* [V5 PATCH 7/8] csiostor: Chelsio FCoE offload driver submission (sources part 4).
From: Naresh Kumar Inna @ 2012-09-24 17:17 UTC (permalink / raw)
To: JBottomley, linux-scsi, dm, leedom; +Cc: netdev, naresh, chethan
In-Reply-To: <1348507072-25625-1-git-send-email-naresh@chelsio.com>
This patch contains code to implement the interrupt handling and the fast
path I/O functionality. The interrupt handling includes allocation of
MSIX vectors, registering and implemeting the interrupt service routines.
The fast path I/O functionality includes posting the I/O request to firmware
via Work Requests, tracking/completing them, and handling task management
requests. SCSI midlayer host template implementation is also covered by
this patch.
Signed-off-by: Naresh Kumar Inna <naresh@chelsio.com>
---
V2:
- Inlined code instead of macro in csio_isr.c
- Use true/false instead if CSIO_TRUE/CSIO_FALSE.
- Use DMA_ defines instead of CSIO_IOREQF_DMA_ defines.
- Replaced large local stack buffer with pre-allocated memory.
- Replaced fast path macros with static functions.
- Removed extra empty lines, needless comment.
V4: Removed needless header file inclusion.
V5:
- Changed over to lockless queuecommand.
- Removed needless header inclusion.
- Re-wrote error paths to reduce lock/unlock ambiguity.
drivers/scsi/csiostor/csio_isr.c | 624 +++++++++
drivers/scsi/csiostor/csio_scsi.c | 2554 +++++++++++++++++++++++++++++++++++++
2 files changed, 3178 insertions(+), 0 deletions(-)
create mode 100644 drivers/scsi/csiostor/csio_isr.c
create mode 100644 drivers/scsi/csiostor/csio_scsi.c
diff --git a/drivers/scsi/csiostor/csio_isr.c b/drivers/scsi/csiostor/csio_isr.c
new file mode 100644
index 0000000..7ee9777
--- /dev/null
+++ b/drivers/scsi/csiostor/csio_isr.c
@@ -0,0 +1,624 @@
+/*
+ * This file is part of the Chelsio FCoE driver for Linux.
+ *
+ * Copyright (c) 2008-2012 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <linux/cpumask.h>
+#include <linux/string.h>
+
+#include "csio_init.h"
+#include "csio_hw.h"
+
+static irqreturn_t
+csio_nondata_isr(int irq, void *dev_id)
+{
+ struct csio_hw *hw = (struct csio_hw *) dev_id;
+ int rv;
+ unsigned long flags;
+
+ if (unlikely(!hw))
+ return IRQ_NONE;
+
+ if (unlikely(pci_channel_offline(hw->pdev))) {
+ CSIO_INC_STATS(hw, n_pcich_offline);
+ return IRQ_NONE;
+ }
+
+ spin_lock_irqsave(&hw->lock, flags);
+ csio_hw_slow_intr_handler(hw);
+ rv = csio_mb_isr_handler(hw);
+
+ if (rv == 0 && !(hw->flags & CSIO_HWF_FWEVT_PENDING)) {
+ hw->flags |= CSIO_HWF_FWEVT_PENDING;
+ spin_unlock_irqrestore(&hw->lock, flags);
+ schedule_work(&hw->evtq_work);
+ return IRQ_HANDLED;
+ }
+ spin_unlock_irqrestore(&hw->lock, flags);
+ return IRQ_HANDLED;
+}
+
+/*
+ * csio_fwevt_handler - Common FW event handler routine.
+ * @hw: HW module.
+ *
+ * This is the ISR for FW events. It is shared b/w MSIX
+ * and INTx handlers.
+ */
+static void
+csio_fwevt_handler(struct csio_hw *hw)
+{
+ int rv;
+ unsigned long flags;
+
+ rv = csio_fwevtq_handler(hw);
+
+ spin_lock_irqsave(&hw->lock, flags);
+ if (rv == 0 && !(hw->flags & CSIO_HWF_FWEVT_PENDING)) {
+ hw->flags |= CSIO_HWF_FWEVT_PENDING;
+ spin_unlock_irqrestore(&hw->lock, flags);
+ schedule_work(&hw->evtq_work);
+ return;
+ }
+ spin_unlock_irqrestore(&hw->lock, flags);
+
+} /* csio_fwevt_handler */
+
+/*
+ * csio_fwevt_isr() - FW events MSIX ISR
+ * @irq:
+ * @dev_id:
+ *
+ * Process WRs on the FW event queue.
+ *
+ */
+static irqreturn_t
+csio_fwevt_isr(int irq, void *dev_id)
+{
+ struct csio_hw *hw = (struct csio_hw *) dev_id;
+
+ if (unlikely(!hw))
+ return IRQ_NONE;
+
+ if (unlikely(pci_channel_offline(hw->pdev))) {
+ CSIO_INC_STATS(hw, n_pcich_offline);
+ return IRQ_NONE;
+ }
+
+ csio_fwevt_handler(hw);
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * csio_fwevt_isr() - INTx wrapper for handling FW events.
+ * @irq:
+ * @dev_id:
+ */
+void
+csio_fwevt_intx_handler(struct csio_hw *hw, void *wr, uint32_t len,
+ struct csio_fl_dma_buf *flb, void *priv)
+{
+ csio_fwevt_handler(hw);
+} /* csio_fwevt_intx_handler */
+
+/*
+ * csio_process_scsi_cmpl - Process a SCSI WR completion.
+ * @hw: HW module.
+ * @wr: The completed WR from the ingress queue.
+ * @len: Length of the WR.
+ * @flb: Freelist buffer array.
+ *
+ */
+static void
+csio_process_scsi_cmpl(struct csio_hw *hw, void *wr, uint32_t len,
+ struct csio_fl_dma_buf *flb, void *cbfn_q)
+{
+ struct csio_ioreq *ioreq;
+ uint8_t *scsiwr;
+ uint8_t subop;
+ void *cmnd;
+ unsigned long flags;
+
+ ioreq = csio_scsi_cmpl_handler(hw, wr, len, flb, NULL, &scsiwr);
+ if (likely(ioreq)) {
+ if (unlikely(*scsiwr == FW_SCSI_ABRT_CLS_WR)) {
+ subop = FW_SCSI_ABRT_CLS_WR_SUB_OPCODE_GET(
+ ((struct fw_scsi_abrt_cls_wr *)
+ scsiwr)->sub_opcode_to_chk_all_io);
+
+ csio_dbg(hw, "%s cmpl recvd ioreq:%p status:%d\n",
+ subop ? "Close" : "Abort",
+ ioreq, ioreq->wr_status);
+
+ spin_lock_irqsave(&hw->lock, flags);
+ if (subop)
+ csio_scsi_closed(ioreq,
+ (struct list_head *)cbfn_q);
+ else
+ csio_scsi_aborted(ioreq,
+ (struct list_head *)cbfn_q);
+ /*
+ * We call scsi_done for I/Os that driver thinks aborts
+ * have timed out. If there is a race caused by FW
+ * completing abort at the exact same time that the
+ * driver has deteced the abort timeout, the following
+ * check prevents calling of scsi_done twice for the
+ * same command: once from the eh_abort_handler, another
+ * from csio_scsi_isr_handler(). This also avoids the
+ * need to check if csio_scsi_cmnd(req) is NULL in the
+ * fast path.
+ */
+ cmnd = csio_scsi_cmnd(ioreq);
+ if (unlikely(cmnd == NULL))
+ list_del_init(&ioreq->sm.sm_list);
+
+ spin_unlock_irqrestore(&hw->lock, flags);
+
+ if (unlikely(cmnd == NULL))
+ csio_put_scsi_ioreq_lock(hw,
+ csio_hw_to_scsim(hw), ioreq);
+ } else {
+ spin_lock_irqsave(&hw->lock, flags);
+ csio_scsi_completed(ioreq, (struct list_head *)cbfn_q);
+ spin_unlock_irqrestore(&hw->lock, flags);
+ }
+ }
+}
+
+/*
+ * csio_scsi_isr_handler() - Common SCSI ISR handler.
+ * @iq: Ingress queue pointer.
+ *
+ * Processes SCSI completions on the SCSI IQ indicated by scm->iq_idx
+ * by calling csio_wr_process_iq_idx. If there are completions on the
+ * isr_cbfn_q, yank them out into a local queue and call their io_cbfns.
+ * Once done, add these completions onto the freelist.
+ * This routine is shared b/w MSIX and INTx.
+ */
+static inline irqreturn_t
+csio_scsi_isr_handler(struct csio_q *iq)
+{
+ struct csio_hw *hw = (struct csio_hw *)iq->owner;
+ LIST_HEAD(cbfn_q);
+ struct list_head *tmp;
+ struct csio_scsim *scm;
+ struct csio_ioreq *ioreq;
+ int isr_completions = 0;
+
+ scm = csio_hw_to_scsim(hw);
+
+ if (unlikely(csio_wr_process_iq(hw, iq, csio_process_scsi_cmpl,
+ &cbfn_q) != 0))
+ return IRQ_NONE;
+
+ /* Call back the completion routines */
+ list_for_each(tmp, &cbfn_q) {
+ ioreq = (struct csio_ioreq *)tmp;
+ isr_completions++;
+ ioreq->io_cbfn(hw, ioreq);
+ /* Release ddp buffer if used for this req */
+ if (unlikely(ioreq->dcopy))
+ csio_put_scsi_ddp_list_lock(hw, scm, &ioreq->gen_list,
+ ioreq->nsge);
+ }
+
+ if (isr_completions) {
+ /* Return the ioreqs back to ioreq->freelist */
+ csio_put_scsi_ioreq_list_lock(hw, scm, &cbfn_q,
+ isr_completions);
+ }
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * csio_scsi_isr() - SCSI MSIX handler
+ * @irq:
+ * @dev_id:
+ *
+ * This is the top level SCSI MSIX handler. Calls csio_scsi_isr_handler()
+ * for handling SCSI completions.
+ */
+static irqreturn_t
+csio_scsi_isr(int irq, void *dev_id)
+{
+ struct csio_q *iq = (struct csio_q *) dev_id;
+ struct csio_hw *hw;
+
+ if (unlikely(!iq))
+ return IRQ_NONE;
+
+ hw = (struct csio_hw *)iq->owner;
+
+ if (unlikely(pci_channel_offline(hw->pdev))) {
+ CSIO_INC_STATS(hw, n_pcich_offline);
+ return IRQ_NONE;
+ }
+
+ csio_scsi_isr_handler(iq);
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * csio_scsi_intx_handler() - SCSI INTx handler
+ * @irq:
+ * @dev_id:
+ *
+ * This is the top level SCSI INTx handler. Calls csio_scsi_isr_handler()
+ * for handling SCSI completions.
+ */
+void
+csio_scsi_intx_handler(struct csio_hw *hw, void *wr, uint32_t len,
+ struct csio_fl_dma_buf *flb, void *priv)
+{
+ struct csio_q *iq = priv;
+
+ csio_scsi_isr_handler(iq);
+
+} /* csio_scsi_intx_handler */
+
+/*
+ * csio_fcoe_isr() - INTx/MSI interrupt service routine for FCoE.
+ * @irq:
+ * @dev_id:
+ *
+ *
+ */
+static irqreturn_t
+csio_fcoe_isr(int irq, void *dev_id)
+{
+ struct csio_hw *hw = (struct csio_hw *) dev_id;
+ struct csio_q *intx_q = NULL;
+ int rv;
+ irqreturn_t ret = IRQ_NONE;
+ unsigned long flags;
+
+ if (unlikely(!hw))
+ return IRQ_NONE;
+
+ if (unlikely(pci_channel_offline(hw->pdev))) {
+ CSIO_INC_STATS(hw, n_pcich_offline);
+ return IRQ_NONE;
+ }
+
+ /* Disable the interrupt for this PCI function. */
+ if (hw->intr_mode == CSIO_IM_INTX)
+ csio_wr_reg32(hw, 0, MYPF_REG(PCIE_PF_CLI));
+
+ /*
+ * The read in the following function will flush the
+ * above write.
+ */
+ if (csio_hw_slow_intr_handler(hw))
+ ret = IRQ_HANDLED;
+
+ /* Get the INTx Forward interrupt IQ. */
+ intx_q = csio_get_q(hw, hw->intr_iq_idx);
+
+ CSIO_DB_ASSERT(intx_q);
+
+ /* IQ handler is not possible for intx_q, hence pass in NULL */
+ if (likely(csio_wr_process_iq(hw, intx_q, NULL, NULL) == 0))
+ ret = IRQ_HANDLED;
+
+ spin_lock_irqsave(&hw->lock, flags);
+ rv = csio_mb_isr_handler(hw);
+ if (rv == 0 && !(hw->flags & CSIO_HWF_FWEVT_PENDING)) {
+ hw->flags |= CSIO_HWF_FWEVT_PENDING;
+ spin_unlock_irqrestore(&hw->lock, flags);
+ schedule_work(&hw->evtq_work);
+ return IRQ_HANDLED;
+ }
+ spin_unlock_irqrestore(&hw->lock, flags);
+
+ return ret;
+}
+
+static void
+csio_add_msix_desc(struct csio_hw *hw)
+{
+ int i;
+ struct csio_msix_entries *entryp = &hw->msix_entries[0];
+ int k = CSIO_EXTRA_VECS;
+ int len = sizeof(entryp->desc) - 1;
+ int cnt = hw->num_sqsets + k;
+
+ /* Non-data vector */
+ memset(entryp->desc, 0, len + 1);
+ snprintf(entryp->desc, len, "csio-%02x:%02x:%x-nondata",
+ CSIO_PCI_BUS(hw), CSIO_PCI_DEV(hw), CSIO_PCI_FUNC(hw));
+
+ entryp++;
+ memset(entryp->desc, 0, len + 1);
+ snprintf(entryp->desc, len, "csio-%02x:%02x:%x-fwevt",
+ CSIO_PCI_BUS(hw), CSIO_PCI_DEV(hw), CSIO_PCI_FUNC(hw));
+ entryp++;
+
+ /* Name SCSI vecs */
+ for (i = k; i < cnt; i++, entryp++) {
+ memset(entryp->desc, 0, len + 1);
+ snprintf(entryp->desc, len, "csio-%02x:%02x:%x-scsi%d",
+ CSIO_PCI_BUS(hw), CSIO_PCI_DEV(hw),
+ CSIO_PCI_FUNC(hw), i - CSIO_EXTRA_VECS);
+ }
+}
+
+int
+csio_request_irqs(struct csio_hw *hw)
+{
+ int rv, i, j, k = 0;
+ struct csio_msix_entries *entryp = &hw->msix_entries[0];
+ struct csio_scsi_cpu_info *info;
+
+ if (hw->intr_mode != CSIO_IM_MSIX) {
+ rv = request_irq(hw->pdev->irq, csio_fcoe_isr,
+ (hw->intr_mode == CSIO_IM_MSI) ?
+ 0 : IRQF_SHARED,
+ KBUILD_MODNAME, hw);
+ if (rv) {
+ if (hw->intr_mode == CSIO_IM_MSI)
+ pci_disable_msi(hw->pdev);
+ csio_err(hw, "Failed to allocate interrupt line.\n");
+ return -EINVAL;
+ }
+
+ goto out;
+ }
+
+ /* Add the MSIX vector descriptions */
+ csio_add_msix_desc(hw);
+
+ rv = request_irq(entryp[k].vector, csio_nondata_isr, 0,
+ entryp[k].desc, hw);
+ if (rv) {
+ csio_err(hw, "IRQ request failed for vec %d err:%d\n",
+ entryp[k].vector, rv);
+ goto err;
+ }
+
+ entryp[k++].dev_id = (void *)hw;
+
+ rv = request_irq(entryp[k].vector, csio_fwevt_isr, 0,
+ entryp[k].desc, hw);
+ if (rv) {
+ csio_err(hw, "IRQ request failed for vec %d err:%d\n",
+ entryp[k].vector, rv);
+ goto err;
+ }
+
+ entryp[k++].dev_id = (void *)hw;
+
+ /* Allocate IRQs for SCSI */
+ for (i = 0; i < hw->num_pports; i++) {
+ info = &hw->scsi_cpu_info[i];
+ for (j = 0; j < info->max_cpus; j++, k++) {
+ struct csio_scsi_qset *sqset = &hw->sqset[i][j];
+ struct csio_q *q = hw->wrm.q_arr[sqset->iq_idx];
+
+ rv = request_irq(entryp[k].vector, csio_scsi_isr, 0,
+ entryp[k].desc, q);
+ if (rv) {
+ csio_err(hw,
+ "IRQ request failed for vec %d err:%d\n",
+ entryp[k].vector, rv);
+ goto err;
+ }
+
+ entryp[k].dev_id = (void *)q;
+
+ } /* for all scsi cpus */
+ } /* for all ports */
+
+out:
+ hw->flags |= CSIO_HWF_HOST_INTR_ENABLED;
+
+ return 0;
+
+err:
+ for (i = 0; i < k; i++) {
+ entryp = &hw->msix_entries[i];
+ free_irq(entryp->vector, entryp->dev_id);
+ }
+ pci_disable_msix(hw->pdev);
+
+ return -EINVAL;
+}
+
+static void
+csio_disable_msix(struct csio_hw *hw, bool free)
+{
+ int i;
+ struct csio_msix_entries *entryp;
+ int cnt = hw->num_sqsets + CSIO_EXTRA_VECS;
+
+ if (free) {
+ for (i = 0; i < cnt; i++) {
+ entryp = &hw->msix_entries[i];
+ free_irq(entryp->vector, entryp->dev_id);
+ }
+ }
+ pci_disable_msix(hw->pdev);
+}
+
+/* Reduce per-port max possible CPUs */
+static void
+csio_reduce_sqsets(struct csio_hw *hw, int cnt)
+{
+ int i;
+ struct csio_scsi_cpu_info *info;
+
+ while (cnt < hw->num_sqsets) {
+ for (i = 0; i < hw->num_pports; i++) {
+ info = &hw->scsi_cpu_info[i];
+ if (info->max_cpus > 1) {
+ info->max_cpus--;
+ hw->num_sqsets--;
+ if (hw->num_sqsets <= cnt)
+ break;
+ }
+ }
+ }
+
+ csio_dbg(hw, "Reduced sqsets to %d\n", hw->num_sqsets);
+}
+
+static int
+csio_enable_msix(struct csio_hw *hw)
+{
+ int rv, i, j, k, n, min, cnt;
+ struct csio_msix_entries *entryp;
+ struct msix_entry *entries;
+ int extra = CSIO_EXTRA_VECS;
+ struct csio_scsi_cpu_info *info;
+
+ min = hw->num_pports + extra;
+ cnt = hw->num_sqsets + extra;
+
+ /* Max vectors required based on #niqs configured in fw */
+ if (hw->flags & CSIO_HWF_USING_SOFT_PARAMS || !csio_is_hw_master(hw))
+ cnt = min_t(uint8_t, hw->cfg_niq, cnt);
+
+ entries = kzalloc(sizeof(struct msix_entry) * cnt, GFP_KERNEL);
+ if (!entries)
+ return -ENOMEM;
+
+ for (i = 0; i < cnt; i++)
+ entries[i].entry = (uint16_t)i;
+
+ csio_dbg(hw, "FW supp #niq:%d, trying %d msix's\n", hw->cfg_niq, cnt);
+
+ while ((rv = pci_enable_msix(hw->pdev, entries, cnt)) >= min)
+ cnt = rv;
+ if (!rv) {
+ if (cnt < (hw->num_sqsets + extra)) {
+ csio_dbg(hw, "Reducing sqsets to %d\n", cnt - extra);
+ csio_reduce_sqsets(hw, cnt - extra);
+ }
+ } else {
+ if (rv > 0) {
+ pci_disable_msix(hw->pdev);
+ csio_info(hw, "Not using MSI-X, remainder:%d\n", rv);
+ }
+
+ kfree(entries);
+ return -ENOMEM;
+ }
+
+ /* Save off vectors */
+ for (i = 0; i < cnt; i++) {
+ entryp = &hw->msix_entries[i];
+ entryp->vector = entries[i].vector;
+ }
+
+ /* Distribute vectors */
+ k = 0;
+ csio_set_nondata_intr_idx(hw, entries[k].entry);
+ csio_set_mb_intr_idx(csio_hw_to_mbm(hw), entries[k++].entry);
+ csio_set_fwevt_intr_idx(hw, entries[k++].entry);
+
+ for (i = 0; i < hw->num_pports; i++) {
+ info = &hw->scsi_cpu_info[i];
+
+ for (j = 0; j < hw->num_scsi_msix_cpus; j++) {
+ n = (j % info->max_cpus) + k;
+ hw->sqset[i][j].intr_idx = entries[n].entry;
+ }
+
+ k += info->max_cpus;
+ }
+
+ kfree(entries);
+ return 0;
+}
+
+void
+csio_intr_enable(struct csio_hw *hw)
+{
+ hw->intr_mode = CSIO_IM_NONE;
+ hw->flags &= ~CSIO_HWF_HOST_INTR_ENABLED;
+
+ /* Try MSIX, then MSI or fall back to INTx */
+ if ((csio_msi == 2) && !csio_enable_msix(hw))
+ hw->intr_mode = CSIO_IM_MSIX;
+ else {
+ /* Max iqs required based on #niqs configured in fw */
+ if (hw->flags & CSIO_HWF_USING_SOFT_PARAMS ||
+ !csio_is_hw_master(hw)) {
+ int extra = CSIO_EXTRA_MSI_IQS;
+
+ if (hw->cfg_niq < (hw->num_sqsets + extra)) {
+ csio_dbg(hw, "Reducing sqsets to %d\n",
+ hw->cfg_niq - extra);
+ csio_reduce_sqsets(hw, hw->cfg_niq - extra);
+ }
+ }
+
+ if ((csio_msi == 1) && !pci_enable_msi(hw->pdev))
+ hw->intr_mode = CSIO_IM_MSI;
+ else
+ hw->intr_mode = CSIO_IM_INTX;
+ }
+
+ csio_dbg(hw, "Using %s interrupt mode.\n",
+ (hw->intr_mode == CSIO_IM_MSIX) ? "MSIX" :
+ ((hw->intr_mode == CSIO_IM_MSI) ? "MSI" : "INTx"));
+}
+
+void
+csio_intr_disable(struct csio_hw *hw, bool free)
+{
+ csio_hw_intr_disable(hw);
+
+ switch (hw->intr_mode) {
+ case CSIO_IM_MSIX:
+ csio_disable_msix(hw, free);
+ break;
+ case CSIO_IM_MSI:
+ if (free)
+ free_irq(hw->pdev->irq, hw);
+ pci_disable_msi(hw->pdev);
+ break;
+ case CSIO_IM_INTX:
+ if (free)
+ free_irq(hw->pdev->irq, hw);
+ break;
+ default:
+ break;
+ }
+ hw->intr_mode = CSIO_IM_NONE;
+ hw->flags &= ~CSIO_HWF_HOST_INTR_ENABLED;
+}
diff --git a/drivers/scsi/csiostor/csio_scsi.c b/drivers/scsi/csiostor/csio_scsi.c
new file mode 100644
index 0000000..1e22bb7
--- /dev/null
+++ b/drivers/scsi/csiostor/csio_scsi.c
@@ -0,0 +1,2554 @@
+/*
+ * This file is part of the Chelsio FCoE driver for Linux.
+ *
+ * Copyright (c) 2008-2012 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/device.h>
+#include <linux/delay.h>
+#include <linux/ctype.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/compiler.h>
+#include <linux/export.h>
+#include <linux/module.h>
+#include <asm/unaligned.h>
+#include <asm/page.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_transport_fc.h>
+
+#include "csio_hw.h"
+#include "csio_lnode.h"
+#include "csio_rnode.h"
+#include "csio_scsi.h"
+#include "csio_init.h"
+
+int csio_scsi_eqsize = 65536;
+int csio_scsi_iqlen = 128;
+int csio_scsi_ioreqs = 2048;
+uint32_t csio_max_scan_tmo;
+uint32_t csio_delta_scan_tmo = 5;
+int csio_lun_qdepth = 32;
+
+static int csio_ddp_descs = 128;
+
+static int csio_do_abrt_cls(struct csio_hw *,
+ struct csio_ioreq *, bool);
+
+static void csio_scsis_uninit(struct csio_ioreq *, enum csio_scsi_ev);
+static void csio_scsis_io_active(struct csio_ioreq *, enum csio_scsi_ev);
+static void csio_scsis_tm_active(struct csio_ioreq *, enum csio_scsi_ev);
+static void csio_scsis_aborting(struct csio_ioreq *, enum csio_scsi_ev);
+static void csio_scsis_closing(struct csio_ioreq *, enum csio_scsi_ev);
+static void csio_scsis_shost_cmpl_await(struct csio_ioreq *, enum csio_scsi_ev);
+
+/*
+ * csio_scsi_match_io - Match an ioreq with the given SCSI level data.
+ * @ioreq: The I/O request
+ * @sld: Level information
+ *
+ * Should be called with lock held.
+ *
+ */
+static bool
+csio_scsi_match_io(struct csio_ioreq *ioreq, struct csio_scsi_level_data *sld)
+{
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(ioreq);
+
+ switch (sld->level) {
+ case CSIO_LEV_LUN:
+ if (scmnd == NULL)
+ return false;
+
+ return ((ioreq->lnode == sld->lnode) &&
+ (ioreq->rnode == sld->rnode) &&
+ ((uint64_t)scmnd->device->lun == sld->oslun));
+
+ case CSIO_LEV_RNODE:
+ return ((ioreq->lnode == sld->lnode) &&
+ (ioreq->rnode == sld->rnode));
+ case CSIO_LEV_LNODE:
+ return (ioreq->lnode == sld->lnode);
+ case CSIO_LEV_ALL:
+ return true;
+ default:
+ return false;
+ }
+}
+
+/*
+ * csio_scsi_gather_active_ios - Gather active I/Os based on level
+ * @scm: SCSI module
+ * @sld: Level information
+ * @dest: The queue where these I/Os have to be gathered.
+ *
+ * Should be called with lock held.
+ */
+static void
+csio_scsi_gather_active_ios(struct csio_scsim *scm,
+ struct csio_scsi_level_data *sld,
+ struct list_head *dest)
+{
+ struct list_head *tmp, *next;
+
+ if (list_empty(&scm->active_q))
+ return;
+
+ /* Just splice the entire active_q into dest */
+ if (sld->level == CSIO_LEV_ALL) {
+ list_splice_tail_init(&scm->active_q, dest);
+ return;
+ }
+
+ list_for_each_safe(tmp, next, &scm->active_q) {
+ if (csio_scsi_match_io((struct csio_ioreq *)tmp, sld)) {
+ list_del_init(tmp);
+ list_add_tail(tmp, dest);
+ }
+ }
+}
+
+static inline bool
+csio_scsi_itnexus_loss_error(uint16_t error)
+{
+ switch (error) {
+ case FW_ERR_LINK_DOWN:
+ case FW_RDEV_NOT_READY:
+ case FW_ERR_RDEV_LOST:
+ case FW_ERR_RDEV_LOGO:
+ case FW_ERR_RDEV_IMPL_LOGO:
+ return 1;
+ }
+ return 0;
+}
+
+static inline void
+csio_scsi_tag(struct scsi_cmnd *scmnd, uint8_t *tag, uint8_t hq,
+ uint8_t oq, uint8_t sq)
+{
+ char stag[2];
+
+ if (scsi_populate_tag_msg(scmnd, stag)) {
+ switch (stag[0]) {
+ case HEAD_OF_QUEUE_TAG:
+ *tag = hq;
+ break;
+ case ORDERED_QUEUE_TAG:
+ *tag = oq;
+ break;
+ default:
+ *tag = sq;
+ break;
+ }
+ } else
+ *tag = 0;
+}
+
+/*
+ * csio_scsi_fcp_cmnd - Frame the SCSI FCP command paylod.
+ * @req: IO req structure.
+ * @addr: DMA location to place the payload.
+ *
+ * This routine is shared between FCP_WRITE, FCP_READ and FCP_CMD requests.
+ */
+static inline void
+csio_scsi_fcp_cmnd(struct csio_ioreq *req, void *addr)
+{
+ struct fcp_cmnd *fcp_cmnd = (struct fcp_cmnd *)addr;
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(req);
+
+ /* Check for Task Management */
+ if (likely(scmnd->SCp.Message == 0)) {
+ int_to_scsilun(scmnd->device->lun, &fcp_cmnd->fc_lun);
+ fcp_cmnd->fc_tm_flags = 0;
+ fcp_cmnd->fc_cmdref = 0;
+ fcp_cmnd->fc_pri_ta = 0;
+
+ memcpy(fcp_cmnd->fc_cdb, scmnd->cmnd, 16);
+ csio_scsi_tag(scmnd, &fcp_cmnd->fc_pri_ta,
+ FCP_PTA_HEADQ, FCP_PTA_ORDERED, FCP_PTA_SIMPLE);
+ fcp_cmnd->fc_dl = cpu_to_be32(scsi_bufflen(scmnd));
+
+ if (req->nsge)
+ if (req->datadir == DMA_TO_DEVICE)
+ fcp_cmnd->fc_flags = FCP_CFL_WRDATA;
+ else
+ fcp_cmnd->fc_flags = FCP_CFL_RDDATA;
+ else
+ fcp_cmnd->fc_flags = 0;
+ } else {
+ memset(fcp_cmnd, 0, sizeof(*fcp_cmnd));
+ int_to_scsilun(scmnd->device->lun, &fcp_cmnd->fc_lun);
+ fcp_cmnd->fc_tm_flags = (uint8_t)scmnd->SCp.Message;
+ }
+}
+
+/*
+ * csio_scsi_init_cmd_wr - Initialize the SCSI CMD WR.
+ * @req: IO req structure.
+ * @addr: DMA location to place the payload.
+ * @size: Size of WR (including FW WR + immed data + rsp SG entry
+ *
+ * Wrapper for populating fw_scsi_cmd_wr.
+ */
+static inline void
+csio_scsi_init_cmd_wr(struct csio_ioreq *req, void *addr, uint32_t size)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_rnode *rn = req->rnode;
+ struct fw_scsi_cmd_wr *wr = (struct fw_scsi_cmd_wr *)addr;
+ struct csio_dma_buf *dma_buf;
+ uint8_t imm = csio_hw_to_scsim(hw)->proto_cmd_len;
+
+ wr->op_immdlen = cpu_to_be32(FW_WR_OP(FW_SCSI_CMD_WR) |
+ FW_SCSI_CMD_WR_IMMDLEN(imm));
+ wr->flowid_len16 = cpu_to_be32(FW_WR_FLOWID(rn->flowid) |
+ FW_WR_LEN16(
+ DIV_ROUND_UP(size, 16)));
+
+ wr->cookie = (uintptr_t) req;
+ wr->iqid = (uint16_t)cpu_to_be16(csio_q_physiqid(hw, req->iq_idx));
+ wr->tmo_val = (uint8_t) req->tmo;
+ wr->r3 = 0;
+ memset(&wr->r5, 0, 8);
+
+ /* Get RSP DMA buffer */
+ dma_buf = &req->dma_buf;
+
+ /* Prepare RSP SGL */
+ wr->rsp_dmalen = cpu_to_be32(dma_buf->len);
+ wr->rsp_dmaaddr = cpu_to_be64(dma_buf->paddr);
+
+ wr->r6 = 0;
+
+ wr->u.fcoe.ctl_pri = 0;
+ wr->u.fcoe.cp_en_class = 0;
+ wr->u.fcoe.r4_lo[0] = 0;
+ wr->u.fcoe.r4_lo[1] = 0;
+
+ /* Frame a FCP command */
+ csio_scsi_fcp_cmnd(req, (void *)((uintptr_t)addr +
+ sizeof(struct fw_scsi_cmd_wr)));
+}
+
+#define CSIO_SCSI_CMD_WR_SZ(_imm) \
+ (sizeof(struct fw_scsi_cmd_wr) + /* WR size */ \
+ ALIGN((_imm), 16)) /* Immed data */
+
+#define CSIO_SCSI_CMD_WR_SZ_16(_imm) \
+ (ALIGN(CSIO_SCSI_CMD_WR_SZ((_imm)), 16))
+
+/*
+ * csio_scsi_cmd - Create a SCSI CMD WR.
