fedora core 6 1.2949 + vserver 2.2.0

[linux-2.6.git] / drivers / infiniband / hw / mthca / mthca_mr.c

diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c
index e995e2a..58012ac 100644 (file)
--- a/drivers/infiniband/hw/mthca/mthca_mr.c
+++ b/drivers/infiniband/hw/mthca/mthca_mr.c
@@ -34,7 +34,6 @@
  */
 
 #include <linux/slab.h>
-#include <linux/init.h>
 #include <linux/errno.h>
 
 #include "mthca_dev.h"
@@ -76,6 +75,8 @@ struct mthca_mpt_entry {
 #define MTHCA_MPT_STATUS_SW 0xF0
 #define MTHCA_MPT_STATUS_HW 0x00
 
+#define SINAI_FMR_KEY_INC 0x1000000
+
 /*
  * Buddy allocator for MTT segments (currently not very efficient
  * since it doesn't keep a free list and just searches linearly
@@ -133,7 +134,7 @@ static void mthca_buddy_free(struct mthca_buddy *buddy, u32 seg, int order)
        spin_unlock(&buddy->lock);
 }
 
-static int __devinit mthca_buddy_init(struct mthca_buddy *buddy, int max_order)
+static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order)
 {
        int i, s;
 
@@ -168,7 +169,7 @@ err_out:
        return -ENOMEM;
 }
 
-static void __devexit mthca_buddy_cleanup(struct mthca_buddy *buddy)
+static void mthca_buddy_cleanup(struct mthca_buddy *buddy)
 {
        int i;
 
@@ -330,6 +331,14 @@ static inline u32 key_to_hw_index(struct mthca_dev *dev, u32 key)
                return tavor_key_to_hw_index(key);
 }
 
+static inline u32 adjust_key(struct mthca_dev *dev, u32 key)
+{
+       if (dev->mthca_flags & MTHCA_FLAG_SINAI_OPT)
+               return ((key << 20) & 0x800000) | (key & 0x7fffff);
+       else
+               return key;
+}
+
 int mthca_mr_alloc(struct mthca_dev *dev, u32 pd, int buffer_size_shift,
                   u64 iova, u64 total_size, u32 access, struct mthca_mr *mr)
 {
@@ -340,13 +349,12 @@ int mthca_mr_alloc(struct mthca_dev *dev, u32 pd, int buffer_size_shift,
        int err;
        u8 status;
 
-       might_sleep();
-
        WARN_ON(buffer_size_shift >= 32);
 
        key = mthca_alloc(&dev->mr_table.mpt_alloc);
        if (key == -1)
                return -ENOMEM;
+       key = adjust_key(dev, key);
        mr->ibmr.rkey = mr->ibmr.lkey = hw_index_to_key(dev, key);
 
        if (mthca_is_memfree(dev)) {
@@ -467,8 +475,6 @@ void mthca_free_mr(struct mthca_dev *dev, struct mthca_mr *mr)
        int err;
        u8 status;
 
-       might_sleep();
-
        err = mthca_HW2SW_MPT(dev, NULL,
                              key_to_hw_index(dev, mr->ibmr.lkey) &
                              (dev->limits.num_mpts - 1),
@@ -495,9 +501,7 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
        int err = -ENOMEM;
        int i;
 
-       might_sleep();
-
-       if (mr->attr.page_size < 12 || mr->attr.page_size >= 32)
+       if (mr->attr.page_shift < 12 || mr->attr.page_shift >= 32)
                return -EINVAL;
 
        /* For Arbel, all MTTs must fit in the same page. */
@@ -510,6 +514,7 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
        key = mthca_alloc(&dev->mr_table.mpt_alloc);
        if (key == -1)
                return -ENOMEM;
+       key = adjust_key(dev, key);
 
        idx = key & (dev->limits.num_mpts - 1);
        mr->ibmr.rkey = mr->ibmr.lkey = hw_index_to_key(dev, key);
@@ -523,7 +528,7 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
                BUG_ON(!mr->mem.arbel.mpt);
        } else
                mr->mem.tavor.mpt = dev->mr_table.tavor_fmr.mpt_base +
-                       sizeof *(mr->mem.tavor.mpt) * idx;
+                       sizeof *(mr->mem.tavor.mpt) * idx;
 
        mr->mtt = __mthca_alloc_mtt(dev, list_len, dev->mr_table.fmr_mtt_buddy);
        if (IS_ERR(mr->mtt))
@@ -549,7 +554,7 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
                                       MTHCA_MPT_FLAG_REGION      |
                                       access);
 
-       mpt_entry->page_size = cpu_to_be32(mr->attr.page_size - 12);
+       mpt_entry->page_size = cpu_to_be32(mr->attr.page_shift - 12);
        mpt_entry->key       = cpu_to_be32(key);
        mpt_entry->pd        = cpu_to_be32(pd);
        memset(&mpt_entry->start, 0,
@@ -617,7 +622,7 @@ static inline int mthca_check_fmr(struct mthca_fmr *fmr, u64 *page_list,
        if (list_len > fmr->attr.max_pages)
                return -EINVAL;
 
-       page_mask = (1 << fmr->attr.page_size) - 1;
+       page_mask = (1 << fmr->attr.page_shift) - 1;
 
