2007-05-09 09:00:38 +08:00
|
|
|
/*
|
|
|
|
* Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
|
2008-07-26 01:32:52 +08:00
|
|
|
* Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
|
2007-05-09 09:00:38 +08:00
|
|
|
*
|
|
|
|
* 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/errno.h>
|
|
|
|
#include <linux/slab.h>
|
2008-09-06 05:04:07 +08:00
|
|
|
#include <linux/mm.h>
|
2011-05-28 04:14:23 +08:00
|
|
|
#include <linux/export.h>
|
2007-05-09 09:00:38 +08:00
|
|
|
#include <linux/bitmap.h>
|
2007-05-16 03:36:30 +08:00
|
|
|
#include <linux/dma-mapping.h>
|
2008-02-11 10:22:57 +08:00
|
|
|
#include <linux/vmalloc.h>
|
2007-05-09 09:00:38 +08:00
|
|
|
|
|
|
|
#include "mlx4.h"
|
|
|
|
|
|
|
|
u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap)
|
|
|
|
{
|
|
|
|
u32 obj;
|
|
|
|
|
|
|
|
spin_lock(&bitmap->lock);
|
|
|
|
|
|
|
|
obj = find_next_zero_bit(bitmap->table, bitmap->max, bitmap->last);
|
|
|
|
if (obj >= bitmap->max) {
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
|
|
|
|
& bitmap->mask;
|
2007-05-09 09:00:38 +08:00
|
|
|
obj = find_first_zero_bit(bitmap->table, bitmap->max);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (obj < bitmap->max) {
|
|
|
|
set_bit(obj, bitmap->table);
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->last = (obj + 1);
|
|
|
|
if (bitmap->last == bitmap->max)
|
|
|
|
bitmap->last = 0;
|
2007-05-09 09:00:38 +08:00
|
|
|
obj |= bitmap->top;
|
|
|
|
} else
|
|
|
|
obj = -1;
|
|
|
|
|
2011-03-23 06:38:45 +08:00
|
|
|
if (obj != -1)
|
|
|
|
--bitmap->avail;
|
|
|
|
|
2007-05-09 09:00:38 +08:00
|
|
|
spin_unlock(&bitmap->lock);
|
|
|
|
|
|
|
|
return obj;
|
|
|
|
}
|
|
|
|
|
mlx4_core: Roll back round robin bitmap allocation commit for CQs, SRQs, and MPTs
Commit f4ec9e9 "mlx4_core: Change bitmap allocator to work in round-robin fashion"
introduced round-robin allocation (via bitmap) for all resources which allocate
via a bitmap.
Round robin allocation is desirable for mcgs, counters, pd's, UARs, and xrcds.
These are simply numbers, with no involvement of ICM memory mapping.
Round robin is required for QPs, since we had a problem with immediate
reuse of a 24-bit QP number (commit f4ec9e9).
However, for other resources which use the bitmap allocator and involve
mapping ICM memory -- MPTs, CQs, SRQs -- round-robin is not desirable.
What happens in these cases is the following:
ICM memory is allocated and mapped in chunks of 256K.
Since the resource allocation index goes up monotonically, the allocator
will eventually require mapping a new chunk. Now, chunks are also unmapped
when their reference count goes back to zero. Thus, if a single app is
running and starts/exits frequently we will have the following situation:
When the app starts, a new chunk must be allocated and mapped.
When the app exits, the chunk reference count goes back to zero, and the
chunk is unmapped and freed. Therefore, the app must pay the cost of allocation
and mapping of ICM memory each time it runs (although the price is paid only when
allocating the initial entry in the new chunk).
For apps which allocate MPTs/SRQs/CQs and which operate as described above,
this presented a performance problem.
We therefore roll back the round-robin allocator modification for MPTs, CQs, SRQs.
Reported-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-08 22:50:17 +08:00
|
|
|
void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj, int use_rr)
|
2007-05-09 09:00:38 +08:00
|
|
|
{
|
mlx4_core: Roll back round robin bitmap allocation commit for CQs, SRQs, and MPTs
Commit f4ec9e9 "mlx4_core: Change bitmap allocator to work in round-robin fashion"
introduced round-robin allocation (via bitmap) for all resources which allocate
via a bitmap.
Round robin allocation is desirable for mcgs, counters, pd's, UARs, and xrcds.
These are simply numbers, with no involvement of ICM memory mapping.
Round robin is required for QPs, since we had a problem with immediate
reuse of a 24-bit QP number (commit f4ec9e9).
However, for other resources which use the bitmap allocator and involve
mapping ICM memory -- MPTs, CQs, SRQs -- round-robin is not desirable.
What happens in these cases is the following:
ICM memory is allocated and mapped in chunks of 256K.
Since the resource allocation index goes up monotonically, the allocator
will eventually require mapping a new chunk. Now, chunks are also unmapped
when their reference count goes back to zero. Thus, if a single app is
running and starts/exits frequently we will have the following situation:
When the app starts, a new chunk must be allocated and mapped.
When the app exits, the chunk reference count goes back to zero, and the
chunk is unmapped and freed. Therefore, the app must pay the cost of allocation
and mapping of ICM memory each time it runs (although the price is paid only when
allocating the initial entry in the new chunk).
For apps which allocate MPTs/SRQs/CQs and which operate as described above,
this presented a performance problem.
We therefore roll back the round-robin allocator modification for MPTs, CQs, SRQs.
Reported-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-08 22:50:17 +08:00
|
|
|
mlx4_bitmap_free_range(bitmap, obj, 1, use_rr);
|
2008-10-11 03:01:37 +08:00
|
|
|
}
|
|
|
|
|
net/mlx4: Change QP allocation scheme
When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields
in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset.
The current Ethernet driver code reserves a Tx QP range with 256b alignment.
This is wrong because if there are more than 64 Tx QPs in use,
QPNs >= base + 65 will have bits 6/7 set.
This problem is not specific for the Ethernet driver, any entity that
tries to reserve more than 64 BF-enabled QPs should fail. Also, using
ranges is not necessary here and is wasteful.
