Merge branch 'tracing/urgent' into tracing/core

Merge reason: Pick up fixes and move base from -rc1 to -rc5.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 402 insertions, 0 deletions

diff --git a/drivers/scsi/bfa/bfa_core.c b/drivers/scsi/bfa/bfa_core.c
new file mode 100644
index 000000000000..44e2d1155c51
--- /dev/null
+++ b/drivers/scsi/bfa/bfa_core.c
@@ -0,0 +1,402 @@
+/*
+ * Copyright (c) 2005-2009 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ *
+ * Linux driver for Brocade Fibre Channel Host Bus Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <bfa.h>
+#include <defs/bfa_defs_pci.h>
+#include <cs/bfa_debug.h>
+#include <bfa_iocfc.h>
+
+#define DEF_CFG_NUM_FABRICS         1
+#define DEF_CFG_NUM_LPORTS          256
+#define DEF_CFG_NUM_CQS             4
+#define DEF_CFG_NUM_IOIM_REQS       (BFA_IOIM_MAX)
+#define DEF_CFG_NUM_TSKIM_REQS      128
+#define DEF_CFG_NUM_FCXP_REQS       64
+#define DEF_CFG_NUM_UF_BUFS         64
+#define DEF_CFG_NUM_RPORTS          1024
+#define DEF_CFG_NUM_ITNIMS          (DEF_CFG_NUM_RPORTS)
+#define DEF_CFG_NUM_TINS            256
+
+#define DEF_CFG_NUM_SGPGS           2048
+#define DEF_CFG_NUM_REQQ_ELEMS      256
+#define DEF_CFG_NUM_RSPQ_ELEMS      64
+#define DEF_CFG_NUM_SBOOT_TGTS      16
+#define DEF_CFG_NUM_SBOOT_LUNS      16
+
+/**
+ * Use this function query the memory requirement of the BFA library.
+ * This function needs to be called before bfa_attach() to get the
+ * memory required of the BFA layer for a given driver configuration.
+ *
+ * This call will fail, if the cap is out of range compared to pre-defined
+ * values within the BFA library
+ *
+ * @param[in] cfg -     pointer to bfa_ioc_cfg_t. Driver layer should indicate
+ *                      its configuration in this structure.
+ *                      The default values for struct bfa_iocfc_cfg_s can be
+ *                      fetched using bfa_cfg_get_default() API.
+ *
+ *                      If cap's boundary check fails, the library will use
+ *                      the default bfa_cap_t values (and log a warning msg).
+ *
+ * @param[out] meminfo - pointer to bfa_meminfo_t. This content
+ *                      indicates the memory type (see bfa_mem_type_t) and
+ *                      amount of memory required.
+ *
+ *                      Driver should allocate the memory, populate the
+ *                      starting address for each block and provide the same
+ *                      structure as input parameter to bfa_attach() call.
+ *
+ * @return void
+ *
+ * Special Considerations: @note
+ */
+void
+bfa_cfg_get_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo)
+{
+        int             i;
+        u32        km_len = 0, dm_len = 0;
+
+        bfa_assert((cfg != NULL) && (meminfo != NULL));
+
+        bfa_os_memset((void *)meminfo, 0, sizeof(struct bfa_meminfo_s));
+        meminfo->meminfo[BFA_MEM_TYPE_KVA - 1].mem_type =
+                BFA_MEM_TYPE_KVA;
+        meminfo->meminfo[BFA_MEM_TYPE_DMA - 1].mem_type =
+                BFA_MEM_TYPE_DMA;
+
+        bfa_iocfc_meminfo(cfg, &km_len, &dm_len);
+
+        for (i = 0; hal_mods[i]; i++)
+                hal_mods[i]->meminfo(cfg, &km_len, &dm_len);
+
+
+        meminfo->meminfo[BFA_MEM_TYPE_KVA - 1].mem_len = km_len;
+        meminfo->meminfo[BFA_MEM_TYPE_DMA - 1].mem_len = dm_len;
+}
+
+/**
+ * Use this function to do attach the driver instance with the BFA
+ * library. This function will not trigger any HW initialization
+ * process (which will be done in bfa_init() call)
+ *
+ * This call will fail, if the cap is out of range compared to
+ * pre-defined values within the BFA library
+ *
+ * @param[out]  bfa     Pointer to bfa_t.
+ * @param[in]   bfad    Opaque handle back to the driver's IOC structure
+ * @param[in]   cfg     Pointer to bfa_ioc_cfg_t. Should be same structure
+ *                      that was used in bfa_cfg_get_meminfo().
+ * @param[in]   meminfo Pointer to bfa_meminfo_t. The driver should
+ *                      use the bfa_cfg_get_meminfo() call to
+ *                      find the memory blocks required, allocate the
+ *                      required memory and provide the starting addresses.