+ * @req: IO req structure.
+ *
+ * Gets a WR slot in the ingress queue and initializes it with SCSI CMD WR.
+ *
+ */
+static inline void
+csio_scsi_cmd(struct csio_ioreq *req)
+{
+ struct csio_wr_pair wrp;
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+ uint32_t size = CSIO_SCSI_CMD_WR_SZ_16(scsim->proto_cmd_len);
+
+ req->drv_status = csio_wr_get(hw, req->eq_idx, size, &wrp);
+ if (unlikely(req->drv_status != 0))
+ return;
+
+ if (wrp.size1 >= size) {
+ /* Initialize WR in one shot */
+ csio_scsi_init_cmd_wr(req, wrp.addr1, size);
+ } else {
+ uint8_t *tmpwr = csio_q_eq_wrap(hw, req->eq_idx);
+
+ /*
+ * Make a temporary copy of the WR and write back
+ * the copy into the WR pair.
+ */
+ csio_scsi_init_cmd_wr(req, (void *)tmpwr, size);
+ memcpy(wrp.addr1, tmpwr, wrp.size1);
+ memcpy(wrp.addr2, tmpwr + wrp.size1, size - wrp.size1);
+ }
+}
+
+/*
+ * csio_scsi_init_ulptx_dsgl - Fill in a ULP_TX_SC_DSGL
+ * @hw: HW module
+ * @req: IO request
+ * @sgl: ULP TX SGL pointer.
+ *
+ */
+static inline void
+csio_scsi_init_ultptx_dsgl(struct csio_hw *hw, struct csio_ioreq *req,
+ struct ulptx_sgl *sgl)
+{
+ struct ulptx_sge_pair *sge_pair = NULL;
+ struct scatterlist *sgel;
+ uint32_t i = 0;
+ uint32_t xfer_len;
+ struct list_head *tmp;
+ struct csio_dma_buf *dma_buf;
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(req);
+
+ sgl->cmd_nsge = htonl(ULPTX_CMD(ULP_TX_SC_DSGL) | ULPTX_MORE |
+ ULPTX_NSGE(req->nsge));
+ /* Now add the data SGLs */
+ if (likely(!req->dcopy)) {
+ scsi_for_each_sg(scmnd, sgel, req->nsge, i) {
+ if (i == 0) {
+ sgl->addr0 = cpu_to_be64(sg_dma_address(sgel));
+ sgl->len0 = cpu_to_be32(sg_dma_len(sgel));
+ sge_pair = (struct ulptx_sge_pair *)(sgl + 1);
+ continue;
+ }
+ if ((i - 1) & 0x1) {
+ sge_pair->addr[1] = cpu_to_be64(
+ sg_dma_address(sgel));
+ sge_pair->len[1] = cpu_to_be32(
+ sg_dma_len(sgel));
+ sge_pair++;
+ } else {
+ sge_pair->addr[0] = cpu_to_be64(
+ sg_dma_address(sgel));
+ sge_pair->len[0] = cpu_to_be32(
+ sg_dma_len(sgel));
+ }
+ }
+ } else {
+ /* Program sg elements with driver's DDP buffer */
+ xfer_len = scsi_bufflen(scmnd);
+ list_for_each(tmp, &req->gen_list) {
+ dma_buf = (struct csio_dma_buf *)tmp;
+ if (i == 0) {
+ sgl->addr0 = cpu_to_be64(dma_buf->paddr);
+ sgl->len0 = cpu_to_be32(
+ min(xfer_len, dma_buf->len));
+ sge_pair = (struct ulptx_sge_pair *)(sgl + 1);
+ } else if ((i - 1) & 0x1) {
+ sge_pair->addr[1] = cpu_to_be64(dma_buf->paddr);
+ sge_pair->len[1] = cpu_to_be32(
+ min(xfer_len, dma_buf->len));
+ sge_pair++;
+ } else {
+ sge_pair->addr[0] = cpu_to_be64(dma_buf->paddr);
+ sge_pair->len[0] = cpu_to_be32(
+ min(xfer_len, dma_buf->len));
+ }
+ xfer_len -= min(xfer_len, dma_buf->len);
+ i++;
+ }
+ }
+}
+
+/*
+ * csio_scsi_init_read_wr - Initialize the READ SCSI WR.
+ * @req: IO req structure.
+ * @wrp: DMA location to place the payload.
+ * @size: Size of WR (including FW WR + immed data + rsp SG entry + data SGL
+ *
+ * Wrapper for populating fw_scsi_read_wr.
+ */
+static inline void
+csio_scsi_init_read_wr(struct csio_ioreq *req, void *wrp, uint32_t size)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_rnode *rn = req->rnode;
+ struct fw_scsi_read_wr *wr = (struct fw_scsi_read_wr *)wrp;
+ struct ulptx_sgl *sgl;
+ struct csio_dma_buf *dma_buf;
+ uint8_t imm = csio_hw_to_scsim(hw)->proto_cmd_len;
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(req);
+
+ wr->op_immdlen = cpu_to_be32(FW_WR_OP(FW_SCSI_READ_WR) |
+ FW_SCSI_READ_WR_IMMDLEN(imm));
+ wr->flowid_len16 = cpu_to_be32(FW_WR_FLOWID(rn->flowid) |
+ FW_WR_LEN16(DIV_ROUND_UP(size, 16)));
+ wr->cookie = (uintptr_t)req;
+ wr->iqid = (uint16_t)cpu_to_be16(csio_q_physiqid(hw, req->iq_idx));
+ wr->tmo_val = (uint8_t)(req->tmo);
+ wr->use_xfer_cnt = 1;
+ wr->xfer_cnt = cpu_to_be32(scsi_bufflen(scmnd));
+ wr->ini_xfer_cnt = cpu_to_be32(scsi_bufflen(scmnd));
+ /* Get RSP DMA buffer */
+ dma_buf = &req->dma_buf;
+
+ /* Prepare RSP SGL */
+ wr->rsp_dmalen = cpu_to_be32(dma_buf->len);
+ wr->rsp_dmaaddr = cpu_to_be64(dma_buf->paddr);
+
+ wr->r4 = 0;
+
+ wr->u.fcoe.ctl_pri = 0;
+ wr->u.fcoe.cp_en_class = 0;
+ wr->u.fcoe.r3_lo[0] = 0;
+ wr->u.fcoe.r3_lo[1] = 0;
+ csio_scsi_fcp_cmnd(req, (void *)((uintptr_t)wrp +
+ sizeof(struct fw_scsi_read_wr)));
+
+ /* Move WR pointer past command and immediate data */
+ sgl = (struct ulptx_sgl *)((uintptr_t)wrp +
+ sizeof(struct fw_scsi_read_wr) + ALIGN(imm, 16));
+
+ /* Fill in the DSGL */
+ csio_scsi_init_ultptx_dsgl(hw, req, sgl);
+}
+
+/*
+ * csio_scsi_init_write_wr - Initialize the WRITE SCSI WR.
+ * @req: IO req structure.
+ * @wrp: DMA location to place the payload.
+ * @size: Size of WR (including FW WR + immed data + rsp SG entry + data SGL
+ *
+ * Wrapper for populating fw_scsi_write_wr.
+ */
+static inline void
+csio_scsi_init_write_wr(struct csio_ioreq *req, void *wrp, uint32_t size)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_rnode *rn = req->rnode;
+ struct fw_scsi_write_wr *wr = (struct fw_scsi_write_wr *)wrp;
+ struct ulptx_sgl *sgl;
+ struct csio_dma_buf *dma_buf;
+ uint8_t imm = csio_hw_to_scsim(hw)->proto_cmd_len;
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(req);
+
+ wr->op_immdlen = cpu_to_be32(FW_WR_OP(FW_SCSI_WRITE_WR) |
+ FW_SCSI_WRITE_WR_IMMDLEN(imm));
+ wr->flowid_len16 = cpu_to_be32(FW_WR_FLOWID(rn->flowid) |
+ FW_WR_LEN16(DIV_ROUND_UP(size, 16)));
+ wr->cookie = (uintptr_t)req;
+ wr->iqid = (uint16_t)cpu_to_be16(csio_q_physiqid(hw, req->iq_idx));
+ wr->tmo_val = (uint8_t)(req->tmo);
+ wr->use_xfer_cnt = 1;
+ wr->xfer_cnt = cpu_to_be32(scsi_bufflen(scmnd));
+ wr->ini_xfer_cnt = cpu_to_be32(scsi_bufflen(scmnd));
+ /* Get RSP DMA buffer */
+ dma_buf = &req->dma_buf;
+
+ /* Prepare RSP SGL */
+ wr->rsp_dmalen = cpu_to_be32(dma_buf->len);
+ wr->rsp_dmaaddr = cpu_to_be64(dma_buf->paddr);
+
+ wr->r4 = 0;
+
+ wr->u.fcoe.ctl_pri = 0;
+ wr->u.fcoe.cp_en_class = 0;
+ wr->u.fcoe.r3_lo[0] = 0;
+ wr->u.fcoe.r3_lo[1] = 0;
+ csio_scsi_fcp_cmnd(req, (void *)((uintptr_t)wrp +
+ sizeof(struct fw_scsi_write_wr)));
+
+ /* Move WR pointer past command and immediate data */
+ sgl = (struct ulptx_sgl *)((uintptr_t)wrp +
+ sizeof(struct fw_scsi_write_wr) + ALIGN(imm, 16));
+
+ /* Fill in the DSGL */
+ csio_scsi_init_ultptx_dsgl(hw, req, sgl);
+}
+
+/* Calculate WR size needed for fw_scsi_read_wr/fw_scsi_write_wr */
+#define CSIO_SCSI_DATA_WRSZ(req, oper, sz, imm) \
+do { \
+ (sz) = sizeof(struct fw_scsi_##oper##_wr) + /* WR size */ \
+ ALIGN((imm), 16) + /* Immed data */ \
+ sizeof(struct ulptx_sgl); /* ulptx_sgl */ \
+ \
+ if (unlikely((req)->nsge > 1)) \
+ (sz) += (sizeof(struct ulptx_sge_pair) * \
+ (ALIGN(((req)->nsge - 1), 2) / 2)); \
+ /* Data SGE */ \
+} while (0)
+
+/*
+ * csio_scsi_read - Create a SCSI READ WR.
+ * @req: IO req structure.
+ *
+ * Gets a WR slot in the ingress queue and initializes it with
+ * SCSI READ WR.
+ *
+ */
+static inline void
+csio_scsi_read(struct csio_ioreq *req)
+{
+ struct csio_wr_pair wrp;
+ uint32_t size;
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+
+ CSIO_SCSI_DATA_WRSZ(req, read, size, scsim->proto_cmd_len);
+ size = ALIGN(size, 16);
+
+ req->drv_status = csio_wr_get(hw, req->eq_idx, size, &wrp);
+ if (likely(req->drv_status == 0)) {
+ if (likely(wrp.size1 >= size)) {
+ /* Initialize WR in one shot */
+ csio_scsi_init_read_wr(req, wrp.addr1, size);
+ } else {
+ uint8_t *tmpwr = csio_q_eq_wrap(hw, req->eq_idx);
+ /*
+ * Make a temporary copy of the WR and write back
+ * the copy into the WR pair.
+ */
+ csio_scsi_init_read_wr(req, (void *)tmpwr, size);
+ memcpy(wrp.addr1, tmpwr, wrp.size1);
+ memcpy(wrp.addr2, tmpwr + wrp.size1, size - wrp.size1);
+ }
+ }
+}
+
+/*
+ * csio_scsi_write - Create a SCSI WRITE WR.
+ * @req: IO req structure.
+ *
+ * Gets a WR slot in the ingress queue and initializes it with
+ * SCSI WRITE WR.
+ *
+ */
+static inline void
+csio_scsi_write(struct csio_ioreq *req)
+{
+ struct csio_wr_pair wrp;
+ uint32_t size;
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+
+ CSIO_SCSI_DATA_WRSZ(req, write, size, scsim->proto_cmd_len);
+ size = ALIGN(size, 16);
+
+ req->drv_status = csio_wr_get(hw, req->eq_idx, size, &wrp);
+ if (likely(req->drv_status == 0)) {
+ if (likely(wrp.size1 >= size)) {
+ /* Initialize WR in one shot */
+ csio_scsi_init_write_wr(req, wrp.addr1, size);
+ } else {
+ uint8_t *tmpwr = csio_q_eq_wrap(hw, req->eq_idx);
+ /*
+ * Make a temporary copy of the WR and write back
+ * the copy into the WR pair.
+ */
+ csio_scsi_init_write_wr(req, (void *)tmpwr, size);
+ memcpy(wrp.addr1, tmpwr, wrp.size1);
+ memcpy(wrp.addr2, tmpwr + wrp.size1, size - wrp.size1);
+ }
+ }
+}
+
+/*
+ * csio_setup_ddp - Setup DDP buffers for Read request.
+ * @req: IO req structure.
+ *
+ * Checks SGLs/Data buffers are virtually contiguous required for DDP.
+ * If contiguous,driver posts SGLs in the WR otherwise post internal
+ * buffers for such request for DDP.
+ */
+static inline void
+csio_setup_ddp(struct csio_scsim *scsim, struct csio_ioreq *req)
+{
+#ifdef __CSIO_DEBUG__
+ struct csio_hw *hw = req->lnode->hwp;
+#endif
+ struct scatterlist *sgel = NULL;
+ struct scsi_cmnd *scmnd = csio_scsi_cmnd(req);
+ uint64_t sg_addr = 0;
+ uint32_t ddp_pagesz = 4096;
+ uint32_t buf_off;
+ struct csio_dma_buf *dma_buf = NULL;
+ uint32_t alloc_len = 0;
+ uint32_t xfer_len = 0;
+ uint32_t sg_len = 0;
+ uint32_t i;
+
+ scsi_for_each_sg(scmnd, sgel, req->nsge, i) {
+ sg_addr = sg_dma_address(sgel);
+ sg_len = sg_dma_len(sgel);
+
+ buf_off = sg_addr & (ddp_pagesz - 1);
+
+ /* Except 1st buffer,all buffer addr have to be Page aligned */
+ if (i != 0 && buf_off) {
+ csio_dbg(hw, "SGL addr not DDP aligned (%llx:%d)\n",
+ sg_addr, sg_len);
+ goto unaligned;
+ }
+
+ /* Except last buffer,all buffer must end on page boundary */
+ if ((i != (req->nsge - 1)) &&
+ ((buf_off + sg_len) & (ddp_pagesz - 1))) {
+ csio_dbg(hw,
+ "SGL addr not ending on page boundary"
+ "(%llx:%d)\n", sg_addr, sg_len);
+ goto unaligned;
+ }
+ }
+
+ /* SGL's are virtually contiguous. HW will DDP to SGLs */
+ req->dcopy = 0;
+ csio_scsi_read(req);
+
+ return;
+
+unaligned:
+ CSIO_INC_STATS(scsim, n_unaligned);
+ /*
+ * For unaligned SGLs, driver will allocate internal DDP buffer.
+ * Once command is completed data from DDP buffer copied to SGLs
+ */
+ req->dcopy = 1;
+
+ /* Use gen_list to store the DDP buffers */
+ INIT_LIST_HEAD(&req->gen_list);
+ xfer_len = scsi_bufflen(scmnd);
+
+ i = 0;
+ /* Allocate ddp buffers for this request */
+ while (alloc_len < xfer_len) {
+ dma_buf = csio_get_scsi_ddp(scsim);
+ if (dma_buf == NULL || i > scsim->max_sge) {
+ req->drv_status = -EBUSY;
+ break;
+ }
+ alloc_len += dma_buf->len;
+ /* Added to IO req */
+ list_add_tail(&dma_buf->list, &req->gen_list);
+ i++;
+ }
+
+ if (!req->drv_status) {
+ /* set number of ddp bufs used */
+ req->nsge = i;
+ csio_scsi_read(req);
+ return;
+ }
+
+ /* release dma descs */
+ if (i > 0)
+ csio_put_scsi_ddp_list(scsim, &req->gen_list, i);
+}
+
+/*
+ * csio_scsi_init_abrt_cls_wr - Initialize an ABORT/CLOSE WR.
+ * @req: IO req structure.
+ * @addr: DMA location to place the payload.
+ * @size: Size of WR
+ * @abort: abort OR close
+ *
+ * Wrapper for populating fw_scsi_cmd_wr.
+ */
+static inline void
+csio_scsi_init_abrt_cls_wr(struct csio_ioreq *req, void *addr, uint32_t size,
+ bool abort)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_rnode *rn = req->rnode;
+ struct fw_scsi_abrt_cls_wr *wr = (struct fw_scsi_abrt_cls_wr *)addr;
+
+ wr->op_immdlen = cpu_to_be32(FW_WR_OP(FW_SCSI_ABRT_CLS_WR));
+ wr->flowid_len16 = cpu_to_be32(FW_WR_FLOWID(rn->flowid) |
+ FW_WR_LEN16(
+ DIV_ROUND_UP(size, 16)));
+
+ wr->cookie = (uintptr_t) req;
+ wr->iqid = (uint16_t)cpu_to_be16(csio_q_physiqid(hw, req->iq_idx));
+ wr->tmo_val = (uint8_t) req->tmo;
+ /* 0 for CHK_ALL_IO tells FW to look up t_cookie */
+ wr->sub_opcode_to_chk_all_io =
+ (FW_SCSI_ABRT_CLS_WR_SUB_OPCODE(abort) |
+ FW_SCSI_ABRT_CLS_WR_CHK_ALL_IO(0));
+ wr->r3[0] = 0;
+ wr->r3[1] = 0;
+ wr->r3[2] = 0;
+ wr->r3[3] = 0;
+ /* Since we re-use the same ioreq for abort as well */
+ wr->t_cookie = (uintptr_t) req;
+}
+
+static inline void
+csio_scsi_abrt_cls(struct csio_ioreq *req, bool abort)
+{
+ struct csio_wr_pair wrp;
+ struct csio_hw *hw = req->lnode->hwp;
+ uint32_t size = ALIGN(sizeof(struct fw_scsi_abrt_cls_wr), 16);
+
+ req->drv_status = csio_wr_get(hw, req->eq_idx, size, &wrp);
+ if (req->drv_status != 0)
+ return;
+
+ if (wrp.size1 >= size) {
+ /* Initialize WR in one shot */
+ csio_scsi_init_abrt_cls_wr(req, wrp.addr1, size, abort);
+ } else {
+ uint8_t *tmpwr = csio_q_eq_wrap(hw, req->eq_idx);
+ /*
+ * Make a temporary copy of the WR and write back
+ * the copy into the WR pair.
+ */
+ csio_scsi_init_abrt_cls_wr(req, (void *)tmpwr, size, abort);
+ memcpy(wrp.addr1, tmpwr, wrp.size1);
+ memcpy(wrp.addr2, tmpwr + wrp.size1, size - wrp.size1);
+ }
+}
+
+/*****************************************************************************/
+/* START: SCSI SM */
+/*****************************************************************************/
+static void
+csio_scsis_uninit(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+
+ switch (evt) {
+ case CSIO_SCSIE_START_IO:
+
+ if (req->nsge) {
+ if (req->datadir == DMA_TO_DEVICE) {
+ req->dcopy = 0;
+ csio_scsi_write(req);
+ } else
+ csio_setup_ddp(scsim, req);
+ } else {
+ csio_scsi_cmd(req);
+ }
+
+ if (likely(req->drv_status == 0)) {
+ /* change state and enqueue on active_q */
+ csio_set_state(&req->sm, csio_scsis_io_active);
+ list_add_tail(&req->sm.sm_list, &scsim->active_q);
+ csio_wr_issue(hw, req->eq_idx, false);
+ CSIO_INC_STATS(scsim, n_active);
+
+ return;
+ }
+ break;
+
+ case CSIO_SCSIE_START_TM:
+ csio_scsi_cmd(req);
+ if (req->drv_status == 0) {
+ /*
+ * NOTE: We collect the affected I/Os prior to issuing
+ * LUN reset, and not after it. This is to prevent
+ * aborting I/Os that get issued after the LUN reset,
+ * but prior to LUN reset completion (in the event that
+ * the host stack has not blocked I/Os to a LUN that is
+ * being reset.
+ */
+ csio_set_state(&req->sm, csio_scsis_tm_active);
+ list_add_tail(&req->sm.sm_list, &scsim->active_q);
+ csio_wr_issue(hw, req->eq_idx, false);
+ CSIO_INC_STATS(scsim, n_tm_active);
+ }
+ return;
+
+ case CSIO_SCSIE_ABORT:
+ case CSIO_SCSIE_CLOSE:
+ /*
+ * NOTE:
+ * We could get here due to :
+ * - a window in the cleanup path of the SCSI module
+ * (csio_scsi_abort_io()). Please see NOTE in this function.
+ * - a window in the time we tried to issue an abort/close
+ * of a request to FW, and the FW completed the request
+ * itself.
+ * Print a message for now, and return INVAL either way.
+ */
+ req->drv_status = -EINVAL;
+ csio_warn(hw, "Trying to abort/close completed IO:%p!\n", req);
+ break;
+
+ default:
+ csio_dbg(hw, "Unhandled event:%d sent to req:%p\n", evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+static void
+csio_scsis_io_active(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+ struct csio_rnode *rn;
+
+ switch (evt) {
+ case CSIO_SCSIE_COMPLETED:
+ CSIO_DEC_STATS(scm, n_active);
+ list_del_init(&req->sm.sm_list);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ /*
+ * In MSIX mode, with multiple queues, the SCSI compeltions
+ * could reach us sooner than the FW events sent to indicate
+ * I-T nexus loss (link down, remote device logo etc). We
+ * dont want to be returning such I/Os to the upper layer
+ * immediately, since we wouldnt have reported the I-T nexus
+ * loss itself. This forces us to serialize such completions
+ * with the reporting of the I-T nexus loss. Therefore, we
+ * internally queue up such up such completions in the rnode.
+ * The reporting of I-T nexus loss to the upper layer is then
+ * followed by the returning of I/Os in this internal queue.
+ * Having another state alongwith another queue helps us take
+ * actions for events such as ABORT received while we are
+ * in this rnode queue.
+ */
+ if (unlikely(req->wr_status != FW_SUCCESS)) {
+ rn = req->rnode;
+ /*
+ * FW says remote device is lost, but rnode
+ * doesnt reflect it.
+ */
+ if (csio_scsi_itnexus_loss_error(req->wr_status) &&
+ csio_is_rnode_ready(rn)) {
+ csio_set_state(&req->sm,
+ csio_scsis_shost_cmpl_await);
+ list_add_tail(&req->sm.sm_list,
+ &rn->host_cmpl_q);
+ }
+ }
+
+ break;
+
+ case CSIO_SCSIE_ABORT:
+ csio_scsi_abrt_cls(req, SCSI_ABORT);
+ if (req->drv_status == 0) {
+ csio_wr_issue(hw, req->eq_idx, false);
+ csio_set_state(&req->sm, csio_scsis_aborting);
+ }
+ break;
+
+ case CSIO_SCSIE_CLOSE:
+ csio_scsi_abrt_cls(req, SCSI_CLOSE);
+ if (req->drv_status == 0) {
+ csio_wr_issue(hw, req->eq_idx, false);
+ csio_set_state(&req->sm, csio_scsis_closing);
+ }
+ break;
+
+ case CSIO_SCSIE_DRVCLEANUP:
+ req->wr_status = FW_HOSTERROR;
+ CSIO_DEC_STATS(scm, n_active);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ default:
+ csio_dbg(hw, "Unhandled event:%d sent to req:%p\n", evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+static void
+csio_scsis_tm_active(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+
+ switch (evt) {
+ case CSIO_SCSIE_COMPLETED:
+ CSIO_DEC_STATS(scm, n_tm_active);
+ list_del_init(&req->sm.sm_list);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+
+ break;
+
+ case CSIO_SCSIE_ABORT:
+ csio_scsi_abrt_cls(req, SCSI_ABORT);
+ if (req->drv_status == 0) {
+ csio_wr_issue(hw, req->eq_idx, false);
+ csio_set_state(&req->sm, csio_scsis_aborting);
+ }
+ break;
+
+
+ case CSIO_SCSIE_CLOSE:
+ csio_scsi_abrt_cls(req, SCSI_CLOSE);
+ if (req->drv_status == 0) {
+ csio_wr_issue(hw, req->eq_idx, false);
+ csio_set_state(&req->sm, csio_scsis_closing);
+ }
+ break;
+
+ case CSIO_SCSIE_DRVCLEANUP:
+ req->wr_status = FW_HOSTERROR;
+ CSIO_DEC_STATS(scm, n_tm_active);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ default:
+ csio_dbg(hw, "Unhandled event:%d sent to req:%p\n", evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+static void
+csio_scsis_aborting(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+
+ switch (evt) {
+ case CSIO_SCSIE_COMPLETED:
+ csio_dbg(hw,
+ "ioreq %p recvd cmpltd (wr_status:%d) "
+ "in aborting st\n", req, req->wr_status);
+ /*
+ * Use -ECANCELED to explicitly tell the ABORTED event that
+ * the original I/O was returned to driver by FW.
+ * We dont really care if the I/O was returned with success by
+ * FW (because the ABORT and completion of the I/O crossed each
+ * other), or any other return value. Once we are in aborting
+ * state, the success or failure of the I/O is unimportant to
+ * us.
+ */
+ req->drv_status = -ECANCELED;
+ break;
+
+ case CSIO_SCSIE_ABORT:
+ CSIO_INC_STATS(scm, n_abrt_dups);
+ break;
+
+ case CSIO_SCSIE_ABORTED:
+
+ csio_dbg(hw, "abort of %p return status:0x%x drv_status:%x\n",
+ req, req->wr_status, req->drv_status);
+ /*
+ * Check if original I/O WR completed before the Abort
+ * completion.
+ */
+ if (req->drv_status != -ECANCELED) {
+ csio_warn(hw,
+ "Abort completed before original I/O,"
+ " req:%p\n", req);
+ CSIO_DB_ASSERT(0);
+ }
+
+ /*
+ * There are the following possible scenarios:
+ * 1. The abort completed successfully, FW returned FW_SUCCESS.
+ * 2. The completion of an I/O and the receipt of
+ * abort for that I/O by the FW crossed each other.
+ * The FW returned FW_EINVAL. The original I/O would have
+ * returned with FW_SUCCESS or any other SCSI error.
+ * 3. The FW couldnt sent the abort out on the wire, as there
+ * was an I-T nexus loss (link down, remote device logged
+ * out etc). FW sent back an appropriate IT nexus loss status
+ * for the abort.
+ * 4. FW sent an abort, but abort timed out (remote device
+ * didnt respond). FW replied back with
+ * FW_SCSI_ABORT_TIMEDOUT.
+ * 5. FW couldnt genuinely abort the request for some reason,
+ * and sent us an error.
+ *
+ * The first 3 scenarios are treated as succesful abort
+ * operations by the host, while the last 2 are failed attempts
+ * to abort. Manipulate the return value of the request
+ * appropriately, so that host can convey these results
+ * back to the upper layer.
+ */
+ if ((req->wr_status == FW_SUCCESS) ||
+ (req->wr_status == FW_EINVAL) ||
+ csio_scsi_itnexus_loss_error(req->wr_status))
+ req->wr_status = FW_SCSI_ABORT_REQUESTED;
+
+ CSIO_DEC_STATS(scm, n_active);
+ list_del_init(&req->sm.sm_list);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ case CSIO_SCSIE_DRVCLEANUP:
+ req->wr_status = FW_HOSTERROR;
+ CSIO_DEC_STATS(scm, n_active);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ case CSIO_SCSIE_CLOSE:
+ /*
+ * We can receive this event from the module
+ * cleanup paths, if the FW forgot to reply to the ABORT WR
+ * and left this ioreq in this state. For now, just ignore
+ * the event. The CLOSE event is sent to this state, as
+ * the LINK may have already gone down.
+ */
+ break;
+
+ default:
+ csio_dbg(hw, "Unhandled event:%d sent to req:%p\n", evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+static void
+csio_scsis_closing(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ struct csio_hw *hw = req->lnode->hwp;
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+
+ switch (evt) {
+ case CSIO_SCSIE_COMPLETED:
+ csio_dbg(hw,
+ "ioreq %p recvd cmpltd (wr_status:%d) "
+ "in closing st\n", req, req->wr_status);
+ /*
+ * Use -ECANCELED to explicitly tell the CLOSED event that
+ * the original I/O was returned to driver by FW.
+ * We dont really care if the I/O was returned with success by
+ * FW (because the CLOSE and completion of the I/O crossed each
+ * other), or any other return value. Once we are in aborting
+ * state, the success or failure of the I/O is unimportant to
+ * us.
+ */
+ req->drv_status = -ECANCELED;
+ break;
+
+ case CSIO_SCSIE_CLOSED:
+ /*
+ * Check if original I/O WR completed before the Close
+ * completion.
+ */
+ if (req->drv_status != -ECANCELED) {
+ csio_fatal(hw,
+ "Close completed before original I/O,"
+ " req:%p\n", req);
+ CSIO_DB_ASSERT(0);
+ }
+
+ /*
+ * Either close succeeded, or we issued close to FW at the
+ * same time FW compelted it to us. Either way, the I/O
+ * is closed.
+ */
+ CSIO_DB_ASSERT((req->wr_status == FW_SUCCESS) ||
+ (req->wr_status == FW_EINVAL));
+ req->wr_status = FW_SCSI_CLOSE_REQUESTED;
+
+ CSIO_DEC_STATS(scm, n_active);
+ list_del_init(&req->sm.sm_list);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ case CSIO_SCSIE_CLOSE:
+ break;
+
+ case CSIO_SCSIE_DRVCLEANUP:
+ req->wr_status = FW_HOSTERROR;
+ CSIO_DEC_STATS(scm, n_active);
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+
+ default:
+ csio_dbg(hw, "Unhandled event:%d sent to req:%p\n", evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+static void
+csio_scsis_shost_cmpl_await(struct csio_ioreq *req, enum csio_scsi_ev evt)
+{
+ switch (evt) {
+ case CSIO_SCSIE_ABORT:
+ case CSIO_SCSIE_CLOSE:
+ /*
+ * Just succeed the abort request, and hope that
+ * the remote device unregister path will cleanup
+ * this I/O to the upper layer within a sane
+ * amount of time.
+ */
+ /*
+ * A close can come in during a LINK DOWN. The FW would have
+ * returned us the I/O back, but not the remote device lost
+ * FW event. In this interval, if the I/O times out at the upper
+ * layer, a close can come in. Take the same action as abort:
+ * return success, and hope that the remote device unregister
+ * path will cleanup this I/O. If the FW still doesnt send
+ * the msg, the close times out, and the upper layer resorts
+ * to the next level of error recovery.
+ */
+ req->drv_status = 0;
+ break;
+ case CSIO_SCSIE_DRVCLEANUP:
+ csio_set_state(&req->sm, csio_scsis_uninit);
+ break;
+ default:
+ csio_dbg(req->lnode->hwp, "Unhandled event:%d sent to req:%p\n",
+ evt, req);
+ CSIO_DB_ASSERT(0);
+ }
+}
+
+/*
+ * csio_scsi_cmpl_handler - WR completion handler for SCSI.