The new mechanism introduced here will support reservation for
"Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs
(when hypervisors support WC in VMs). The flow we use is:
1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation,
and request "BF enabled QPs" if BF is supported for the function
2. In the ALLOC_RES FW command, change param1 to:
a. param1[23:0] - number of QPs
b. param1[31-24] - flags controlling QPs reservation
Bit 31 refers to Eth blueflame supported QPs. Those QPs must have
bits 6 and 7 unset in order to be used in Ethernet.
Bits 24-30 of the flags are currently reserved.
When a function tries to allocate a QP, it states the required attributes
for this QP. Those attributes are considered "best-effort". If an attribute,
such as Ethernet BF enabled QP, is a must-have attribute, the function has
to check that attribute is supported before trying to do the allocation.
In a lower layer of the code, mlx4_qp_reserve_range masks out the bits
which are unsupported. If SRIOV is used, the PF validates those attributes
and masks out unsupported attributes as well. In order to notify VFs which
attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's
mailbox is filled by the PF, which notifies which QP allocation attributes
it supports.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 16:57:54 +08:00
|
|
|
static unsigned long find_aligned_range(unsigned long *bitmap,
|
|
|
|
u32 start, u32 nbits,
|
|
|
|
int len, int align, u32 skip_mask)
|
|
|
|
{
|
|
|
|
unsigned long end, i;
|
|
|
|
|
|
|
|
again:
|
|
|
|
start = ALIGN(start, align);
|
|
|
|
|
|
|
|
while ((start < nbits) && (test_bit(start, bitmap) ||
|
|
|
|
(start & skip_mask)))
|
|
|
|
start += align;
|
|
|
|
|
|
|
|
if (start >= nbits)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
end = start+len;
|
|
|
|
if (end > nbits)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
for (i = start + 1; i < end; i++) {
|
|
|
|
if (test_bit(i, bitmap) || ((u32)i & skip_mask)) {
|
|
|
|
start = i + 1;
|
|
|
|
goto again;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return start;
|
|
|
|
}
|
|
|
|
|
|
|
|
u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt,
|
|
|
|
int align, u32 skip_mask)
|
2008-10-11 03:01:37 +08:00
|
|
|
{
|
2010-08-28 03:08:13 +08:00
|
|
|
u32 obj;
|
2008-10-11 03:01:37 +08:00
|
|
|
|
net/mlx4: Change QP allocation scheme
When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields
in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset.
The current Ethernet driver code reserves a Tx QP range with 256b alignment.
This is wrong because if there are more than 64 Tx QPs in use,
QPNs >= base + 65 will have bits 6/7 set.
This problem is not specific for the Ethernet driver, any entity that
tries to reserve more than 64 BF-enabled QPs should fail. Also, using
ranges is not necessary here and is wasteful.
The new mechanism introduced here will support reservation for
"Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs
(when hypervisors support WC in VMs). The flow we use is:
1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation,
and request "BF enabled QPs" if BF is supported for the function
2. In the ALLOC_RES FW command, change param1 to:
a. param1[23:0] - number of QPs
b. param1[31-24] - flags controlling QPs reservation
Bit 31 refers to Eth blueflame supported QPs. Those QPs must have
bits 6 and 7 unset in order to be used in Ethernet.
Bits 24-30 of the flags are currently reserved.
When a function tries to allocate a QP, it states the required attributes
for this QP. Those attributes are considered "best-effort". If an attribute,
such as Ethernet BF enabled QP, is a must-have attribute, the function has
to check that attribute is supported before trying to do the allocation.
In a lower layer of the code, mlx4_qp_reserve_range masks out the bits
which are unsupported. If SRIOV is used, the PF validates those attributes
and masks out unsupported attributes as well. In order to notify VFs which
attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's
mailbox is filled by the PF, which notifies which QP allocation attributes
it supports.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 16:57:54 +08:00
|
|
|
if (likely(cnt == 1 && align == 1 && !skip_mask))
|
2008-10-11 03:01:37 +08:00
|
|
|
return mlx4_bitmap_alloc(bitmap);
|
|
|
|
|
|
|
|
spin_lock(&bitmap->lock);
|
|
|
|
|
net/mlx4: Change QP allocation scheme
When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields
in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset.
The current Ethernet driver code reserves a Tx QP range with 256b alignment.
This is wrong because if there are more than 64 Tx QPs in use,
QPNs >= base + 65 will have bits 6/7 set.
This problem is not specific for the Ethernet driver, any entity that
tries to reserve more than 64 BF-enabled QPs should fail. Also, using
ranges is not necessary here and is wasteful.
The new mechanism introduced here will support reservation for
"Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs
(when hypervisors support WC in VMs). The flow we use is:
1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation,
and request "BF enabled QPs" if BF is supported for the function
2. In the ALLOC_RES FW command, change param1 to:
a. param1[23:0] - number of QPs
b. param1[31-24] - flags controlling QPs reservation
Bit 31 refers to Eth blueflame supported QPs. Those QPs must have
bits 6 and 7 unset in order to be used in Ethernet.
Bits 24-30 of the flags are currently reserved.
When a function tries to allocate a QP, it states the required attributes
for this QP. Those attributes are considered "best-effort". If an attribute,
such as Ethernet BF enabled QP, is a must-have attribute, the function has
to check that attribute is supported before trying to do the allocation.
In a lower layer of the code, mlx4_qp_reserve_range masks out the bits
which are unsupported. If SRIOV is used, the PF validates those attributes
and masks out unsupported attributes as well. In order to notify VFs which
attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's
mailbox is filled by the PF, which notifies which QP allocation attributes
it supports.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 16:57:54 +08:00
|
|
|
obj = find_aligned_range(bitmap->table, bitmap->last,
|
|
|
|
bitmap->max, cnt, align, skip_mask);
|
2008-10-11 03:01:37 +08:00
|
|
|
if (obj >= bitmap->max) {
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
|
|
|
|
& bitmap->mask;
|
net/mlx4: Change QP allocation scheme
When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields
in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset.
The current Ethernet driver code reserves a Tx QP range with 256b alignment.
