forge/linux-pinenote
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
linux-pinenote/drivers/infiniband/hw/mthca/mthca_profile.c

286 lines
9.2 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
[PATCH] IB: Add copyright notices Make some lawyers happy and add copyright notices for people who forgot to include them when they actually touched the code. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-08-10 23:03:10 -07:00
* Copyright (c) 2005 Mellanox Technologies. All rights reserved.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*
* 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/module.h>
#include <linux/moduleparam.h>
[PATCH] fix missing includes I recently picked up my older work to remove unnecessary #includes of sched.h, starting from a patch by Dave Jones to not include sched.h from module.h. This reduces the number of indirect includes of sched.h by ~300. Another ~400 pointless direct includes can be removed after this disentangling (patch to follow later). However, quite a few indirect includes need to be fixed up for this. In order to feed the patches through -mm with as little disturbance as possible, I've split out the fixes I accumulated up to now (complete for i386 and x86_64, more archs to follow later) and post them before the real patch. This way this large part of the patch is kept simple with only adding #includes, and all hunks are independent of each other. So if any hunk rejects or gets in the way of other patches, just drop it. My scripts will pick it up again in the next round. Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 15:03:48 -08:00
#include <linux/string.h>
#include <linux/slab.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#include "mthca_profile.h"
enum {
MTHCA_RES_QP,
MTHCA_RES_EEC,
MTHCA_RES_SRQ,
MTHCA_RES_CQ,
MTHCA_RES_EQP,
MTHCA_RES_EEEC,
MTHCA_RES_EQ,
MTHCA_RES_RDB,
MTHCA_RES_MCG,
MTHCA_RES_MPT,
MTHCA_RES_MTT,
MTHCA_RES_UAR,
MTHCA_RES_UDAV,
MTHCA_RES_UARC,
MTHCA_RES_NUM
};
enum {
MTHCA_NUM_EQS = 32,
MTHCA_NUM_PDS = 1 << 15
};
IB/mthca: Avoid integer overflow when dealing with profile size mthca_make_profile() returns the size in bytes of the HCA context layout it creates, or a negative value if an error occurs. However, the return value is declared as u64 and the memfree initialization path casts this value to int to test if it is negative. This makes it think incorrectly than an error has occurred if the context size happens to be bigger than 2GB, since this turns into a negative int. Fix this by having mthca_make_profile() return an s64 and testing for an error by checking whether this 64-bit value itself is negative. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16 21:01:13 -07:00
s64 mthca_make_profile(struct mthca_dev *dev,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
struct mthca_profile *request,
struct mthca_dev_lim *dev_lim,
struct mthca_init_hca_param *init_hca)
{
struct mthca_resource {
u64 size;
u64 start;
int type;
int num;
int log_num;
};
u64 mem_base, mem_avail;
IB/mthca: Avoid integer overflow when dealing with profile size mthca_make_profile() returns the size in bytes of the HCA context layout it creates, or a negative value if an error occurs. However, the return value is declared as u64 and the memfree initialization path casts this value to int to test if it is negative. This makes it think incorrectly than an error has occurred if the context size happens to be bigger than 2GB, since this turns into a negative int. Fix this by having mthca_make_profile() return an s64 and testing for an error by checking whether this 64-bit value itself is negative. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16 21:01:13 -07:00
s64 total_size = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