+ * @hw: HW module.
+ * @wr: The completed WR from the ingress queue.
+ * @len: Length of the WR.
+ * @flb: Freelist buffer array.
+ * @priv: Private object
+ * @scsiwr: Pointer to SCSI WR.
+ *
+ * This is the WR completion handler called per completion from the
+ * ISR. It is called with lock held. It walks past the RSS and CPL message
+ * header where the actual WR is present.
+ * It then gets the status, WR handle (ioreq pointer) and the len of
+ * the WR, based on WR opcode. Only on a non-good status is the entire
+ * WR copied into the WR cache (ioreq->fw_wr).
+ * The ioreq corresponding to the WR is returned to the caller.
+ * NOTE: The SCSI queue doesnt allocate a freelist today, hence
+ * no freelist buffer is expected.
+ */
+struct csio_ioreq *
+csio_scsi_cmpl_handler(struct csio_hw *hw, void *wr, uint32_t len,
+ struct csio_fl_dma_buf *flb, void *priv, uint8_t **scsiwr)
+{
+ struct csio_ioreq *ioreq = NULL;
+ struct cpl_fw6_msg *cpl;
+ uint8_t *tempwr;
+ uint8_t status;
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+
+ /* skip RSS header */
+ cpl = (struct cpl_fw6_msg *)((uintptr_t)wr + sizeof(__be64));
+
+ if (unlikely(cpl->opcode != CPL_FW6_MSG)) {
+ csio_warn(hw, "Error: Invalid CPL msg %x recvd on SCSI q\n",
+ cpl->opcode);
+ CSIO_INC_STATS(scm, n_inval_cplop);
+ return NULL;
+ }
+
+ tempwr = (uint8_t *)(cpl->data);
+ status = csio_wr_status(tempwr);
+ *scsiwr = tempwr;
+
+ if (likely((*tempwr == FW_SCSI_READ_WR) ||
+ (*tempwr == FW_SCSI_WRITE_WR) ||
+ (*tempwr == FW_SCSI_CMD_WR))) {
+ ioreq = (struct csio_ioreq *)((uintptr_t)
+ (((struct fw_scsi_read_wr *)tempwr)->cookie));
+ CSIO_DB_ASSERT(virt_addr_valid(ioreq));
+
+ ioreq->wr_status = status;
+
+ return ioreq;
+ }
+
+ if (*tempwr == FW_SCSI_ABRT_CLS_WR) {
+ ioreq = (struct csio_ioreq *)((uintptr_t)
+ (((struct fw_scsi_abrt_cls_wr *)tempwr)->cookie));
+ CSIO_DB_ASSERT(virt_addr_valid(ioreq));
+
+ ioreq->wr_status = status;
+ return ioreq;
+ }
+
+ csio_warn(hw, "WR with invalid opcode in SCSI IQ: %x\n", *tempwr);
+ CSIO_INC_STATS(scm, n_inval_scsiop);
+ return NULL;
+}
+
+/*
+ * csio_scsi_cleanup_io_q - Cleanup the given queue.
+ * @scm: SCSI module.
+ * @q: Queue to be cleaned up.
+ *
+ * Called with lock held. Has to exit with lock held.
+ */
+void
+csio_scsi_cleanup_io_q(struct csio_scsim *scm, struct list_head *q)
+{
+ struct csio_hw *hw = scm->hw;
+ struct csio_ioreq *ioreq;
+ struct list_head *tmp, *next;
+ struct scsi_cmnd *scmnd;
+
+ /* Call back the completion routines of the active_q */
+ list_for_each_safe(tmp, next, q) {
+ ioreq = (struct csio_ioreq *)tmp;
+ csio_scsi_drvcleanup(ioreq);
+ list_del_init(&ioreq->sm.sm_list);
+ scmnd = csio_scsi_cmnd(ioreq);
+ spin_unlock_irq(&hw->lock);
+
+ /*
+ * Upper layers may have cleared this command, hence this
+ * check to avoid accessing stale references.
+ */
+ if (scmnd != NULL)
+ ioreq->io_cbfn(hw, ioreq);
+
+ spin_lock_irq(&scm->freelist_lock);
+ csio_put_scsi_ioreq(scm, ioreq);
+ spin_unlock_irq(&scm->freelist_lock);
+
+ spin_lock_irq(&hw->lock);
+ }
+}
+
+#define CSIO_SCSI_ABORT_Q_POLL_MS 2000
+
+static void
+csio_abrt_cls(struct csio_ioreq *ioreq, struct scsi_cmnd *scmnd)
+{
+ struct csio_lnode *ln = ioreq->lnode;
+ struct csio_hw *hw = ln->hwp;
+ int ready = 0;
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+ int rv;
+
+ if (csio_scsi_cmnd(ioreq) != scmnd) {
+ CSIO_INC_STATS(scsim, n_abrt_race_comp);
+ return;
+ }
+
+ ready = csio_is_lnode_ready(ln);
+
+ rv = csio_do_abrt_cls(hw, ioreq, (ready ? SCSI_ABORT : SCSI_CLOSE));
+ if (rv != 0) {
+ if (ready)
+ CSIO_INC_STATS(scsim, n_abrt_busy_error);
+ else
+ CSIO_INC_STATS(scsim, n_cls_busy_error);
+ }
+}
+
+/*
+ * csio_scsi_abort_io_q - Abort all I/Os on given queue
+ * @scm: SCSI module.
+ * @q: Queue to abort.
+ * @tmo: Timeout in ms
+ *
+ * Attempt to abort all I/Os on given queue, and wait for a max
+ * of tmo milliseconds for them to complete. Returns success
+ * if all I/Os are aborted. Else returns -ETIMEDOUT.
+ * Should be entered with lock held. Exits with lock held.
+ * NOTE:
+ * Lock has to be held across the loop that aborts I/Os, since dropping the lock
+ * in between can cause the list to be corrupted. As a result, the caller
+ * of this function has to ensure that the number of I/os to be aborted
+ * is finite enough to not cause lock-held-for-too-long issues.
+ */
+static int
+csio_scsi_abort_io_q(struct csio_scsim *scm, struct list_head *q, uint32_t tmo)
+{
+ struct csio_hw *hw = scm->hw;
+ struct list_head *tmp, *next;
+ int count = DIV_ROUND_UP(tmo, CSIO_SCSI_ABORT_Q_POLL_MS);
+ struct scsi_cmnd *scmnd;
+
+ if (list_empty(q))
+ return 0;
+
+ csio_dbg(hw, "Aborting SCSI I/Os\n");
+
+ /* Now abort/close I/Os in the queue passed */
+ list_for_each_safe(tmp, next, q) {
+ scmnd = csio_scsi_cmnd((struct csio_ioreq *)tmp);
+ csio_abrt_cls((struct csio_ioreq *)tmp, scmnd);
+ }
+
+ /* Wait till all active I/Os are completed/aborted/closed */
+ while (!list_empty(q) && count--) {
+ spin_unlock_irq(&hw->lock);
+ msleep(CSIO_SCSI_ABORT_Q_POLL_MS);
+ spin_lock_irq(&hw->lock);
+ }
+
+ /* all aborts completed */
+ if (list_empty(q))
+ return 0;
+
+ return -ETIMEDOUT;
+}
+
+/*
+ * csio_scsim_cleanup_io - Cleanup all I/Os in SCSI module.
+ * @scm: SCSI module.
+ * @abort: abort required.
+ * Called with lock held, should exit with lock held.
+ * Can sleep when waiting for I/Os to complete.
+ */
+int
+csio_scsim_cleanup_io(struct csio_scsim *scm, bool abort)
+{
+ struct csio_hw *hw = scm->hw;
+ int rv = 0;
+ int count = DIV_ROUND_UP(60 * 1000, CSIO_SCSI_ABORT_Q_POLL_MS);
+
+ /* No I/Os pending */
+ if (list_empty(&scm->active_q))
+ return 0;
+
+ /* Wait until all active I/Os are completed */
+ while (!list_empty(&scm->active_q) && count--) {
+ spin_unlock_irq(&hw->lock);
+ msleep(CSIO_SCSI_ABORT_Q_POLL_MS);
+ spin_lock_irq(&hw->lock);
+ }
+
+ /* all I/Os completed */
+ if (list_empty(&scm->active_q))
+ return 0;
+
+ /* Else abort */
+ if (abort) {
+ rv = csio_scsi_abort_io_q(scm, &scm->active_q, 30000);
+ if (rv == 0)
+ return rv;
+ csio_dbg(hw, "Some I/O aborts timed out, cleaning up..\n");
+ }
+
+ csio_scsi_cleanup_io_q(scm, &scm->active_q);
+
+ CSIO_DB_ASSERT(list_empty(&scm->active_q));
+
+ return rv;
+}
+
+/*
+ * csio_scsim_cleanup_io_lnode - Cleanup all I/Os of given lnode.
+ * @scm: SCSI module.
+ * @lnode: lnode
+ *
+ * Called with lock held, should exit with lock held.
+ * Can sleep (with dropped lock) when waiting for I/Os to complete.
+ */
+int
+csio_scsim_cleanup_io_lnode(struct csio_scsim *scm, struct csio_lnode *ln)
+{
+ struct csio_hw *hw = scm->hw;
+ struct csio_scsi_level_data sld;
+ int rv;
+ int count = DIV_ROUND_UP(60 * 1000, CSIO_SCSI_ABORT_Q_POLL_MS);
+
+ csio_dbg(hw, "Gathering all SCSI I/Os on lnode %p\n", ln);
+
+ sld.level = CSIO_LEV_LNODE;
+ sld.lnode = ln;
+ INIT_LIST_HEAD(&ln->cmpl_q);
+ csio_scsi_gather_active_ios(scm, &sld, &ln->cmpl_q);
+
+ /* No I/Os pending on this lnode */
+ if (list_empty(&ln->cmpl_q))
+ return 0;
+
+ /* Wait until all active I/Os on this lnode are completed */
+ while (!list_empty(&ln->cmpl_q) && count--) {
+ spin_unlock_irq(&hw->lock);
+ msleep(CSIO_SCSI_ABORT_Q_POLL_MS);
+ spin_lock_irq(&hw->lock);
+ }
+
+ /* all I/Os completed */
+ if (list_empty(&ln->cmpl_q))
+ return 0;
+
+ csio_dbg(hw, "Some I/Os pending on ln:%p, aborting them..\n", ln);
+
+ /* I/Os are pending, abort them */
+ rv = csio_scsi_abort_io_q(scm, &ln->cmpl_q, 30000);
+ if (rv != 0) {
+ csio_dbg(hw, "Some I/O aborts timed out, cleaning up..\n");
+ csio_scsi_cleanup_io_q(scm, &ln->cmpl_q);
+ }
+
+ CSIO_DB_ASSERT(list_empty(&ln->cmpl_q));
+
+ return rv;
+}
+
+static ssize_t
+csio_show_hw_state(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+
+ if (csio_is_hw_ready(hw))
+ return snprintf(buf, PAGE_SIZE, "ready\n");
+ else
+ return snprintf(buf, PAGE_SIZE, "not ready\n");
+}
+
+/* Device reset */
+static ssize_t
+csio_device_reset(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t count)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+
+ if (*buf != '1')
+ return -EINVAL;
+
+ /* Delete NPIV lnodes */
+ csio_lnodes_exit(hw, 1);
+
+ /* Block upper IOs */
+ csio_lnodes_block_request(hw);
+
+ spin_lock_irq(&hw->lock);
+ csio_hw_reset(hw);
+ spin_unlock_irq(&hw->lock);
+
+ /* Unblock upper IOs */
+ csio_lnodes_unblock_request(hw);
+ return count;
+}
+
+/* disable port */
+static ssize_t
+csio_disable_port(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t count)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+ bool disable;
+
+ if (*buf == '1' || *buf == '0')
+ disable = (*buf == '1') ? true : false;
+ else
+ return -EINVAL;
+
+ /* Block upper IOs */
+ csio_lnodes_block_by_port(hw, ln->portid);
+
+ spin_lock_irq(&hw->lock);
+ csio_disable_lnodes(hw, ln->portid, disable);
+ spin_unlock_irq(&hw->lock);
+
+ /* Unblock upper IOs */
+ csio_lnodes_unblock_by_port(hw, ln->portid);
+ return count;
+}
+
+/* Show debug level */
+static ssize_t
+csio_show_dbg_level(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+
+ return snprintf(buf, PAGE_SIZE, "%x\n", ln->params.log_level);
+}
+
+/* Store debug level */
+static ssize_t
+csio_store_dbg_level(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t count)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+ uint32_t dbg_level = 0;
+
+ if (!isdigit(buf[0]))
+ return -EINVAL;
+
+ if (sscanf(buf, "%i", &dbg_level))
+ return -EINVAL;
+
+ ln->params.log_level = dbg_level;
+ hw->params.log_level = dbg_level;
+
+ return 0;
+}
+
+static DEVICE_ATTR(hw_state, S_IRUGO, csio_show_hw_state, NULL);
+static DEVICE_ATTR(device_reset, S_IRUGO | S_IWUSR, NULL, csio_device_reset);
+static DEVICE_ATTR(disable_port, S_IRUGO | S_IWUSR, NULL, csio_disable_port);
+static DEVICE_ATTR(dbg_level, S_IRUGO | S_IWUSR, csio_show_dbg_level,
+ csio_store_dbg_level);
+
+static struct device_attribute *csio_fcoe_lport_attrs[] = {
+ &dev_attr_hw_state,
+ &dev_attr_device_reset,
+ &dev_attr_disable_port,
+ &dev_attr_dbg_level,
+ NULL,
+};
+
+static ssize_t
+csio_show_num_reg_rnodes(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct csio_lnode *ln = shost_priv(class_to_shost(dev));
+
+ return snprintf(buf, PAGE_SIZE, "%d\n", ln->num_reg_rnodes);
+}
+
+static DEVICE_ATTR(num_reg_rnodes, S_IRUGO, csio_show_num_reg_rnodes, NULL);
+
+static struct device_attribute *csio_fcoe_vport_attrs[] = {
+ &dev_attr_num_reg_rnodes,
+ &dev_attr_dbg_level,
+ NULL,
+};
+
+static inline uint32_t
+csio_scsi_copy_to_sgl(struct csio_hw *hw, struct csio_ioreq *req)
+{
+ struct scsi_cmnd *scmnd = (struct scsi_cmnd *)csio_scsi_cmnd(req);
+ struct scatterlist *sg;
+ uint32_t bytes_left;
+ uint32_t bytes_copy;
+ uint32_t buf_off = 0;
+ uint32_t start_off = 0;
+ uint32_t sg_off = 0;
+ void *sg_addr;
+ void *buf_addr;
+ struct csio_dma_buf *dma_buf;
+
+ bytes_left = scsi_bufflen(scmnd);
+ sg = scsi_sglist(scmnd);
+ dma_buf = (struct csio_dma_buf *)csio_list_next(&req->gen_list);
+
+ /* Copy data from driver buffer to SGs of SCSI CMD */
+ while (bytes_left > 0 && sg && dma_buf) {
+ if (buf_off >= dma_buf->len) {
+ buf_off = 0;
+ dma_buf = (struct csio_dma_buf *)
+ csio_list_next(dma_buf);
+ continue;
+ }
+
+ if (start_off >= sg->length) {
+ start_off -= sg->length;
+ sg = sg_next(sg);
+ continue;
+ }
+
+ buf_addr = dma_buf->vaddr + buf_off;
+ sg_off = sg->offset + start_off;
+ bytes_copy = min((dma_buf->len - buf_off),
+ sg->length - start_off);
+ bytes_copy = min((uint32_t)(PAGE_SIZE - (sg_off & ~PAGE_MASK)),
+ bytes_copy);
+
+ sg_addr = kmap_atomic(sg_page(sg) + (sg_off >> PAGE_SHIFT));
+ if (!sg_addr) {
+ csio_err(hw, "failed to kmap sg:%p of ioreq:%p\n",
+ sg, req);
+ break;
+ }
+
+ csio_dbg(hw, "copy_to_sgl:sg_addr %p sg_off %d buf %p len %d\n",
+ sg_addr, sg_off, buf_addr, bytes_copy);
+ memcpy(sg_addr + (sg_off & ~PAGE_MASK), buf_addr, bytes_copy);
+ kunmap_atomic(sg_addr);
+
+ start_off += bytes_copy;
+ buf_off += bytes_copy;
+ bytes_left -= bytes_copy;
+ }
+
+ if (bytes_left > 0)
+ return DID_ERROR;
+ else
+ return DID_OK;
+}
+
+/*
+ * csio_scsi_err_handler - SCSI error handler.
+ * @hw: HW module.
+ * @req: IO request.
+ *
+ */
+static inline void
+csio_scsi_err_handler(struct csio_hw *hw, struct csio_ioreq *req)
+{
+ struct scsi_cmnd *cmnd = (struct scsi_cmnd *)csio_scsi_cmnd(req);
+ struct csio_scsim *scm = csio_hw_to_scsim(hw);
+ struct fcp_resp_with_ext *fcp_resp;
+ struct fcp_resp_rsp_info *rsp_info;
+ struct csio_dma_buf *dma_buf;
+ uint8_t flags, scsi_status = 0;
+ uint32_t host_status = DID_OK;
+ uint32_t rsp_len = 0, sns_len = 0;
+ struct csio_rnode *rn = (struct csio_rnode *)(cmnd->device->hostdata);
+
+
+ switch (req->wr_status) {
+ case FW_HOSTERROR:
+ if (unlikely(!csio_is_hw_ready(hw)))
+ return;
+
+ host_status = DID_ERROR;
+ CSIO_INC_STATS(scm, n_hosterror);
+
+ break;
+ case FW_SCSI_RSP_ERR:
+ dma_buf = &req->dma_buf;
+ fcp_resp = (struct fcp_resp_with_ext *)dma_buf->vaddr;
+ rsp_info = (struct fcp_resp_rsp_info *)(fcp_resp + 1);
+ flags = fcp_resp->resp.fr_flags;
+ scsi_status = fcp_resp->resp.fr_status;
+
+ if (flags & FCP_RSP_LEN_VAL) {
+ rsp_len = be32_to_cpu(fcp_resp->ext.fr_rsp_len);
+ if ((rsp_len != 0 && rsp_len != 4 && rsp_len != 8) ||
+ (rsp_info->rsp_code != FCP_TMF_CMPL)) {
+ host_status = DID_ERROR;
+ goto out;
+ }
+ }
+
+ if ((flags & FCP_SNS_LEN_VAL) && fcp_resp->ext.fr_sns_len) {
+ sns_len = be32_to_cpu(fcp_resp->ext.fr_sns_len);
+ if (sns_len > SCSI_SENSE_BUFFERSIZE)
+ sns_len = SCSI_SENSE_BUFFERSIZE;
+
+ memcpy(cmnd->sense_buffer,
+ &rsp_info->_fr_resvd[0] + rsp_len, sns_len);
+ CSIO_INC_STATS(scm, n_autosense);
+ }
+
+ scsi_set_resid(cmnd, 0);
+
+ /* Under run */
+ if (flags & FCP_RESID_UNDER) {
+ scsi_set_resid(cmnd,
+ be32_to_cpu(fcp_resp->ext.fr_resid));
+
+ if (!(flags & FCP_SNS_LEN_VAL) &&
+ (scsi_status == SAM_STAT_GOOD) &&
+ ((scsi_bufflen(cmnd) - scsi_get_resid(cmnd))
+ < cmnd->underflow))
+ host_status = DID_ERROR;
+ } else if (flags & FCP_RESID_OVER)
+ host_status = DID_ERROR;
+
+ CSIO_INC_STATS(scm, n_rsperror);
+ break;
+
+ case FW_SCSI_OVER_FLOW_ERR:
+ csio_warn(hw,
+ "Over-flow error,cmnd:0x%x expected len:0x%x"
+ " resid:0x%x\n", cmnd->cmnd[0],
+ scsi_bufflen(cmnd), scsi_get_resid(cmnd));
+ host_status = DID_ERROR;
+ CSIO_INC_STATS(scm, n_ovflerror);
+ break;
+
+ case FW_SCSI_UNDER_FLOW_ERR:
+ csio_warn(hw,
+ "Under-flow error,cmnd:0x%x expected"
+ " len:0x%x resid:0x%x lun:0x%x ssn:0x%x\n",
+ cmnd->cmnd[0], scsi_bufflen(cmnd),
+ scsi_get_resid(cmnd), cmnd->device->lun,
+ rn->flowid);
+ host_status = DID_ERROR;
+ CSIO_INC_STATS(scm, n_unflerror);
+ break;
+
+ case FW_SCSI_ABORT_REQUESTED:
+ case FW_SCSI_ABORTED:
+ case FW_SCSI_CLOSE_REQUESTED:
+ csio_dbg(hw, "Req %p cmd:%p op:%x %s\n", req, cmnd,
+ cmnd->cmnd[0],
+ (req->wr_status == FW_SCSI_CLOSE_REQUESTED) ?
+ "closed" : "aborted");
+ /*
+ * csio_eh_abort_handler checks this value to
+ * succeed or fail the abort request.
+ */
+ host_status = DID_REQUEUE;
+ if (req->wr_status == FW_SCSI_CLOSE_REQUESTED)
+ CSIO_INC_STATS(scm, n_closed);
+ else
+ CSIO_INC_STATS(scm, n_aborted);
+ break;
+
+ case FW_SCSI_ABORT_TIMEDOUT:
+ /* FW timed out the abort itself */
+ csio_dbg(hw, "FW timed out abort req:%p cmnd:%p status:%x\n",
+ req, cmnd, req->wr_status);
+ host_status = DID_ERROR;
+ CSIO_INC_STATS(scm, n_abrt_timedout);
+ break;
+
+ case FW_RDEV_NOT_READY:
+ /*
+ * In firmware, a RDEV can get into this state
+ * temporarily, before moving into dissapeared/lost
+ * state. So, the driver should complete the request equivalent
+ * to device-disappeared!
+ */
+ CSIO_INC_STATS(scm, n_rdev_nr_error);
+ host_status = DID_ERROR;
+ break;
+
+ case FW_ERR_RDEV_LOST:
+ CSIO_INC_STATS(scm, n_rdev_lost_error);
+ host_status = DID_ERROR;
+ break;
+
+ case FW_ERR_RDEV_LOGO:
+ CSIO_INC_STATS(scm, n_rdev_logo_error);
+ host_status = DID_ERROR;
+ break;
+
+ case FW_ERR_RDEV_IMPL_LOGO:
+ host_status = DID_ERROR;
+ break;
+
+ case FW_ERR_LINK_DOWN:
+ CSIO_INC_STATS(scm, n_link_down_error);
+ host_status = DID_ERROR;
+ break;
+
+ case FW_FCOE_NO_XCHG:
+ CSIO_INC_STATS(scm, n_no_xchg_error);
+ host_status = DID_ERROR;
+ break;
+
+ default:
+ csio_err(hw, "Unknown SCSI FW WR status:%d req:%p cmnd:%p\n",
+ req->wr_status, req, cmnd);
+ CSIO_DB_ASSERT(0);
+
+ CSIO_INC_STATS(scm, n_unknown_error);
+ host_status = DID_ERROR;
+ break;
+ }
+
+out:
+ if (req->nsge > 0)
+ scsi_dma_unmap(cmnd);
+
+ cmnd->result = (((host_status) << 16) | scsi_status);
+ cmnd->scsi_done(cmnd);
+
+ /* Wake up waiting threads */
+ csio_scsi_cmnd(req) = NULL;
+ complete_all(&req->cmplobj);
+}
+
+/*
+ * csio_scsi_cbfn - SCSI callback function.
+ * @hw: HW module.
+ * @req: IO request.
+ *
+ */
+static void
+csio_scsi_cbfn(struct csio_hw *hw, struct csio_ioreq *req)
+{
+ struct scsi_cmnd *cmnd = (struct scsi_cmnd *)csio_scsi_cmnd(req);
+ uint8_t scsi_status = SAM_STAT_GOOD;
+ uint32_t host_status = DID_OK;
+
+ if (likely(req->wr_status == FW_SUCCESS)) {
+ if (req->nsge > 0) {
+ scsi_dma_unmap(cmnd);
+ if (req->dcopy)
+ host_status = csio_scsi_copy_to_sgl(hw, req);
+ }
+
+ cmnd->result = (((host_status) << 16) | scsi_status);
+ cmnd->scsi_done(cmnd);
+ csio_scsi_cmnd(req) = NULL;
+ CSIO_INC_STATS(csio_hw_to_scsim(hw), n_tot_success);
+ } else {
+ /* Error handling */
+ csio_scsi_err_handler(hw, req);
+ }
+}
+
+/**
+ * csio_queuecommand - Entry point to kickstart an I/O request.
+ * @host: The scsi_host pointer.
+ * @cmnd: The I/O request from ML.
+ *
+ * This routine does the following:
+ * - Checks for HW and Rnode module readiness.
+ * - Gets a free ioreq structure (which is already initialized
+ * to uninit during its allocation).
+ * - Maps SG elements.
+ * - Initializes ioreq members.
+ * - Kicks off the SCSI state machine for this IO.
+ * - Returns busy status on error.
+ */
+static int
+csio_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmnd)
+{
+ struct csio_lnode *ln = shost_priv(host);
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+ struct csio_rnode *rn = (struct csio_rnode *)(cmnd->device->hostdata);
+ struct csio_ioreq *ioreq = NULL;
+ unsigned long flags;
+ int nsge = 0;
+ int rv = SCSI_MLQUEUE_HOST_BUSY, nr;
+ int retval;
+ int cpu;
+ struct csio_scsi_qset *sqset;
+ struct fc_rport *rport = starget_to_rport(scsi_target(cmnd->device));
+
+ if (!blk_rq_cpu_valid(cmnd->request))
+ cpu = smp_processor_id();
+ else
+ cpu = cmnd->request->cpu;
+
+ sqset = &hw->sqset[ln->portid][cpu];
+
+ nr = fc_remote_port_chkready(rport);
+ if (nr) {
+ cmnd->result = nr;
+ CSIO_INC_STATS(scsim, n_rn_nr_error);
+ goto err_done;
+ }
+
+ if (unlikely(!csio_is_hw_ready(hw))) {
+ cmnd->result = (DID_REQUEUE << 16);
+ CSIO_INC_STATS(scsim, n_hw_nr_error);
+ goto err_done;
+ }
+
+ /* Get req->nsge, if there are SG elements to be mapped */
+ nsge = scsi_dma_map(cmnd);
+ if (unlikely(nsge < 0)) {
+ CSIO_INC_STATS(scsim, n_dmamap_error);
+ goto err;
+ }
+
+ /* Do we support so many mappings? */
+ if (unlikely(nsge > scsim->max_sge)) {
+ csio_warn(hw,
+ "More SGEs than can be supported."
+ " SGEs: %d, Max SGEs: %d\n", nsge, scsim->max_sge);
+ CSIO_INC_STATS(scsim, n_unsupp_sge_error);
+ goto err_dma_unmap;
+ }
+
+ /* Get a free ioreq structure - SM is already set to uninit */
+ ioreq = csio_get_scsi_ioreq_lock(hw, scsim);
+ if (!ioreq) {
+ csio_err(hw, "Out of I/O request elements. Active #:%d\n",
+ scsim->stats.n_active);
+ CSIO_INC_STATS(scsim, n_no_req_error);
+ goto err_dma_unmap;
+ }
+
+ ioreq->nsge = nsge;
+ ioreq->lnode = ln;
+ ioreq->rnode = rn;
+ ioreq->iq_idx = sqset->iq_idx;
+ ioreq->eq_idx = sqset->eq_idx;
+ ioreq->wr_status = 0;
+ ioreq->drv_status = 0;
+ csio_scsi_cmnd(ioreq) = (void *)cmnd;
+ ioreq->tmo = 0;
+ ioreq->datadir = cmnd->sc_data_direction;
+
+ if (cmnd->sc_data_direction == DMA_TO_DEVICE) {
+ CSIO_INC_STATS(ln, n_output_requests);
+ ln->stats.n_output_bytes += scsi_bufflen(cmnd);
+ } else if (cmnd->sc_data_direction == DMA_FROM_DEVICE) {
+ CSIO_INC_STATS(ln, n_input_requests);
+ ln->stats.n_input_bytes += scsi_bufflen(cmnd);
+ } else
+ CSIO_INC_STATS(ln, n_control_requests);
+
+ /* Set cbfn */
+ ioreq->io_cbfn = csio_scsi_cbfn;
+
+ /* Needed during abort */
+ cmnd->host_scribble = (unsigned char *)ioreq;
+ cmnd->SCp.Message = 0;
+
+ /* Kick off SCSI IO SM on the ioreq */
+ spin_lock_irqsave(&hw->lock, flags);
+ retval = csio_scsi_start_io(ioreq);
+ spin_unlock_irqrestore(&hw->lock, flags);
+
+ if (retval != 0) {
+ csio_err(hw, "ioreq: %p couldnt be started, status:%d\n",
+ ioreq, retval);
+ CSIO_INC_STATS(scsim, n_busy_error);
+ goto err_put_req;
+ }
+
+ return 0;
+
+err_put_req:
+ csio_put_scsi_ioreq_lock(hw, scsim, ioreq);
+err_dma_unmap:
+ if (nsge > 0)
+ scsi_dma_unmap(cmnd);
+err:
+ return rv;
+
+err_done:
+ cmnd->scsi_done(cmnd);
+ return 0;
+}
+
+static int
+csio_do_abrt_cls(struct csio_hw *hw, struct csio_ioreq *ioreq, bool abort)
+{
+ int rv;
+ int cpu = smp_processor_id();
+ struct csio_lnode *ln = ioreq->lnode;
+ struct csio_scsi_qset *sqset = &hw->sqset[ln->portid][cpu];
+
+ ioreq->tmo = CSIO_SCSI_ABRT_TMO_MS;
+ /*
+ * Use current processor queue for posting the abort/close, but retain
+ * the ingress queue ID of the original I/O being aborted/closed - we
+ * need the abort/close completion to be received on the same queue
+ * as the original I/O.