This is wrong because if there are more than 64 Tx QPs in use,
QPNs >= base + 65 will have bits 6/7 set.
This problem is not specific for the Ethernet driver, any entity that
tries to reserve more than 64 BF-enabled QPs should fail. Also, using
ranges is not necessary here and is wasteful.
The new mechanism introduced here will support reservation for
"Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs
(when hypervisors support WC in VMs). The flow we use is:
1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation,
and request "BF enabled QPs" if BF is supported for the function
2. In the ALLOC_RES FW command, change param1 to:
a. param1[23:0] - number of QPs
b. param1[31-24] - flags controlling QPs reservation
Bit 31 refers to Eth blueflame supported QPs. Those QPs must have
bits 6 and 7 unset in order to be used in Ethernet.
Bits 24-30 of the flags are currently reserved.
When a function tries to allocate a QP, it states the required attributes
for this QP. Those attributes are considered "best-effort". If an attribute,
such as Ethernet BF enabled QP, is a must-have attribute, the function has
to check that attribute is supported before trying to do the allocation.
In a lower layer of the code, mlx4_qp_reserve_range masks out the bits
which are unsupported. If SRIOV is used, the PF validates those attributes
and masks out unsupported attributes as well. In order to notify VFs which
attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's
mailbox is filled by the PF, which notifies which QP allocation attributes
it supports.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 16:57:54 +08:00
|
|
|
obj = find_aligned_range(bitmap->table, 0, bitmap->max,
|
|
|
|
cnt, align, skip_mask);
|
2008-10-11 03:01:37 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
if (obj < bitmap->max) {
|
2010-08-28 03:08:13 +08:00
|
|
|
bitmap_set(bitmap->table, obj, cnt);
|
2008-10-11 03:01:37 +08:00
|
|
|
if (obj == bitmap->last) {
|
|
|
|
bitmap->last = (obj + cnt);
|
|
|
|
if (bitmap->last >= bitmap->max)
|
|
|
|
bitmap->last = 0;
|
|
|
|
}
|
|
|
|
obj |= bitmap->top;
|
|
|
|
} else
|
|
|
|
obj = -1;
|
|
|
|
|
2011-03-23 06:38:45 +08:00
|
|
|
if (obj != -1)
|
|
|
|
bitmap->avail -= cnt;
|
|
|
|
|
2008-10-11 03:01:37 +08:00
|
|
|
spin_unlock(&bitmap->lock);
|
|
|
|
|
|
|
|
return obj;
|
|
|
|
}
|
|
|
|
|
2011-03-23 06:38:45 +08:00
|
|
|
u32 mlx4_bitmap_avail(struct mlx4_bitmap *bitmap)
|
|
|
|
{
|
|
|
|
return bitmap->avail;
|
|
|
|
}
|
|
|
|
|
2014-12-11 16:57:56 +08:00
|
|
|
static u32 mlx4_bitmap_masked_value(struct mlx4_bitmap *bitmap, u32 obj)
|
|
|
|
{
|
|
|
|
return obj & (bitmap->max + bitmap->reserved_top - 1);
|
|
|
|
}
|
|
|
|
|
mlx4_core: Roll back round robin bitmap allocation commit for CQs, SRQs, and MPTs
Commit f4ec9e9 "mlx4_core: Change bitmap allocator to work in round-robin fashion"
introduced round-robin allocation (via bitmap) for all resources which allocate
via a bitmap.
Round robin allocation is desirable for mcgs, counters, pd's, UARs, and xrcds.
These are simply numbers, with no involvement of ICM memory mapping.
Round robin is required for QPs, since we had a problem with immediate
reuse of a 24-bit QP number (commit f4ec9e9).
However, for other resources which use the bitmap allocator and involve
mapping ICM memory -- MPTs, CQs, SRQs -- round-robin is not desirable.
What happens in these cases is the following:
ICM memory is allocated and mapped in chunks of 256K.
Since the resource allocation index goes up monotonically, the allocator
will eventually require mapping a new chunk. Now, chunks are also unmapped
when their reference count goes back to zero. Thus, if a single app is
running and starts/exits frequently we will have the following situation:
When the app starts, a new chunk must be allocated and mapped.
When the app exits, the chunk reference count goes back to zero, and the
chunk is unmapped and freed. Therefore, the app must pay the cost of allocation
and mapping of ICM memory each time it runs (although the price is paid only when
allocating the initial entry in the new chunk).
For apps which allocate MPTs/SRQs/CQs and which operate as described above,
this presented a performance problem.
We therefore roll back the round-robin allocator modification for MPTs, CQs, SRQs.
Reported-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-08 22:50:17 +08:00
|
|
|
void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt,
|
|
|
|
int use_rr)
|
2008-10-11 03:01:37 +08:00
|
|
|
{
|
2008-10-23 01:25:29 +08:00
|
|
|
obj &= bitmap->max + bitmap->reserved_top - 1;
|
2007-05-09 09:00:38 +08:00
|
|
|
|
|
|
|
spin_lock(&bitmap->lock);
|
mlx4_core: Roll back round robin bitmap allocation commit for CQs, SRQs, and MPTs
Commit f4ec9e9 "mlx4_core: Change bitmap allocator to work in round-robin fashion"
introduced round-robin allocation (via bitmap) for all resources which allocate
via a bitmap.
Round robin allocation is desirable for mcgs, counters, pd's, UARs, and xrcds.
These are simply numbers, with no involvement of ICM memory mapping.
Round robin is required for QPs, since we had a problem with immediate
reuse of a 24-bit QP number (commit f4ec9e9).
However, for other resources which use the bitmap allocator and involve
mapping ICM memory -- MPTs, CQs, SRQs -- round-robin is not desirable.
What happens in these cases is the following:
ICM memory is allocated and mapped in chunks of 256K.
Since the resource allocation index goes up monotonically, the allocator
will eventually require mapping a new chunk. Now, chunks are also unmapped
when their reference count goes back to zero. Thus, if a single app is
running and starts/exits frequently we will have the following situation:
When the app starts, a new chunk must be allocated and mapped.