+ * @param[in]   pcidev  pointer to struct bfa_pcidev_s
+ *
+ * @return
+ * void
+ *
+ * Special Considerations:
+ *
+ * @note
+ *
+ */
+void
+bfa_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
+               struct bfa_meminfo_s *meminfo, struct bfa_pcidev_s *pcidev)
+{
+        int             i;
+        struct bfa_mem_elem_s *melem;
+
+        bfa->fcs = BFA_FALSE;
+
+        bfa_assert((cfg != NULL) && (meminfo != NULL));
+
+        /**
+         * initialize all memory pointers for iterative allocation
+         */
+        for (i = 0; i < BFA_MEM_TYPE_MAX; i++) {
+                melem = meminfo->meminfo + i;
+                melem->kva_curp = melem->kva;
+                melem->dma_curp = melem->dma;
+        }
+
+        bfa_iocfc_attach(bfa, bfad, cfg, meminfo, pcidev);
+
+        for (i = 0; hal_mods[i]; i++)
+                hal_mods[i]->attach(bfa, bfad, cfg, meminfo, pcidev);
+
+}
+
+/**
+ * Use this function to delete a BFA IOC. IOC should be stopped (by
+ * calling bfa_stop()) before this function call.
+ *
+ * @param[in] bfa - pointer to bfa_t.
+ *
+ * @return
+ * void
+ *
+ * Special Considerations:
+ *
+ * @note
+ */
+void
+bfa_detach(struct bfa_s *bfa)
+{
+        int     i;
+
+        for (i = 0; hal_mods[i]; i++)
+                hal_mods[i]->detach(bfa);
+
+        bfa_iocfc_detach(bfa);
+}
+
+
+void
+bfa_init_trc(struct bfa_s *bfa, struct bfa_trc_mod_s *trcmod)
+{
+        bfa->trcmod = trcmod;
+}
+
+
+void
+bfa_init_log(struct bfa_s *bfa, struct bfa_log_mod_s *logmod)
+{
+        bfa->logm = logmod;
+}
+
+
+void
+bfa_init_aen(struct bfa_s *bfa, struct bfa_aen_s *aen)
+{
+        bfa->aen = aen;
+}
+
+void
+bfa_init_plog(struct bfa_s *bfa, struct bfa_plog_s *plog)
+{
+        bfa->plog = plog;
+}
+
+/**
+ * Initialize IOC.
+ *
+ * This function will return immediately, when the IOC initialization is
+ * completed, the bfa_cb_init() will be called.
+ *
+ * @param[in]   bfa     instance
+ *
+ * @return void
+ *
+ * Special Considerations:
+ *
+ * @note
+ * When this function returns, the driver should register the interrupt service
+ * routine(s) and enable the device interrupts. If this is not done,
+ * bfa_cb_init() will never get called
+ */
+void
+bfa_init(struct bfa_s *bfa)
+{
+        bfa_iocfc_init(bfa);
+}
+
+/**
+ * Use this function initiate the IOC configuration setup. This function
+ * will return immediately.
+ *
+ * @param[in]   bfa     instance
+ *
+ * @return None
+ */
+void
+bfa_start(struct bfa_s *bfa)
+{
+        bfa_iocfc_start(bfa);
+}
+
+/**
+ * Use this function quiese the IOC. This function will return immediately,
+ * when the IOC is actually stopped, the bfa_cb_stop() will be called.
+ *
+ * @param[in]   bfa - pointer to bfa_t.
+ *
+ * @return None
+ *
+ * Special Considerations:
+ * bfa_cb_stop() could be called before or after bfa_stop() returns.
+ *
+ * @note
+ * In case of any failure, we could handle it automatically by doing a
+ * reset and then succeed the bfa_stop() call.
+ */
+void
+bfa_stop(struct bfa_s *bfa)
+{
+        bfa_iocfc_stop(bfa);
+}
+
+void
+bfa_comp_deq(struct bfa_s *bfa, struct list_head *comp_q)
+{
+        INIT_LIST_HEAD(comp_q);
+        list_splice_tail_init(&bfa->comp_q, comp_q);
+}
+
+void
+bfa_comp_process(struct bfa_s *bfa, struct list_head *comp_q)
+{
+        struct list_head        *qe;
+        struct list_head        *qen;
+        struct bfa_cb_qe_s   *hcb_qe;
+
+        list_for_each_safe(qe, qen, comp_q) {
+                hcb_qe = (struct bfa_cb_qe_s *) qe;
+                hcb_qe->cbfn(hcb_qe->cbarg, BFA_TRUE);
+        }
+}
+
+void
+bfa_comp_free(struct bfa_s *bfa, struct list_head *comp_q)
+{
+        struct list_head        *qe;
+        struct bfa_cb_qe_s   *hcb_qe;
+
+        while (!list_empty(comp_q)) {
+                bfa_q_deq(comp_q, &qe);
+                hcb_qe = (struct bfa_cb_qe_s *) qe;
+                hcb_qe->cbfn(hcb_qe->cbarg, BFA_FALSE);
+        }
+}
+
+void
+bfa_attach_fcs(struct bfa_s *bfa)
+{
+        bfa->fcs = BFA_TRUE;
+}
+
+/**
+ * Periodic timer heart beat from driver
+ */
+void
+bfa_timer_tick(struct bfa_s *bfa)
+{
+        bfa_timer_beat(&bfa->timer_mod);
+}
+
+#ifndef BFA_BIOS_BUILD
+/**
+ * Return the list of PCI vendor/device id lists supported by this
+ * BFA instance.