        /* We are getting page lists, so va must be page aligned. */
        if (iova & page_mask)
@@ -665,7 +670,7 @@ int mthca_tavor_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
        }
 
        mpt_entry.lkey   = cpu_to_be32(key);
-       mpt_entry.length = cpu_to_be64(list_len * (1ull << fmr->attr.page_size));
+       mpt_entry.length = cpu_to_be64(list_len * (1ull << fmr->attr.page_shift));
        mpt_entry.start  = cpu_to_be64(iova);
 
        __raw_writel((__force u32) mpt_entry.lkey, &fmr->mem.tavor.mpt->key);
@@ -693,7 +698,10 @@ int mthca_arbel_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
        ++fmr->maps;
 
        key = arbel_key_to_hw_index(fmr->ibmr.lkey);
-       key += dev->limits.num_mpts;
+       if (dev->mthca_flags & MTHCA_FLAG_SINAI_OPT)
+               key += SINAI_FMR_KEY_INC;
+       else
+               key += dev->limits.num_mpts;
        fmr->ibmr.lkey = fmr->ibmr.rkey = arbel_hw_index_to_key(key);
 
        *(u8 *) fmr->mem.arbel.mpt = MTHCA_MPT_STATUS_SW;
@@ -706,7 +714,7 @@ int mthca_arbel_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
 
        fmr->mem.arbel.mpt->key    = cpu_to_be32(key);
        fmr->mem.arbel.mpt->lkey   = cpu_to_be32(key);
-       fmr->mem.arbel.mpt->length = cpu_to_be64(list_len * (1ull << fmr->attr.page_size));
+       fmr->mem.arbel.mpt->length = cpu_to_be64(list_len * (1ull << fmr->attr.page_shift));
        fmr->mem.arbel.mpt->start  = cpu_to_be64(iova);
 
        wmb();
@@ -743,6 +751,7 @@ void mthca_arbel_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr)
 
        key = arbel_key_to_hw_index(fmr->ibmr.lkey);
        key &= dev->limits.num_mpts - 1;
+       key = adjust_key(dev, key);
        fmr->ibmr.lkey = fmr->ibmr.rkey = arbel_hw_index_to_key(key);
 
        fmr->maps = 0;
@@ -750,8 +759,9 @@ void mthca_arbel_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr)
        *(u8 *) fmr->mem.arbel.mpt = MTHCA_MPT_STATUS_SW;
 }
 
-int __devinit mthca_init_mr_table(struct mthca_dev *dev)
+int mthca_init_mr_table(struct mthca_dev *dev)
 {
+       unsigned long addr;
        int err, i;
 
        err = mthca_alloc_init(&dev->mr_table.mpt_alloc,
@@ -766,6 +776,9 @@ int __devinit mthca_init_mr_table(struct mthca_dev *dev)
        else
                dev->mthca_flags |= MTHCA_FLAG_FMR;
 
+       if (dev->mthca_flags & MTHCA_FLAG_SINAI_OPT)
+               mthca_dbg(dev, "Memory key throughput optimization activated.\n");
+
        err = mthca_buddy_init(&dev->mr_table.mtt_buddy,
                               fls(dev->limits.num_mtt_segs - 1));
 
@@ -784,9 +797,12 @@ int __devinit mthca_init_mr_table(struct mthca_dev *dev)
                        goto err_fmr_mpt;
                }
 
+               addr = pci_resource_start(dev->pdev, 4) +
+                       ((pci_resource_len(dev->pdev, 4) - 1) &
+                        dev->mr_table.mpt_base);
+
                dev->mr_table.tavor_fmr.mpt_base =
-                       ioremap(dev->mr_table.mpt_base,
-                               (1 << i) * sizeof (struct mthca_mpt_entry));
+                       ioremap(addr, (1 << i) * sizeof(struct mthca_mpt_entry));
 
                if (!dev->mr_table.tavor_fmr.mpt_base) {
                        mthca_warn(dev, "MPT ioremap for FMR failed.\n");
@@ -794,9 +810,12 @@ int __devinit mthca_init_mr_table(struct mthca_dev *dev)
                        goto err_fmr_mpt;
                }
 
+               addr = pci_resource_start(dev->pdev, 4) +
+                       ((pci_resource_len(dev->pdev, 4) - 1) &
+                        dev->mr_table.mtt_base);
+
                dev->mr_table.tavor_fmr.mtt_base =
-                       ioremap(dev->mr_table.mtt_base,
-                               (1 << i) * MTHCA_MTT_SEG_SIZE);
+                       ioremap(addr, (1 << i) * MTHCA_MTT_SEG_SIZE);
                if (!dev->mr_table.tavor_fmr.mtt_base) {
                        mthca_warn(dev, "MTT ioremap for FMR failed.\n");
                        err = -ENOMEM;
@@ -813,7 +832,7 @@ int __devinit mthca_init_mr_table(struct mthca_dev *dev)
                        goto err_reserve_fmr;
 
                dev->mr_table.fmr_mtt_buddy =
-                       &dev->mr_table.tavor_fmr.mtt_buddy;
+                       &dev->mr_table.tavor_fmr.mtt_buddy;
        } else
                dev->mr_table.fmr_mtt_buddy = &dev->mr_table.mtt_buddy;
 
@@ -854,7 +873,7 @@ err_mtt_buddy:
        return err;
 }
 
-void __devexit mthca_cleanup_mr_table(struct mthca_dev *dev)
+void mthca_cleanup_mr_table(struct mthca_dev *dev)
 {
        /* XXX check if any MRs are still allocated? */
        if (dev->limits.fmr_reserved_mtts)