When the app exits, the chunk reference count goes back to zero, and the
chunk is unmapped and freed. Therefore, the app must pay the cost of allocation
and mapping of ICM memory each time it runs (although the price is paid only when
allocating the initial entry in the new chunk).
For apps which allocate MPTs/SRQs/CQs and which operate as described above,
this presented a performance problem.
We therefore roll back the round-robin allocator modification for MPTs, CQs, SRQs.
Reported-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-08 22:50:17 +08:00
|
|
|
if (!use_rr) {
|
|
|
|
bitmap->last = min(bitmap->last, obj);
|
|
|
|
bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
|
|
|
|
& bitmap->mask;
|
|
|
|
}
|
2010-08-28 03:08:13 +08:00
|
|
|
bitmap_clear(bitmap->table, obj, cnt);
|
2011-03-23 06:38:45 +08:00
|
|
|
bitmap->avail += cnt;
|
2007-05-09 09:00:38 +08:00
|
|
|
spin_unlock(&bitmap->lock);
|
|
|
|
}
|
|
|
|
|
2008-10-23 01:25:29 +08:00
|
|
|
int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
|
|
|
|
u32 reserved_bot, u32 reserved_top)
|
2007-05-09 09:00:38 +08:00
|
|
|
{
|
|
|
|
/* num must be a power of 2 */
|
|
|
|
if (num != roundup_pow_of_two(num))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
bitmap->last = 0;
|
|
|
|
bitmap->top = 0;
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->max = num - reserved_top;
|
2007-05-09 09:00:38 +08:00
|
|
|
bitmap->mask = mask;
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->reserved_top = reserved_top;
|
2011-03-23 06:38:45 +08:00
|
|
|
bitmap->avail = num - reserved_top - reserved_bot;
|
2014-12-11 16:57:56 +08:00
|
|
|
bitmap->effective_len = bitmap->avail;
|
2007-05-09 09:00:38 +08:00
|
|
|
spin_lock_init(&bitmap->lock);
|
2008-10-23 01:25:29 +08:00
|
|
|
bitmap->table = kzalloc(BITS_TO_LONGS(bitmap->max) *
|
|
|
|
sizeof (long), GFP_KERNEL);
|
2007-05-09 09:00:38 +08:00
|
|
|
if (!bitmap->table)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
2010-08-28 03:08:13 +08:00
|
|
|
bitmap_set(bitmap->table, 0, reserved_bot);
|
2007-05-09 09:00:38 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void mlx4_bitmap_cleanup(struct mlx4_bitmap *bitmap)
|
|
|
|
{
|
|
|
|
kfree(bitmap->table);
|
|
|
|
}
|
|
|
|
|
2014-12-11 16:57:56 +08:00
|
|
|
struct mlx4_zone_allocator {
|
|
|
|
struct list_head entries;
|
|
|
|
struct list_head prios;
|
|
|
|
u32 last_uid;
|
|
|
|
u32 mask;
|
|
|
|
/* protect the zone_allocator from concurrent accesses */
|
|
|
|
spinlock_t lock;
|
|
|
|
enum mlx4_zone_alloc_flags flags;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct mlx4_zone_entry {
|
|
|
|
struct list_head list;
|
|
|
|
struct list_head prio_list;
|
|
|
|
u32 uid;
|
|
|
|
struct mlx4_zone_allocator *allocator;
|
|
|
|
struct mlx4_bitmap *bitmap;
|
|
|
|
int use_rr;
|
|
|
|
int priority;
|
|
|
|
int offset;
|
|
|
|
enum mlx4_zone_flags flags;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct mlx4_zone_allocator *mlx4_zone_allocator_create(enum mlx4_zone_alloc_flags flags)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_allocator *zones = kmalloc(sizeof(*zones), GFP_KERNEL);
|
|
|
|
|
|
|
|
if (NULL == zones)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
INIT_LIST_HEAD(&zones->entries);
|
|
|
|
INIT_LIST_HEAD(&zones->prios);
|
|
|
|
spin_lock_init(&zones->lock);
|
|
|
|
zones->last_uid = 0;
|
|
|
|
zones->mask = 0;
|
|
|
|
zones->flags = flags;
|
|
|
|
|
|
|
|
return zones;
|
|
|
|
}
|
|
|
|
|
|
|
|
int mlx4_zone_add_one(struct mlx4_zone_allocator *zone_alloc,
|
|
|
|
struct mlx4_bitmap *bitmap,
|
|
|
|
u32 flags,
|
|
|
|
int priority,
|
|
|
|
int offset,
|
|
|
|
u32 *puid)
|
|
|
|
{
|
|
|
|
u32 mask = mlx4_bitmap_masked_value(bitmap, (u32)-1);
|
|
|
|
struct mlx4_zone_entry *it;
|
|
|
|
struct mlx4_zone_entry *zone = kmalloc(sizeof(*zone), GFP_KERNEL);
|
|
|
|
|
|
|
|
if (NULL == zone)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
zone->flags = flags;
|
|
|
|
zone->bitmap = bitmap;
|
|
|
|
zone->use_rr = (flags & MLX4_ZONE_USE_RR) ? MLX4_USE_RR : 0;
|
|
|
|
zone->priority = priority;
|
|
|
|
zone->offset = offset;
|
|
|
|
|
|
|
|
spin_lock(&zone_alloc->lock);
|
|
|
|
|
|
|
|
zone->uid = zone_alloc->last_uid++;
|
|
|
|
zone->allocator = zone_alloc;
|
|
|
|
|
|
|
|
if (zone_alloc->mask < mask)
|
|
|
|
zone_alloc->mask = mask;
|
|
|
|
|
|
|
|
list_for_each_entry(it, &zone_alloc->prios, prio_list)
|
|
|
|
if (it->priority >= priority)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (&it->prio_list == &zone_alloc->prios || it->priority > priority)