+ */
+ ioreq->eq_idx = sqset->eq_idx;
+
+ if (abort == SCSI_ABORT)
+ rv = csio_scsi_abort(ioreq);
+ else
+ rv = csio_scsi_close(ioreq);
+
+ return rv;
+}
+
+static int
+csio_eh_abort_handler(struct scsi_cmnd *cmnd)
+{
+ struct csio_ioreq *ioreq;
+ struct csio_lnode *ln = shost_priv(cmnd->device->host);
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+ int ready = 0, ret;
+ unsigned long tmo = 0;
+ int rv;
+ struct csio_rnode *rn = (struct csio_rnode *)(cmnd->device->hostdata);
+
+ ret = fc_block_scsi_eh(cmnd);
+ if (ret)
+ return ret;
+
+ ioreq = (struct csio_ioreq *)cmnd->host_scribble;
+ if (!ioreq)
+ return SUCCESS;
+
+ if (!rn)
+ return FAILED;
+
+ csio_dbg(hw,
+ "Request to abort ioreq:%p cmd:%p cdb:%08llx"
+ " ssni:0x%x lun:%d iq:0x%x\n",
+ ioreq, cmnd, *((uint64_t *)cmnd->cmnd), rn->flowid,
+ cmnd->device->lun, csio_q_physiqid(hw, ioreq->iq_idx));
+
+ if (((struct scsi_cmnd *)csio_scsi_cmnd(ioreq)) != cmnd) {
+ CSIO_INC_STATS(scsim, n_abrt_race_comp);
+ return SUCCESS;
+ }
+
+ ready = csio_is_lnode_ready(ln);
+ tmo = CSIO_SCSI_ABRT_TMO_MS;
+
+ spin_lock_irq(&hw->lock);
+ rv = csio_do_abrt_cls(hw, ioreq, (ready ? SCSI_ABORT : SCSI_CLOSE));
+ spin_unlock_irq(&hw->lock);
+
+ if (rv != 0) {
+ if (rv == -EINVAL) {
+ /* Return success, if abort/close request issued on
+ * already completed IO
+ */
+ return SUCCESS;
+ }
+ if (ready)
+ CSIO_INC_STATS(scsim, n_abrt_busy_error);
+ else
+ CSIO_INC_STATS(scsim, n_cls_busy_error);
+
+ goto inval_scmnd;
+ }
+
+ /* Wait for completion */
+ init_completion(&ioreq->cmplobj);
+ wait_for_completion_timeout(&ioreq->cmplobj, msecs_to_jiffies(tmo));
+
+ /* FW didnt respond to abort within our timeout */
+ if (((struct scsi_cmnd *)csio_scsi_cmnd(ioreq)) == cmnd) {
+
+ csio_err(hw, "Abort timed out -- req: %p\n", ioreq);
+ CSIO_INC_STATS(scsim, n_abrt_timedout);
+
+inval_scmnd:
+ if (ioreq->nsge > 0)
+ scsi_dma_unmap(cmnd);
+
+ spin_lock_irq(&hw->lock);
+ csio_scsi_cmnd(ioreq) = NULL;
+ spin_unlock_irq(&hw->lock);
+
+ cmnd->result = (DID_ERROR << 16);
+ cmnd->scsi_done(cmnd);
+
+ return FAILED;
+ }
+
+ /* FW successfully aborted the request */
+ if (host_byte(cmnd->result) == DID_REQUEUE) {
+ csio_info(hw,
+ "Aborted SCSI command to (%d:%d) serial#:0x%lx\n",
+ cmnd->device->id, cmnd->device->lun,
+ cmnd->serial_number);
+ return SUCCESS;
+ } else {
+ csio_info(hw,
+ "Failed to abort SCSI command, (%d:%d) serial#:0x%lx\n",
+ cmnd->device->id, cmnd->device->lun,
+ cmnd->serial_number);
+ return FAILED;
+ }
+}
+
+/*
+ * csio_tm_cbfn - TM callback function.
+ * @hw: HW module.
+ * @req: IO request.
+ *
+ * Cache the result in 'cmnd', since ioreq will be freed soon
+ * after we return from here, and the waiting thread shouldnt trust
+ * the ioreq contents.
+ */
+static void
+csio_tm_cbfn(struct csio_hw *hw, struct csio_ioreq *req)
+{
+ struct scsi_cmnd *cmnd = (struct scsi_cmnd *)csio_scsi_cmnd(req);
+ struct csio_dma_buf *dma_buf;
+ uint8_t flags = 0;
+ struct fcp_resp_with_ext *fcp_resp;
+ struct fcp_resp_rsp_info *rsp_info;
+
+ csio_dbg(hw, "req: %p in csio_tm_cbfn status: %d\n",
+ req, req->wr_status);
+
+ /* Cache FW return status */
+ cmnd->SCp.Status = req->wr_status;
+
+ /* Special handling based on FCP response */
+
+ /*
+ * FW returns us this error, if flags were set. FCP4 says
+ * FCP_RSP_LEN_VAL in flags shall be set for TM completions.
+ * So if a target were to set this bit, we expect that the
+ * rsp_code is set to FCP_TMF_CMPL for a successful TM
+ * completion. Any other rsp_code means TM operation failed.
+ * If a target were to just ignore setting flags, we treat
+ * the TM operation as success, and FW returns FW_SUCCESS.
+ */
+ if (req->wr_status == FW_SCSI_RSP_ERR) {
+ dma_buf = &req->dma_buf;
+ fcp_resp = (struct fcp_resp_with_ext *)dma_buf->vaddr;
+ rsp_info = (struct fcp_resp_rsp_info *)(fcp_resp + 1);
+
+ flags = fcp_resp->resp.fr_flags;
+
+ /* Modify return status if flags indicate success */
+ if (flags & FCP_RSP_LEN_VAL)
+ if (rsp_info->rsp_code == FCP_TMF_CMPL)
+ cmnd->SCp.Status = FW_SUCCESS;
+
+ csio_dbg(hw, "TM FCP rsp code: %d\n", rsp_info->rsp_code);
+ }
+
+ /* Wake up the TM handler thread */
+ csio_scsi_cmnd(req) = NULL;
+}
+
+static int
+csio_eh_lun_reset_handler(struct scsi_cmnd *cmnd)
+{
+ struct csio_lnode *ln = shost_priv(cmnd->device->host);
+ struct csio_hw *hw = csio_lnode_to_hw(ln);
+ struct csio_scsim *scsim = csio_hw_to_scsim(hw);
+ struct csio_rnode *rn = (struct csio_rnode *)(cmnd->device->hostdata);
+ struct csio_ioreq *ioreq = NULL;
+ struct csio_scsi_qset *sqset;
+ unsigned long flags;
+ int retval;
+ int count, ret;
+ LIST_HEAD(local_q);
+ struct csio_scsi_level_data sld;
+
+ if (!rn)
+ goto fail;
+
+ csio_dbg(hw, "Request to reset LUN:%d (ssni:0x%x tgtid:%d)\n",
+ cmnd->device->lun, rn->flowid, rn->scsi_id);
+
+ if (!csio_is_lnode_ready(ln)) {
+ csio_err(hw,
+ "LUN reset cannot be issued on non-ready"
+ " local node vnpi:0x%x (LUN:%d)\n",
+ ln->vnp_flowid, cmnd->device->lun);
+ goto fail;
+ }
+
+ /* Lnode is ready, now wait on rport node readiness */
+ ret = fc_block_scsi_eh(cmnd);
+ if (ret)
+ return ret;
+
+ /*
+ * If we have blocked in the previous call, at this point, either the
+ * remote node has come back online, or device loss timer has fired
+ * and the remote node is destroyed. Allow the LUN reset only for
+ * the former case, since LUN reset is a TMF I/O on the wire, and we
+ * need a valid session to issue it.
+ */
+ if (fc_remote_port_chkready(rn->rport)) {
+ csio_err(hw,
+ "LUN reset cannot be issued on non-ready"
+ " remote node ssni:0x%x (LUN:%d)\n",
+ rn->flowid, cmnd->device->lun);
+ goto fail;
+ }
+
+ /* Get a free ioreq structure - SM is already set to uninit */
+ ioreq = csio_get_scsi_ioreq_lock(hw, scsim);
+
+ if (!ioreq) {
+ csio_err(hw, "Out of IO request elements. Active # :%d\n",
+ scsim->stats.n_active);
+ goto fail;
+ }
+
+ sqset = &hw->sqset[ln->portid][smp_processor_id()];
+ ioreq->nsge = 0;
+ ioreq->lnode = ln;
+ ioreq->rnode = rn;
+ ioreq->iq_idx = sqset->iq_idx;
+ ioreq->eq_idx = sqset->eq_idx;
+
+ csio_scsi_cmnd(ioreq) = cmnd;
+ cmnd->host_scribble = (unsigned char *)ioreq;
+ cmnd->SCp.Status = 0;
+
+ cmnd->SCp.Message = FCP_TMF_LUN_RESET;
+ ioreq->tmo = CSIO_SCSI_LUNRST_TMO_MS / 1000;
+
+ /*
+ * FW times the LUN reset for ioreq->tmo, so we got to wait a little
+ * longer (10s for now) than that to allow FW to return the timed
+ * out command.
+ */
+ count = DIV_ROUND_UP((ioreq->tmo + 10) * 1000, CSIO_SCSI_TM_POLL_MS);
+
+ /* Set cbfn */
+ ioreq->io_cbfn = csio_tm_cbfn;
+
+ /* Save of the ioreq info for later use */
+ sld.level = CSIO_LEV_LUN;
+ sld.lnode = ioreq->lnode;
+ sld.rnode = ioreq->rnode;
+ sld.oslun = (uint64_t)cmnd->device->lun;
+
+ spin_lock_irqsave(&hw->lock, flags);
+ /* Kick off TM SM on the ioreq */
+ retval = csio_scsi_start_tm(ioreq);
+ spin_unlock_irqrestore(&hw->lock, flags);
+
+ if (retval != 0) {
+ csio_err(hw, "Failed to issue LUN reset, req:%p, status:%d\n",
+ ioreq, retval);
+ goto fail_ret_ioreq;
+ }
+
+ csio_dbg(hw, "Waiting max %d secs for LUN reset completion\n",
+ count * (CSIO_SCSI_TM_POLL_MS / 1000));
+ /* Wait for completion */
+ while ((((struct scsi_cmnd *)csio_scsi_cmnd(ioreq)) == cmnd)
+ && count--)
+ msleep(CSIO_SCSI_TM_POLL_MS);
+
+ /* LUN reset timed-out */
+ if (((struct scsi_cmnd *)csio_scsi_cmnd(ioreq)) == cmnd) {
+ csio_err(hw, "LUN reset (%d:%d) timed out\n",
+ cmnd->device->id, cmnd->device->lun);
+
+ spin_lock_irq(&hw->lock);
+ csio_scsi_drvcleanup(ioreq);
+ list_del_init(&ioreq->sm.sm_list);
+ spin_unlock_irq(&hw->lock);
+
+ goto fail_ret_ioreq;
+ }
+
+ /* LUN reset returned, check cached status */
+ if (cmnd->SCp.Status != FW_SUCCESS) {
+ csio_err(hw, "LUN reset failed (%d:%d), status: %d\n",
+ cmnd->device->id, cmnd->device->lun, cmnd->SCp.Status);
+ goto fail;
+ }
+
+ /* LUN reset succeeded, Start aborting affected I/Os */
+ /*
+ * Since the host guarantees during LUN reset that there
+ * will not be any more I/Os to that LUN, until the LUN reset
+ * completes, we gather pending I/Os after the LUN reset.
+ */
+ spin_lock_irq(&hw->lock);
+ csio_scsi_gather_active_ios(scsim, &sld, &local_q);
+
+ retval = csio_scsi_abort_io_q(scsim, &local_q, 30000);
+ spin_unlock_irq(&hw->lock);
+
+ /* Aborts may have timed out */
+ if (retval != 0) {
+ csio_err(hw,
+ "Attempt to abort I/Os during LUN reset of %d"
+ " returned %d\n", cmnd->device->lun, retval);
+ /* Return I/Os back to active_q */
+ spin_lock_irq(&hw->lock);
+ list_splice_tail_init(&local_q, &scsim->active_q);
+ spin_unlock_irq(&hw->lock);
+ goto fail;
+ }
+
+ CSIO_INC_STATS(rn, n_lun_rst);
+
+ csio_info(hw, "LUN reset occurred (%d:%d)\n",
+ cmnd->device->id, cmnd->device->lun);
+
+ return SUCCESS;
+
+fail_ret_ioreq:
+ csio_put_scsi_ioreq_lock(hw, scsim, ioreq);
+fail:
+ CSIO_INC_STATS(rn, n_lun_rst_fail);
+ return FAILED;
+}
+
+static int
+csio_slave_alloc(struct scsi_device *sdev)
+{
+ struct fc_rport *rport = starget_to_rport(scsi_target(sdev));
+
+ if (!rport || fc_remote_port_chkready(rport))
+ return -ENXIO;
+
+ sdev->hostdata = *((struct csio_lnode **)(rport->dd_data));
+
+ return 0;
+}
+
+static int
+csio_slave_configure(struct scsi_device *sdev)
+{
+ if (sdev->tagged_supported)
+ scsi_activate_tcq(sdev, csio_lun_qdepth);
+ else
+ scsi_deactivate_tcq(sdev, csio_lun_qdepth);
+
+ return 0;
+}
+
+static void
+csio_slave_destroy(struct scsi_device *sdev)
+{
+ sdev->hostdata = NULL;
+}
+
+static int
+csio_scan_finished(struct Scsi_Host *shost, unsigned long time)
+{
+ struct csio_lnode *ln = shost_priv(shost);
+ int rv = 1;
+
+ spin_lock_irq(shost->host_lock);
+ if (!ln->hwp || csio_list_deleted(&ln->sm.sm_list))
+ goto out;
+
+ rv = csio_scan_done(ln, jiffies, time, csio_max_scan_tmo * HZ,
+ csio_delta_scan_tmo * HZ);
+out:
+ spin_unlock_irq(shost->host_lock);
+
+ return rv;
+}
+
+struct scsi_host_template csio_fcoe_shost_template = {
+ .module = THIS_MODULE,
+ .name = CSIO_DRV_DESC,
+ .proc_name = KBUILD_MODNAME,
+ .queuecommand = csio_queuecommand,
+ .eh_abort_handler = csio_eh_abort_handler,
+ .eh_device_reset_handler = csio_eh_lun_reset_handler,
+ .slave_alloc = csio_slave_alloc,
+ .slave_configure = csio_slave_configure,
+ .slave_destroy = csio_slave_destroy,
+ .scan_finished = csio_scan_finished,
+ .this_id = -1,
+ .sg_tablesize = CSIO_SCSI_MAX_SGE,
+ .cmd_per_lun = CSIO_MAX_CMD_PER_LUN,
+ .use_clustering = ENABLE_CLUSTERING,
+ .shost_attrs = csio_fcoe_lport_attrs,
+ .max_sectors = CSIO_MAX_SECTOR_SIZE,
+};
+
+struct scsi_host_template csio_fcoe_shost_vport_template = {
+ .module = THIS_MODULE,
+ .name = CSIO_DRV_DESC,
+ .proc_name = KBUILD_MODNAME,
+ .queuecommand = csio_queuecommand,
+ .eh_abort_handler = csio_eh_abort_handler,
+ .eh_device_reset_handler = csio_eh_lun_reset_handler,
+ .slave_alloc = csio_slave_alloc,
+ .slave_configure = csio_slave_configure,
+ .slave_destroy = csio_slave_destroy,
+ .scan_finished = csio_scan_finished,
+ .this_id = -1,
+ .sg_tablesize = CSIO_SCSI_MAX_SGE,
+ .cmd_per_lun = CSIO_MAX_CMD_PER_LUN,
+ .use_clustering = ENABLE_CLUSTERING,
+ .shost_attrs = csio_fcoe_vport_attrs,
+ .max_sectors = CSIO_MAX_SECTOR_SIZE,
+};
+
+/*
+ * csio_scsi_alloc_ddp_bufs - Allocate buffers for DDP of unaligned SGLs.
+ * @scm: SCSI Module
+ * @hw: HW device.
+ * @buf_size: buffer size
+ * @num_buf : Number of buffers.
+ *
+ * This routine allocates DMA buffers required for SCSI Data xfer, if
+ * each SGL buffer for a SCSI Read request posted by SCSI midlayer are
+ * not virtually contiguous.
+ */
+static int
+csio_scsi_alloc_ddp_bufs(struct csio_scsim *scm, struct csio_hw *hw,
+ int buf_size, int num_buf)
+{
+ int n = 0;
+ struct list_head *tmp;
+ struct csio_dma_buf *ddp_desc = NULL;
+ uint32_t unit_size = 0;
+
+ if (!num_buf)
+ return 0;
+
+ if (!buf_size)
+ return -EINVAL;
+
+ INIT_LIST_HEAD(&scm->ddp_freelist);
+
+ /* Align buf size to page size */
+ buf_size = (buf_size + PAGE_SIZE - 1) & PAGE_MASK;
+ /* Initialize dma descriptors */
+ for (n = 0; n < num_buf; n++) {
+ /* Set unit size to request size */
+ unit_size = buf_size;
+ ddp_desc = kzalloc(sizeof(struct csio_dma_buf), GFP_KERNEL);
+ if (!ddp_desc) {
+ csio_err(hw,
+ "Failed to allocate ddp descriptors,"
+ " Num allocated = %d.\n",
+ scm->stats.n_free_ddp);
+ goto no_mem;
+ }
+
+ /* Allocate Dma buffers for DDP */
+ ddp_desc->vaddr = pci_alloc_consistent(hw->pdev, unit_size,
+ &ddp_desc->paddr);
+ if (!ddp_desc->vaddr) {
+ csio_err(hw,
+ "SCSI response DMA buffer (ddp) allocation"
+ " failed!\n");
+ kfree(ddp_desc);
+ goto no_mem;
+ }
+
+ ddp_desc->len = unit_size;
+
+ /* Added it to scsi ddp freelist */
+ list_add_tail(&ddp_desc->list, &scm->ddp_freelist);
+ CSIO_INC_STATS(scm, n_free_ddp);
+ }
+
+ return 0;
+no_mem:
+ /* release dma descs back to freelist and free dma memory */
+ list_for_each(tmp, &scm->ddp_freelist) {
+ ddp_desc = (struct csio_dma_buf *) tmp;
+ tmp = csio_list_prev(tmp);
+ pci_free_consistent(hw->pdev, ddp_desc->len, ddp_desc->vaddr,
+ ddp_desc->paddr);
+ list_del_init(&ddp_desc->list);
+ kfree(ddp_desc);
+ }
+ scm->stats.n_free_ddp = 0;
+
+ return -ENOMEM;
+}
+
+/*
+ * csio_scsi_free_ddp_bufs - free DDP buffers of unaligned SGLs.
+ * @scm: SCSI Module
+ * @hw: HW device.
+ *
+ * This routine frees ddp buffers.
+ */
+static void
+csio_scsi_free_ddp_bufs(struct csio_scsim *scm, struct csio_hw *hw)
+{
+ struct list_head *tmp;
+ struct csio_dma_buf *ddp_desc;
+
+ /* release dma descs back to freelist and free dma memory */
+ list_for_each(tmp, &scm->ddp_freelist) {
+ ddp_desc = (struct csio_dma_buf *) tmp;
+ tmp = csio_list_prev(tmp);
+ pci_free_consistent(hw->pdev, ddp_desc->len, ddp_desc->vaddr,
+ ddp_desc->paddr);
+ list_del_init(&ddp_desc->list);
+ kfree(ddp_desc);
+ }
+ scm->stats.n_free_ddp = 0;
+}
+
+/**
+ * csio_scsim_init - Initialize SCSI Module
+ * @scm: SCSI Module
+ * @hw: HW module
+ *
+ */
+int
+csio_scsim_init(struct csio_scsim *scm, struct csio_hw *hw)
+{
+ int i;
+ struct csio_ioreq *ioreq;
+ struct csio_dma_buf *dma_buf;
+
+ INIT_LIST_HEAD(&scm->active_q);
+ scm->hw = hw;
+
+ scm->proto_cmd_len = sizeof(struct fcp_cmnd);
+ scm->proto_rsp_len = CSIO_SCSI_RSP_LEN;
+ scm->max_sge = CSIO_SCSI_MAX_SGE;
+
+ spin_lock_init(&scm->freelist_lock);
+
+ /* Pre-allocate ioreqs and initialize them */
+ INIT_LIST_HEAD(&scm->ioreq_freelist);
+ for (i = 0; i < csio_scsi_ioreqs; i++) {
+
+ ioreq = kzalloc(sizeof(struct csio_ioreq), GFP_KERNEL);
+ if (!ioreq) {
+ csio_err(hw,
+ "I/O request element allocation failed, "
+ " Num allocated = %d.\n",
+ scm->stats.n_free_ioreq);
+
+ goto free_ioreq;
+ }
+
+ /* Allocate Dma buffers for Response Payload */
+ dma_buf = &ioreq->dma_buf;
+ dma_buf->vaddr = pci_pool_alloc(hw->scsi_pci_pool, GFP_KERNEL,
+ &dma_buf->paddr);
+ if (!dma_buf->vaddr) {
+ csio_err(hw,
+ "SCSI response DMA buffer allocation"
+ " failed!\n");
+ kfree(ioreq);
+ goto free_ioreq;
+ }
+
+ dma_buf->len = scm->proto_rsp_len;
+
+ /* Set state to uninit */
+ csio_init_state(&ioreq->sm, csio_scsis_uninit);
+ INIT_LIST_HEAD(&ioreq->gen_list);
+ init_completion(&ioreq->cmplobj);
+
+ list_add_tail(&ioreq->sm.sm_list, &scm->ioreq_freelist);
+ CSIO_INC_STATS(scm, n_free_ioreq);
+ }
+
+ if (csio_scsi_alloc_ddp_bufs(scm, hw, PAGE_SIZE, csio_ddp_descs))
+ goto free_ioreq;
+
+ return 0;
+
+free_ioreq:
+ /*
+ * Free up existing allocations, since an error
+ * from here means we are returning for good
+ */
+ while (!list_empty(&scm->ioreq_freelist)) {
+ struct csio_sm *tmp;
+
+ tmp = list_first_entry(&scm->ioreq_freelist,
+ struct csio_sm, sm_list);
+ list_del_init(&tmp->sm_list);
+ ioreq = (struct csio_ioreq *)tmp;
+
+ dma_buf = &ioreq->dma_buf;
+ pci_pool_free(hw->scsi_pci_pool, dma_buf->vaddr,
+ dma_buf->paddr);
+
+ kfree(ioreq);
+ }
+
+ scm->stats.n_free_ioreq = 0;
+
+ return -ENOMEM;
+}
+
+/**
+ * csio_scsim_exit: Uninitialize SCSI Module
+ * @scm: SCSI Module
+ *
+ */
+void
+csio_scsim_exit(struct csio_scsim *scm)
+{
+ struct csio_ioreq *ioreq;
+ struct csio_dma_buf *dma_buf;
+
+ while (!list_empty(&scm->ioreq_freelist)) {
+ struct csio_sm *tmp;
+
+ tmp = list_first_entry(&scm->ioreq_freelist,
+ struct csio_sm, sm_list);
+ list_del_init(&tmp->sm_list);
+ ioreq = (struct csio_ioreq *)tmp;
+
+ dma_buf = &ioreq->dma_buf;
+ pci_pool_free(scm->hw->scsi_pci_pool, dma_buf->vaddr,
+ dma_buf->paddr);
+
+ kfree(ioreq);
+ }
+
+ scm->stats.n_free_ioreq = 0;
+
+ csio_scsi_free_ddp_bufs(scm, scm->hw);
+}
--
1.7.1
^ permalink raw reply related
* Re: kernel BUG at kernel/timer.c:748!
From: Eric Dumazet @ 2012-09-24 17:31 UTC (permalink / raw)
To: Dave Jones; +Cc: Yuchung Cheng, Julian Anastasov, netdev
In-Reply-To: <20120924171152.GA26454@redhat.com>
On Mon, 2012-09-24 at 13:11 -0400, Dave Jones wrote:
> Great, I'll give this a shot.
>
> Any idea why this only just started triggering ?
> (Ie, do we need this for stable too?)
Seems this is a very old bug.
I guess your trinity tool gets better ?
I used following program to trigger the bug :
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in.h>
#include <stdio.h>
#include <string.h>
int main(int argc, char *argv[])
{
int on = 1;
struct sockaddr_in addr;
int raw_sock = socket(AF_INET, SOCK_RAW, IPPROTO_TCP);
memset(&addr, 0, sizeof(addr));
addr.sin_family = AF_INET;
addr.sin_addr.s_addr = htonl(0x7f000001);
addr.sin_port = 123;
connect(raw_sock, (struct sockaddr *)&addr, sizeof(addr));
setsockopt(raw_sock, SOL_SOCKET, SO_KEEPALIVE, &on, sizeof(on));
return 0;
}
^ permalink raw reply
* [PATCH net-next] x86: bpf_jit_comp: add XOR instruction for BPF JIT
From: Daniel Borkmann @ 2012-09-24 17:34 UTC (permalink / raw)
To: davem; +Cc: eric.dumazet, netdev
This patch is a follow-up for patch "filter: add XOR instruction for use
with X/K" that implements BPF x86 JIT parts for the BPF XOR operation.
Signed-off-by: Daniel Borkmann <daniel.borkmann@tik.ee.ethz.ch>
---
arch/x86/net/bpf_jit_comp.c | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 106c578..520d2bd 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -335,9 +335,18 @@ void bpf_jit_compile(struct sk_filter *fp)
EMIT1_off32(0x0d, K); /* or imm32,%eax */
break;
case BPF_S_ANC_ALU_XOR_X: /* A ^= X; */
+ case BPF_S_ALU_XOR_X:
seen |= SEEN_XREG;
EMIT2(0x31, 0xd8); /* xor %ebx,%eax */
break;
+ case BPF_S_ALU_XOR_K: /* A ^= K; */
+ if (K == 0)
+ break;
+ if (is_imm8(K))
+ EMIT3(0x83, 0xf0, K); /* xor imm8,%eax */
+ else
+ EMIT1_off32(0x35, K); /* xor imm32,%eax */
+ break;
case BPF_S_ALU_LSH_X: /* A <<= X; */
seen |= SEEN_XREG;
EMIT4(0x89, 0xd9, 0xd3, 0xe0); /* mov %ebx,%ecx; shl %cl,%eax */
^ permalink raw reply related
* Re: New commands to configure IOV features
From: Don Dutile @ 2012-09-24 17:37 UTC (permalink / raw)
To: Yuval Mintz
Cc: Yinghai Lu, Rose, Gregory V, Ben Hutchings, Bjorn Helgaas,
davem@davemloft.net, netdev@vger.kernel.org, Ariel Elior,
Eilon Greenstein, linux-pci
In-Reply-To: <505F2F99.9090302@broadcom.com>
On 09/23/2012 11:49 AM, Yuval Mintz wrote:
>>>> It provides 3 sysfs files (enable, disable, num_max_vfs);
>>>> callbacks for enable& disable.
>>>
>>> why using three? only pass max_vfs should be enough...
>>>
>>> aka pass 0 mean disabling, pass other value mean enabling.
>>>
>> could do that. but I wouldn't use 'max_vfs'; I would recommend
>> 'num_vfs', as max implies, er, um, the maximum, and what one
>> wants is to be able to enable a number from 1->max_vfs.
>> 'max_vfs' will be provided by another file.
>>
>
> If we're at it, how do you suggest the configurations of the VFs
> resources/properties should be made?
> Notice I'm asking about properties which relate to all PCI devices,
> E.g., number of interrupts assigned to each VF (for VF RSS).
>
This issue/question is indep of VFs & SRIOV.
It's a generic problem for PFs and how many (msi) intr's a device
is given, based on its request, i.e., system provides 8 MSI per cpu
per device ... how are those 8 assigned/used by the device? ...
how to free 4 of the 8 if one device doesn't use them, and another
device could use them.. cache/memory affinity to intr's.... the list
goes on...
> (For properties which relate only to networking devices, we could
> use standard network API such as netlink).
>
> Perhaps given a configurable property that applies to all PCI devices,
A VF is 'just another PCI device', so configurable properties
for all PCI devices should encompass VF/SRIOV devices equally.
> once SRIOV is enabled and the VFs are created, there would be an
> additional sysfs node under the VFs through which the user could
> configure that property.
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pci" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH net-next] x86: bpf_jit_comp: add XOR instruction for BPF JIT
From: Eric Dumazet @ 2012-09-24 17:44 UTC (permalink / raw)
To: Daniel Borkmann; +Cc: davem, netdev
In-Reply-To: <20120924173450.GA28969@thinkbox>
On Mon, 2012-09-24 at 19:34 +0200, Daniel Borkmann wrote:
> This patch is a follow-up for patch "filter: add XOR instruction for use
> with X/K" that implements BPF x86 JIT parts for the BPF XOR operation.
>
> Signed-off-by: Daniel Borkmann <daniel.borkmann@tik.ee.ethz.ch>
> ---
> arch/x86/net/bpf_jit_comp.c | 9 +++++++++
> 1 files changed, 9 insertions(+), 0 deletions(-)
Acked-by: Eric Dumazet <edumazet@google.com>
^ permalink raw reply
* Re: [net-next 0/6][pull request] Intel Wired LAN Driver Updates
From: David Miller @ 2012-09-24 17:48 UTC (permalink / raw)
To: jeffrey.t.kirsher; +Cc: netdev, gospo, sassmann
In-Reply-To: <1348477207-27030-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 24 Sep 2012 02:00:01 -0700
> This series contains updates to ixgbe and ixgbevf (mainly ixgbevf).
>
> The following are changes since commit c9d2ea96ca3bbc85264803ff6bd66eb3bbefdb77:
> netlink: Rearrange netlink_kernel_cfg to save space on 64-bit.
> and are available in the git repository at:
> git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master
>
> Alexander Duyck (3):
> ixgbevf: Fix code for handling timeout
> ixgbevf: Add fix to VF to handle multi-descriptor buffers
> ixgbevf: Return error on failure to enable VLAN
>
> Greg Rose (2):
> ixgbe: Do not read the spoofed packets counter when not in IOV mode
> ixgbevf: Fix AIM (Adaptive Interrupt Moderation)
>
> Narendra K (1):
> ixgbevf - Remove unused parameter in ixgbevf_receive_skb
Pulled, thanks Jeff.
^ permalink raw reply
* ixgbe unstable performance at 1Gb/s
From: Charles @ 2012-09-24 17:47 UTC (permalink / raw)
To: netdev
Hi,
I hope I am posting on the right mailing-list. If not, sorry; please redirect me
to the right place. Thanks.
I have a new motherboard with integrated Intel X540 10GBase-T. For now, I want
to use it at 1Gb/s.
The bandwidth is only of ~300 Mbit/s (with Iperf). It's actually very unstable
(always varies between 100 to 800 Mbit/s during the transfer).
If I do "ethtool -K eth1 ntuple on", things are a bit better but still very
unstable bandwidth. If I add the kernel parameter "pci=nomsi", everything is
fine (and normal, with the same performance, ~950 MBit/s, as a 1Gbit/s NIC).
I also run the set_irq_affinity.sh script, w/o noticing a difference.
The only weird thing I have found is with ethtoll -S eth1:
tx_queue_0_packets: 304262
tx_queue_1_packets: 0
tx_queue_2_packets: 0
tx_queue_3_packets: 1
tx_queue_4_packets: 502939
tx_queue_5_packets: 677967
tx_queue_6_packets: 459869
tx_queue_7_packets: 0
tx_queue_8_packets: 0
tx_queue_9_packets: 0
tx_queue_10_packets: 632626
tx_queue_11_packets: 1719870
tx_queue_12_packets: 0
tx_queue_13_packets: 449811
tx_queue_14_packets: 0
.......
I am using ArchLinux with kernel 3.5.4. Please ask for more details.
Any help would be appreciated. Thanks
Charles
^ permalink raw reply
* Re: kernel BUG at kernel/timer.c:748!
From: Dave Jones @ 2012-09-24 18:11 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Yuchung Cheng, Julian Anastasov, netdev
In-Reply-To: <1348507878.26828.1271.camel@edumazet-glaptop>
On Mon, Sep 24, 2012 at 07:31:18PM +0200, Eric Dumazet wrote:
> On Mon, 2012-09-24 at 13:11 -0400, Dave Jones wrote:
>
> > Great, I'll give this a shot.
> >
> > Any idea why this only just started triggering ?
> > (Ie, do we need this for stable too?)
>
> Seems this is a very old bug.
Indeed. As far as I can tell, this goes back to 2.1.8,
back in November 1996.
> I guess your trinity tool gets better ?
Possible. Or it may be that it was finding other bugs first
that are now fixed, that prevented us getting far enough along
to try this.