struct mthca_resource *profile;
struct mthca_resource tmp;
int i, j;
[IB] kzalloc() conversions Replace kmalloc()+memset(,0,) with kzalloc(), for a net savings of 35 source lines and about 500 bytes of text. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 07:23:14 -08:00
profile = kzalloc(MTHCA_RES_NUM * sizeof *profile, GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (!profile)
return -ENOMEM;
profile[MTHCA_RES_QP].size = dev_lim->qpc_entry_sz;
profile[MTHCA_RES_EEC].size = dev_lim->eec_entry_sz;
profile[MTHCA_RES_SRQ].size = dev_lim->srq_entry_sz;
profile[MTHCA_RES_CQ].size = dev_lim->cqc_entry_sz;
profile[MTHCA_RES_EQP].size = dev_lim->eqpc_entry_sz;
profile[MTHCA_RES_EEEC].size = dev_lim->eeec_entry_sz;
profile[MTHCA_RES_EQ].size = dev_lim->eqc_entry_sz;
profile[MTHCA_RES_RDB].size = MTHCA_RDB_ENTRY_SIZE;
profile[MTHCA_RES_MCG].size = MTHCA_MGM_ENTRY_SIZE;
profile[MTHCA_RES_MPT].size = dev_lim->mpt_entry_sz;
IB/mthca: Add module parameter for number of MTTs per segment The current MTT allocator uses kmalloc() to allocate a buffer for its buddy allocator, and thus is limited in the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of MTT entries that each segment represents, allowing more memory to be registered with the same number of segments. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-05-27 14:36:16 -07:00
profile[MTHCA_RES_MTT].size = dev->limits.mtt_seg_size;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
profile[MTHCA_RES_UAR].size = dev_lim->uar_scratch_entry_sz;
profile[MTHCA_RES_UDAV].size = MTHCA_AV_SIZE;
profile[MTHCA_RES_UARC].size = request->uarc_size;
profile[MTHCA_RES_QP].num = request->num_qp;
[PATCH] IB/mthca: Add SRQ implementation Add mthca support for shared receive queues (SRQs), including userspace SRQs. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-08-19 10:59:31 -07:00
profile[MTHCA_RES_SRQ].num = request->num_srq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
profile[MTHCA_RES_EQP].num = request->num_qp;
profile[MTHCA_RES_RDB].num = request->num_qp * request->rdb_per_qp;
profile[MTHCA_RES_CQ].num = request->num_cq;
profile[MTHCA_RES_EQ].num = MTHCA_NUM_EQS;
profile[MTHCA_RES_MCG].num = request->num_mcg;
profile[MTHCA_RES_MPT].num = request->num_mpt;
profile[MTHCA_RES_MTT].num = request->num_mtt;
profile[MTHCA_RES_UAR].num = request->num_uar;
profile[MTHCA_RES_UARC].num = request->num_uar;
profile[MTHCA_RES_UDAV].num = request->num_udav;
for (i = 0; i < MTHCA_RES_NUM; ++i) {
profile[i].type = i;
profile[i].log_num = max(ffs(profile[i].num) - 1, 0);
profile[i].size *= profile[i].num;
[PATCH] IB/mthca: encapsulate mem-free check into mthca_is_memfree() Clean up mem-free mode support by introducing mthca_is_memfree() function, which encapsulates the logic of deciding if a device is mem-free. Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:32 -07:00
if (mthca_is_memfree(dev))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
profile[i].size = max(profile[i].size, (u64) PAGE_SIZE);
}
[PATCH] IB/mthca: encapsulate mem-free check into mthca_is_memfree() Clean up mem-free mode support by introducing mthca_is_memfree() function, which encapsulates the logic of deciding if a device is mem-free. Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:32 -07:00
if (mthca_is_memfree(dev)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
mem_base = 0;
mem_avail = dev_lim->hca.arbel.max_icm_sz;
} else {
mem_base = dev->ddr_start;
mem_avail = dev->fw.tavor.fw_start - dev->ddr_start;
}
/*
* Sort the resources in decreasing order of size. Since they
* all have sizes that are powers of 2, we'll be able to keep
* resources aligned to their size and pack them without gaps
* using the sorted order.