|
|
|
|
list_add_tail(&zone->prio_list, &it->prio_list);
|
|
|
|
list_add_tail(&zone->list, &it->list);
|
|
|
|
|
|
|
|
spin_unlock(&zone_alloc->lock);
|
|
|
|
|
|
|
|
*puid = zone->uid;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Should be called under a lock */
|
|
|
|
static int __mlx4_zone_remove_one_entry(struct mlx4_zone_entry *entry)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_allocator *zone_alloc = entry->allocator;
|
|
|
|
|
|
|
|
if (!list_empty(&entry->prio_list)) {
|
|
|
|
/* Check if we need to add an alternative node to the prio list */
|
|
|
|
if (!list_is_last(&entry->list, &zone_alloc->entries)) {
|
|
|
|
struct mlx4_zone_entry *next = list_first_entry(&entry->list,
|
|
|
|
typeof(*next),
|
|
|
|
list);
|
|
|
|
|
|
|
|
if (next->priority == entry->priority)
|
|
|
|
list_add_tail(&next->prio_list, &entry->prio_list);
|
|
|
|
}
|
|
|
|
|
|
|
|
list_del(&entry->prio_list);
|
|
|
|
}
|
|
|
|
|
|
|
|
list_del(&entry->list);
|
|
|
|
|
|
|
|
if (zone_alloc->flags & MLX4_ZONE_ALLOC_FLAGS_NO_OVERLAP) {
|
|
|
|
u32 mask = 0;
|
|
|
|
struct mlx4_zone_entry *it;
|
|
|
|
|
|
|
|
list_for_each_entry(it, &zone_alloc->prios, prio_list) {
|
|
|
|
u32 cur_mask = mlx4_bitmap_masked_value(it->bitmap, (u32)-1);
|
|
|
|
|
|
|
|
if (mask < cur_mask)
|
|
|
|
mask = cur_mask;
|
|
|
|
}
|
|
|
|
zone_alloc->mask = mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void mlx4_zone_allocator_destroy(struct mlx4_zone_allocator *zone_alloc)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone, *tmp;
|
|
|
|
|
|
|
|
spin_lock(&zone_alloc->lock);
|
|
|
|
|
|
|
|
list_for_each_entry_safe(zone, tmp, &zone_alloc->entries, list) {
|
|
|
|
list_del(&zone->list);
|
|
|
|
list_del(&zone->prio_list);
|
|
|
|
kfree(zone);
|
|
|
|
}
|
|
|
|
|
|
|
|
spin_unlock(&zone_alloc->lock);
|
|
|
|
kfree(zone_alloc);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Should be called under a lock */
|
|
|
|
static u32 __mlx4_alloc_from_zone(struct mlx4_zone_entry *zone, int count,
|
|
|
|
int align, u32 skip_mask, u32 *puid)
|
|
|
|
{
|
|
|
|
u32 uid;
|
|
|
|
u32 res;
|
|
|
|
struct mlx4_zone_allocator *zone_alloc = zone->allocator;
|
|
|
|
struct mlx4_zone_entry *curr_node;
|
|
|
|
|
|
|
|
res = mlx4_bitmap_alloc_range(zone->bitmap, count,
|
|
|
|
align, skip_mask);
|
|
|
|
|
|
|
|
if (res != (u32)-1) {
|
|
|
|
res += zone->offset;
|
|
|
|
uid = zone->uid;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
list_for_each_entry(curr_node, &zone_alloc->prios, prio_list) {
|
|
|
|
if (unlikely(curr_node->priority == zone->priority))
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zone->flags & MLX4_ZONE_ALLOW_ALLOC_FROM_LOWER_PRIO) {
|
|
|
|
struct mlx4_zone_entry *it = curr_node;
|
|
|
|
|
|
|
|
list_for_each_entry_continue_reverse(it, &zone_alloc->entries, list) {
|
|
|
|
res = mlx4_bitmap_alloc_range(it->bitmap, count,
|
|
|
|
align, skip_mask);
|
|
|
|
if (res != (u32)-1) {
|
|
|
|
res += it->offset;
|
|
|
|
uid = it->uid;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zone->flags & MLX4_ZONE_ALLOW_ALLOC_FROM_EQ_PRIO) {
|
|
|
|
struct mlx4_zone_entry *it = curr_node;
|
|
|
|
|
|
|
|
list_for_each_entry_from(it, &zone_alloc->entries, list) {
|
|
|
|
if (unlikely(it == zone))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (unlikely(it->priority != curr_node->priority))
|
|
|
|
break;
|
|
|
|
|
|
|
|
res = mlx4_bitmap_alloc_range(it->bitmap, count,
|
|
|
|
align, skip_mask);
|
|
|
|
if (res != (u32)-1) {
|
|
|
|
res += it->offset;
|
|
|
|
uid = it->uid;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zone->flags & MLX4_ZONE_FALLBACK_TO_HIGHER_PRIO) {
|
|
|
|
if (list_is_last(&curr_node->prio_list, &zone_alloc->prios))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
curr_node = list_first_entry(&curr_node->prio_list,
|
|
|
|
typeof(*curr_node),
|
|
|
|
prio_list);
|
|
|
|
|
|
|
|
list_for_each_entry_from(curr_node, &zone_alloc->entries, list) {
|
|
|
|
res = mlx4_bitmap_alloc_range(curr_node->bitmap, count,
|
|
|
|
align, skip_mask);
|
|
|
|
if (res != (u32)-1) {
|
|
|
|
res += curr_node->offset;
|
|
|
|
uid = curr_node->uid;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
|
|
|
if (NULL != puid && res != (u32)-1)
|
|
|
|
*puid = uid;
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Should be called under a lock */
|
|
|
|
static void __mlx4_free_from_zone(struct mlx4_zone_entry *zone, u32 obj,
|
|
|
|
u32 count)
|
|
|
|
{
|
|
|
|
