status) != MLX5_CMD_STAT_OK) - err = -EREMOTEIO; - + err = cmd_status_err(err, out); return mlx5_cmd_check(dev, err, in, out); } EXPORT_SYMBOL(mlx5_cmd_exec_polling); @@ -1980,8 +1988,10 @@ EXPORT_SYMBOL(mlx5_cmd_cleanup_async_ctx); static void mlx5_cmd_exec_cb_handler(int status, void *_work) { struct mlx5_async_work *work = _work; - struct mlx5_async_ctx *ctx = work->ctx; + struct mlx5_async_ctx *ctx; + ctx = work->ctx; + status = cmd_status_err(status, work->out); work->user_callback(status, work); if (atomic_dec_and_test(&ctx->num_inflight)) wake_up(&ctx->wait); @@ -1995,6 +2005,7 @@ int mlx5_cmd_exec_cb(struct mlx5_async_ctx *ctx, void *in, int in_size, work->ctx = ctx; work->user_callback = callback; + work->out = out; if (WARN_ON(!atomic_inc_not_zero(&ctx->num_inflight))) return -EIO; ret = cmd_exec(ctx->dev, in, in_size, out, out_size, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 1b9bec8fa870..432151d7d0bf 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -955,6 +955,7 @@ typedef void (*mlx5_async_cbk_t)(int status, struct mlx5_async_work *context); struct mlx5_async_work { struct mlx5_async_ctx *ctx; mlx5_async_cbk_t user_callback; + void *out; /* pointer to the cmd output buffer */ }; void mlx5_cmd_init_async_ctx(struct mlx5_core_dev *dev, @@ -963,7 +964,7 @@ void mlx5_cmd_cleanup_async_ctx(struct mlx5_async_ctx *ctx); int mlx5_cmd_exec_cb(struct mlx5_async_ctx *ctx, void *in, int in_size, void *out, int out_size, mlx5_async_cbk_t callback, struct mlx5_async_work *work); - +void mlx5_cmd_out_err(struct mlx5_core_dev *dev, u16 opcode, u16 op_mod, void *out); int mlx5_cmd_do(struct mlx5_core_dev *dev, void *in, int in_size, void *out, int out_size); int mlx5_cmd_check(struct mlx5_core_dev *dev, int err, void *in, void *out); int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,