> I used following program to trigger the bug :
I'm sure reproducers like that will be appreciated by anyone
running QA tests on older kernels.
tangent: Is there any kind of networking correctness suite
other than fuzz testers like isic etc ?
thanks,
Dave
^ permalink raw reply
* Re: [PATCH v4] lxt PHY: Support for the buggy LXT973 rev A2
From: Richard Cochran @ 2012-09-24 18:30 UTC (permalink / raw)
To: Christophe Leroy; +Cc: David S Miller, netdev, linux-kernel
In-Reply-To: <201209241400.q8OE0w38011790@localhost.localdomain>
On Mon, Sep 24, 2012 at 04:00:58PM +0200, Christophe Leroy wrote:
> diff -u a/drivers/net/phy/lxt.c b/drivers/net/phy/lxt.c
> --- a/drivers/net/phy/lxt.c 2012-09-23 03:08:48.000000000 +0200
> +++ b/drivers/net/phy/lxt.c 2012-09-23 03:18:00.000000000 +0200
...
> @@ -175,6 +292,16 @@
> .driver = { .owner = THIS_MODULE,},
> }, {
> .phy_id = 0x00137a10,
> + .name = "LXT973-A2",
> + .phy_id_mask = 0xffffffff,
> + .features = PHY_BASIC_FEATURES,
> + .flags = 0,
> + .probe = lxt973_probe,
> + .config_aneg = lxt973_config_aneg,
> + .read_status = lxt973a2_read_status,
I like this way of matching the A2 chips much better than what you had
before. But are you sure this will work correctly?
What do A3 chips have in the last nibble of phy_id?
> + .driver = { .owner = THIS_MODULE,},
> +}, {
> + .phy_id = 0x00137a10,
> .name = "LXT973",
> .phy_id_mask = 0xfffffff0,
> .features = PHY_BASIC_FEATURES,
Thanks,
Richard
^ permalink raw reply
* Re: [PATCH net-next v3] net: use a per task frag allocator
From: Vijay Subramanian @ 2012-09-24 18:34 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, netdev, Ben Hutchings, Alexander Duyck
In-Reply-To: <1348477482.26828.235.camel@edumazet-glaptop>
>
> Successfully tested on various ethernet devices.
> (ixgbe, igb, bnx2x, tg3, mellanox mlx4)
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Ben Hutchings <bhutchings@solarflare.com>
> Cc: Vijay Subramanian <subramanian.vijay@gmail.com>
> Cc: Alexander Duyck <alexander.h.duyck@intel.com>
> ---
> v3: net/sched/em_meta.c bit pointed out by Vijay
> v2: uses existing page_frag structure to hold page/offset/size
> convert linear_to_page()/__ip_append_data()/ip6_append_data()
> remove @page and @off fields from struct inet_cork
> move the destructor from tcp_v4_destroy_sock() to sk_common_release
With netperf (64KB send buffer) and with e1000e driver, I get 13%
improvement with v3 of this patch.
When I run netserver and netperf on separate cpus with taskset, I get
7% improvement.
Tested-by: Vijay Subramanian <subramanian.vijay@gmail.com>
FYI, checkpatch complained about this patch with 4 lines over 80
characters (which probably be ignored I think) and one
trailing space error.
ERROR: trailing whitespace
#443: FILE: net/ipv4/raw.c:139:
+^Ipr_err("icmp_filter skb_transport_offset %d data-head %ld len %d/%d\n", $
Vijay
^ permalink raw reply
* Re: [RFC PATCHv2 bridge 7/7] bridge: Add the ability to show dump the vlan map from a bridge port
From: Ben Hutchings @ 2012-09-24 18:39 UTC (permalink / raw)
To: vyasevic; +Cc: Stephen Hemminger, David Miller, netdev
In-Reply-To: <50606500.1040102@redhat.com>
On Mon, 2012-09-24 at 09:49 -0400, Vlad Yasevich wrote:
> On 09/22/2012 04:05 PM, Stephen Hemminger wrote:
> > On Sat, 22 Sep 2012 13:27:47 -0400 (EDT)
> > David Miller <davem@davemloft.net> wrote:
> >
> >> From: Ben Hutchings <bhutchings@solarflare.com>
> >> Date: Sat, 22 Sep 2012 18:15:32 +0100
> >>
> >>> On Wed, 2012-09-19 at 08:42 -0400, Vlad Yasevich wrote:
> >>>> Using the RTM_GETLINK dump the vlan map of a given bridge port.
> >>> [...]
> >>>
> >>> This enlarges the RTM_GETLINK response quite a bit. I think perhaps
> >>> this should be optional, like IFLA_VFINFO_LIST is now.
> >>
> >> Completely agreed.
> >
> > Since most users won't use it, it should be not necessary to include
> > it if the map is blank.
> >
>
> Response is included only for AF_BRIDGE calls (proably used by STP
> implementation) and only when vlan filtering is configured.
>
> I do see the point though and can add a filter for this.
I thought the per-AF stuff was included by default (or at least what
most clients ask for). If not then this is probably fine.
IFLA_VFINFO_LIST was causing glibc's queries to fail and that must be
avoided.
Ben.
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* [PATCH net-next 2/3] igmp: export symbol ip_mc_leave_group
From: Stephen Hemminger @ 2012-09-24 18:43 UTC (permalink / raw)
To: David Miller, Chris Wright; +Cc: netdev
In-Reply-To: <20120924184304.727711327@vyatta.com>
[-- Attachment #1: igmp-leave.patch --]
[-- Type: text/plain, Size: 427 bytes --]
Needed for VXLAN.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
--- a/net/ipv4/igmp.c 2012-09-17 17:15:06.747860247 -0700
+++ b/net/ipv4/igmp.c 2012-09-17 17:16:33.554984978 -0700
@@ -1904,6 +1904,7 @@ int ip_mc_leave_group(struct sock *sk, s
rtnl_unlock();
return ret;
}
+EXPORT_SYMBOL(ip_mc_leave_group);
int ip_mc_source(int add, int omode, struct sock *sk, struct
ip_mreq_source *mreqs, int ifindex)
^ permalink raw reply
* [PATCH net-next 0/3] VXLAN driver
From: Stephen Hemminger @ 2012-09-24 18:43 UTC (permalink / raw)
To: David Miller, Chris Wright; +Cc: netdev
This patchset implements the Virtual eXtensible Local Area Network
tunnel. The code comes from GRE, Bridge, and the OVS Vxlan. The code
has only been tested with the TAP based vxlan
https://github.com/upa/vxlan.git
Thanks to Chris for the idea.
^ permalink raw reply
* [PATCH net-next 1/3] netlink: add attributes to fdb interface
From: Stephen Hemminger @ 2012-09-24 18:43 UTC (permalink / raw)
To: David Miller, Chris Wright; +Cc: netdev
In-Reply-To: <20120924184304.727711327@vyatta.com>
[-- Attachment #1: fdb-attr.patch --]
[-- Type: text/plain, Size: 4186 bytes --]
Later changes need to be able to refer to neighbour attributes
when doing fdb_add.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +-
drivers/net/macvlan.c | 2 +-
include/linux/netdevice.h | 4 +++-
net/bridge/br_fdb.c | 3 ++-
net/bridge/br_private.h | 2 +-
net/core/rtnetlink.c | 6 ++++--
6 files changed, 12 insertions(+), 7 deletions(-)
--- a/drivers/net/macvlan.c 2012-09-17 11:24:34.000000000 -0700
+++ b/drivers/net/macvlan.c 2012-09-17 11:28:13.999165490 -0700
@@ -546,7 +546,7 @@ static int macvlan_vlan_rx_kill_vid(stru
return 0;
}
-static int macvlan_fdb_add(struct ndmsg *ndm,
+static int macvlan_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr,
u16 flags)
--- a/include/linux/netdevice.h 2012-09-17 11:24:34.000000000 -0700
+++ b/include/linux/netdevice.h 2012-09-17 11:28:13.999165490 -0700
@@ -906,7 +906,8 @@ struct netdev_fcoe_hbainfo {
* feature set might be less than what was returned by ndo_fix_features()).
* Must return >0 or -errno if it changed dev->features itself.
*
- * int (*ndo_fdb_add)(struct ndmsg *ndm, struct net_device *dev,
+ * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
+ * struct net_device *dev,
* const unsigned char *addr, u16 flags)
* Adds an FDB entry to dev for addr.
* int (*ndo_fdb_del)(struct ndmsg *ndm, struct net_device *dev,
@@ -1016,6 +1017,7 @@ struct net_device_ops {
void (*ndo_neigh_destroy)(struct neighbour *n);
int (*ndo_fdb_add)(struct ndmsg *ndm,
+ struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr,
u16 flags);
--- a/net/bridge/br_fdb.c 2012-09-17 11:24:34.000000000 -0700
+++ b/net/bridge/br_fdb.c 2012-09-17 11:28:13.999165490 -0700
@@ -608,7 +608,8 @@ static int fdb_add_entry(struct net_brid
}
/* Add new permanent fdb entry with RTM_NEWNEIGH */
-int br_fdb_add(struct ndmsg *ndm, struct net_device *dev,
+int br_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
+ struct net_device *dev,
const unsigned char *addr, u16 nlh_flags)
{
struct net_bridge_port *p;
--- a/net/bridge/br_private.h 2012-09-17 11:24:34.000000000 -0700
+++ b/net/bridge/br_private.h 2012-09-17 11:28:13.999165490 -0700
@@ -364,7 +364,7 @@ extern void br_fdb_update(struct net_bri
extern int br_fdb_delete(struct ndmsg *ndm,
struct net_device *dev,
const unsigned char *addr);
-extern int br_fdb_add(struct ndmsg *nlh,
+extern int br_fdb_add(struct ndmsg *nlh, struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr,
u16 nlh_flags);
--- a/net/core/rtnetlink.c 2012-09-17 11:24:34.000000000 -0700
+++ b/net/core/rtnetlink.c 2012-09-17 11:28:14.003165450 -0700
@@ -2090,7 +2090,8 @@ static int rtnl_fdb_add(struct sk_buff *
if ((!ndm->ndm_flags || ndm->ndm_flags & NTF_MASTER) &&
(dev->priv_flags & IFF_BRIDGE_PORT)) {
master = dev->master;
- err = master->netdev_ops->ndo_fdb_add(ndm, dev, addr,
+ err = master->netdev_ops->ndo_fdb_add(ndm, tb,
+ dev, addr,
nlh->nlmsg_flags);
if (err)
goto out;
@@ -2100,7 +2101,8 @@ static int rtnl_fdb_add(struct sk_buff *
/* Embedded bridge, macvlan, and any other device support */
if ((ndm->ndm_flags & NTF_SELF) && dev->netdev_ops->ndo_fdb_add) {
- err = dev->netdev_ops->ndo_fdb_add(ndm, dev, addr,
+ err = dev->netdev_ops->ndo_fdb_add(ndm, tb,
+ dev, addr,
nlh->nlmsg_flags);
if (!err) {
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c 2012-09-17 11:25:15.000000000 -0700
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c 2012-09-17 11:31:03.325464947 -0700
@@ -6888,7 +6888,7 @@ static int ixgbe_set_features(struct net
return 0;
}
-static int ixgbe_ndo_fdb_add(struct ndmsg *ndm,
+static int ixgbe_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr,
u16 flags)
^ permalink raw reply
* [PATCH net-next 3/3] vxlan: virtual extensible lan
From: Stephen Hemminger @ 2012-09-24 18:43 UTC (permalink / raw)
To: David Miller, Chris Wright; +Cc: netdev
In-Reply-To: <20120924184304.727711327@vyatta.com>
[-- Attachment #1: vxlan.patch --]
[-- Type: text/plain, Size: 33509 bytes --]
This is an implementation of Virtual eXtensible Local Area Network
as described in draft RFC:
http://tools.ietf.org/html/draft-mahalingam-dutt-dcops-vxlan-02
The driver integrates a Virtual Tunnel Endpoint (VTEP) functionality
that learns MAC to IP address mapping.
This implementation has not been tested for Interoperation with
other equipment.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
---
Documentation/networking/vxlan.txt | 36 +
drivers/net/Kconfig | 13
drivers/net/Makefile | 1
drivers/net/vxlan.c | 1180 +++++++++++++++++++++++++++++++++++++
include/linux/if_link.h | 14
5 files changed, 1244 insertions(+)
--- a/drivers/net/Kconfig 2012-09-24 10:56:57.080291529 -0700
+++ b/drivers/net/Kconfig 2012-09-24 11:08:02.865416523 -0700
@@ -149,6 +149,19 @@ config MACVTAP
To compile this driver as a module, choose M here: the module
will be called macvtap.
+config VXLAN
+ tristate "Virtual eXtensible Local Area Network (VXLAN)"
+ depends on EXPERIMENTAL
+ ---help---
+ This allows one to create vxlan virtual interfaces that provide
+ Layer 2 Networks over Layer 3 Networks. VXLAN is often used
+ to tunnel virtual network infrastructure in virtualized environments.
+ For more information see:
+ http://tools.ietf.org/html/draft-mahalingam-dutt-dcops-vxlan-02
+
+ To compile this driver as a module, choose M here: the module
+ will be called macvlan.
+
config NETCONSOLE
tristate "Network console logging support"
---help---
--- a/drivers/net/Makefile 2012-09-24 10:56:57.080291529 -0700
+++ b/drivers/net/Makefile 2012-09-24 11:08:02.865416523 -0700
@@ -21,6 +21,7 @@ obj-$(CONFIG_NET_TEAM) += team/
obj-$(CONFIG_TUN) += tun.o
obj-$(CONFIG_VETH) += veth.o
obj-$(CONFIG_VIRTIO_NET) += virtio_net.o
+obj-$(CONFIG_VXLAN) += vxlan.o
#
# Networking Drivers
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/drivers/net/vxlan.c 2012-09-24 11:08:02.869416484 -0700
@@ -0,0 +1,1171 @@
+/*
+ * VXLAN: Virtual eXtensiable Local Area Network
+ *
+ * Copyright (c) 2012 Vyatta Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * TODO
+ * - use IANA UDP port number (when defined)
+ * - IPv6 (not in RFC)
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/skbuff.h>
+#include <linux/rculist.h>
+#include <linux/netdevice.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/udp.h>
+#include <linux/igmp.h>
+#include <linux/etherdevice.h>
+#include <linux/if_ether.h>
+#include <linux/version.h>
+#include <linux/hash.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/udp.h>
+#include <net/rtnetlink.h>
+#include <net/route.h>
+#include <net/dsfield.h>
+#include <net/inet_ecn.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+
+#define VXLAN_VERSION "0.0"
+
+#define VNI_HASH_BITS 10
+#define VNI_HASH_SIZE (1<<VNI_HASH_BITS)
+#define FDB_HASH_BITS 8
+#define FDB_HASH_SIZE (1<<FDB_HASH_BITS)
+
+#define VXLAN_N_VID (1u << 24)
+#define VXLAN_VID_MASK (VXLAN_N_VID - 1)
+
+#define FDB_AGE_INTERVAL (10 * HZ) /* rescan interval */
+#define FDB_AGE_TIME (300 * HZ) /* drop if not used in 5 min */
+
+#define VXLAN_FLAGS 0x08000000 /* struct vxlanhdr.vx_flags required value. */
+
+/* VXLAN protocol header */
+struct vxlanhdr {
+ __be32 vx_flags;
+ __be32 vx_vni;
+};
+
+#define VXLAN_HEADROOM (sizeof(struct iphdr) \
+ + sizeof(struct udphdr) \
+ + sizeof(struct vxlanhdr))
+
+/* UDP port for VXLAN traffic. */
+static unsigned int vxlan_port __read_mostly = 8472;
+module_param_named(port, vxlan_port, uint, 0);
+MODULE_PARM_DESC(vxlan_port, "Destination UDP port");
+
+/* per-net private data for this module */
+static unsigned int vxlan_net_id;
+struct vxlan_net {
+ struct socket *sock; /* UDP encap socket */
+ struct hlist_head vni_list[VNI_HASH_SIZE];
+};
+
+/* Forwarding table entry */
+struct vxlan_fdb {
+ struct hlist_node hlist; /* linked list of entries */
+ struct rcu_head rcu;
+ unsigned long updated; /* jiffies */
+ unsigned long used;
+#define VXLAN_FDB_PERM 0x1
+ u32 flags;
+ __be32 remote_ip;
+ u8 eth_addr[ETH_ALEN];
+};
+
+/* Per-cpu network traffic stats */
+struct vxlan_stats {
+ u64 rx_packets;
+ u64 rx_bytes;
+ u64 tx_packets;
+ u64 tx_bytes;
+ struct u64_stats_sync syncp;
+};
+
+/* Pseudo network device */
+struct vxlan_dev {
+ struct hlist_node hlist;
+ struct net_device *dev;
+ struct vxlan_stats __percpu *stats;
+ __u32 vni; /* virtual network id */
+ __be32 gaddr; /* multicast group */
+ __be32 saddr; /* source address */
+ unsigned int link; /* link to multicast over */
+ __u8 tos; /* TOS override */
+ __u8 ttl;
+ bool learn;
+
+ struct timer_list age_timer;
+ spinlock_t hash_lock;
+ struct hlist_head fdb_head[FDB_HASH_SIZE];
+};
+
+/* salt for hash table */
+static u32 vxlan_salt __read_mostly;
+
+static inline struct hlist_head *vni_head(struct net *net, u32 id)
+{
+ struct vxlan_net *vn = net_generic(net, vxlan_net_id);
+
+ return &vn->vni_list[hash_32(id, VNI_HASH_BITS)];
+}
+
+/* Look up VNI in a per net namespace table */
+static struct vxlan_dev *vxlan_find_vni(struct net *net, u32 id)
+{
+ struct vxlan_dev *vxlan;
+ struct hlist_node *node;
+
+ hlist_for_each_entry_rcu(vxlan, node, vni_head(net, id), hlist) {
+ if (vxlan->vni == id)
+ return vxlan;
+ }
+
+ return NULL;
+}
+
+/* Fill in neighbour message in skbuff. */
+static int vxlan_fdb_info(struct sk_buff *skb, struct vxlan_dev *vxlan,
+ const struct vxlan_fdb *fdb,
+ u32 portid, u32 seq, int type, unsigned int flags)
+{
+ unsigned long now = jiffies;
+ struct nda_cacheinfo ci;
+ struct nlmsghdr *nlh;
+ struct ndmsg *ndm;
+
+ nlh = nlmsg_put(skb, portid, seq, type, sizeof(*ndm), flags);
+ if (nlh == NULL)
+ return -EMSGSIZE;
+
+ ndm = nlmsg_data(nlh);
+ memset(ndm, 0, sizeof(*ndm));
+ ndm->ndm_family = AF_BRIDGE;
+
+ if (fdb->flags & VXLAN_FDB_PERM)
+ ndm->ndm_state = NUD_PERMANENT;
+ else if (time_before_eq(fdb->used + FDB_AGE_TIME, now))
+ ndm->ndm_state = NUD_STALE;
+ else
+ ndm->ndm_state = NUD_REACHABLE;
+
+ if (nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->eth_addr))
+ goto nla_put_failure;
+
+ if (nla_put_be32(skb, NDA_DST, fdb->remote_ip))
+ goto nla_put_failure;
+
+ ci.ndm_used = jiffies_to_clock_t(now - fdb->used);
+ ci.ndm_confirmed = 0;
+ ci.ndm_updated = jiffies_to_clock_t(now - fdb->updated);
+ ci.ndm_refcnt = 0;
+
+ if (nla_put(skb, NDA_CACHEINFO, sizeof(ci), &ci))
+ goto nla_put_failure;
+
+ return nlmsg_end(skb, nlh);
+
+nla_put_failure:
+ nlmsg_cancel(skb, nlh);
+ return -EMSGSIZE;
+}
+
+static inline size_t vxlan_nlmsg_size(void)
+{
+ return NLMSG_ALIGN(sizeof(struct ndmsg))
+ + nla_total_size(ETH_ALEN) /* NDA_LLADDR */
+ + nla_total_size(sizeof(__be32)) /* NDA_DST */
+ + nla_total_size(sizeof(struct nda_cacheinfo));
+}
+
+static void vxlan_fdb_notify(struct vxlan_dev *vxlan,
+ const struct vxlan_fdb *fdb, int type)
+{
+ struct net *net = dev_net(vxlan->dev);
+ struct sk_buff *skb;
+ int err = -ENOBUFS;
+
+ skb = nlmsg_new(vxlan_nlmsg_size(), GFP_ATOMIC);
+ if (skb == NULL)
+ goto errout;
+
+ err = vxlan_fdb_info(skb, vxlan, fdb, 0, 0, type, 0);
+ if (err < 0) {
+ /* -EMSGSIZE implies BUG in vxlan_nlmsg_size() */
+ WARN_ON(err == -EMSGSIZE);
+ kfree_skb(skb);
+ goto errout;
+ }
+
+ rtnl_notify(skb, net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
+ return;
+errout:
+ if (err < 0)
+ rtnl_set_sk_err(net, RTNLGRP_NEIGH, err);
+}
+
+/* Hash Ethernet address */
+static u32 eth_hash(const unsigned char *addr)
+{
+ /* could be optimized for unaligned access */
+ u32 a = addr[5] << 8 | addr[4];
+ u32 b = addr[3] << 24 | addr[2] << 16 | addr[1] << 8 | addr[0];
+
+ return jhash_2words(a, b, vxlan_salt);
+}
+
+/* Hash chain to use given mac address */
+static inline struct hlist_head *vxlan_fdb_head(struct vxlan_dev *vxlan,
+ const u8 *mac)
+{
+ return &vxlan->fdb_head[hash_32(eth_hash(mac), FDB_HASH_BITS)];
+}
+
+/* Look up Ethernet address in forwarding table */
+static struct vxlan_fdb *vxlan_find_mac(struct vxlan_dev *vxlan,
+ const u8 *mac)
+
+{
+ struct hlist_head *head = vxlan_fdb_head(vxlan, mac);
+ struct vxlan_fdb *f;
+ struct hlist_node *node;
+
+ hlist_for_each_entry_rcu(f, node, head, hlist) {
+ if (compare_ether_addr(mac, f->eth_addr) == 0)
+ return f;
+ }
+
+ return NULL;
+}
+
+/* Add new entry to forwarding table -- assumes lock held */
+static int vxlan_fdb_create(struct vxlan_dev *vxlan,
+ const u8 *mac, __be32 ip,
+ u32 flags)
+{
+ struct vxlan_fdb *f;
+
+ if (unlikely(vxlan_find_mac(vxlan, mac))) {
+ netdev_dbg(vxlan->dev,
+ "lost race to create %pM\n", mac);
+ return -EEXIST;
+ }
+
+ netdev_dbg(vxlan->dev, "add %pM -> %pI4\n", mac, &ip);
+
+ f = kmalloc(sizeof(*f), GFP_ATOMIC);
+ if (!f)
+ return -ENOMEM;
+
+ f->remote_ip = ip;
+ memcpy(f->eth_addr, mac, ETH_ALEN);
+ hlist_add_head_rcu(&f->hlist,
+ vxlan_fdb_head(vxlan, mac));
+ f->updated = f->used = jiffies;
+ f->flags = flags;
+
+ vxlan_fdb_notify(vxlan, f, RTM_NEWNEIGH);
+
+ return 0;
+}
+
+static void vxlan_fdb_destroy(struct vxlan_dev *vxlan, struct vxlan_fdb *f)
+{
+ netdev_dbg(vxlan->dev,
+ "delete %pM\n", f->eth_addr);
+
+ vxlan_fdb_notify(vxlan, f, RTM_DELNEIGH);
+
+ hlist_del_rcu(&f->hlist);
+ kfree_rcu(f, rcu);
+}
+
+/* Add static entry (via netlink) */
+static int vxlan_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
+ struct net_device *dev,
+ const unsigned char *addr, u16 flags)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ __be32 ip;
+ int err;
+
+ if (tb[NDA_DST] == NULL)
+ return -EINVAL;
+
+ if (nla_len(tb[NDA_DST]) != sizeof(__be32))
+ return -EAFNOSUPPORT;
+
+ ip = nla_get_be32(tb[NDA_DST]);
+
+ spin_lock_bh(&vxlan->hash_lock);
+ err = vxlan_fdb_create(vxlan, addr, ip, VXLAN_FDB_PERM);
+ spin_unlock_bh(&vxlan->hash_lock);
+
+ return err;
+}
+
+/* Delete entry (via netlink) */
+static int vxlan_fdb_delete(struct ndmsg *ndm, struct net_device *dev,
+ const unsigned char *addr)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct vxlan_fdb *f;
+ int err = -ENOENT;
+
+ spin_lock_bh(&vxlan->hash_lock);
+ f = vxlan_find_mac(vxlan, addr);
+ if (f) {
+ vxlan_fdb_destroy(vxlan, f);
+ err = 0;
+ }
+ spin_unlock_bh(&vxlan->hash_lock);
+
+ return err;
+}
+
+/* Dump forwarding table */
+static int vxlan_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb,
+ struct net_device *dev, int idx)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ unsigned int h;
+
+ for (h = 0; h < FDB_HASH_SIZE; ++h) {
+ struct vxlan_fdb *f;
+ struct hlist_node *n;
+
+ hlist_for_each_entry_rcu(f, n, &vxlan->fdb_head[h], hlist) {
+ if (idx < cb->args[0])
+ goto skip;
+
+ if (vxlan_fdb_info(skb, vxlan, f,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNEIGH,
+ NLM_F_MULTI) < 0)
+ break;
+skip:
+ ++idx;
+ }
+ }
+
+ return idx;
+}
+
+/* Watch incoming packets to learn mapping between Ethernet address
+ * and Tunnel endpoint.
+ */
+static void vxlan_snoop(struct net_device *dev,
+ __be32 src_ip, const u8 *src_mac)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct vxlan_fdb *f;
+ int err;
+
+ f = vxlan_find_mac(vxlan, src_mac);
+ if (likely(f)) {
+ f->used = jiffies;
+ if (likely(f->remote_ip == src_ip))
+ return;
+
+ f->remote_ip = src_ip;
+ f->updated = jiffies;
+
+ if (net_ratelimit())
+ netdev_info(dev,
+ "%pM migrated from %pI4 to %pI4\n",
+ src_mac, &f->remote_ip, &src_ip);
+ } else {
+ /* learned new entry */
+ spin_lock(&vxlan->hash_lock);
+ err = vxlan_fdb_create(vxlan, src_mac, src_ip, 0);
+ spin_unlock(&vxlan->hash_lock);
+ }
+}
+
+
+/* See if multicast group is already in use by other ID */
+static bool vxlan_group_used(struct vxlan_net *vn,
+ const struct vxlan_dev *this)
+{
+ const struct vxlan_dev *vxlan;
+ struct hlist_node *node;
+ unsigned h;
+
+ for (h = 0; h < VNI_HASH_SIZE; ++h)
+ hlist_for_each_entry(vxlan, node, &vn->vni_list[h], hlist) {
+ if (vxlan == this)
+ continue;
+
+ if (!netif_running(vxlan->dev))
+ continue;
+
+ if (vxlan->gaddr == this->gaddr)
+ return true;
+ }
+
+ return false;
+}
+
+/* kernel equivalent to IP_ADD_MEMBERSHIP */
+static int vxlan_join_group(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
+ struct sock *sk = vn->sock->sk;
+ struct ip_mreqn mreq = {
+ .imr_multiaddr.s_addr = vxlan->gaddr,
+ };
+ int err;
+
+ if (vxlan_group_used(vn, vxlan))
+ return 0;
+
+ /* Need to drop RTNL to call multicast join */
+ rtnl_unlock();
+ lock_sock(sk);
+ err = ip_mc_join_group(sk, &mreq);
+ release_sock(sk);
+ rtnl_lock();
+
+ return err;
+}
+
+
+/* kernel equivalent to IP_DROP_MEMBERSHIP */
+static int vxlan_leave_group(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
+ int err = 0;
+ struct sock *sk = vn->sock->sk;
+ struct ip_mreqn mreq = {
+ .imr_multiaddr.s_addr = vxlan->gaddr,
+ };
+
+ if (vxlan_group_used(vn, vxlan))
+ return 0;
+
+ /* Need to drop RTNL to call multicast leave */
+ rtnl_unlock();
+ lock_sock(sk);
+ err = ip_mc_leave_group(sk, &mreq);
+ release_sock(sk);
+ rtnl_lock();
+
+ return err;
+}
+
+/* Propogate ECN from outer IP header to tunneled packet */
+static inline void vxlan_ecn_decap(const struct iphdr *iph, struct sk_buff *skb)
+{
+ if (INET_ECN_is_ce(iph->tos)) {
+ if (skb->protocol == htons(ETH_P_IP))
+ IP_ECN_set_ce(ip_hdr(skb));
+ else if (skb->protocol == htons(ETH_P_IPV6))
+ IP6_ECN_set_ce(ipv6_hdr(skb));
+ }
+}
+
+/* Callback from net/ipv4/udp.c to receive packets */
+static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
+{
+ struct iphdr *oip = ip_hdr(skb);
+ struct vxlanhdr *vxh;
+ struct vxlan_dev *vxlan;
+ struct vxlan_stats *stats;
+ __u32 vni;
+ __be32 saddr = oip->saddr; /* source address for learning */
+ __be32 daddr = oip->daddr; /* destination for checking */
+
+ /* pop off outer UDP header */
+ __skb_pull(skb, sizeof(struct udphdr));
+
+ /* Need Vxlan and inner Ethernet header to be present */
+ if (!pskb_may_pull(skb,
+ sizeof(struct vxlanhdr) + sizeof(struct ethhdr)))
+ goto error;
+
+ /* Drop packets with reserved bits set */
+ vxh = (struct vxlanhdr *) skb->data;
+ if (vxh->vx_flags != htonl(VXLAN_FLAGS) ||
+ (vxh->vx_vni & htonl(0xff)))
+ goto error;
+
+ __skb_pull(skb, sizeof(struct vxlanhdr));
+
+ /* Is this VNI defined? */
+ vni = ntohl(vxh->vx_vni) >> 8;
+ vxlan = vxlan_find_vni(sock_net(sk), vni);
+ if (!vxlan)
+ goto drop;
+
+ /* Ignore packets if device is not up */
+ if (!netif_running(vxlan->dev))
+ goto drop;
+
+ /* Re-examine inner Ethernet packet */
+ skb->protocol = eth_type_trans(skb, vxlan->dev);
+ skb->ip_summed = CHECKSUM_NONE;
+
+ /* Ignore packet loops (and multicast echo) */
+ if (compare_ether_addr(eth_hdr(skb)->h_source,
+ vxlan->dev->dev_addr) == 0)
+ goto drop;
+
+ /* Check for multicast group configuration errors */
+ if (IN_MULTICAST(ntohl(daddr)) &&
+ daddr != vxlan->gaddr) {
+ if (net_ratelimit())
+ netdev_notice(vxlan->dev,
+ "group address %pI4 does not match\n",
+ &daddr);
+ goto drop;
+ }
+
+ if (vxlan->learn)
+ vxlan_snoop(skb->dev, saddr, eth_hdr(skb)->h_source);
+
+ stats = this_cpu_ptr(vxlan->stats);
+ u64_stats_update_begin(&stats->syncp);
+ stats->rx_packets++;
+ stats->rx_bytes += skb->len;
+ u64_stats_update_end(&stats->syncp);
+
+ __skb_tunnel_rx(skb, vxlan->dev);
+ skb_reset_network_header(skb);
+ vxlan_ecn_decap(oip, skb);
+
+ netif_rx(skb);
+
+ return 0;
+error:
+ /* Put UDP header back */
+ __skb_push(skb, sizeof(struct udphdr));
+
+ return 1;
+drop:
+ /* Consume bad packet */
+ kfree_skb(skb);
+ return 0;
+}
+
+/* Extract dsfield from inner protocol */
+static inline u8 vxlan_get_dsfield(const struct iphdr *iph,
+ const struct sk_buff *skb)
+{
+ if (skb->protocol == htons(ETH_P_IP))
+ return iph->tos;
+ else if (skb->protocol == htons(ETH_P_IPV6))
+ return ipv6_get_dsfield((const struct ipv6hdr *)iph);
+ else
+ return 0;
+}
+
+/* Propogate ECN bits out */
+static inline u8 vxlan_ecn_encap(u8 tos,
+ const struct iphdr *iph,
+ const struct sk_buff *skb)
+{
+ u8 inner = vxlan_get_dsfield(iph, skb);
+
+ return INET_ECN_encapsulate(tos, inner);
+}
+
+/* Transmit local packets over Vxlan
+ *
+ * Outer IP header inherits ECN and DF from inner header.