mlx4_bitmap_free_range(zone->bitmap, obj - zone->offset, count, zone->use_rr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Should be called under a lock */
|
|
|
|
static struct mlx4_zone_entry *__mlx4_find_zone_by_uid(
|
|
|
|
struct mlx4_zone_allocator *zones, u32 uid)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
|
|
|
|
list_for_each_entry(zone, &zones->entries, list) {
|
|
|
|
if (zone->uid == uid)
|
|
|
|
return zone;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
struct mlx4_bitmap *mlx4_zone_get_bitmap(struct mlx4_zone_allocator *zones, u32 uid)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
struct mlx4_bitmap *bitmap;
|
|
|
|
|
|
|
|
spin_lock(&zones->lock);
|
|
|
|
|
|
|
|
zone = __mlx4_find_zone_by_uid(zones, uid);
|
|
|
|
|
|
|
|
bitmap = zone == NULL ? NULL : zone->bitmap;
|
|
|
|
|
|
|
|
spin_unlock(&zones->lock);
|
|
|
|
|
|
|
|
return bitmap;
|
|
|
|
}
|
|
|
|
|
|
|
|
int mlx4_zone_remove_one(struct mlx4_zone_allocator *zones, u32 uid)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
int res;
|
|
|
|
|
|
|
|
spin_lock(&zones->lock);
|
|
|
|
|
|
|
|
zone = __mlx4_find_zone_by_uid(zones, uid);
|
|
|
|
|
|
|
|
if (NULL == zone) {
|
|
|
|
res = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
res = __mlx4_zone_remove_one_entry(zone);
|
|
|
|
|
|
|
|
out:
|
|
|
|
spin_unlock(&zones->lock);
|
|
|
|
kfree(zone);
|
|
|
|
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Should be called under a lock */
|
|
|
|
static struct mlx4_zone_entry *__mlx4_find_zone_by_uid_unique(
|
|
|
|
struct mlx4_zone_allocator *zones, u32 obj)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone, *zone_candidate = NULL;
|
|
|
|
u32 dist = (u32)-1;
|
|
|
|
|
|
|
|
/* Search for the smallest zone that this obj could be
|
|
|
|
* allocated from. This is done in order to handle
|
|
|
|
* situations when small bitmaps are allocated from bigger
|
|
|
|
* bitmaps (and the allocated space is marked as reserved in
|
|
|
|
* the bigger bitmap.
|
|
|
|
*/
|
|
|
|
list_for_each_entry(zone, &zones->entries, list) {
|
|
|
|
if (obj >= zone->offset) {
|
|
|
|
u32 mobj = (obj - zone->offset) & zones->mask;
|
|
|
|
|
|
|
|
if (mobj < zone->bitmap->max) {
|
|
|
|
u32 curr_dist = zone->bitmap->effective_len;
|
|
|
|
|
|
|
|
if (curr_dist < dist) {
|
|
|
|
dist = curr_dist;
|
|
|
|
zone_candidate = zone;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return zone_candidate;
|
|
|
|
}
|
|
|
|
|
|
|
|
u32 mlx4_zone_alloc_entries(struct mlx4_zone_allocator *zones, u32 uid, int count,
|
|
|
|
int align, u32 skip_mask, u32 *puid)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
int res = -1;
|
|
|
|
|
|
|
|
spin_lock(&zones->lock);
|
|
|
|
|
|
|
|
zone = __mlx4_find_zone_by_uid(zones, uid);
|
|
|
|
|
|
|
|
if (NULL == zone)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
res = __mlx4_alloc_from_zone(zone, count, align, skip_mask, puid);
|
|
|
|
|
|
|
|
out:
|
|
|
|
spin_unlock(&zones->lock);
|
|
|
|
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
u32 mlx4_zone_free_entries(struct mlx4_zone_allocator *zones, u32 uid, u32 obj, u32 count)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
int res = 0;
|
|
|
|
|
|
|
|
spin_lock(&zones->lock);
|
|
|
|
|
|
|
|
zone = __mlx4_find_zone_by_uid(zones, uid);
|
|
|
|
|
|
|
|
if (NULL == zone) {
|
|
|
|
res = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
__mlx4_free_from_zone(zone, obj, count);
|
|
|
|
|
|
|
|
out:
|
|
|
|
spin_unlock(&zones->lock);
|
|
|
|
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
u32 mlx4_zone_free_entries_unique(struct mlx4_zone_allocator *zones, u32 obj, u32 count)
|
|
|
|
{
|
|
|
|
struct mlx4_zone_entry *zone;
|
|
|
|
int res;
|
|
|
|
|
|
|
|
if (!(zones->flags & MLX4_ZONE_ALLOC_FLAGS_NO_OVERLAP))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
spin_lock(&zones->lock);
|
|
|
|
|
|
|
|
zone = __mlx4_find_zone_by_uid_unique(zones, obj);
|
|
|
|
|
|
|
|
if (NULL == zone) {
|
|
|
|
res = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
__mlx4_free_from_zone(zone, obj, count);
|
|
|
|
res = 0;
|
|
|
|
|
|
|
|
out:
|
|
|
|
spin_unlock(&zones->lock);
|
|
|
|
|
|
|
|
return res;
|
|
|
|
}
|
2007-05-09 09:00:38 +08:00
|
|
|
/*
|
|
|
|
* Handling for queue buffers -- we allocate a bunch of memory and
|
|
|
|
* register it in a memory region at HCA virtual address 0. If the
|
|
|
|
* requested size is > max_direct, we split the allocation into
|
|
|
|
* multiple pages, so we don't require too much contiguous memory.