+ */
+void
+bfa_get_pciids(struct bfa_pciid_s **pciids, int *npciids)
+{
+        static struct bfa_pciid_s __pciids[] = {
+                {BFA_PCI_VENDOR_ID_BROCADE, BFA_PCI_DEVICE_ID_FC_8G2P},
+                {BFA_PCI_VENDOR_ID_BROCADE, BFA_PCI_DEVICE_ID_FC_8G1P},
+                {BFA_PCI_VENDOR_ID_BROCADE, BFA_PCI_DEVICE_ID_CT},
+        };
+
+        *npciids = sizeof(__pciids) / sizeof(__pciids[0]);
+        *pciids = __pciids;
+}
+
+/**
+ * Use this function query the default struct bfa_iocfc_cfg_s value (compiled
+ * into BFA layer). The OS driver can then turn back and overwrite entries that
+ * have been configured by the user.
+ *
+ * @param[in] cfg - pointer to bfa_ioc_cfg_t
+ *
+ * @return
+ *      void
+ *
+ * Special Considerations:
+ *      note
+ */
+void
+bfa_cfg_get_default(struct bfa_iocfc_cfg_s *cfg)
+{
+        cfg->fwcfg.num_fabrics = DEF_CFG_NUM_FABRICS;
+        cfg->fwcfg.num_lports = DEF_CFG_NUM_LPORTS;
+        cfg->fwcfg.num_rports = DEF_CFG_NUM_RPORTS;
+        cfg->fwcfg.num_ioim_reqs = DEF_CFG_NUM_IOIM_REQS;
+        cfg->fwcfg.num_tskim_reqs = DEF_CFG_NUM_TSKIM_REQS;
+        cfg->fwcfg.num_fcxp_reqs = DEF_CFG_NUM_FCXP_REQS;
+        cfg->fwcfg.num_uf_bufs = DEF_CFG_NUM_UF_BUFS;
+        cfg->fwcfg.num_cqs = DEF_CFG_NUM_CQS;
+
+        cfg->drvcfg.num_reqq_elems = DEF_CFG_NUM_REQQ_ELEMS;
+        cfg->drvcfg.num_rspq_elems = DEF_CFG_NUM_RSPQ_ELEMS;
+        cfg->drvcfg.num_sgpgs = DEF_CFG_NUM_SGPGS;
+        cfg->drvcfg.num_sboot_tgts = DEF_CFG_NUM_SBOOT_TGTS;
+        cfg->drvcfg.num_sboot_luns = DEF_CFG_NUM_SBOOT_LUNS;
+        cfg->drvcfg.path_tov = BFA_FCPIM_PATHTOV_DEF;
+        cfg->drvcfg.ioc_recover = BFA_FALSE;
+        cfg->drvcfg.delay_comp = BFA_FALSE;
+
+}
+
+void
+bfa_cfg_get_min(struct bfa_iocfc_cfg_s *cfg)
+{
+        bfa_cfg_get_default(cfg);
+        cfg->fwcfg.num_ioim_reqs   = BFA_IOIM_MIN;
+        cfg->fwcfg.num_tskim_reqs  = BFA_TSKIM_MIN;
+        cfg->fwcfg.num_fcxp_reqs   = BFA_FCXP_MIN;
+        cfg->fwcfg.num_uf_bufs     = BFA_UF_MIN;
+        cfg->fwcfg.num_rports      = BFA_RPORT_MIN;
+
+        cfg->drvcfg.num_sgpgs      = BFA_SGPG_MIN;
+        cfg->drvcfg.num_reqq_elems = BFA_REQQ_NELEMS_MIN;
+        cfg->drvcfg.num_rspq_elems = BFA_RSPQ_NELEMS_MIN;
+        cfg->drvcfg.min_cfg        = BFA_TRUE;
+}
+
+void
+bfa_get_attr(struct bfa_s *bfa, struct bfa_ioc_attr_s *ioc_attr)
+{
+        bfa_ioc_get_attr(&bfa->ioc, ioc_attr);
+}
+
+/**
+ * Retrieve firmware trace information on IOC failure.
+ */
+bfa_status_t
+bfa_debug_fwsave(struct bfa_s *bfa, void *trcdata, int *trclen)
+{
+        return bfa_ioc_debug_fwsave(&bfa->ioc, trcdata, trclen);
+}
+
+/**
+ *              Fetch firmware trace data.
+ *
+ * @param[in]           bfa                     BFA instance
+ * @param[out]          trcdata         Firmware trace buffer
+ * @param[in,out]       trclen          Firmware trace buffer len
+ *
+ * @retval BFA_STATUS_OK                        Firmware trace is fetched.
+ * @retval BFA_STATUS_INPROGRESS        Firmware trace fetch is in progress.
+ */
+bfa_status_t
+bfa_debug_fwtrc(struct bfa_s *bfa, void *trcdata, int *trclen)
+{
+        return bfa_ioc_debug_fwtrc(&bfa->ioc, trcdata, trclen);
+}
+#endif