+ * Outer UDP destination is the VXLAN assigned port.
+ * source port is based on hash of flow if available
+ * otherwise use a random value
+ */
+static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct rtable *rt;
+ const struct ethhdr *eth;
+ const struct iphdr *old_iph;
+ struct iphdr *iph;
+ struct vxlanhdr *vxh;
+ struct udphdr *uh;
+ struct flowi4 fl4;
+ struct vxlan_fdb *f;
+ unsigned int pkt_len = skb->len;
+ unsigned int mtu;
+ u32 hash;
+ __be32 dst;
+ __be16 df = 0;
+ __u8 tos, ttl;
+ int err;
+
+ /* Need space for new headers (invalidates iph ptr) */
+ if (skb_cow_head(skb, VXLAN_HEADROOM))
+ goto drop;
+
+ eth = (void *)skb->data;
+ old_iph = ip_hdr(skb);
+
+ if (!is_multicast_ether_addr(eth->h_dest) &&
+ (f = vxlan_find_mac(vxlan, eth->h_dest)))
+ dst = f->remote_ip;
+ else if (vxlan->gaddr) {
+ dst = vxlan->gaddr;
+ } else
+ goto drop;
+
+ ttl = vxlan->ttl;
+ if (!ttl && IN_MULTICAST(ntohl(dst)))
+ ttl = 1;
+
+ tos = vxlan->tos;
+ if (tos == 1)
+ tos = vxlan_get_dsfield(old_iph, skb);
+
+ hash = skb_get_rxhash(skb);
+
+ rt = ip_route_output_gre(dev_net(dev), &fl4, dst,
+ vxlan->saddr, vxlan->vni,
+ RT_TOS(tos), vxlan->link);
+ if (IS_ERR(rt)) {
+ netdev_dbg(dev, "no route to %pI4\n", &dst);
+ dev->stats.tx_carrier_errors++;
+ goto tx_error;
+ }
+
+ if (rt->dst.dev == dev) {
+ netdev_dbg(dev, "circular route to %pI4\n", &dst);
+ ip_rt_put(rt);
+ dev->stats.collisions++;
+ goto tx_error;
+ }
+
+ mtu = dst_mtu(&rt->dst) - VXLAN_HEADROOM;
+ /* Do PMTU */
+ if (skb->protocol == htons(ETH_P_IP)) {
+ df |= old_iph->frag_off & htons(IP_DF);
+ if (df && mtu < pkt_len) {
+ icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
+ htonl(mtu));
+ ip_rt_put(rt);
+ goto tx_error;
+ }
+ }
+#if IS_ENABLED(CONFIG_IPV6)
+ else if (skb->protocol == htons(ETH_P_IPV6)) {
+ if (mtu >= IPV6_MIN_MTU && mtu < pkt_len) {
+ icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
+ ip_rt_put(rt);
+ goto tx_error;
+ }
+ }
+#endif
+
+ memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+ IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
+ IPSKB_REROUTED);
+ skb_dst_drop(skb);
+ skb_dst_set(skb, &rt->dst);
+
+ vxh = (struct vxlanhdr *) __skb_push(skb, sizeof(*vxh));
+ vxh->vx_flags = htonl(VXLAN_FLAGS);
+ vxh->vx_vni = htonl(vxlan->vni << 8);
+
+ __skb_push(skb, sizeof(*uh));
+ skb_reset_transport_header(skb);
+ uh = udp_hdr(skb);
+
+ uh->dest = htons(vxlan_port);
+ uh->source = hash ? :random32();
+
+ uh->len = htons(skb->len);
+ uh->check = 0;
+
+ __skb_push(skb, sizeof(*iph));
+ skb_reset_network_header(skb);
+ iph = ip_hdr(skb);
+ iph->version = 4;
+ iph->ihl = sizeof(struct iphdr) >> 2;
+ iph->frag_off = df;
+ iph->protocol = IPPROTO_UDP;
+ iph->tos = tos;
+ iph->daddr = fl4.daddr;
+ iph->saddr = fl4.saddr;
+ iph->ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
+
+ /* See __IPTUNNEL_XMIT */
+ skb->ip_summed = CHECKSUM_NONE;
+ ip_select_ident(iph, &rt->dst, NULL);
+
+ err = ip_local_out(skb);
+ if (likely(net_xmit_eval(err) == 0)) {
+ struct vxlan_stats *stats = this_cpu_ptr(vxlan->stats);
+
+ u64_stats_update_begin(&stats->syncp);
+ stats->tx_packets++;
+ stats->tx_bytes += pkt_len;
+ u64_stats_update_end(&stats->syncp);
+ } else {
+ dev->stats.tx_errors++;
+ dev->stats.tx_aborted_errors++;
+ }
+ return NETDEV_TX_OK;
+
+drop:
+ dev->stats.tx_dropped++;
+ goto tx_free;
+
+tx_error:
+ dev->stats.tx_errors++;
+tx_free:
+ dev_kfree_skb(skb);
+ return NETDEV_TX_OK;
+}
+
+/* Walk the forwarding table and purge stale entries */
+static void vxlan_cleanup(unsigned long arg)
+{
+ struct vxlan_dev *vxlan = (struct vxlan_dev *) arg;
+ unsigned long next_timer = jiffies + FDB_AGE_INTERVAL;
+ unsigned int h;
+
+ if (!netif_running(vxlan->dev))
+ return;
+
+ spin_lock_bh(&vxlan->hash_lock);
+ for (h = 0; h < FDB_HASH_SIZE; ++h) {
+ struct hlist_node *p, *n;
+ hlist_for_each_safe(p, n, &vxlan->fdb_head[h]) {
+ struct vxlan_fdb *f
+ = container_of(p, struct vxlan_fdb, hlist);
+ unsigned long timeout = f->used + FDB_AGE_TIME;
+
+ if (f->flags & VXLAN_FDB_PERM)
+ continue;
+
+ if (time_before_eq(timeout, jiffies)) {
+ netdev_dbg(vxlan->dev,
+ "garbage collect %pM\n",
+ f->eth_addr);
+ vxlan_fdb_destroy(vxlan, f);
+ } else if (time_before(timeout, next_timer))
+ next_timer = timeout;
+ }
+ }
+ spin_unlock_bh(&vxlan->hash_lock);
+
+ mod_timer(&vxlan->age_timer, next_timer);
+}
+
+/* Setup stats when device is created */
+static int vxlan_init(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+
+ vxlan->stats = alloc_percpu(struct vxlan_stats);
+ if (!vxlan->stats)
+ return -ENOMEM;
+
+ return 0;
+}
+
+/* Start ageing timer and join group when device is brought up */
+static int vxlan_open(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ int err;
+
+ if (vxlan->gaddr) {
+ err = vxlan_join_group(dev);
+ if (err)
+ return err;
+ }
+
+ mod_timer(&vxlan->age_timer, jiffies + FDB_AGE_INTERVAL);
+ return 0;
+}
+
+/* Purge the forwarding table */
+static void vxlan_flush(struct vxlan_dev *vxlan)
+{
+ unsigned h;
+
+ spin_lock_bh(&vxlan->hash_lock);
+ for (h = 0; h < FDB_HASH_SIZE; ++h) {
+ struct hlist_node *p, *n;
+ hlist_for_each_safe(p, n, &vxlan->fdb_head[h]) {
+ struct vxlan_fdb *f
+ = container_of(p, struct vxlan_fdb, hlist);
+ vxlan_fdb_destroy(vxlan, f);
+ }
+ }
+ spin_unlock_bh(&vxlan->hash_lock);
+}
+
+/* Cleanup timer and forwarding table on shutdown */
+static int vxlan_stop(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+
+ if (vxlan->gaddr)
+ vxlan_leave_group(dev);
+
+ del_timer_sync(&vxlan->age_timer);
+
+ vxlan_flush(vxlan);
+
+ return 0;
+}
+
+/* Merge per-cpu statistics */
+static struct rtnl_link_stats64 *vxlan_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *stats)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct vxlan_stats tmp, sum = { 0 };
+ unsigned int cpu;
+
+ for_each_possible_cpu(cpu) {
+ unsigned int start;
+ const struct vxlan_stats *stats
+ = per_cpu_ptr(vxlan->stats, cpu);
+
+ do {
+ start = u64_stats_fetch_begin_bh(&stats->syncp);
+ memcpy(&tmp, stats, sizeof(tmp));
+ } while (u64_stats_fetch_retry_bh(&stats->syncp, start));
+
+ sum.tx_bytes += tmp.tx_bytes;
+ sum.tx_packets += tmp.tx_packets;
+ sum.rx_bytes += tmp.rx_bytes;
+ sum.rx_packets += tmp.rx_packets;
+ }
+
+ stats->tx_bytes = sum.tx_bytes;
+ stats->tx_packets = sum.tx_packets;
+ stats->rx_bytes = sum.rx_bytes;
+ stats->rx_packets = sum.rx_packets;
+
+ stats->tx_dropped = dev->stats.tx_dropped;
+ stats->tx_errors = dev->stats.tx_errors;
+ stats->tx_carrier_errors = dev->stats.tx_carrier_errors;
+ stats->collisions = dev->stats.collisions;
+
+ return stats;
+}
+
+/* Stub, nothing needs to be done. */
+static void vxlan_set_multicast_list(struct net_device *dev)
+{
+}
+
+static const struct net_device_ops vxlan_netdev_ops = {
+ .ndo_init = vxlan_init,
+ .ndo_open = vxlan_open,
+ .ndo_stop = vxlan_stop,
+ .ndo_start_xmit = vxlan_xmit,
+ .ndo_get_stats64 = vxlan_stats64,
+ .ndo_set_rx_mode = vxlan_set_multicast_list,
+ .ndo_change_mtu = eth_change_mtu,
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_set_mac_address = eth_mac_addr,
+ .ndo_fdb_add = vxlan_fdb_add,
+ .ndo_fdb_del = vxlan_fdb_delete,
+ .ndo_fdb_dump = vxlan_fdb_dump,
+};
+
+/* Info for udev, that this is a virtual tunnel endpoint */
+static struct device_type vxlan_type = {
+ .name = "vxlan",
+};
+
+static void vxlan_free(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+
+ free_percpu(vxlan->stats);
+ free_netdev(dev);
+}
+
+/* Initialize the device structure. */
+static void vxlan_setup(struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ unsigned h;
+
+ eth_hw_addr_random(dev);
+ ether_setup(dev);
+
+ dev->netdev_ops = &vxlan_netdev_ops;
+ dev->destructor = vxlan_free;
+ SET_NETDEV_DEVTYPE(dev, &vxlan_type);
+
+ dev->tx_queue_len = 0;
+ dev->features |= NETIF_F_LLTX;
+ dev->features |= NETIF_F_NETNS_LOCAL;
+ dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+
+ spin_lock_init(&vxlan->hash_lock);
+
+ init_timer_deferrable(&vxlan->age_timer);
+ vxlan->age_timer.function = vxlan_cleanup;
+ vxlan->age_timer.data = (unsigned long) vxlan;
+
+ vxlan->dev = dev;
+
+ for (h = 0; h < FDB_HASH_SIZE; ++h)
+ INIT_HLIST_HEAD(&vxlan->fdb_head[h]);
+}
+
+static const struct nla_policy vxlan_policy[IFLA_VXLAN_MAX + 1] = {
+ [IFLA_VXLAN_ID] = { .type = NLA_U32 },
+ [IFLA_VXLAN_GROUP] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
+ [IFLA_VXLAN_LINK] = { .type = NLA_U32 },
+ [IFLA_VXLAN_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
+ [IFLA_VXLAN_TOS] = { .type = NLA_U8 },
+ [IFLA_VXLAN_LEARNING] = { .type = NLA_U8 },
+};
+
+static int vxlan_validate(struct nlattr *tb[], struct nlattr *data[])
+{
+ if (tb[IFLA_ADDRESS]) {
+ if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN) {
+ pr_debug("invalid link address (not ethernet)\n");
+ return -EINVAL;
+ }
+
+ if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS]))) {
+ pr_debug("invalid all zero ethernet address\n");
+ return -EADDRNOTAVAIL;
+ }
+ }
+
+ if (!data)
+ return -EINVAL;
+
+ if (data[IFLA_VXLAN_ID]) {
+ __u32 id = nla_get_u32(data[IFLA_VXLAN_ID]);
+ if (id >= VXLAN_VID_MASK)
+ return -ERANGE;
+ }
+
+ if (data[IFLA_VXLAN_GROUP]) {
+ __be32 gaddr = nla_get_be32(data[IFLA_VXLAN_GROUP]);
+ if (!IN_MULTICAST(ntohl(gaddr))) {
+ pr_debug("group address is not IPv4 multicast\n");
+ return -EADDRNOTAVAIL;
+ }
+ }
+
+ return 0;
+}
+
+static int vxlan_newlink(struct net *net, struct net_device *dev,
+ struct nlattr *tb[], struct nlattr *data[])
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ __u32 vni;
+ int err;
+
+ if (!data[IFLA_VXLAN_ID])
+ return -EINVAL;
+
+ vni = nla_get_u32(data[IFLA_VXLAN_ID]);
+ if (vxlan_find_vni(net, vni)) {
+ pr_info("duplicate VNI %u\n", vni);
+ return -EEXIST;
+ }
+ vxlan->vni = vni;
+
+ if (data[IFLA_VXLAN_GROUP])
+ vxlan->gaddr = nla_get_be32(data[IFLA_VXLAN_GROUP]);
+
+ if (data[IFLA_VXLAN_LOCAL])
+ vxlan->saddr = nla_get_be32(data[IFLA_VXLAN_LOCAL]);
+
+ if (data[IFLA_VXLAN_LINK])
+ vxlan->link = nla_get_u32(data[IFLA_VXLAN_LINK]);
+
+ if (data[IFLA_VXLAN_TOS])
+ vxlan->tos = nla_get_u8(data[IFLA_VXLAN_TOS]);
+
+ if (!data[IFLA_VXLAN_LEARNING] || nla_get_u8(data[IFLA_VXLAN_LEARNING]))
+ vxlan->learn = true;
+
+ err = register_netdevice(dev);
+ if (!err)
+ hlist_add_head_rcu(&vxlan->hlist, vni_head(net, vxlan->vni));
+
+ return err;
+}
+
+static void vxlan_dellink(struct net_device *dev, struct list_head *head)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+
+ hlist_del_rcu(&vxlan->hlist);
+
+ unregister_netdevice_queue(dev, head);
+}
+
+static size_t vxlan_get_size(const struct net_device *dev)
+{
+
+ return nla_total_size(sizeof(__u32)) + /* IFLA_VXLAN_ID */
+ nla_total_size(sizeof(__be32)) +/* IFLA_VXLAN_GROUP */
+ nla_total_size(sizeof(__u32)) + /* IFLA_VXLAN_LINK */
+ nla_total_size(sizeof(__be32))+ /* IFLA_VXLAN_LOCAL */
+ nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_TTL */
+ nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_TOS */
+ nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_LEARNING */
+ 0;
+}
+
+static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
+{
+ const struct vxlan_dev *vxlan = netdev_priv(dev);
+
+ if (nla_put_u32(skb, IFLA_VXLAN_ID, vxlan->vni))
+ goto nla_put_failure;
+
+ if (vxlan->gaddr && nla_put_u32(skb, IFLA_VXLAN_GROUP, vxlan->gaddr))
+ goto nla_put_failure;
+
+ if (vxlan->link && nla_put_u32(skb, IFLA_VXLAN_LINK, vxlan->link))
+ goto nla_put_failure;
+
+ if (vxlan->saddr && nla_put_u32(skb, IFLA_VXLAN_LOCAL, vxlan->saddr))
+ goto nla_put_failure;
+
+ if (nla_put_u8(skb, IFLA_VXLAN_TTL, vxlan->ttl) ||
+ nla_put_u8(skb, IFLA_VXLAN_TOS, vxlan->tos) ||
+ nla_put_u8(skb, IFLA_VXLAN_LEARNING, vxlan->learn))
+ goto nla_put_failure;
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
+static struct rtnl_link_ops vxlan_link_ops __read_mostly = {
+ .kind = "vxlan",
+ .maxtype = IFLA_VXLAN_MAX,
+ .policy = vxlan_policy,
+ .priv_size = sizeof(struct vxlan_dev),
+ .setup = vxlan_setup,
+ .validate = vxlan_validate,
+ .newlink = vxlan_newlink,
+ .dellink = vxlan_dellink,
+ .get_size = vxlan_get_size,
+ .fill_info = vxlan_fill_info,
+};
+
+static __net_init int vxlan_init_net(struct net *net)
+{
+ struct vxlan_net *vn = net_generic(net, vxlan_net_id);
+ struct sock *sk;
+ struct sockaddr_in addr;
+ int rc;
+ unsigned h;
+
+ /* Create UDP socket for encapsulation receive. */
+ rc = sock_create_kern(AF_INET, SOCK_DGRAM, IPPROTO_UDP, &vn->sock);
+ if (rc < 0) {
+ pr_debug("UDP socket create failed\n");
+ return rc;
+ }
+
+ addr.sin_family = AF_INET;
+ addr.sin_addr.s_addr = INADDR_ANY;
+ addr.sin_port = htons(vxlan_port);
+
+ rc = kernel_bind(vn->sock, (struct sockaddr *) &addr, sizeof(addr));
+ if (rc < 0) {
+ pr_debug("bind for port %u failed %d\n", vxlan_port, rc);
+ sock_release(vn->sock);
+ vn->sock = NULL;
+ return rc;
+ }
+
+ /* Disable multicast loopback */
+ sk = vn->sock->sk;
+ inet_sk(sk)->mc_loop = 0;
+
+ /* Mark socket as an encapsulation socket. */
+ udp_sk(sk)->encap_type = UDP_ENCAP_L2TPINUDP;
+ udp_sk(sk)->encap_rcv = vxlan_udp_encap_recv;
+ udp_encap_enable();
+
+ for (h = 0; h < VNI_HASH_SIZE; ++h)
+ INIT_HLIST_HEAD(&vn->vni_list[h]);
+
+ return 0;
+}
+
+static __net_exit void vxlan_exit_net(struct net *net)
+{
+ struct vxlan_net *vn = net_generic(net, vxlan_net_id);
+
+ if (vn->sock) {
+ sock_release(vn->sock);
+ vn->sock = NULL;
+ }
+}
+
+static struct pernet_operations vxlan_net_ops = {
+ .init = vxlan_init_net,
+ .exit = vxlan_exit_net,
+ .id = &vxlan_net_id,
+ .size = sizeof(struct vxlan_net),
+};
+
+static int __init vxlan_init_module(void)
+{
+ int rc;
+
+ get_random_bytes(&vxlan_salt, sizeof(vxlan_salt));
+
+ rc = register_pernet_device(&vxlan_net_ops);
+ if (rc)
+ goto out1;
+
+ rc = rtnl_link_register(&vxlan_link_ops);
+ if (rc)
+ goto out2;
+
+ return 0;
+
+out2:
+ unregister_pernet_device(&vxlan_net_ops);
+out1:
+ return rc;
+}
+module_init(vxlan_init_module);
+
+static void __exit vxlan_cleanup_module(void)
+{
+ rtnl_link_unregister(&vxlan_link_ops);
+ unregister_pernet_device(&vxlan_net_ops);
+}
+module_exit(vxlan_cleanup_module);
+
+MODULE_LICENSE("GPL");
+MODULE_VERSION(VXLAN_VERSION);
+MODULE_AUTHOR("Stephen Hemminger <shemminger@vyatta.com>");
+MODULE_ALIAS_RTNL_LINK("vxlan");
--- a/include/linux/if_link.h 2012-09-24 10:56:57.080291529 -0700
+++ b/include/linux/if_link.h 2012-09-24 11:08:02.869416484 -0700
@@ -272,6 +272,20 @@ enum macvlan_mode {
#define MACVLAN_FLAG_NOPROMISC 1
+/* VXLAN section */
+enum {
+ IFLA_VXLAN_UNSPEC,
+ IFLA_VXLAN_ID,
+ IFLA_VXLAN_GROUP,
+ IFLA_VXLAN_LINK,
+ IFLA_VXLAN_LOCAL,
+ IFLA_VXLAN_TTL,
+ IFLA_VXLAN_TOS,
+ IFLA_VXLAN_LEARNING,
+ __IFLA_VXLAN_MAX
+};
+#define IFLA_VXLAN_MAX (__IFLA_VXLAN_MAX - 1)
+
/* SR-IOV virtual function management section */
enum {
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/Documentation/networking/vxlan.txt 2012-09-24 11:08:02.869416484 -0700
@@ -0,0 +1,36 @@
+Virtual eXtensible Local Area Networking documentation
+======================================================
+
+The VXLAN protocol is a tunnelling protocol that is designed to
+solve the problem of limited number of available VLAN's (4096).
+With VXLAN identifier is expanded to 24 bits.
+
+It is a draft RFC standard, that is implemented by Cisco Nexus,
+Vmware and Brocade. The protocol runs over UDP using a single
+destination port (still not standardized by IANA).
+This document describes the Linux kernel tunnel device,
+there is also an implantation of VXLAN for Openvswitch.
+
+Unlike most tunnels, a VXLAN is a 1 to N network, not just point
+to point. A VXLAN device can either dynamically learn the IP address
+of the other end, in a manner similar to a learning bridge, or the
+forwarding entries can be configured statically.
+
+The management of vxlan is done in a similar fashion to it's
+too closest neighbors GRE and VLAN. Configuring VXLAN requires
+the version of iproute2 that matches the kernel release
+where VXLAN was first merged upstream.
+
+1. Create vxlan device
+ # ip li add vxlan0 type vxlan id 42 group 239.1.1.1 dev eth1
+
+This creates a new device (vxlan0). The device uses the
+the multicast group 239.1.1.1 over eth1 to handle packets where
+no entry is in the forwarding table.
+
+2. Delete vxlan device
+ # ip link delete vxlan0
+
+3. Show vxlan info
+ # ip -d show vxlan0
+
^ permalink raw reply
* [PATCH 01/16] netfilter: fix IPv6 NAT dependencies in Kconfig
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Pablo Neira Ayuso <pablo@netfilter.org>
* NF_NAT_IPV6 requires IP6_NF_IPTABLES
* IP6_NF_TARGET_MASQUERADE, IP6_NF_TARGET_NETMAP, IP6_NF_TARGET_REDIRECT
and IP6_NF_TARGET_NPT require NF_NAT_IPV6.
This change just mirrors what IPv4 does in Kconfig, for consistency.
Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
net/ipv6/netfilter/Kconfig | 110 ++++++++++++++++++++++----------------------
1 file changed, 55 insertions(+), 55 deletions(-)
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 3b73254..d8f276b 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -25,18 +25,6 @@ config NF_CONNTRACK_IPV6
To compile it as a module, choose M here. If unsure, say N.
-config NF_NAT_IPV6
- tristate "IPv6 NAT"
- depends on NF_CONNTRACK_IPV6
- depends on NETFILTER_ADVANCED
- select NF_NAT
- help
- The IPv6 NAT option allows masquerading, port forwarding and other
- forms of full Network Address Port Translation. It is controlled by
- the `nat' table in ip6tables, see the man page for ip6tables(8).
-
- To compile it as a module, choose M here. If unsure, say N.
-
config IP6_NF_IPTABLES
tristate "IP6 tables support (required for filtering)"
depends on INET && IPV6
@@ -144,48 +132,6 @@ config IP6_NF_TARGET_HL
(e.g. when running oldconfig). It selects
CONFIG_NETFILTER_XT_TARGET_HL.
-config IP6_NF_TARGET_MASQUERADE
- tristate "MASQUERADE target support"
- depends on NF_NAT_IPV6
- help
- Masquerading is a special case of NAT: all outgoing connections are
- changed to seem to come from a particular interface's address, and
- if the interface goes down, those connections are lost. This is
- only useful for dialup accounts with dynamic IP address (ie. your IP
- address will be different on next dialup).
-
- To compile it as a module, choose M here. If unsure, say N.
-
-config IP6_NF_TARGET_NETMAP
- tristate "NETMAP target support"
- depends on NF_NAT_IPV6
- help
- NETMAP is an implementation of static 1:1 NAT mapping of network
- addresses. It maps the network address part, while keeping the host
- address part intact.
-
- To compile it as a module, choose M here. If unsure, say N.
-
-config IP6_NF_TARGET_REDIRECT
- tristate "REDIRECT target support"
- depends on NF_NAT_IPV6
- help
- REDIRECT is a special case of NAT: all incoming connections are
- mapped onto the incoming interface's address, causing the packets to
- come to the local machine instead of passing through. This is
- useful for transparent proxies.
-
- To compile it as a module, choose M here. If unsure, say N.
-
-config IP6_NF_TARGET_NPT
- tristate "NPT (Network Prefix translation) target support"
- depends on NETFILTER_ADVANCED
- help
- This option adds the `SNPT' and `DNPT' target, which perform
- stateless IPv6-to-IPv6 Network Prefix Translation per RFC 6296.
-
- To compile it as a module, choose M here. If unsure, say N.
-
config IP6_NF_FILTER
tristate "Packet filtering"
default m if NETFILTER_ADVANCED=n
@@ -235,9 +181,63 @@ config IP6_NF_SECURITY
help
This option adds a `security' table to iptables, for use
with Mandatory Access Control (MAC) policy.
-
+
If unsure, say N.
+config NF_NAT_IPV6
+ tristate "IPv6 NAT"
+ depends on NF_CONNTRACK_IPV6
+ depends on NETFILTER_ADVANCED
+ select NF_NAT
+ help
+ The IPv6 NAT option allows masquerading, port forwarding and other
+ forms of full Network Address Port Translation. It is controlled by
+ the `nat' table in ip6tables, see the man page for ip6tables(8).
+
+ To compile it as a module, choose M here. If unsure, say N.
+
+if NF_NAT_IPV6
+
+config IP6_NF_TARGET_MASQUERADE
+ tristate "MASQUERADE target support"
+ help
+ Masquerading is a special case of NAT: all outgoing connections are
+ changed to seem to come from a particular interface's address, and
+ if the interface goes down, those connections are lost. This is
+ only useful for dialup accounts with dynamic IP address (ie. your IP
+ address will be different on next dialup).
+
+ To compile it as a module, choose M here. If unsure, say N.
+
+config IP6_NF_TARGET_NETMAP
+ tristate "NETMAP target support"
+ help
+ NETMAP is an implementation of static 1:1 NAT mapping of network
+ addresses. It maps the network address part, while keeping the host
+ address part intact.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
+config IP6_NF_TARGET_REDIRECT
+ tristate "REDIRECT target support"
+ help
+ REDIRECT is a special case of NAT: all incoming connections are
+ mapped onto the incoming interface's address, causing the packets to
+ come to the local machine instead of passing through. This is
+ useful for transparent proxies.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
+config IP6_NF_TARGET_NPT
+ tristate "NPT (Network Prefix translation) target support"
+ help
+ This option adds the `SNPT' and `DNPT' target, which perform
+ stateless IPv6-to-IPv6 Network Prefix Translation per RFC 6296.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
+endif # NF_NAT_IPV6
+
endif # IP6_NF_IPTABLES
endmenu
--
1.7.10.4
^ permalink raw reply related
* [PATCH 04/16] netfilter: combine ipt_NETMAP and ip6t_NETMAP
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Jan Engelhardt <jengelh@inai.de>
Combine more modules since the actual code is so small anyway that the
kmod metadata and the module in its loaded state totally outweighs the
combined actual code size.
IP_NF_TARGET_NETMAP becomes a compat option; IP6_NF_TARGET_NETMAP
is completely eliminated since it has not see a release yet.
Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
net/ipv4/netfilter/Kconfig | 11 ++-
net/ipv4/netfilter/Makefile | 1 -
net/ipv4/netfilter/ipt_NETMAP.c | 101 -----------------------
net/ipv6/netfilter/Kconfig | 9 ---
net/ipv6/netfilter/Makefile | 1 -
net/ipv6/netfilter/ip6t_NETMAP.c | 94 ----------------------
net/netfilter/Kconfig | 10 +++
net/netfilter/Makefile | 1 +
net/netfilter/xt_NETMAP.c | 165 ++++++++++++++++++++++++++++++++++++++
9 files changed, 181 insertions(+), 212 deletions(-)
delete mode 100644 net/ipv4/netfilter/ipt_NETMAP.c
delete mode 100644 net/ipv6/netfilter/ip6t_NETMAP.c
create mode 100644 net/netfilter/xt_NETMAP.c
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 131e537..6f14008 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -172,12 +172,11 @@ config IP_NF_TARGET_MASQUERADE
config IP_NF_TARGET_NETMAP
tristate "NETMAP target support"
depends on NETFILTER_ADVANCED
- help
- NETMAP is an implementation of static 1:1 NAT mapping of network
- addresses. It maps the network address part, while keeping the host
- address part intact.
-
- To compile it as a module, choose M here. If unsure, say N.
+ select NETFILTER_XT_TARGET_NETMAP
+ ---help---
+ This is a backwards-compat option for the user's convenience
+ (e.g. when running oldconfig). It selects
+ CONFIG_NETFILTER_XT_TARGET_NETMAP.
config IP_NF_TARGET_REDIRECT
tristate "REDIRECT target support"
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index b7dd189..f4446c5 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -45,7 +45,6 @@ obj-$(CONFIG_IP_NF_MATCH_RPFILTER) += ipt_rpfilter.o
obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o
obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o
obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o
-obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o
obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o
obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o
obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
deleted file mode 100644
index 85028dc..0000000
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ /dev/null
@@ -1,101 +0,0 @@
-/* NETMAP - static NAT mapping of IP network addresses (1:1).
- * The mapping can be applied to source (POSTROUTING),
- * destination (PREROUTING), or both (with separate rules).