*/
for (i = MTHCA_RES_NUM; i > 0; --i)
for (j = 1; j < i; ++j) {
if (profile[j].size > profile[j - 1].size) {
tmp = profile[j];
profile[j] = profile[j - 1];
profile[j - 1] = tmp;
}
}
for (i = 0; i < MTHCA_RES_NUM; ++i) {
if (profile[i].size) {
profile[i].start = mem_base + total_size;
total_size += profile[i].size;
}
if (total_size > mem_avail) {
mthca_err(dev, "Profile requires 0x%llx bytes; "
IB/mthca: Optimize large messages on Sinai HCAs Sinai (one-port PCI Express) HCAs get improved throughput for messages bigger than 80 KB in DDR mode if memory keys are formatted in a specific way. The enhancement only works if the memory key table is smaller than 2^24 entries. For larger tables, the enhancement is off and a warning is printed (to avoid silent performance loss). Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Michael Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-03-02 12:40:46 -08:00
"won't fit in 0x%llx bytes of context memory.\n",
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
(unsigned long long) total_size,
(unsigned long long) mem_avail);
kfree(profile);
return -ENOMEM;
}
if (profile[i].size)
mthca_dbg(dev, "profile[%2d]--%2d/%2d @ 0x%16llx "
"(size 0x%8llx)\n",
i, profile[i].type, profile[i].log_num,
(unsigned long long) profile[i].start,
(unsigned long long) profile[i].size);
}
[PATCH] IB/mthca: encapsulate mem-free check into mthca_is_memfree() Clean up mem-free mode support by introducing mthca_is_memfree() function, which encapsulates the logic of deciding if a device is mem-free. Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:32 -07:00
if (mthca_is_memfree(dev))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
mthca_dbg(dev, "HCA context memory: reserving %d KB\n",
(int) (total_size >> 10));
else
mthca_dbg(dev, "HCA memory: allocated %d KB/%d KB (%d KB free)\n",
(int) (total_size >> 10), (int) (mem_avail >> 10),
(int) ((mem_avail - total_size) >> 10));
for (i = 0; i < MTHCA_RES_NUM; ++i) {
switch (profile[i].type) {
case MTHCA_RES_QP:
dev->limits.num_qps = profile[i].num;
init_hca->qpc_base = profile[i].start;
init_hca->log_num_qps = profile[i].log_num;
break;
case MTHCA_RES_EEC:
dev->limits.num_eecs = profile[i].num;
init_hca->eec_base = profile[i].start;
init_hca->log_num_eecs = profile[i].log_num;
break;
case MTHCA_RES_SRQ:
dev->limits.num_srqs = profile[i].num;
init_hca->srqc_base = profile[i].start;
init_hca->log_num_srqs = profile[i].log_num;
break;
case MTHCA_RES_CQ:
dev->limits.num_cqs = profile[i].num;
init_hca->cqc_base = profile[i].start;
init_hca->log_num_cqs = profile[i].log_num;
break;
case MTHCA_RES_EQP:
init_hca->eqpc_base = profile[i].start;
break;
case MTHCA_RES_EEEC:
init_hca->eeec_base = profile[i].start;
break;
case MTHCA_RES_EQ:
dev->limits.num_eqs = profile[i].num;
init_hca->eqc_base = profile[i].start;
init_hca->log_num_eqs = profile[i].log_num;
break;
case MTHCA_RES_RDB:
for (dev->qp_table.rdb_shift = 0;
[PATCH] IB/mthca: fix calculation of RDB shift Fix calculation of rdb_shift by using original number of QPs, not their slot in profile[] (which will be rearranged when we sort it). Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:15 -07:00
request->num_qp << dev->qp_table.rdb_shift < profile[i].num;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
++dev->qp_table.rdb_shift)
; /* nothing */
dev->qp_table.rdb_base = (u32) profile[i].start;
init_hca->rdb_base = profile[i].start;
break;
case MTHCA_RES_MCG:
dev->limits.num_mgms = profile[i].num >> 1;
dev->limits.num_amgms = profile[i].num >> 1;
init_hca->mc_base = profile[i].start;
init_hca->log_mc_entry_sz = ffs(MTHCA_MGM_ENTRY_SIZE) - 1;
init_hca->log_mc_table_sz = profile[i].log_num;
init_hca->mc_hash_sz = 1 << (profile[i].log_num - 1);
break;
case MTHCA_RES_MPT:
[PATCH] IB/mthca: add fast memory region implementation Implement fast memory regions (FMRs), where the driver writes directly into the HCA's translation tables rather than requiring a firmware command. For Tavor, MTTs for FMR are separate from regular MTTs, and are reserved at driver initialization. This is done to limit the amount of virtual memory needed to map the MTTs. For Arbel, there's no such limitation, and all MTTs and MPTs may be used for FMR or for regular MR. Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:30 -07:00
dev->limits.num_mpts = profile[i].num;
dev->mr_table.mpt_base = profile[i].start;