|
|
|
|
*/
|
|
|
|
|
|
|
|
int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct,
|
2014-05-11 20:15:12 +08:00
|
|
|
struct mlx4_buf *buf, gfp_t gfp)
|
2007-05-09 09:00:38 +08:00
|
|
|
{
|
|
|
|
dma_addr_t t;
|
|
|
|
|
|
|
|
if (size <= max_direct) {
|
|
|
|
buf->nbufs = 1;
|
|
|
|
buf->npages = 1;
|
|
|
|
buf->page_shift = get_order(size) + PAGE_SHIFT;
|
2015-01-25 22:59:35 +08:00
|
|
|
buf->direct.buf = dma_alloc_coherent(&dev->persist->pdev->dev,
|
2014-05-11 20:15:12 +08:00
|
|
|
size, &t, gfp);
|
2008-02-07 13:17:59 +08:00
|
|
|
if (!buf->direct.buf)
|
2007-05-09 09:00:38 +08:00
|
|
|
return -ENOMEM;
|
|
|
|
|
2008-02-07 13:17:59 +08:00
|
|
|
buf->direct.map = t;
|
2007-05-09 09:00:38 +08:00
|
|
|
|
|
|
|
while (t & ((1 << buf->page_shift) - 1)) {
|
|
|
|
--buf->page_shift;
|
|
|
|
buf->npages *= 2;
|
|
|
|
}
|
|
|
|
|
2008-02-07 13:17:59 +08:00
|
|
|
memset(buf->direct.buf, 0, size);
|
2007-05-09 09:00:38 +08:00
|
|
|
} else {
|
|
|
|
int i;
|
|
|
|
|
2011-01-11 09:42:06 +08:00
|
|
|
buf->direct.buf = NULL;
|
2007-05-09 09:00:38 +08:00
|
|
|
buf->nbufs = (size + PAGE_SIZE - 1) / PAGE_SIZE;
|
|
|
|
buf->npages = buf->nbufs;
|
|
|
|
buf->page_shift = PAGE_SHIFT;
|
2010-08-11 15:02:48 +08:00
|
|
|
buf->page_list = kcalloc(buf->nbufs, sizeof(*buf->page_list),
|
2014-05-11 20:15:12 +08:00
|
|
|
gfp);
|
2008-02-07 13:17:59 +08:00
|
|
|
if (!buf->page_list)
|
2007-05-09 09:00:38 +08:00
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
for (i = 0; i < buf->nbufs; ++i) {
|
2008-02-07 13:17:59 +08:00
|
|
|
buf->page_list[i].buf =
|
2015-01-25 22:59:35 +08:00
|
|
|
dma_alloc_coherent(&dev->persist->pdev->dev,
|
|
|
|
PAGE_SIZE,
|
2014-05-11 20:15:12 +08:00
|
|
|
&t, gfp);
|
2008-02-07 13:17:59 +08:00
|
|
|
if (!buf->page_list[i].buf)
|
2007-05-09 09:00:38 +08:00
|
|
|
goto err_free;
|
|
|
|
|
2008-02-07 13:17:59 +08:00
|
|
|
buf->page_list[i].map = t;
|
2007-05-09 09:00:38 +08:00
|
|
|
|
2008-02-07 13:17:59 +08:00
|
|
|
memset(buf->page_list[i].buf, 0, PAGE_SIZE);
|
2007-05-09 09:00:38 +08:00
|
|
|
}
|
2008-01-28 16:40:51 +08:00
|
|
|
|
|
|
|
if (BITS_PER_LONG == 64) {
|
|
|
|
struct page **pages;
|
2014-05-11 20:15:12 +08:00
|
|
|
pages = kmalloc(sizeof *pages * buf->nbufs, gfp);
|
2008-01-28 16:40:51 +08:00
|
|
|
if (!pages)
|
|
|
|
goto err_free;
|
|
|
|
for (i = 0; i < buf->nbufs; ++i)
|
2008-02-07 13:17:59 +08:00
|
|
|
pages[i] = virt_to_page(buf->page_list[i].buf);
|
|
|
|
buf->direct.buf = vmap(pages, buf->nbufs, VM_MAP, PAGE_KERNEL);
|
2008-01-28 16:40:51 +08:00
|
|
|
kfree(pages);
|
2008-02-07 13:17:59 +08:00
|
|
|
if (!buf->direct.buf)
|
2008-01-28 16:40:51 +08:00
|
|
|
goto err_free;
|
|
|
|
}
|
2007-05-09 09:00:38 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
err_free:
|
|
|
|
mlx4_buf_free(dev, size, buf);
|
|
|
|
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_buf_alloc);
|
|
|
|
|
|
|
|
void mlx4_buf_free(struct mlx4_dev *dev, int size, struct mlx4_buf *buf)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (buf->nbufs == 1)
|
2015-01-25 22:59:35 +08:00
|
|
|
dma_free_coherent(&dev->persist->pdev->dev, size,
|
|
|
|
buf->direct.buf,
|
2008-02-07 13:17:59 +08:00
|
|
|
buf->direct.map);
|
2007-05-09 09:00:38 +08:00
|
|
|
else {
|
2015-02-09 18:10:41 +08:00
|
|
|
if (BITS_PER_LONG == 64)
|
2008-02-07 13:17:59 +08:00
|
|
|
vunmap(buf->direct.buf);
|
2008-01-28 16:40:51 +08:00
|
|
|
|
2007-05-09 09:00:38 +08:00
|
|
|
for (i = 0; i < buf->nbufs; ++i)
|
2008-02-07 13:17:59 +08:00
|
|
|
if (buf->page_list[i].buf)
|
2015-01-25 22:59:35 +08:00
|
|
|
dma_free_coherent(&dev->persist->pdev->dev,
|
|
|
|
PAGE_SIZE,
|
2008-02-07 13:17:59 +08:00
|
|
|
buf->page_list[i].buf,
|
|
|
|
buf->page_list[i].map);
|
|
|
|
kfree(buf->page_list);
|
2007-05-09 09:00:38 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_buf_free);
|
2008-04-24 02:55:45 +08:00
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
static struct mlx4_db_pgdir *mlx4_alloc_db_pgdir(struct device *dma_device,
|
|
|
|
gfp_t gfp)
|
2008-04-24 02:55:45 +08:00
|
|
|
{
|
|
|
|
struct mlx4_db_pgdir *pgdir;
|
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
pgdir = kzalloc(sizeof *pgdir, gfp);
|
2008-04-24 02:55:45 +08:00
|
|
|
if (!pgdir)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
bitmap_fill(pgdir->order1, MLX4_DB_PER_PAGE / 2);
|
|
|
|
pgdir->bits[0] = pgdir->order0;
|
|
|
|
pgdir->bits[1] = pgdir->order1;
|
|
|
|
pgdir->db_page = dma_alloc_coherent(dma_device, PAGE_SIZE,
|
2014-05-11 20:15:12 +08:00
|
|
|
&pgdir->db_dma, gfp);
|
2008-04-24 02:55:45 +08:00
|
|
|
if (!pgdir->db_page) {
|
|
|
|
kfree(pgdir);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return pgdir;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int mlx4_alloc_db_from_pgdir(struct mlx4_db_pgdir *pgdir,
|
|
|
|
struct mlx4_db *db, int order)
|
|
|
|
{
|
|
|
|
int o;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (o = order; o <= 1; ++o) {
|
|
|
|
i = find_first_bit(pgdir->bits[o], MLX4_DB_PER_PAGE >> o);
|
|
|
|
if (i < MLX4_DB_PER_PAGE >> o)
|
|
|
|
goto found;
|
|
|
|
}
|
|
|
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
found:
|
|
|
|
clear_bit(i, pgdir->bits[o]);
|
|
|
|
|
|
|
|
i <<= o;
|
|
|
|
|
|
|
|
if (o > order)
|
|
|
|
set_bit(i ^ 1, pgdir->bits[order]);
|
|
|
|
|
|
|
|
db->u.pgdir = pgdir;
|
|
|
|
db->index = i;
|
|
|
|
db->db = pgdir->db_page + db->index;
|
|
|
|
db->dma = pgdir->db_dma + db->index * 4;
|
|
|
|
db->order = order;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order, gfp_t gfp)
|
2008-04-24 02:55:45 +08:00
|
|
|
{
|
|
|
|
struct mlx4_priv *priv = mlx4_priv(dev);
|
|
|
|
struct mlx4_db_pgdir *pgdir;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
mutex_lock(&priv->pgdir_mutex);
|
|
|
|
|
|
|
|
list_for_each_entry(pgdir, &priv->pgdir_list, list)
|
|
|
|
if (!mlx4_alloc_db_from_pgdir(pgdir, db, order))
|
|
|
|
goto out;
|
|
|
|
|
2015-01-25 22:59:35 +08:00
|
|
|
pgdir = mlx4_alloc_db_pgdir(&dev->persist->pdev->dev, gfp);
|
2008-04-24 02:55:45 +08:00
|
|
|
if (!pgdir) {
|
|
|
|
ret = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
list_add(&pgdir->list, &priv->pgdir_list);
|
|
|
|
|
|
|
|
/* This should never fail -- we just allocated an empty page: */
|
|
|
|
WARN_ON(mlx4_alloc_db_from_pgdir(pgdir, db, order));
|
|
|
|
|
|
|
|
out:
|
|
|
|
mutex_unlock(&priv->pgdir_mutex);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_db_alloc);
|
|
|
|
|
|
|
|
void mlx4_db_free(struct mlx4_dev *dev, struct mlx4_db *db)
|
|
|
|
{
|
|
|
|
struct mlx4_priv *priv = mlx4_priv(dev);
|
|
|
|
int o;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
mutex_lock(&priv->pgdir_mutex);
|
|
|
|
|
|
|
|
o = db->order;
|
|
|
|
i = db->index;
|
|
|
|
|
|
|
|
if (db->order == 0 && test_bit(i ^ 1, db->u.pgdir->order0)) {
|
|
|
|
clear_bit(i ^ 1, db->u.pgdir->order0);
|
|
|
|
++o;
|
|
|
|
}
|
|
|
|
i >>= o;
|
|
|
|
set_bit(i, db->u.pgdir->bits[o]);
|
|
|
|
|
|
|
|
if (bitmap_full(db->u.pgdir->order1, MLX4_DB_PER_PAGE / 2)) {
|
2015-01-25 22:59:35 +08:00
|
|
|
dma_free_coherent(&dev->persist->pdev->dev, PAGE_SIZE,
|
2008-04-24 02:55:45 +08:00
|
|
|
db->u.pgdir->db_page, db->u.pgdir->db_dma);
|
|
|
|
list_del(&db->u.pgdir->list);
|
|
|
|
kfree(db->u.pgdir);
|
|
|
|
}
|
|
|
|
|
|
|
|
mutex_unlock(&priv->pgdir_mutex);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_db_free);
|
2008-04-26 05:27:08 +08:00
|
|
|
|
|
|
|
int mlx4_alloc_hwq_res(struct mlx4_dev *dev, struct mlx4_hwq_resources *wqres,
|
|
|
|
int size, int max_direct)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
err = mlx4_db_alloc(dev, &wqres->db, 1, GFP_KERNEL);
|
2008-04-26 05:27:08 +08:00
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
*wqres->db.db = 0;
|
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
err = mlx4_buf_alloc(dev, size, max_direct, &wqres->buf, GFP_KERNEL);
|
2008-04-26 05:27:08 +08:00
|
|
|
if (err)
|
|
|
|
goto err_db;
|
|
|
|
|
|
|
|
err = mlx4_mtt_init(dev, wqres->buf.npages, wqres->buf.page_shift,
|
|
|
|
&wqres->mtt);
|
|
|
|
if (err)
|
|
|
|
goto err_buf;
|
|
|
|
|
2014-05-11 20:15:12 +08:00
|
|
|
err = mlx4_buf_write_mtt(dev, &wqres->mtt, &wqres->buf, GFP_KERNEL);
|
2008-04-26 05:27:08 +08:00
|
|
|
if (err)
|
|
|
|
goto err_mtt;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
err_mtt:
|
|
|
|
mlx4_mtt_cleanup(dev, &wqres->mtt);
|
|
|
|
err_buf:
|
|
|
|
mlx4_buf_free(dev, size, &wqres->buf);
|
|
|
|
err_db:
|
|
|
|
mlx4_db_free(dev, &wqres->db);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_alloc_hwq_res);
|
|
|
|
|
|
|
|
void mlx4_free_hwq_res(struct mlx4_dev *dev, struct mlx4_hwq_resources *wqres,
|
|
|
|
int size)
|
|
|
|
{
|
|
|
|
mlx4_mtt_cleanup(dev, &wqres->mtt);
|
|
|
|
mlx4_buf_free(dev, size, &wqres->buf);
|
|
|
|
mlx4_db_free(dev, &wqres->db);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(mlx4_free_hwq_res);
|