- */
-
-/* (C) 2000-2001 Svenning Soerensen <svenning@post5.tele.dk>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-#include <linux/ip.h>
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <linux/netfilter.h>
-#include <linux/netfilter_ipv4.h>
-#include <linux/netfilter/x_tables.h>
-#include <net/netfilter/nf_nat.h>
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Svenning Soerensen <svenning@post5.tele.dk>");
-MODULE_DESCRIPTION("Xtables: 1:1 NAT mapping of IPv4 subnets");
-
-static int netmap_tg_check(const struct xt_tgchk_param *par)
-{
- const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
-
- if (!(mr->range[0].flags & NF_NAT_RANGE_MAP_IPS)) {
- pr_debug("bad MAP_IPS.\n");
- return -EINVAL;
- }
- if (mr->rangesize != 1) {
- pr_debug("bad rangesize %u.\n", mr->rangesize);
- return -EINVAL;
- }
- return 0;
-}
-
-static unsigned int
-netmap_tg(struct sk_buff *skb, const struct xt_action_param *par)
-{
- struct nf_conn *ct;
- enum ip_conntrack_info ctinfo;
- __be32 new_ip, netmask;
- const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
- struct nf_nat_range newrange;
-
- NF_CT_ASSERT(par->hooknum == NF_INET_PRE_ROUTING ||
- par->hooknum == NF_INET_POST_ROUTING ||
- par->hooknum == NF_INET_LOCAL_OUT ||
- par->hooknum == NF_INET_LOCAL_IN);
- ct = nf_ct_get(skb, &ctinfo);
-
- netmask = ~(mr->range[0].min_ip ^ mr->range[0].max_ip);
-
- if (par->hooknum == NF_INET_PRE_ROUTING ||
- par->hooknum == NF_INET_LOCAL_OUT)
- new_ip = ip_hdr(skb)->daddr & ~netmask;
- else
- new_ip = ip_hdr(skb)->saddr & ~netmask;
- new_ip |= mr->range[0].min_ip & netmask;
-
- memset(&newrange.min_addr, 0, sizeof(newrange.min_addr));
- memset(&newrange.max_addr, 0, sizeof(newrange.max_addr));
- newrange.flags = mr->range[0].flags | NF_NAT_RANGE_MAP_IPS;
- newrange.min_addr.ip = new_ip;
- newrange.max_addr.ip = new_ip;
- newrange.min_proto = mr->range[0].min;
- newrange.max_proto = mr->range[0].max;
-
- /* Hand modified range to generic setup. */
- return nf_nat_setup_info(ct, &newrange, HOOK2MANIP(par->hooknum));
-}
-
-static struct xt_target netmap_tg_reg __read_mostly = {
- .name = "NETMAP",
- .family = NFPROTO_IPV4,
- .target = netmap_tg,
- .targetsize = sizeof(struct nf_nat_ipv4_multi_range_compat),
- .table = "nat",
- .hooks = (1 << NF_INET_PRE_ROUTING) |
- (1 << NF_INET_POST_ROUTING) |
- (1 << NF_INET_LOCAL_OUT) |
- (1 << NF_INET_LOCAL_IN),
- .checkentry = netmap_tg_check,
- .me = THIS_MODULE
-};
-
-static int __init netmap_tg_init(void)
-{
- return xt_register_target(&netmap_tg_reg);
-}
-
-static void __exit netmap_tg_exit(void)
-{
- xt_unregister_target(&netmap_tg_reg);
-}
-
-module_init(netmap_tg_init);
-module_exit(netmap_tg_exit);
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index d8f276b..007bb45 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -209,15 +209,6 @@ config IP6_NF_TARGET_MASQUERADE
To compile it as a module, choose M here. If unsure, say N.
-config IP6_NF_TARGET_NETMAP
- tristate "NETMAP target support"
- help
- NETMAP is an implementation of static 1:1 NAT mapping of network
- addresses. It maps the network address part, while keeping the host
- address part intact.
-
- To compile it as a module, choose M here. If unsure, say N.
-
config IP6_NF_TARGET_REDIRECT
tristate "REDIRECT target support"
help
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index 5752132..de8e0d1 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -35,7 +35,6 @@ obj-$(CONFIG_IP6_NF_MATCH_RT) += ip6t_rt.o
# targets
obj-$(CONFIG_IP6_NF_TARGET_MASQUERADE) += ip6t_MASQUERADE.o
-obj-$(CONFIG_IP6_NF_TARGET_NETMAP) += ip6t_NETMAP.o
obj-$(CONFIG_IP6_NF_TARGET_NPT) += ip6t_NPT.o
obj-$(CONFIG_IP6_NF_TARGET_REDIRECT) += ip6t_REDIRECT.o
obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o
diff --git a/net/ipv6/netfilter/ip6t_NETMAP.c b/net/ipv6/netfilter/ip6t_NETMAP.c
deleted file mode 100644
index 4f3bf36..0000000
--- a/net/ipv6/netfilter/ip6t_NETMAP.c
+++ /dev/null
@@ -1,94 +0,0 @@
-/*
- * Copyright (c) 2011 Patrick McHardy <kaber@trash.net>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * Based on Svenning Soerensen's IPv4 NETMAP target. Development of IPv6
- * NAT funded by Astaro.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/ipv6.h>
-#include <linux/netfilter.h>
-#include <linux/netfilter_ipv6.h>
-#include <linux/netfilter/x_tables.h>
-#include <net/netfilter/nf_nat.h>
-
-static unsigned int
-netmap_tg6(struct sk_buff *skb, const struct xt_action_param *par)
-{
- const struct nf_nat_range *range = par->targinfo;
- struct nf_nat_range newrange;
- struct nf_conn *ct;
- enum ip_conntrack_info ctinfo;
- union nf_inet_addr new_addr, netmask;
- unsigned int i;
-
- ct = nf_ct_get(skb, &ctinfo);
- for (i = 0; i < ARRAY_SIZE(range->min_addr.ip6); i++)
- netmask.ip6[i] = ~(range->min_addr.ip6[i] ^
- range->max_addr.ip6[i]);
-
- if (par->hooknum == NF_INET_PRE_ROUTING ||
- par->hooknum == NF_INET_LOCAL_OUT)
- new_addr.in6 = ipv6_hdr(skb)->daddr;
- else
- new_addr.in6 = ipv6_hdr(skb)->saddr;
-
- for (i = 0; i < ARRAY_SIZE(new_addr.ip6); i++) {
- new_addr.ip6[i] &= ~netmask.ip6[i];
- new_addr.ip6[i] |= range->min_addr.ip6[i] &
- netmask.ip6[i];
- }
-
- newrange.flags = range->flags | NF_NAT_RANGE_MAP_IPS;
- newrange.min_addr = new_addr;
- newrange.max_addr = new_addr;
- newrange.min_proto = range->min_proto;
- newrange.max_proto = range->max_proto;
-
- return nf_nat_setup_info(ct, &newrange, HOOK2MANIP(par->hooknum));
-}
-
-static int netmap_tg6_checkentry(const struct xt_tgchk_param *par)
-{
- const struct nf_nat_range *range = par->targinfo;
-
- if (!(range->flags & NF_NAT_RANGE_MAP_IPS))
- return -EINVAL;
- return 0;
-}
-
-static struct xt_target netmap_tg6_reg __read_mostly = {
- .name = "NETMAP",
- .family = NFPROTO_IPV6,
- .target = netmap_tg6,
- .targetsize = sizeof(struct nf_nat_range),
- .table = "nat",
- .hooks = (1 << NF_INET_PRE_ROUTING) |
- (1 << NF_INET_POST_ROUTING) |
- (1 << NF_INET_LOCAL_OUT) |
- (1 << NF_INET_LOCAL_IN),
- .checkentry = netmap_tg6_checkentry,
- .me = THIS_MODULE,
-};
-
-static int __init netmap_tg6_init(void)
-{
- return xt_register_target(&netmap_tg6_reg);
-}
-
-static void netmap_tg6_exit(void)
-{
- xt_unregister_target(&netmap_tg6_reg);
-}
-
-module_init(netmap_tg6_init);
-module_exit(netmap_tg6_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Xtables: 1:1 NAT mapping of IPv6 subnets");
-MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 3f4b3b4..ad0e0da 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -648,6 +648,16 @@ config NETFILTER_XT_TARGET_MARK
(e.g. when running oldconfig). It selects
CONFIG_NETFILTER_XT_MARK (combined mark/MARK module).
+config NETFILTER_XT_TARGET_NETMAP
+ tristate '"NETMAP" target support'
+ depends on NF_NAT
+ ---help---
+ NETMAP is an implementation of static 1:1 NAT mapping of network
+ addresses. It maps the network address part, while keeping the host
+ address part intact.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
config NETFILTER_XT_TARGET_NFLOG
tristate '"NFLOG" target support'
default m if NETFILTER_ADVANCED=n
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 0baa3f1..600d28b 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -83,6 +83,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_HL) += xt_HL.o
obj-$(CONFIG_NETFILTER_XT_TARGET_HMARK) += xt_HMARK.o
obj-$(CONFIG_NETFILTER_XT_TARGET_LED) += xt_LED.o
obj-$(CONFIG_NETFILTER_XT_TARGET_LOG) += xt_LOG.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_NETMAP) += xt_NETMAP.o
obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o
obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o
obj-$(CONFIG_NETFILTER_XT_TARGET_RATEEST) += xt_RATEEST.o
diff --git a/net/netfilter/xt_NETMAP.c b/net/netfilter/xt_NETMAP.c
new file mode 100644
index 0000000..b253e07
--- /dev/null
+++ b/net/netfilter/xt_NETMAP.c
@@ -0,0 +1,165 @@
+/*
+ * (C) 2000-2001 Svenning Soerensen <svenning@post5.tele.dk>
+ * Copyright (c) 2011 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/ip.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/ipv6.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter_ipv4.h>
+#include <linux/netfilter_ipv6.h>
+#include <linux/netfilter/x_tables.h>
+#include <net/netfilter/nf_nat.h>
+
+static unsigned int
+netmap_tg6(struct sk_buff *skb, const struct xt_action_param *par)
+{
+ const struct nf_nat_range *range = par->targinfo;
+ struct nf_nat_range newrange;
+ struct nf_conn *ct;
+ enum ip_conntrack_info ctinfo;
+ union nf_inet_addr new_addr, netmask;
+ unsigned int i;
+
+ ct = nf_ct_get(skb, &ctinfo);
+ for (i = 0; i < ARRAY_SIZE(range->min_addr.ip6); i++)
+ netmask.ip6[i] = ~(range->min_addr.ip6[i] ^
+ range->max_addr.ip6[i]);
+
+ if (par->hooknum == NF_INET_PRE_ROUTING ||
+ par->hooknum == NF_INET_LOCAL_OUT)
+ new_addr.in6 = ipv6_hdr(skb)->daddr;
+ else
+ new_addr.in6 = ipv6_hdr(skb)->saddr;
+
+ for (i = 0; i < ARRAY_SIZE(new_addr.ip6); i++) {
+ new_addr.ip6[i] &= ~netmask.ip6[i];
+ new_addr.ip6[i] |= range->min_addr.ip6[i] &
+ netmask.ip6[i];
+ }
+
+ newrange.flags = range->flags | NF_NAT_RANGE_MAP_IPS;
+ newrange.min_addr = new_addr;
+ newrange.max_addr = new_addr;
+ newrange.min_proto = range->min_proto;
+ newrange.max_proto = range->max_proto;
+
+ return nf_nat_setup_info(ct, &newrange, HOOK2MANIP(par->hooknum));
+}
+
+static int netmap_tg6_checkentry(const struct xt_tgchk_param *par)
+{
+ const struct nf_nat_range *range = par->targinfo;
+
+ if (!(range->flags & NF_NAT_RANGE_MAP_IPS))
+ return -EINVAL;
+ return 0;
+}
+
+static unsigned int
+netmap_tg4(struct sk_buff *skb, const struct xt_action_param *par)
+{
+ struct nf_conn *ct;
+ enum ip_conntrack_info ctinfo;
+ __be32 new_ip, netmask;
+ const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
+ struct nf_nat_range newrange;
+
+ NF_CT_ASSERT(par->hooknum == NF_INET_PRE_ROUTING ||
+ par->hooknum == NF_INET_POST_ROUTING ||
+ par->hooknum == NF_INET_LOCAL_OUT ||
+ par->hooknum == NF_INET_LOCAL_IN);
+ ct = nf_ct_get(skb, &ctinfo);
+
+ netmask = ~(mr->range[0].min_ip ^ mr->range[0].max_ip);
+
+ if (par->hooknum == NF_INET_PRE_ROUTING ||
+ par->hooknum == NF_INET_LOCAL_OUT)
+ new_ip = ip_hdr(skb)->daddr & ~netmask;
+ else
+ new_ip = ip_hdr(skb)->saddr & ~netmask;
+ new_ip |= mr->range[0].min_ip & netmask;
+
+ memset(&newrange.min_addr, 0, sizeof(newrange.min_addr));
+ memset(&newrange.max_addr, 0, sizeof(newrange.max_addr));
+ newrange.flags = mr->range[0].flags | NF_NAT_RANGE_MAP_IPS;
+ newrange.min_addr.ip = new_ip;
+ newrange.max_addr.ip = new_ip;
+ newrange.min_proto = mr->range[0].min;
+ newrange.max_proto = mr->range[0].max;
+
+ /* Hand modified range to generic setup. */
+ return nf_nat_setup_info(ct, &newrange, HOOK2MANIP(par->hooknum));
+}
+
+static int netmap_tg4_check(const struct xt_tgchk_param *par)
+{
+ const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
+
+ if (!(mr->range[0].flags & NF_NAT_RANGE_MAP_IPS)) {
+ pr_debug("bad MAP_IPS.\n");
+ return -EINVAL;
+ }
+ if (mr->rangesize != 1) {
+ pr_debug("bad rangesize %u.\n", mr->rangesize);
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static struct xt_target netmap_tg_reg[] __read_mostly = {
+ {
+ .name = "NETMAP",
+ .family = NFPROTO_IPV6,
+ .revision = 0,
+ .target = netmap_tg6,
+ .targetsize = sizeof(struct nf_nat_range),
+ .table = "nat",
+ .hooks = (1 << NF_INET_PRE_ROUTING) |
+ (1 << NF_INET_POST_ROUTING) |
+ (1 << NF_INET_LOCAL_OUT) |
+ (1 << NF_INET_LOCAL_IN),
+ .checkentry = netmap_tg6_checkentry,
+ .me = THIS_MODULE,
+ },
+ {
+ .name = "NETMAP",
+ .family = NFPROTO_IPV4,
+ .revision = 0,
+ .target = netmap_tg4,
+ .targetsize = sizeof(struct nf_nat_ipv4_multi_range_compat),
+ .table = "nat",
+ .hooks = (1 << NF_INET_PRE_ROUTING) |
+ (1 << NF_INET_POST_ROUTING) |
+ (1 << NF_INET_LOCAL_OUT) |
+ (1 << NF_INET_LOCAL_IN),
+ .checkentry = netmap_tg4_check,
+ .me = THIS_MODULE,
+ },
+};
+
+static int __init netmap_tg_init(void)
+{
+ return xt_register_targets(netmap_tg_reg, ARRAY_SIZE(netmap_tg_reg));
+}
+
+static void netmap_tg_exit(void)
+{
+ xt_unregister_targets(netmap_tg_reg, ARRAY_SIZE(netmap_tg_reg));
+}
+
+module_init(netmap_tg_init);
+module_exit(netmap_tg_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Xtables: 1:1 NAT mapping of subnets");
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
+MODULE_ALIAS("ip6t_NETMAP");
+MODULE_ALIAS("ipt_NETMAP");
--
1.7.10.4
^ permalink raw reply related
* [PATCH 07/16] netfilter: ipset: Check and reject crazy /0 input parameters
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
bitmap:ip and bitmap:ip,mac type did not reject such a crazy range
when created and using such a set results in a kernel crash.
The hash types just silently ignored such parameters.
Reject invalid /0 input parameters explicitely.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
net/netfilter/ipset/ip_set_bitmap_ip.c | 10 ++++++----
net/netfilter/ipset/ip_set_bitmap_ipmac.c | 5 +++--
net/netfilter/ipset/ip_set_hash_ip.c | 2 +-
net/netfilter/ipset/ip_set_hash_ipport.c | 2 +-
net/netfilter/ipset/ip_set_hash_ipportip.c | 2 +-
net/netfilter/ipset/ip_set_hash_ipportnet.c | 2 +-
6 files changed, 13 insertions(+), 10 deletions(-)
diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c
index 7e1b061..02184b5 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ip.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ip.c
@@ -284,7 +284,7 @@ bitmap_ip_uadt(struct ip_set *set, struct nlattr *tb[],
} else if (tb[IPSET_ATTR_CIDR]) {
u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (cidr > 32)
+ if (!cidr || cidr > 32)
return -IPSET_ERR_INVALID_CIDR;
ip_set_mask_from_to(ip, ip_to, cidr);
} else
@@ -454,7 +454,8 @@ static int
bitmap_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
{
struct bitmap_ip *map;
- u32 first_ip, last_ip, hosts, elements;
+ u32 first_ip, last_ip, hosts;
+ u64 elements;
u8 netmask = 32;
int ret;
@@ -497,7 +498,7 @@ bitmap_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
if (netmask == 32) {
hosts = 1;
- elements = last_ip - first_ip + 1;
+ elements = (u64)last_ip - first_ip + 1;
} else {
u8 mask_bits;
u32 mask;
@@ -515,7 +516,8 @@ bitmap_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
if (elements > IPSET_BITMAP_MAX_RANGE + 1)
return -IPSET_ERR_BITMAP_RANGE_SIZE;
- pr_debug("hosts %u, elements %u\n", hosts, elements);
+ pr_debug("hosts %u, elements %llu\n",
+ hosts, (unsigned long long)elements);
map = kzalloc(sizeof(*map), GFP_KERNEL);
if (!map)
diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
index d7eaf10..6819d3c 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
@@ -557,7 +557,8 @@ static int
bitmap_ipmac_create(struct ip_set *set, struct nlattr *tb[],
u32 flags)
{
- u32 first_ip, last_ip, elements;
+ u32 first_ip, last_ip;
+ u64 elements;
struct bitmap_ipmac *map;
int ret;
@@ -588,7 +589,7 @@ bitmap_ipmac_create(struct ip_set *set, struct nlattr *tb[],
} else
return -IPSET_ERR_PROTOCOL;
- elements = last_ip - first_ip + 1;
+ elements = (u64)last_ip - first_ip + 1;
if (elements > IPSET_BITMAP_MAX_RANGE + 1)
return -IPSET_ERR_BITMAP_RANGE_SIZE;
diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c
index 42bccd5..bc8f76e 100644
--- a/net/netfilter/ipset/ip_set_hash_ip.c
+++ b/net/netfilter/ipset/ip_set_hash_ip.c
@@ -179,7 +179,7 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[],
} else if (tb[IPSET_ATTR_CIDR]) {
u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (cidr > 32)
+ if (!cidr || cidr > 32)
return -IPSET_ERR_INVALID_CIDR;
ip_set_mask_from_to(ip, ip_to, cidr);
} else
diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c
index e0ce0dae..6760fd4 100644
--- a/net/netfilter/ipset/ip_set_hash_ipport.c
+++ b/net/netfilter/ipset/ip_set_hash_ipport.c
@@ -217,7 +217,7 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[],
} else if (tb[IPSET_ATTR_CIDR]) {
u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (cidr > 32)
+ if (!cidr || cidr > 32)
return -IPSET_ERR_INVALID_CIDR;
ip_set_mask_from_to(ip, ip_to, cidr);
} else
diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c
index c864bf4..ac09bec 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportip.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportip.c
@@ -225,7 +225,7 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[],
} else if (tb[IPSET_ATTR_CIDR]) {
u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (cidr > 32)
+ if (!cidr || cidr > 32)
return -IPSET_ERR_INVALID_CIDR;
ip_set_mask_from_to(ip, ip_to, cidr);
} else
diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c
index 2c704bb..242814e 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportnet.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c
@@ -290,7 +290,7 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[],
} else if (tb[IPSET_ATTR_CIDR]) {
u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (cidr > 32)
+ if (!cidr || cidr > 32)
return -IPSET_ERR_INVALID_CIDR;
ip_set_mask_from_to(ip, ip_to, cidr);
}
--
1.7.10.4
^ permalink raw reply related
* [PATCH 08/16] netfilter: ipset: Rewrite cidr book keeping to handle /0
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
include/linux/netfilter/ipset/ip_set_ahash.h | 104 ++++++++++++++------------
1 file changed, 55 insertions(+), 49 deletions(-)
diff --git a/include/linux/netfilter/ipset/ip_set_ahash.h b/include/linux/netfilter/ipset/ip_set_ahash.h
index b114d35..495acdf 100644
--- a/include/linux/netfilter/ipset/ip_set_ahash.h
+++ b/include/linux/netfilter/ipset/ip_set_ahash.h
@@ -137,50 +137,59 @@ htable_bits(u32 hashsize)
#endif
#define SET_HOST_MASK(family) (family == AF_INET ? 32 : 128)
+#ifdef IP_SET_HASH_WITH_MULTI
+#define NETS_LENGTH(family) (SET_HOST_MASK(family) + 1)
+#else
+#define NETS_LENGTH(family) SET_HOST_MASK(family)
+#endif
/* Network cidr size book keeping when the hash stores different
* sized networks */
static void
-add_cidr(struct ip_set_hash *h, u8 cidr, u8 host_mask)
+add_cidr(struct ip_set_hash *h, u8 cidr, u8 nets_length)
{
- u8 i;
-
- ++h->nets[cidr-1].nets;
-
- pr_debug("add_cidr added %u: %u\n", cidr, h->nets[cidr-1].nets);
+ int i, j;
- if (h->nets[cidr-1].nets > 1)
- return;
-
- /* New cidr size */
- for (i = 0; i < host_mask && h->nets[i].cidr; i++) {
- /* Add in increasing prefix order, so larger cidr first */
- if (h->nets[i].cidr < cidr)
- swap(h->nets[i].cidr, cidr);
+ /* Add in increasing prefix order, so larger cidr first */
+ for (i = 0, j = -1; i < nets_length && h->nets[i].nets; i++) {
+ if (j != -1)
+ continue;
+ else if (h->nets[i].cidr < cidr)
+ j = i;
+ else if (h->nets[i].cidr == cidr) {
+ h->nets[i].nets++;
+ return;
+ }
+ }
+ if (j != -1) {
+ for (; i > j; i--) {
+ h->nets[i].cidr = h->nets[i - 1].cidr;
+ h->nets[i].nets = h->nets[i - 1].nets;
+ }
}
- if (i < host_mask)
- h->nets[i].cidr = cidr;
+ h->nets[i].cidr = cidr;
+ h->nets[i].nets = 1;
}
static void
-del_cidr(struct ip_set_hash *h, u8 cidr, u8 host_mask)
+del_cidr(struct ip_set_hash *h, u8 cidr, u8 nets_length)
{
- u8 i;
-
- --h->nets[cidr-1].nets;
+ u8 i, j;
- pr_debug("del_cidr deleted %u: %u\n", cidr, h->nets[cidr-1].nets);
+ for (i = 0; i < nets_length - 1 && h->nets[i].cidr != cidr; i++)
+ ;
+ h->nets[i].nets--;
- if (h->nets[cidr-1].nets != 0)
+ if (h->nets[i].nets != 0)
return;
- /* All entries with this cidr size deleted, so cleanup h->cidr[] */
- for (i = 0; i < host_mask - 1 && h->nets[i].cidr; i++) {
- if (h->nets[i].cidr == cidr)
- h->nets[i].cidr = cidr = h->nets[i+1].cidr;
+ for (j = i; j < nets_length - 1 && h->nets[j].nets; j++) {
+ h->nets[j].cidr = h->nets[j + 1].cidr;
+ h->nets[j].nets = h->nets[j + 1].nets;
}
- h->nets[i - 1].cidr = 0;
}
+#else
+#define NETS_LENGTH(family) 0
#endif
/* Destroy the hashtable part of the set */
@@ -202,14 +211,14 @@ ahash_destroy(struct htable *t)
/* Calculate the actual memory size of the set data */
static size_t
-ahash_memsize(const struct ip_set_hash *h, size_t dsize, u8 host_mask)
+ahash_memsize(const struct ip_set_hash *h, size_t dsize, u8 nets_length)
{
u32 i;
struct htable *t = h->table;
size_t memsize = sizeof(*h)
+ sizeof(*t)
#ifdef IP_SET_HASH_WITH_NETS
- + sizeof(struct ip_set_hash_nets) * host_mask
+ + sizeof(struct ip_set_hash_nets) * nets_length
#endif
+ jhash_size(t->htable_bits) * sizeof(struct hbucket);
@@ -238,7 +247,7 @@ ip_set_hash_flush(struct ip_set *set)
}
#ifdef IP_SET_HASH_WITH_NETS
memset(h->nets, 0, sizeof(struct ip_set_hash_nets)
- * SET_HOST_MASK(set->family));
+ * NETS_LENGTH(set->family));
#endif
h->elements = 0;
}
@@ -271,9 +280,6 @@ ip_set_hash_destroy(struct ip_set *set)
(jhash2((u32 *)(data), HKEY_DATALEN/sizeof(u32), initval) \
& jhash_mask(htable_bits))
-#define CONCAT(a, b, c) a##b##c
-#define TOKEN(a, b, c) CONCAT(a, b, c)
-
/* Type/family dependent function prototypes */
#define type_pf_data_equal TOKEN(TYPE, PF, _data_equal)
@@ -478,7 +484,7 @@ type_pf_add(struct ip_set *set, void *value, u32 timeout, u32 flags)
}
#ifdef IP_SET_HASH_WITH_NETS
- add_cidr(h, CIDR(d->cidr), HOST_MASK);
+ add_cidr(h, CIDR(d->cidr), NETS_LENGTH(set->family));
#endif
h->elements++;
out:
@@ -513,7 +519,7 @@ type_pf_del(struct ip_set *set, void *value, u32 timeout, u32 flags)
n->pos--;
h->elements--;
#ifdef IP_SET_HASH_WITH_NETS
- del_cidr(h, CIDR(d->cidr), HOST_MASK);
+ del_cidr(h, CIDR(d->cidr), NETS_LENGTH(set->family));
#endif
if (n->pos + AHASH_INIT_SIZE < n->size) {
void *tmp = kzalloc((n->size - AHASH_INIT_SIZE)
@@ -546,10 +552,10 @@ type_pf_test_cidrs(struct ip_set *set, struct type_pf_elem *d, u32 timeout)
const struct type_pf_elem *data;
int i, j = 0;
u32 key, multi = 0;
- u8 host_mask = SET_HOST_MASK(set->family);
+ u8 nets_length = NETS_LENGTH(set->family);
pr_debug("test by nets\n");
- for (; j < host_mask && h->nets[j].cidr && !multi; j++) {
+ for (; j < nets_length && h->nets[j].nets && !multi; j++) {
type_pf_data_netmask(d, h->nets[j].cidr);
key = HKEY(d, h->initval, t->htable_bits);
n = hbucket(t, key);
@@ -604,7 +610,7 @@ type_pf_head(struct ip_set *set, struct sk_buff *skb)
memsize = ahash_memsize(h, with_timeout(h->timeout)
? sizeof(struct type_pf_telem)
: sizeof(struct type_pf_elem),
- set->family == AF_INET ? 32 : 128);
+ NETS_LENGTH(set->family));
read_unlock_bh(&set->lock);
nested = ipset_nest_start(skb, IPSET_ATTR_DATA);
@@ -783,7 +789,7 @@ type_pf_elem_tadd(struct hbucket *n, const struct type_pf_elem *value,
/* Delete expired elements from the hashtable */
static void
-type_pf_expire(struct ip_set_hash *h)
+type_pf_expire(struct ip_set_hash *h, u8 nets_length)
{
struct htable *t = h->table;
struct hbucket *n;
@@ -798,7 +804,7 @@ type_pf_expire(struct ip_set_hash *h)
if (type_pf_data_expired(data)) {
pr_debug("expired %u/%u\n", i, j);
#ifdef IP_SET_HASH_WITH_NETS
- del_cidr(h, CIDR(data->cidr), HOST_MASK);
+ del_cidr(h, CIDR(data->cidr), nets_length);
#endif
if (j != n->pos - 1)
/* Not last one */
@@ -839,7 +845,7 @@ type_pf_tresize(struct ip_set *set, bool retried)
if (!retried) {
i = h->elements;
write_lock_bh(&set->lock);
- type_pf_expire(set->data);
+ type_pf_expire(set->data, NETS_LENGTH(set->family));
write_unlock_bh(&set->lock);
if (h->elements < i)
return 0;
@@ -904,7 +910,7 @@ type_pf_tadd(struct ip_set *set, void *value, u32 timeout, u32 flags)
if (h->elements >= h->maxelem)
/* FIXME: when set is full, we slow down here */
- type_pf_expire(h);
+ type_pf_expire(h, NETS_LENGTH(set->family));
if (h->elements >= h->maxelem) {
if (net_ratelimit())
pr_warning("Set %s is full, maxelem %u reached\n",
@@ -933,8 +939,8 @@ type_pf_tadd(struct ip_set *set, void *value, u32 timeout, u32 flags)
if (j != AHASH_MAX(h) + 1) {
data = ahash_tdata(n, j);
#ifdef IP_SET_HASH_WITH_NETS
- del_cidr(h, CIDR(data->cidr), HOST_MASK);
- add_cidr(h, CIDR(d->cidr), HOST_MASK);
+ del_cidr(h, CIDR(data->cidr), NETS_LENGTH(set->family));
+ add_cidr(h, CIDR(d->cidr), NETS_LENGTH(set->family));
#endif
type_pf_data_copy(data, d);
type_pf_data_timeout_set(data, timeout);
@@ -952,7 +958,7 @@ type_pf_tadd(struct ip_set *set, void *value, u32 timeout, u32 flags)
}
#ifdef IP_SET_HASH_WITH_NETS
- add_cidr(h, CIDR(d->cidr), HOST_MASK);
+ add_cidr(h, CIDR(d->cidr), NETS_LENGTH(set->family));
#endif
h->elements++;
out:
@@ -986,7 +992,7 @@ type_pf_tdel(struct ip_set *set, void *value, u32 timeout, u32 flags)
n->pos--;
h->elements--;
#ifdef IP_SET_HASH_WITH_NETS
- del_cidr(h, CIDR(d->cidr), HOST_MASK);
+ del_cidr(h, CIDR(d->cidr), NETS_LENGTH(set->family));
#endif
if (n->pos + AHASH_INIT_SIZE < n->size) {
void *tmp = kzalloc((n->size - AHASH_INIT_SIZE)
@@ -1016,9 +1022,9 @@ type_pf_ttest_cidrs(struct ip_set *set, struct type_pf_elem *d, u32 timeout)
struct hbucket *n;
int i, j = 0;
u32 key, multi = 0;
- u8 host_mask = SET_HOST_MASK(set->family);
+ u8 nets_length = NETS_LENGTH(set->family);
- for (; j < host_mask && h->nets[j].cidr && !multi; j++) {
+ for (; j < nets_length && h->nets[j].nets && !multi; j++) {
type_pf_data_netmask(d, h->nets[j].cidr);
key = HKEY(d, h->initval, t->htable_bits);
n = hbucket(t, key);
@@ -1147,7 +1153,7 @@ type_pf_gc(unsigned long ul_set)
pr_debug("called\n");
write_lock_bh(&set->lock);
- type_pf_expire(h);
+ type_pf_expire(h, NETS_LENGTH(set->family));
write_unlock_bh(&set->lock);
h->gc.expires = jiffies + IPSET_GC_PERIOD(h->timeout) * HZ;
--
1.7.10.4
^ permalink raw reply related
* [PATCH 09/16] netfilter: ipset: Add /0 network support to hash:net,iface type
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Now it is possible to setup a single hash:net,iface type of set and
a single ip6?tables match which covers all egress/ingress filtering.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
net/netfilter/ipset/ip_set_hash_netiface.c | 44 +++++++++++++---------------
1 file changed, 21 insertions(+), 23 deletions(-)
diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c
index 632693f..e7c671d 100644
--- a/net/netfilter/ipset/ip_set_hash_netiface.c
+++ b/net/netfilter/ipset/ip_set_hash_netiface.c
@@ -140,7 +140,7 @@ struct hash_netiface4_elem_hashed {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
};
#define HKEY_DATALEN sizeof(struct hash_netiface4_elem_hashed)
@@ -151,7 +151,7 @@ struct hash_netiface4_elem {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
const char *iface;
};
@@ -161,7 +161,7 @@ struct hash_netiface4_telem {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
const char *iface;
unsigned long timeout;
};
@@ -181,18 +181,14 @@ hash_netiface4_data_equal(const struct hash_netiface4_elem *ip1,
static inline bool
hash_netiface4_data_isnull(const struct hash_netiface4_elem *elem)
{
- return elem->cidr == 0;
+ return elem->elem == 0;
}
static inline void
hash_netiface4_data_copy(struct hash_netiface4_elem *dst,
const struct hash_netiface4_elem *src)
{
- dst->ip = src->ip;
- dst->cidr = src->cidr;
- dst->physdev = src->physdev;
- dst->iface = src->iface;
- dst->nomatch = src->nomatch;
+ memcpy(dst, src, sizeof(*dst));
}
static inline void
@@ -217,7 +213,7 @@ hash_netiface4_data_netmask(struct hash_netiface4_elem *elem, u8 cidr)
static inline void
hash_netiface4_data_zero_out(struct hash_netiface4_elem *elem)
{
- elem->cidr = 0;
+ elem->elem = 0;
}
static bool
@@ -288,7 +284,8 @@ hash_netiface4_kadt(struct ip_set *set, const struct sk_buff *skb,
struct ip_set_hash *h = set->data;
ipset_adtfn adtfn = set->variant->adt[adt];
struct hash_netiface4_elem data = {
- .cidr = h->nets[0].cidr ? h->nets[0].cidr : HOST_MASK
+ .cidr = h->nets[0].cidr ? h->nets[0].cidr : HOST_MASK,
+ .elem = 1,
};
int ret;
@@ -339,7 +336,7 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[],
{
struct ip_set_hash *h = set->data;
ipset_adtfn adtfn = set->variant->adt[adt];
- struct hash_netiface4_elem data = { .cidr = HOST_MASK };