init_hca->mpt_base = profile[i].start;
init_hca->log_mpt_sz = profile[i].log_num;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
break;
case MTHCA_RES_MTT:
dev->limits.num_mtt_segs = profile[i].num;
dev->mr_table.mtt_base = profile[i].start;
init_hca->mtt_base = profile[i].start;
IB/mthca: Add module parameter for number of MTTs per segment The current MTT allocator uses kmalloc() to allocate a buffer for its buddy allocator, and thus is limited in the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of MTT entries that each segment represents, allowing more memory to be registered with the same number of segments. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-05-27 14:36:16 -07:00
init_hca->mtt_seg_sz = ffs(dev->limits.mtt_seg_size) - 7;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
break;
case MTHCA_RES_UAR:
dev->limits.num_uars = profile[i].num;
init_hca->uar_scratch_base = profile[i].start;
break;
case MTHCA_RES_UDAV:
dev->av_table.ddr_av_base = profile[i].start;
dev->av_table.num_ddr_avs = profile[i].num;
break;
case MTHCA_RES_UARC:
dev->uar_table.uarc_size = request->uarc_size;
dev->uar_table.uarc_base = profile[i].start;
init_hca->uarc_base = profile[i].start;
init_hca->log_uarc_sz = ffs(request->uarc_size) - 13;
init_hca->log_uar_sz = ffs(request->num_uar) - 1;
break;
default:
break;
}
}
/*
* PDs don't take any HCA memory, but we assign them as part
* of the HCA profile anyway.
*/
dev->limits.num_pds = MTHCA_NUM_PDS;
IB/mthca: Optimize large messages on Sinai HCAs Sinai (one-port PCI Express) HCAs get improved throughput for messages bigger than 80 KB in DDR mode if memory keys are formatted in a specific way. The enhancement only works if the memory key table is smaller than 2^24 entries. For larger tables, the enhancement is off and a warning is printed (to avoid silent performance loss). Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Michael Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-03-02 12:40:46 -08:00
if (dev->mthca_flags & MTHCA_FLAG_SINAI_OPT &&
init_hca->log_mpt_sz > 23) {
mthca_warn(dev, "MPT table too large (requested size 2^%d >= 2^24)\n",
init_hca->log_mpt_sz);
mthca_warn(dev, "Disabling memory key throughput optimization.\n");
dev->mthca_flags &= ~MTHCA_FLAG_SINAI_OPT;
}
[PATCH] IB/mthca: add fast memory region implementation Implement fast memory regions (FMRs), where the driver writes directly into the HCA's translation tables rather than requiring a firmware command. For Tavor, MTTs for FMR are separate from regular MTTs, and are reserved at driver initialization. This is done to limit the amount of virtual memory needed to map the MTTs. For Arbel, there's no such limitation, and all MTTs and MPTs may be used for FMR or for regular MR. Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:30 -07:00
/*
* For Tavor, FMRs use ioremapped PCI memory. For 32 bit
* systems it may use too much vmalloc space to map all MTT
* memory, so we reserve some MTTs for FMR access, taking them
* out of the MR pool. They don't use additional memory, but
* we assign them as part of the HCA profile anyway.
*/
IB/mthca: Merge MR and FMR space on 64-bit systems For Tavor, we currently reserve separate MPT and MTT space for FMRs to avoid abusing the vmalloc space on 32 bit kernels. No such problem exists on 64 bit kernels so let's not do it there. This way we have a shared pool for MR and FMR resources, used on demand. This will also make it possible to write MTTs for regular regions directly from driver. Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-02-10 23:13:12 +02:00
if (mthca_is_memfree(dev) || BITS_PER_LONG == 64)
[PATCH] IB/mthca: add fast memory region implementation Implement fast memory regions (FMRs), where the driver writes directly into the HCA's translation tables rather than requiring a firmware command. For Tavor, MTTs for FMR are separate from regular MTTs, and are reserved at driver initialization. This is done to limit the amount of virtual memory needed to map the MTTs. For Arbel, there's no such limitation, and all MTTs and MPTs may be used for FMR or for regular MR. Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-16 15:26:30 -07:00
dev->limits.fmr_reserved_mtts = 0;
else
dev->limits.fmr_reserved_mtts = request->fmr_reserved_mtts;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
kfree(profile);
return total_size;
}
Reference in a new issue Copy permalink