+ struct hash_netiface4_elem data = { .cidr = HOST_MASK, .elem = 1 };
u32 ip = 0, ip_to, last;
u32 timeout = h->timeout;
char iface[IFNAMSIZ];
@@ -360,7 +357,7 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[],
if (tb[IPSET_ATTR_CIDR]) {
data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (!data.cidr || data.cidr > HOST_MASK)
+ if (data.cidr > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
}
@@ -389,7 +386,6 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[],
if (adt == IPSET_ADD && (cadt_flags & IPSET_FLAG_NOMATCH))
flags |= (cadt_flags << 16);
}
-
if (adt == IPSET_TEST || !tb[IPSET_ATTR_IP_TO]) {
data.ip = htonl(ip & ip_set_hostmask(data.cidr));
ret = adtfn(set, &data, timeout, flags);
@@ -442,7 +438,7 @@ struct hash_netiface6_elem_hashed {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
};
#define HKEY_DATALEN sizeof(struct hash_netiface6_elem_hashed)
@@ -452,7 +448,7 @@ struct hash_netiface6_elem {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
const char *iface;
};
@@ -461,7 +457,7 @@ struct hash_netiface6_telem {
u8 physdev;
u8 cidr;
u8 nomatch;
- u8 padding;
+ u8 elem;
const char *iface;
unsigned long timeout;
};
@@ -481,7 +477,7 @@ hash_netiface6_data_equal(const struct hash_netiface6_elem *ip1,
static inline bool
hash_netiface6_data_isnull(const struct hash_netiface6_elem *elem)
{
- return elem->cidr == 0;
+ return elem->elem == 0;
}
static inline void
@@ -506,7 +502,7 @@ hash_netiface6_data_match(const struct hash_netiface6_elem *elem)
static inline void
hash_netiface6_data_zero_out(struct hash_netiface6_elem *elem)
{
- elem->cidr = 0;
+ elem->elem = 0;
}
static inline void
@@ -590,7 +586,8 @@ hash_netiface6_kadt(struct ip_set *set, const struct sk_buff *skb,
struct ip_set_hash *h = set->data;
ipset_adtfn adtfn = set->variant->adt[adt];
struct hash_netiface6_elem data = {
- .cidr = h->nets[0].cidr ? h->nets[0].cidr : HOST_MASK
+ .cidr = h->nets[0].cidr ? h->nets[0].cidr : HOST_MASK,
+ .elem = 1,
};
int ret;
@@ -637,7 +634,7 @@ hash_netiface6_uadt(struct ip_set *set, struct nlattr *tb[],
{
struct ip_set_hash *h = set->data;
ipset_adtfn adtfn = set->variant->adt[adt];
- struct hash_netiface6_elem data = { .cidr = HOST_MASK };
+ struct hash_netiface6_elem data = { .cidr = HOST_MASK, .elem = 1 };
u32 timeout = h->timeout;
char iface[IFNAMSIZ];
int ret;
@@ -659,7 +656,7 @@ hash_netiface6_uadt(struct ip_set *set, struct nlattr *tb[],
if (tb[IPSET_ATTR_CIDR])
data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (!data.cidr || data.cidr > HOST_MASK)
+ if (data.cidr > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
ip6_netmask(&data.ip, data.cidr);
@@ -777,7 +774,8 @@ static struct ip_set_type hash_netiface_type __read_mostly = {
.dimension = IPSET_DIM_TWO,
.family = NFPROTO_UNSPEC,
.revision_min = 0,
- .revision_max = 1, /* nomatch flag support added */
+ /* = 1, nomatch flag support added */
+ .revision_max = 2, /* /0 support added */
.create = hash_netiface_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
--
1.7.10.4
^ permalink raw reply related
* [PATCH 10/16] netfilter: ipset: Include supported revisions in module description
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
include/linux/netfilter/ipset/ip_set.h | 6 ++++++
net/netfilter/ipset/ip_set_bitmap_ip.c | 9 ++++++---
net/netfilter/ipset/ip_set_bitmap_ipmac.c | 9 ++++++---
net/netfilter/ipset/ip_set_bitmap_port.c | 9 ++++++---
net/netfilter/ipset/ip_set_hash_ip.c | 9 ++++++---
net/netfilter/ipset/ip_set_hash_ipport.c | 9 ++++++---
net/netfilter/ipset/ip_set_hash_ipportip.c | 9 ++++++---
net/netfilter/ipset/ip_set_hash_ipportnet.c | 13 ++++++++-----
net/netfilter/ipset/ip_set_hash_net.c | 11 +++++++----
net/netfilter/ipset/ip_set_hash_netiface.c | 11 +++++++----
net/netfilter/ipset/ip_set_hash_netport.c | 13 ++++++++-----
net/netfilter/ipset/ip_set_list_set.c | 9 ++++++---
12 files changed, 78 insertions(+), 39 deletions(-)
diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h
index 2edc64c..1d954c6 100644
--- a/include/linux/netfilter/ipset/ip_set.h
+++ b/include/linux/netfilter/ipset/ip_set.h
@@ -206,9 +206,15 @@ enum ip_set_kopt {
#include <linux/netlink.h>
#include <linux/netfilter.h>
#include <linux/netfilter/x_tables.h>
+#include <linux/stringify.h>
#include <linux/vmalloc.h>
#include <net/netlink.h>
+#define _IP_SET_MODULE_DESC(a, b, c) \
+ MODULE_DESCRIPTION(a " type of IP sets, revisions " b "-" c)
+#define IP_SET_MODULE_DESC(a, b, c) \
+ _IP_SET_MODULE_DESC(a, __stringify(b), __stringify(c))
+
/* Set features */
enum ip_set_feature {
IPSET_TYPE_IP_FLAG = 0,
diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c
index 02184b5..4a92fd4 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ip.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ip.c
@@ -27,9 +27,12 @@
#define IP_SET_BITMAP_TIMEOUT
#include <linux/netfilter/ipset/ip_set_timeout.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 0
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("bitmap:ip type of IP sets");
+IP_SET_MODULE_DESC("bitmap:ip", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_bitmap:ip");
/* Type structure */
@@ -556,8 +559,8 @@ static struct ip_set_type bitmap_ip_type __read_mostly = {
.features = IPSET_TYPE_IP,
.dimension = IPSET_DIM_ONE,
.family = NFPROTO_IPV4,
- .revision_min = 0,
- .revision_max = 0,
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = bitmap_ip_create,
.create_policy = {
[IPSET_ATTR_IP] = { .type = NLA_NESTED },
diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
index 6819d3c..645c9d1 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
@@ -26,9 +26,12 @@
#include <linux/netfilter/ipset/ip_set_timeout.h>
#include <linux/netfilter/ipset/ip_set_bitmap.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 0
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("bitmap:ip,mac type of IP sets");
+IP_SET_MODULE_DESC("bitmap:ip,mac", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_bitmap:ip,mac");
enum {
@@ -630,8 +633,8 @@ static struct ip_set_type bitmap_ipmac_type = {
.features = IPSET_TYPE_IP | IPSET_TYPE_MAC,
.dimension = IPSET_DIM_TWO,
.family = NFPROTO_IPV4,
- .revision_min = 0,
- .revision_max = 0,
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = bitmap_ipmac_create,
.create_policy = {
[IPSET_ATTR_IP] = { .type = NLA_NESTED },
diff --git a/net/netfilter/ipset/ip_set_bitmap_port.c b/net/netfilter/ipset/ip_set_bitmap_port.c
index b9f1fce..e6b2db7 100644
--- a/net/netfilter/ipset/ip_set_bitmap_port.c
+++ b/net/netfilter/ipset/ip_set_bitmap_port.c
@@ -22,9 +22,12 @@
#define IP_SET_BITMAP_TIMEOUT
#include <linux/netfilter/ipset/ip_set_timeout.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 0
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("bitmap:port type of IP sets");
+IP_SET_MODULE_DESC("bitmap:port", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_bitmap:port");
/* Type structure */
@@ -487,8 +490,8 @@ static struct ip_set_type bitmap_port_type = {
.features = IPSET_TYPE_PORT,
.dimension = IPSET_DIM_ONE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- .revision_max = 0,
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = bitmap_port_create,
.create_policy = {
[IPSET_ATTR_PORT] = { .type = NLA_U16 },
diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c
index bc8f76e..ec3dba5 100644
--- a/net/netfilter/ipset/ip_set_hash_ip.c
+++ b/net/netfilter/ipset/ip_set_hash_ip.c
@@ -24,9 +24,12 @@
#include <linux/netfilter/ipset/ip_set_timeout.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 0
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:ip type of IP sets");
+IP_SET_MODULE_DESC("hash:ip", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:ip");
/* Type specific function prefix */
@@ -452,8 +455,8 @@ static struct ip_set_type hash_ip_type __read_mostly = {
.features = IPSET_TYPE_IP,
.dimension = IPSET_DIM_ONE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- .revision_max = 0,
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_ip_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c
index 6760fd4..0171f75 100644
--- a/net/netfilter/ipset/ip_set_hash_ipport.c
+++ b/net/netfilter/ipset/ip_set_hash_ipport.c
@@ -25,9 +25,12 @@
#include <linux/netfilter/ipset/ip_set_getport.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 1 /* SCTP and UDPLITE support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:ip,port type of IP sets");
+IP_SET_MODULE_DESC("hash:ip,port", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:ip,port");
/* Type specific function prefix */
@@ -523,8 +526,8 @@ static struct ip_set_type hash_ipport_type __read_mostly = {
.features = IPSET_TYPE_IP | IPSET_TYPE_PORT,
.dimension = IPSET_DIM_TWO,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- .revision_max = 1, /* SCTP and UDPLITE support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_ipport_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c
index ac09bec..6344ef5 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportip.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportip.c
@@ -25,9 +25,12 @@
#include <linux/netfilter/ipset/ip_set_getport.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 1 /* SCTP and UDPLITE support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:ip,port,ip type of IP sets");
+IP_SET_MODULE_DESC("hash:ip,port,ip", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:ip,port,ip");
/* Type specific function prefix */
@@ -541,8 +544,8 @@ static struct ip_set_type hash_ipportip_type __read_mostly = {
.features = IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2,
.dimension = IPSET_DIM_THREE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- .revision_max = 1, /* SCTP and UDPLITE support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_ipportip_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c
index 242814e..8ee9168 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportnet.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c
@@ -25,9 +25,14 @@
#include <linux/netfilter/ipset/ip_set_getport.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+/* 1 SCTP and UDPLITE support added */
+/* 2 Range as input support for IPv4 added */
+#define REVISION_MAX 3 /* nomatch flag support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:ip,port,net type of IP sets");
+IP_SET_MODULE_DESC("hash:ip,port,net", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:ip,port,net");
/* Type specific function prefix */
@@ -695,10 +700,8 @@ static struct ip_set_type hash_ipportnet_type __read_mostly = {
.features = IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2,
.dimension = IPSET_DIM_THREE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- /* 1 SCTP and UDPLITE support added */
- /* 2 Range as input support for IPv4 added */
- .revision_max = 3, /* nomatch flag support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_ipportnet_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c
index d676093..014ff72 100644
--- a/net/netfilter/ipset/ip_set_hash_net.c
+++ b/net/netfilter/ipset/ip_set_hash_net.c
@@ -23,9 +23,13 @@
#include <linux/netfilter/ipset/ip_set_timeout.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+/* 1 Range as input support for IPv4 added */
+#define REVISION_MAX 2 /* nomatch flag support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:net type of IP sets");
+IP_SET_MODULE_DESC("hash:net", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:net");
/* Type specific function prefix */
@@ -535,9 +539,8 @@ static struct ip_set_type hash_net_type __read_mostly = {
.features = IPSET_TYPE_IP,
.dimension = IPSET_DIM_ONE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- /* = 1 Range as input support for IPv4 added */
- .revision_max = 2, /* nomatch flag support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_net_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c
index e7c671d..a5c8491 100644
--- a/net/netfilter/ipset/ip_set_hash_netiface.c
+++ b/net/netfilter/ipset/ip_set_hash_netiface.c
@@ -24,9 +24,13 @@
#include <linux/netfilter/ipset/ip_set_timeout.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+/* 1 nomatch flag support added */
+#define REVISION_MAX 2 /* /0 support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:net,iface type of IP sets");
+IP_SET_MODULE_DESC("hash:net,iface", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:net,iface");
/* Interface name rbtree */
@@ -773,9 +777,8 @@ static struct ip_set_type hash_netiface_type __read_mostly = {
.features = IPSET_TYPE_IP | IPSET_TYPE_IFACE,
.dimension = IPSET_DIM_TWO,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- /* = 1, nomatch flag support added */
- .revision_max = 2, /* /0 support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_netiface_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c
index 3ec27fc..7ca357a 100644
--- a/net/netfilter/ipset/ip_set_hash_netport.c
+++ b/net/netfilter/ipset/ip_set_hash_netport.c
@@ -24,9 +24,14 @@
#include <linux/netfilter/ipset/ip_set_getport.h>
#include <linux/netfilter/ipset/ip_set_hash.h>
+#define REVISION_MIN 0
+/* 1 SCTP and UDPLITE support added */
+/* 2 Range as input support for IPv4 added */
+#define REVISION_MAX 3 /* nomatch flag support added */
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("hash:net,port type of IP sets");
+IP_SET_MODULE_DESC("hash:net,port", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_hash:net,port");
/* Type specific function prefix */
@@ -648,10 +653,8 @@ static struct ip_set_type hash_netport_type __read_mostly = {
.features = IPSET_TYPE_IP | IPSET_TYPE_PORT,
.dimension = IPSET_DIM_TWO,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- /* 1 SCTP and UDPLITE support added */
- /* 2, Range as input support for IPv4 added */
- .revision_max = 3, /* nomatch flag support added */
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = hash_netport_create,
.create_policy = {
[IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 },
diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c
index 6cb1225..8371c2b 100644
--- a/net/netfilter/ipset/ip_set_list_set.c
+++ b/net/netfilter/ipset/ip_set_list_set.c
@@ -16,9 +16,12 @@
#include <linux/netfilter/ipset/ip_set_timeout.h>
#include <linux/netfilter/ipset/ip_set_list.h>
+#define REVISION_MIN 0
+#define REVISION_MAX 0
+
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("list:set type of IP sets");
+IP_SET_MODULE_DESC("list:set", REVISION_MIN, REVISION_MAX);
MODULE_ALIAS("ip_set_list:set");
/* Member elements without and with timeout */
@@ -579,8 +582,8 @@ static struct ip_set_type list_set_type __read_mostly = {
.features = IPSET_TYPE_NAME | IPSET_DUMP_LAST,
.dimension = IPSET_DIM_ONE,
.family = NFPROTO_UNSPEC,
- .revision_min = 0,
- .revision_max = 0,
+ .revision_min = REVISION_MIN,
+ .revision_max = REVISION_MAX,
.create = list_set_create,
.create_policy = {
[IPSET_ATTR_SIZE] = { .type = NLA_U32 },
--
1.7.10.4
^ permalink raw reply related
* [PATCH 13/16] netfilter: xt_time: add support to ignore day transition
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Florian Westphal <fw@strlen.de>
Currently, if you want to do something like:
"match Monday, starting 23:00, for two hours"
You need two rules, one for Mon 23:00 to 0:00 and one for Tue 0:00-1:00.
The rule: --weekdays Mo --timestart 23:00 --timestop 01:00
looks correct, but it will first match on monday from midnight to 1 a.m.
and then again for another hour from 23:00 onwards.
This permits userspace to explicitly ignore the day transition and
match for a single, continuous time period instead.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
include/linux/netfilter/xt_time.h | 5 +++++
net/netfilter/xt_time.c | 24 +++++++++++++++++++++++-
2 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/include/linux/netfilter/xt_time.h b/include/linux/netfilter/xt_time.h
index 7c37fac..0958860 100644
--- a/include/linux/netfilter/xt_time.h
+++ b/include/linux/netfilter/xt_time.h
@@ -17,6 +17,9 @@ enum {
/* Match against local time (instead of UTC) */
XT_TIME_LOCAL_TZ = 1 << 0,
+ /* treat timestart > timestop (e.g. 23:00-01:00) as single period */
+ XT_TIME_CONTIGUOUS = 1 << 1,
+
/* Shortcuts */
XT_TIME_ALL_MONTHDAYS = 0xFFFFFFFE,
XT_TIME_ALL_WEEKDAYS = 0xFE,
@@ -24,4 +27,6 @@ enum {
XT_TIME_MAX_DAYTIME = 24 * 60 * 60 - 1,
};
+#define XT_TIME_ALL_FLAGS (XT_TIME_LOCAL_TZ|XT_TIME_CONTIGUOUS)
+
#endif /* _XT_TIME_H */
diff --git a/net/netfilter/xt_time.c b/net/netfilter/xt_time.c
index c48975f..0ae55a3 100644
--- a/net/netfilter/xt_time.c
+++ b/net/netfilter/xt_time.c
@@ -42,6 +42,7 @@ static const u_int16_t days_since_leapyear[] = {
*/
enum {
DSE_FIRST = 2039,
+ SECONDS_PER_DAY = 86400,
};
static const u_int16_t days_since_epoch[] = {
/* 2039 - 2030 */
@@ -78,7 +79,7 @@ static inline unsigned int localtime_1(struct xtm *r, time_t time)
unsigned int v, w;
/* Each day has 86400s, so finding the hour/minute is actually easy. */
- v = time % 86400;
+ v = time % SECONDS_PER_DAY;
r->second = v % 60;
w = v / 60;
r->minute = w % 60;
@@ -199,6 +200,18 @@ time_mt(const struct sk_buff *skb, struct xt_action_param *par)
if (packet_time < info->daytime_start &&
packet_time > info->daytime_stop)
return false;
+
+ /** if user asked to ignore 'next day', then e.g.
+ * '1 PM Wed, August 1st' should be treated
+ * like 'Tue 1 PM July 31st'.
+ *
+ * This also causes
+ * 'Monday, "23:00 to 01:00", to match for 2 hours, starting
+ * Monday 23:00 to Tuesday 01:00.
+ */
+ if ((info->flags & XT_TIME_CONTIGUOUS) &&
+ packet_time <= info->daytime_stop)
+ stamp -= SECONDS_PER_DAY;
}
localtime_2(¤t_time, stamp);
@@ -227,6 +240,15 @@ static int time_mt_check(const struct xt_mtchk_param *par)
return -EDOM;
}
+ if (info->flags & ~XT_TIME_ALL_FLAGS) {
+ pr_info("unknown flags 0x%x\n", info->flags & ~XT_TIME_ALL_FLAGS);
+ return -EINVAL;
+ }
+
+ if ((info->flags & XT_TIME_CONTIGUOUS) &&
+ info->daytime_start < info->daytime_stop)
+ return -EINVAL;
+
return 0;
}
--
1.7.10.4
^ permalink raw reply related
* [PATCH 00/16] netfilter updates for net-next
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
From: Pablo Neira Ayuso <pablo@netfilter.org>
Hi David,
This patchset contains updates for your net-next tree, they are:
* Mostly fixes for the recently pushed IPv6 NAT support:
- Fix crash while removing nf_nat modules from Patrick McHardy.
- Fix unbalanced rcu_read_unlock from Ulrich Weber.
- Merge NETMAP and REDIRECT into one single xt_target module, from
Jan Engelhardt.
- Fix Kconfig for IPv6 NAT, which allows inconsistent configurations,
from myself.
* Updates for ipset, all of the from Jozsef Kadlecsik:
- Add the new "nomatch" option to obtain reverse set matching.
- Support for /0 CIDR in hash:net,iface set type.
- One non-critical fix for a rare crash due to pass really
wrong configuration parameters.
- Coding style cleanups.
- Sparse fixes.
- Add set revision supported via modinfo.i
* One extension for the xt_time match, to support matching during
the transition between two days with one single rule, from
Florian Westphal.
* Fix maximum packet length supported by nfnetlink_queue and add
NFQA_CAP_LEN attribute, from myself.
You can notice that this batch contains a couple of fixes that may
go to 3.6-rc but I don't consider them critical to push them:
* The ipset fix for the /0 cidr case, which is triggered with one
inconsistent command line invocation of ipset.
* The nfnetlink_queue maximum packet length supported since it requires
the new NFQA_CAP_LEN attribute to provide a full workaround for the
described problem.
You can pull these changes from:
git://1984.lsi.us.es/nf-next master
Thanks!
Florian Westphal (1):
netfilter: xt_time: add support to ignore day transition
Jan Engelhardt (2):
netfilter: combine ipt_NETMAP and ip6t_NETMAP
netfilter: combine ipt_REDIRECT and ip6t_REDIRECT
Jozsef Kadlecsik (7):
netfilter: ipset: Fix sparse warnings "incorrect type in assignment"
netfilter: ipset: Check and reject crazy /0 input parameters
netfilter: ipset: Rewrite cidr book keeping to handle /0
netfilter: ipset: Add /0 network support to hash:net,iface type
netfilter: ipset: Include supported revisions in module description
netfilter: ipset: Coding style fixes
netfilter: ipset: Support to match elements marked with "nomatch"
Pablo Neira Ayuso (4):
netfilter: fix IPv6 NAT dependencies in Kconfig
netfilter: nf_ct_ftp: add sequence tracking pickup facility for injected entries
netfilter: nfnetlink_queue: fix maximum packet length to userspace
netfilter: nfnetlink_queue: add NFQA_CAP_LEN attribute
Patrick McHardy (1):
netfilter: nf_nat: fix oops when unloading protocol modules
Ulrich Weber (1):
netfilter: nf_nat: remove obsolete rcu_read_unlock call
include/linux/netfilter/ipset/ip_set.h | 15 +-
include/linux/netfilter/ipset/ip_set_ahash.h | 106 +++++++-------
include/linux/netfilter/nf_conntrack_ftp.h | 6 +-
include/linux/netfilter/nfnetlink_queue.h | 1 +
include/linux/netfilter/xt_time.h | 5 +
net/ipv4/netfilter/Kconfig | 23 ++--
net/ipv4/netfilter/Makefile | 2 -
net/ipv4/netfilter/ipt_NETMAP.c | 101 --------------
net/ipv4/netfilter/ipt_REDIRECT.c | 113 ---------------
net/ipv6/netfilter/Kconfig | 91 +++++-------
net/ipv6/netfilter/Makefile | 2 -
net/ipv6/netfilter/ip6t_NETMAP.c | 94 -------------
net/ipv6/netfilter/ip6t_REDIRECT.c | 98 -------------
net/netfilter/Kconfig | 21 +++
net/netfilter/Makefile | 2 +
net/netfilter/ipset/ip_set_bitmap_ip.c | 19 ++-
net/netfilter/ipset/ip_set_bitmap_ipmac.c | 18 ++-
net/netfilter/ipset/ip_set_bitmap_port.c | 9 +-
net/netfilter/ipset/ip_set_core.c | 15 +-
net/netfilter/ipset/ip_set_hash_ip.c | 15 +-
net/netfilter/ipset/ip_set_hash_ipport.c | 24 ++--
net/netfilter/ipset/ip_set_hash_ipportip.c | 24 ++--
net/netfilter/ipset/ip_set_hash_ipportnet.c | 47 ++++---
net/netfilter/ipset/ip_set_hash_net.c | 25 ++--
net/netfilter/ipset/ip_set_hash_netiface.c | 66 ++++-----
net/netfilter/ipset/ip_set_hash_netport.c | 36 ++---
net/netfilter/ipset/ip_set_list_set.c | 9 +-
net/netfilter/nf_conntrack_core.c | 2 +
net/netfilter/nf_conntrack_ftp.c | 21 +++
net/netfilter/nf_conntrack_netlink.c | 4 +-
net/netfilter/nf_nat_core.c | 6 +-
net/netfilter/nfnetlink_cthelper.c | 3 +
net/netfilter/nfnetlink_queue_core.c | 19 ++-
net/netfilter/xt_NETMAP.c | 165 ++++++++++++++++++++++
net/netfilter/xt_REDIRECT.c | 190 ++++++++++++++++++++++++++
net/netfilter/xt_set.c | 22 +++
net/netfilter/xt_time.c | 24 +++-
37 files changed, 773 insertions(+), 670 deletions(-)
delete mode 100644 net/ipv4/netfilter/ipt_NETMAP.c
delete mode 100644 net/ipv4/netfilter/ipt_REDIRECT.c
delete mode 100644 net/ipv6/netfilter/ip6t_NETMAP.c
delete mode 100644 net/ipv6/netfilter/ip6t_REDIRECT.c
create mode 100644 net/netfilter/xt_NETMAP.c
create mode 100644 net/netfilter/xt_REDIRECT.c
--
1.7.10.4
^ permalink raw reply
* [PATCH 02/16] netfilter: nf_nat: fix oops when unloading protocol modules
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Patrick McHardy <kaber@trash.net>
When unloading a protocol module nf_ct_iterate_cleanup() is used to
remove all conntracks using the protocol from the bysource hash and
clean their NAT sections. Since the conntrack isn't actually killed,
the NAT callback is invoked twice, once for each direction, which
causes an oops when trying to delete it from the bysource hash for
the second time.
The same oops can also happen when removing both an L3 and L4 protocol
since the cleanup function doesn't check whether the conntrack has
already been cleaned up.
Pid: 4052, comm: modprobe Not tainted 3.6.0-rc3-test-nat-unload-fix+ #32 Red Hat KVM
RIP: 0010:[<ffffffffa002c303>] [<ffffffffa002c303>] nf_nat_proto_clean+0x73/0xd0 [nf_nat]
RSP: 0018:ffff88007808fe18 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff8800728550c0 RCX: ffff8800756288b0
RDX: dead000000200200 RSI: ffff88007808fe88 RDI: ffffffffa002f208
RBP: ffff88007808fe28 R08: ffff88007808e000 R09: 0000000000000000
R10: dead000000200200 R11: dead000000100100 R12: ffffffff81c6dc00
R13: ffff8800787582b8 R14: ffff880078758278 R15: ffff88007808fe88
FS: 00007f515985d700(0000) GS:ffff88007cd00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 00007f515986a000 CR3: 000000007867a000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process modprobe (pid: 4052, threadinfo ffff88007808e000, task ffff8800756288b0)
Stack:
ffff88007808fe68 ffffffffa002c290 ffff88007808fe78 ffffffff815614e3
ffffffff00000000 00000aeb00000246 ffff88007808fe68 ffffffff81c6dc00
ffff88007808fe88 ffffffffa00358a0 0000000000000000 000000000040f5b0
Call Trace:
[<ffffffffa002c290>] ? nf_nat_net_exit+0x50/0x50 [nf_nat]
[<ffffffff815614e3>] nf_ct_iterate_cleanup+0xc3/0x170
[<ffffffffa002c55a>] nf_nat_l3proto_unregister+0x8a/0x100 [nf_nat]
[<ffffffff812a0303>] ? compat_prepare_timeout+0x13/0xb0
[<ffffffffa0035848>] nf_nat_l3proto_ipv4_exit+0x10/0x23 [nf_nat_ipv4]
...
To fix this,
- check whether the conntrack has already been cleaned up in
nf_nat_proto_clean
- change nf_ct_iterate_cleanup() to only invoke the callback function
once for each conntrack (IP_CT_DIR_ORIGINAL).
The second change doesn't affect other callers since when conntracks are
actually killed, both directions are removed from the hash immediately
and the callback is already only invoked once. If it is not killed, the
second callback invocation will always return the same decision not to
kill it.
Reported-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
net/netfilter/nf_conntrack_core.c | 2 ++
net/netfilter/nf_nat_core.c | 2 ++
2 files changed, 4 insertions(+)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index dcb2791..0f241be 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1224,6 +1224,8 @@ get_next_corpse(struct net *net, int (*iter)(struct nf_conn *i, void *data),
spin_lock_bh(&nf_conntrack_lock);
for (; *bucket < net->ct.htable_size; (*bucket)++) {
hlist_nulls_for_each_entry(h, n, &net->ct.hash[*bucket], hnnode) {
+ if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL)
+ continue;
ct = nf_ct_tuplehash_to_ctrack(h);
if (iter(ct, data))
goto found;
diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c
index 1816ad3..65cf694 100644
--- a/net/netfilter/nf_nat_core.c
+++ b/net/netfilter/nf_nat_core.c
@@ -481,6 +481,8 @@ static int nf_nat_proto_clean(struct nf_conn *i, void *data)
if (!nat)
return 0;
+ if (!(i->status & IPS_SRC_NAT_DONE))
+ return 0;
if ((clean->l3proto && nf_ct_l3num(i) != clean->l3proto) ||
(clean->l4proto && nf_ct_protonum(i) != clean->l4proto))
return 0;
--
1.7.10.4
^ permalink raw reply related
* [PATCH 03/16] netfilter: nf_nat: remove obsolete rcu_read_unlock call
From: pablo @ 2012-09-24 19:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: davem, netdev
In-Reply-To: <1348514369-3272-1-git-send-email-pablo@netfilter.org>
From: Ulrich Weber <ulrich.weber@sophos.com>
hlist walk in find_appropriate_src() is not protected anymore by rcu_read_lock(),
so rcu_read_unlock() is unnecessary if in_range() matches.
This bug was added in (c7232c9 netfilter: add protocol independent NAT core).
Signed-off-by: Ulrich Weber <ulrich.weber@sophos.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
net/netfilter/nf_nat_core.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c
index 65cf694..5f2f910 100644
--- a/net/netfilter/nf_nat_core.c
+++ b/net/netfilter/nf_nat_core.c
@@ -201,10 +201,8 @@ find_appropriate_src(struct net *net, u16 zone,
&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
result->dst = tuple->dst;
- if (in_range(l3proto, l4proto, result, range)) {
- rcu_read_unlock();
+ if (in_range(l3proto, l4proto, result, range))
return 1;
- }
}
}
return 0;
--
1.7.10.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox