openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
linux/drivers/scsi/megaraid.c

5068 lines
114 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!
2005-04-17 06:20:36 +08:00
/*
*
* Linux MegaRAID device driver
*
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
* Copyright (c) 2002 LSI Logic Corporation.
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!
2005-04-17 06:20:36 +08:00
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* Copyright (c) 2002 Red Hat, Inc. All rights reserved.
* - fixes
* - speed-ups (list handling fixes, issued_list, optimizations.)
* - lots of cleanups.
*
* Copyright (c) 2003 Christoph Hellwig <hch@lst.de>
* - new-style, hotplug-aware pci probing and scsi registration
*
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
* Version : v2.00.4 Mon Nov 14 14:02:43 EST 2005 - Seokmann Ju
* <Seokmann.Ju@lsil.com>
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!
2005-04-17 06:20:36 +08:00
*
* Description: Linux device driver for LSI Logic MegaRAID controller
*
* Supported controllers: MegaRAID 418, 428, 438, 466, 762, 467, 471, 490, 493
* 518, 520, 531, 532
*
* This driver is supported by LSI Logic, with assistance from Red Hat, Dell,
* and others. Please send updates to the mailing list
* linux-scsi@vger.kernel.org .
*
*/
#include <linux/mm.h>
#include <linux/fs.h>
#include <linux/blkdev.h>
#include <asm/uaccess.h>
#include <asm/io.h>
[SCSI] remove scsi_cmnd->state We never look at it except for the old megaraid driver that abuses it for sending internal commands. That usage can be fixed easily because those internal commands are single-threaded by a mutex and we can easily use a completion there. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-19 19:42:05 +08:00
#include <linux/completion.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!
2005-04-17 06:20:36 +08:00
#include <linux/delay.h>
#include <linux/proc_fs.h>
#include <linux/reboot.h>
#include <linux/module.h>
#include <linux/list.h>
#include <linux/interrupt.h>
#include <linux/pci.h>
#include <linux/init.h>
[PATCH] Replace 0xff.. with correct DMA_xBIT_MASK Replace all occurences of 0xff.. in calls to function pci_set_dma_mask() and pci_set_consistant_dma_mask() with the corresponding DMA_xBIT_MASK from linux/dma-mapping.h. Signed-off-by: Matthias Gehre <M.Gehre@gmx.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-28 17:56:48 +08:00
#include <linux/dma-mapping.h>
scsi: autoconvert trivial BKL users to private mutex All these files use the big kernel lock in a trivial way to serialize their private file operations, typically resulting from an earlier semi-automatic pushdown from VFS. None of these drivers appears to want to lock against other code, and they all use the BKL as the top-level lock in their file operations, meaning that there is no lock-order inversion problem. Consequently, we can remove the BKL completely, replacing it with a per-file mutex in every case. Using a scripted approach means we can avoid typos. file=$1 name=$2 if grep -q lock_kernel ${file} ; then if grep -q 'include.*linux.mutex.h' ${file} ; then sed -i '/include.*<linux\/smp_lock.h>/d' ${file} else sed -i 's/include.*<linux\/smp_lock.h>.*$/include <linux\/mutex.h>/g' ${file} fi sed -i ${file} \ -e "/^#include.*linux.mutex.h/,$ { 1,/^\(static\|int\|long\)/ { /^\(static\|int\|long\)/istatic DEFINE_MUTEX(${name}_mutex); } }" \ -e "s/\(un\)*lock_kernel\>[ ]*()/mutex_\1lock(\&${name}_mutex)/g" \ -e '/[ ]*cycle_kernel_lock();/d' else sed -i -e '/include.*\<smp_lock.h\>/d' ${file} \ -e '/cycle_kernel_lock()/d' fi Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: linux-scsi@vger.kernel.org Cc: "James E.J. Bottomley" <James.Bottomley@suse.de>
2010-06-02 20:28:52 +08:00
#include <linux/mutex.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#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!
2005-04-17 06:20:36 +08:00
#include <scsi/scsicam.h>
#include "scsi.h"
#include <scsi/scsi_host.h>
#include "megaraid.h"
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
#define MEGARAID_MODULE_VERSION "2.00.4"
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!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
MODULE_AUTHOR ("sju@lsil.com");
MODULE_DESCRIPTION ("LSI Logic MegaRAID legacy driver");
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!
2005-04-17 06:20:36 +08:00
MODULE_LICENSE ("GPL");
MODULE_VERSION(MEGARAID_MODULE_VERSION);
scsi: autoconvert trivial BKL users to private mutex All these files use the big kernel lock in a trivial way to serialize their private file operations, typically resulting from an earlier semi-automatic pushdown from VFS. None of these drivers appears to want to lock against other code, and they all use the BKL as the top-level lock in their file operations, meaning that there is no lock-order inversion problem. Consequently, we can remove the BKL completely, replacing it with a per-file mutex in every case. Using a scripted approach means we can avoid typos. file=$1 name=$2 if grep -q lock_kernel ${file} ; then if grep -q 'include.*linux.mutex.h' ${file} ; then sed -i '/include.*<linux\/smp_lock.h>/d' ${file} else sed -i 's/include.*<linux\/smp_lock.h>.*$/include <linux\/mutex.h>/g' ${file} fi sed -i ${file} \ -e "/^#include.*linux.mutex.h/,$ { 1,/^\(static\|int\|long\)/ { /^\(static\|int\|long\)/istatic DEFINE_MUTEX(${name}_mutex); } }" \ -e "s/\(un\)*lock_kernel\>[ ]*()/mutex_\1lock(\&${name}_mutex)/g" \ -e '/[ ]*cycle_kernel_lock();/d' else sed -i -e '/include.*\<smp_lock.h\>/d' ${file} \ -e '/cycle_kernel_lock()/d' fi Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: linux-scsi@vger.kernel.org Cc: "James E.J. Bottomley" <James.Bottomley@suse.de>
2010-06-02 20:28:52 +08:00
static DEFINE_MUTEX(megadev_mutex);
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!
2005-04-17 06:20:36 +08:00
static unsigned int max_cmd_per_lun = DEF_CMD_PER_LUN;
module_param(max_cmd_per_lun, uint, 0);
MODULE_PARM_DESC(max_cmd_per_lun, "Maximum number of commands which can be issued to a single LUN (default=DEF_CMD_PER_LUN=63)");
static unsigned short int max_sectors_per_io = MAX_SECTORS_PER_IO;
module_param(max_sectors_per_io, ushort, 0);
MODULE_PARM_DESC(max_sectors_per_io, "Maximum number of sectors per I/O request (default=MAX_SECTORS_PER_IO=128)");
static unsigned short int max_mbox_busy_wait = MBOX_BUSY_WAIT;
module_param(max_mbox_busy_wait, ushort, 0);
MODULE_PARM_DESC(max_mbox_busy_wait, "Maximum wait for mailbox in microseconds if busy (default=MBOX_BUSY_WAIT=10)");
[SCSI] megaraid: fix MMIO casts megaraid's MMIO RD*/WR* macros directly call readl() and writel() with an 'unsigned long' argument. This throws a warning, but is otherwise OK because the 'unsigned long' is really the result of ioremap(). This setup is also OK because the variable can hold an ioremap cookie /or/ a PCI I/O port (PIO). However, to fix the warning thrown when readl() and writel() are passed an unsigned long cookie, I introduce 'void __iomem *mmio_base', holding the same value as 'base'. This will silence the warnings, and also cause an oops whenever these MMIO-only functions are ever accidentally passed an I/O address. Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-12-04 09:49:23 +08:00
#define RDINDOOR(adapter) readl((adapter)->mmio_base + 0x20)
#define RDOUTDOOR(adapter) readl((adapter)->mmio_base + 0x2C)
#define WRINDOOR(adapter,value) writel(value, (adapter)->mmio_base + 0x20)
#define WROUTDOOR(adapter,value) writel(value, (adapter)->mmio_base + 0x2C)
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!
2005-04-17 06:20:36 +08:00
/*
* Global variables
*/
static int hba_count;
static adapter_t *hba_soft_state[MAX_CONTROLLERS];
static struct proc_dir_entry *mega_proc_dir_entry;
/* For controller re-ordering */
static struct mega_hbas mega_hbas[MAX_CONTROLLERS];
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
static long
megadev_unlocked_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
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!
2005-04-17 06:20:36 +08:00
/*
* The File Operations structure for the serial/ioctl interface of the driver
*/
[PATCH] mark struct file_operations const 6 Many struct file_operations in the kernel can be "const". Marking them const moves these to the .rodata section, which avoids false sharing with potential dirty data. In addition it'll catch accidental writes at compile time to these shared resources. Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-12 16:55:34 +08:00
static const struct file_operations megadev_fops = {
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!
2005-04-17 06:20:36 +08:00
.owner = THIS_MODULE,
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
.unlocked_ioctl = megadev_unlocked_ioctl,
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!
2005-04-17 06:20:36 +08:00
.open = megadev_open,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-16 00:52:59 +08:00
.llseek = noop_llseek,
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!
2005-04-17 06:20:36 +08:00
};
/*
* Array to structures for storing the information about the controllers. This
* information is sent to the user level applications, when they do an ioctl
* for this information.
*/
static struct mcontroller mcontroller[MAX_CONTROLLERS];
/* The current driver version */
static u32 driver_ver = 0x02000000;
/* major number used by the device for character interface */
static int major;
#define IS_RAID_CH(hba, ch) (((hba)->mega_ch_class >> (ch)) & 0x01)
/*
* Debug variable to print some diagnostic messages
*/
static int trace_level;
/**
* mega_setup_mailbox()
* @adapter - pointer to our soft state
*
* Allocates a 8 byte aligned memory for the handshake mailbox.
*/
static int
mega_setup_mailbox(adapter_t *adapter)
{
unsigned long align;
adapter->una_mbox64 = pci_alloc_consistent(adapter->dev,
sizeof(mbox64_t), &adapter->una_mbox64_dma);
if( !adapter->una_mbox64 ) return -1;
adapter->mbox = &adapter->una_mbox64->mbox;
adapter->mbox = (mbox_t *)((((unsigned long) adapter->mbox) + 15) &
(~0UL ^ 0xFUL));
adapter->mbox64 = (mbox64_t *)(((unsigned long)adapter->mbox) - 8);
align = ((void *)adapter->mbox) - ((void *)&adapter->una_mbox64->mbox);
adapter->mbox_dma = adapter->una_mbox64_dma + 8 + align;
/*
* Register the mailbox if the controller is an io-mapped controller
*/
if( adapter->flag & BOARD_IOMAP ) {
[SCSI] megaraid: outb_p extermination From conversations with the maintainers the _p isn't needed so kill it. That removes the last non ISA _p user from the SCSI layer to my knowledge. Signed-off-by: Alan Cox <alan@redhat.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-02-08 23:34:24 +08:00
outb(adapter->mbox_dma & 0xFF,
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!
2005-04-17 06:20:36 +08:00
adapter->host->io_port + MBOX_PORT0);
[SCSI] megaraid: outb_p extermination From conversations with the maintainers the _p isn't needed so kill it. That removes the last non ISA _p user from the SCSI layer to my knowledge. Signed-off-by: Alan Cox <alan@redhat.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-02-08 23:34:24 +08:00
outb((adapter->mbox_dma >> 8) & 0xFF,
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!
2005-04-17 06:20:36 +08:00
adapter->host->io_port + MBOX_PORT1);
[SCSI] megaraid: outb_p extermination From conversations with the maintainers the _p isn't needed so kill it. That removes the last non ISA _p user from the SCSI layer to my knowledge. Signed-off-by: Alan Cox <alan@redhat.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-02-08 23:34:24 +08:00
outb((adapter->mbox_dma >> 16) & 0xFF,
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!
2005-04-17 06:20:36 +08:00
adapter->host->io_port + MBOX_PORT2);
[SCSI] megaraid: outb_p extermination From conversations with the maintainers the _p isn't needed so kill it. That removes the last non ISA _p user from the SCSI layer to my knowledge. Signed-off-by: Alan Cox <alan@redhat.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-02-08 23:34:24 +08:00
outb((adapter->mbox_dma >> 24) & 0xFF,
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!
2005-04-17 06:20:36 +08:00
adapter->host->io_port + MBOX_PORT3);
[SCSI] megaraid: outb_p extermination From conversations with the maintainers the _p isn't needed so kill it. That removes the last non ISA _p user from the SCSI layer to my knowledge. Signed-off-by: Alan Cox <alan@redhat.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-02-08 23:34:24 +08:00
outb(ENABLE_MBOX_BYTE,
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!
2005-04-17 06:20:36 +08:00
adapter->host->io_port + ENABLE_MBOX_REGION);
irq_ack(adapter);
irq_enable(adapter);
}
return 0;
}
/*
* mega_query_adapter()
* @adapter - pointer to our soft state
*
* Issue the adapter inquiry commands to the controller and find out
* information and parameter about the devices attached
*/
static int
mega_query_adapter(adapter_t *adapter)
{
dma_addr_t prod_info_dma_handle;
mega_inquiry3 *inquiry3;
u8 raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
int retval;
/* Initialize adapter inquiry mailbox */
mbox = (mbox_t *)raw_mbox;
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
memset(&mbox->m_out, 0, sizeof(raw_mbox));
/*
* Try to issue Inquiry3 command
* if not succeeded, then issue MEGA_MBOXCMD_ADAPTERINQ command and
* update enquiry3 structure
*/
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
inquiry3 = (mega_inquiry3 *)adapter->mega_buffer;
raw_mbox[0] = FC_NEW_CONFIG; /* i.e. mbox->cmd=0xA1 */
raw_mbox[2] = NC_SUBOP_ENQUIRY3; /* i.e. 0x0F */
raw_mbox[3] = ENQ3_GET_SOLICITED_FULL; /* i.e. 0x02 */
/* Issue a blocking command to the card */
if ((retval = issue_scb_block(adapter, raw_mbox))) {
/* the adapter does not support 40ld */
mraid_ext_inquiry *ext_inq;
mraid_inquiry *inq;
dma_addr_t dma_handle;
ext_inq = pci_alloc_consistent(adapter->dev,
sizeof(mraid_ext_inquiry), &dma_handle);
if( ext_inq == NULL ) return -1;
inq = &ext_inq->raid_inq;
mbox->m_out.xferaddr = (u32)dma_handle;
/*issue old 0x04 command to adapter */
mbox->m_out.cmd = MEGA_MBOXCMD_ADPEXTINQ;
issue_scb_block(adapter, raw_mbox);
/*
* update Enquiry3 and ProductInfo structures with
* mraid_inquiry structure
*/
mega_8_to_40ld(inq, inquiry3,
(mega_product_info *)&adapter->product_info);
pci_free_consistent(adapter->dev, sizeof(mraid_ext_inquiry),
ext_inq, dma_handle);
} else { /*adapter supports 40ld */
adapter->flag |= BOARD_40LD;
/*
* get product_info, which is static information and will be
* unchanged
*/
prod_info_dma_handle = pci_map_single(adapter->dev, (void *)
&adapter->product_info,
sizeof(mega_product_info), PCI_DMA_FROMDEVICE);
mbox->m_out.xferaddr = prod_info_dma_handle;
raw_mbox[0] = FC_NEW_CONFIG; /* i.e. mbox->cmd=0xA1 */
raw_mbox[2] = NC_SUBOP_PRODUCT_INFO; /* i.e. 0x0E */
if ((retval = issue_scb_block(adapter, raw_mbox)))
printk(KERN_WARNING
"megaraid: Product_info cmd failed with error: %d\n",
retval);
pci_unmap_single(adapter->dev, prod_info_dma_handle,
sizeof(mega_product_info), PCI_DMA_FROMDEVICE);
}
/*
* kernel scans the channels from 0 to <= max_channel
*/
adapter->host->max_channel =
adapter->product_info.nchannels + NVIRT_CHAN -1;
adapter->host->max_id = 16; /* max targets per channel */
adapter->host->max_lun = 7; /* Upto 7 luns for non disk devices */
adapter->host->cmd_per_lun = max_cmd_per_lun;
adapter->numldrv = inquiry3->num_ldrv;
adapter->max_cmds = adapter->product_info.max_commands;
if(adapter->max_cmds > MAX_COMMANDS)
adapter->max_cmds = MAX_COMMANDS;
adapter->host->can_queue = adapter->max_cmds - 1;
/*
* Get the maximum number of scatter-gather elements supported by this
* firmware
*/
mega_get_max_sgl(adapter);
adapter->host->sg_tablesize = adapter->sglen;
/* use HP firmware and bios version encoding */
if (adapter->product_info.subsysvid == HP_SUBSYS_VID) {
sprintf (adapter->fw_version, "%c%d%d.%d%d",
adapter->product_info.fw_version[2],
adapter->product_info.fw_version[1] >> 8,
adapter->product_info.fw_version[1] & 0x0f,
adapter->product_info.fw_version[0] >> 8,
adapter->product_info.fw_version[0] & 0x0f);
sprintf (adapter->bios_version, "%c%d%d.%d%d",
adapter->product_info.bios_version[2],
adapter->product_info.bios_version[1] >> 8,
adapter->product_info.bios_version[1] & 0x0f,
adapter->product_info.bios_version[0] >> 8,
adapter->product_info.bios_version[0] & 0x0f);
} else {
memcpy(adapter->fw_version,
(char *)adapter->product_info.fw_version, 4);
adapter->fw_version[4] = 0;
memcpy(adapter->bios_version,
(char *)adapter->product_info.bios_version, 4);
adapter->bios_version[4] = 0;
}
printk(KERN_NOTICE "megaraid: [%s:%s] detected %d logical drives.\n",
adapter->fw_version, adapter->bios_version, adapter->numldrv);
/*
* Do we support extended (>10 bytes) cdbs
*/
adapter->support_ext_cdb = mega_support_ext_cdb(adapter);
if (adapter->support_ext_cdb)
printk(KERN_NOTICE "megaraid: supports extended CDBs.\n");
return 0;
}
/**
* mega_runpendq()
* @adapter - pointer to our soft state
*
* Runs through the list of pending requests.
*/
static inline void
mega_runpendq(adapter_t *adapter)
{
if(!list_empty(&adapter->pending_list))
__mega_runpendq(adapter);
}
/*
* megaraid_queue()
* @scmd - Issue this scsi command
* @done - the callback hook into the scsi mid-layer
*
* The command queuing entry point for the mid-layer.
*/
static int
megaraid_queue(Scsi_Cmnd *scmd, void (*done)(Scsi_Cmnd *))
{
adapter_t *adapter;
scb_t *scb;
int busy=0;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
unsigned long flags;
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!
2005-04-17 06:20:36 +08:00
adapter = (adapter_t *)scmd->device->host->hostdata;
scmd->scsi_done = done;
/*
* Allocate and build a SCB request
* busy flag will be set if mega_build_cmd() command could not
* allocate scb. We will return non-zero status in that case.
* NOTE: scb can be null even though certain commands completed
* successfully, e.g., MODE_SENSE and TEST_UNIT_READY, we would
* return 0 in that case.
*/
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
spin_lock_irqsave(&adapter->lock, flags);
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!
2005-04-17 06:20:36 +08:00
scb = mega_build_cmd(adapter, scmd, &busy);
[PATCH] fix megaraid.c locking This fixes locking in megaraid.c, namely: (1) make sure megaraid_queue release the adapter lock by changing the code to have a single return (2) remove the errornous scsi_assign_lock call Testing by Burton Windle. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Burton Windle <bwindle@fint.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-30 04:36:16 +08:00
if (!scb)
goto out;
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!
2005-04-17 06:20:36 +08:00
[PATCH] fix megaraid.c locking This fixes locking in megaraid.c, namely: (1) make sure megaraid_queue release the adapter lock by changing the code to have a single return (2) remove the errornous scsi_assign_lock call Testing by Burton Windle. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Burton Windle <bwindle@fint.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-30 04:36:16 +08:00
scb->state |= SCB_PENDQ;
list_add_tail(&scb->list, &adapter->pending_list);
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!
2005-04-17 06:20:36 +08:00
[PATCH] fix megaraid.c locking This fixes locking in megaraid.c, namely: (1) make sure megaraid_queue release the adapter lock by changing the code to have a single return (2) remove the errornous scsi_assign_lock call Testing by Burton Windle. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Burton Windle <bwindle@fint.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-30 04:36:16 +08:00
/*
* Check if the HBA is in quiescent state, e.g., during a
* delete logical drive opertion. If it is, don't run
* the pending_list.
*/
if (atomic_read(&adapter->quiescent) == 0)
mega_runpendq(adapter);
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!
2005-04-17 06:20:36 +08:00
[PATCH] fix megaraid.c locking This fixes locking in megaraid.c, namely: (1) make sure megaraid_queue release the adapter lock by changing the code to have a single return (2) remove the errornous scsi_assign_lock call Testing by Burton Windle. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Burton Windle <bwindle@fint.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-30 04:36:16 +08:00
busy = 0;
out:
spin_unlock_irqrestore(&adapter->lock, flags);
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!
2005-04-17 06:20:36 +08:00
return busy;
}
/**
* mega_allocate_scb()
* @adapter - pointer to our soft state
* @cmd - scsi command from the mid-layer
*
* Allocate a SCB structure. This is the central structure for controller
* commands.
*/
static inline scb_t *
mega_allocate_scb(adapter_t *adapter, Scsi_Cmnd *cmd)
{
struct list_head *head = &adapter->free_list;
scb_t *scb;
/* Unlink command from Free List */
if( !list_empty(head) ) {
scb = list_entry(head->next, scb_t, list);
list_del_init(head->next);
scb->state = SCB_ACTIVE;
scb->cmd = cmd;
scb->dma_type = MEGA_DMA_TYPE_NONE;
return scb;
}
return NULL;
}
/**
* mega_get_ldrv_num()
* @adapter - pointer to our soft state
* @cmd - scsi mid layer command
* @channel - channel on the controller
*
* Calculate the logical drive number based on the information in scsi command
* and the channel number.
*/
static inline int
mega_get_ldrv_num(adapter_t *adapter, Scsi_Cmnd *cmd, int channel)
{
int tgt;
int ldrv_num;
tgt = cmd->device->id;
if ( tgt > adapter->this_id )
tgt--; /* we do not get inquires for initiator id */
ldrv_num = (channel * 15) + tgt;
/*
* If we have a logical drive with boot enabled, project it first
*/
if( adapter->boot_ldrv_enabled ) {
if( ldrv_num == 0 ) {
ldrv_num = adapter->boot_ldrv;
}
else {
if( ldrv_num <= adapter->boot_ldrv ) {
ldrv_num--;
}
}
}
/*
* If "delete logical drive" feature is enabled on this controller.
* Do only if at least one delete logical drive operation was done.
*
* Also, after logical drive deletion, instead of logical drive number,
* the value returned should be 0x80+logical drive id.
*
* These is valid only for IO commands.
*/
if (adapter->support_random_del && adapter->read_ldidmap )
switch (cmd->cmnd[0]) {
case READ_6: /* fall through */
case WRITE_6: /* fall through */
case READ_10: /* fall through */
case WRITE_10:
ldrv_num += 0x80;
}
return ldrv_num;
}
/**
* mega_build_cmd()
* @adapter - pointer to our soft state
* @cmd - Prepare using this scsi command
* @busy - busy flag if no resources
*
* Prepares a command and scatter gather list for the controller. This routine
* also finds out if the commands is intended for a logical drive or a
* physical device and prepares the controller command accordingly.
*
* We also re-order the logical drives and physical devices based on their
* boot settings.
*/
static scb_t *
mega_build_cmd(adapter_t *adapter, Scsi_Cmnd *cmd, int *busy)
{
mega_ext_passthru *epthru;
mega_passthru *pthru;
scb_t *scb;
mbox_t *mbox;
long seg;
char islogical;
int max_ldrv_num;
int channel = 0;
int target = 0;
int ldrv_num = 0; /* logical drive number */
/*
* filter the internal and ioctl commands
*/
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
if((cmd->cmnd[0] == MEGA_INTERNAL_CMD))
return (scb_t *)cmd->host_scribble;
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!
2005-04-17 06:20:36 +08:00
/*
* We know what channels our logical drives are on - mega_find_card()
*/
islogical = adapter->logdrv_chan[cmd->device->channel];
/*
* The theory: If physical drive is chosen for boot, all the physical
* devices are exported before the logical drives, otherwise physical
* devices are pushed after logical drives, in which case - Kernel sees
* the physical devices on virtual channel which is obviously converted
* to actual channel on the HBA.
*/
if( adapter->boot_pdrv_enabled ) {
if( islogical ) {
/* logical channel */
channel = cmd->device->channel -
adapter->product_info.nchannels;
}
else {
/* this is physical channel */
channel = cmd->device->channel;
target = cmd->device->id;
/*
* boot from a physical disk, that disk needs to be
* exposed first IF both the channels are SCSI, then
* booting from the second channel is not allowed.
*/
if( target == 0 ) {
target = adapter->boot_pdrv_tgt;
}
else if( target == adapter->boot_pdrv_tgt ) {
target = 0;
}
}
}
else {
if( islogical ) {
/* this is the logical channel */
channel = cmd->device->channel;
}
else {
/* physical channel */
channel = cmd->device->channel - NVIRT_CHAN;
target = cmd->device->id;
}
}
if(islogical) {
/* have just LUN 0 for each target on virtual channels */
if (cmd->device->lun) {
cmd->result = (DID_BAD_TARGET << 16);
cmd->scsi_done(cmd);
return NULL;
}
ldrv_num = mega_get_ldrv_num(adapter, cmd, channel);
max_ldrv_num = (adapter->flag & BOARD_40LD) ?
MAX_LOGICAL_DRIVES_40LD : MAX_LOGICAL_DRIVES_8LD;
/*
* max_ldrv_num increases by 0x80 if some logical drive was
* deleted.
*/
if(adapter->read_ldidmap)
max_ldrv_num += 0x80;
if(ldrv_num > max_ldrv_num ) {
cmd->result = (DID_BAD_TARGET << 16);
cmd->scsi_done(cmd);
return NULL;
}
}
else {
if( cmd->device->lun > 7) {
/*
* Do not support lun >7 for physically accessed
* devices
*/
cmd->result = (DID_BAD_TARGET << 16);
cmd->scsi_done(cmd);
return NULL;
}
}
/*
*
* Logical drive commands
*
*/
if(islogical) {
switch (cmd->cmnd[0]) {
case TEST_UNIT_READY:
#if MEGA_HAVE_CLUSTERING
/*
* Do we support clustering and is the support enabled
* If no, return success always
*/
if( !adapter->has_cluster ) {
cmd->result = (DID_OK << 16);
cmd->scsi_done(cmd);
return NULL;
}
if(!(scb = mega_allocate_scb(adapter, cmd))) {
*busy = 1;
return NULL;
}
scb->raw_mbox[0] = MEGA_CLUSTER_CMD;
scb->raw_mbox[2] = MEGA_RESERVATION_STATUS;
scb->raw_mbox[3] = ldrv_num;
scb->dma_direction = PCI_DMA_NONE;
return scb;
#else
cmd->result = (DID_OK << 16);
cmd->scsi_done(cmd);
return NULL;
#endif
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
case MODE_SENSE: {
char *buf;
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
struct scatterlist *sg;
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
sg = scsi_sglist(cmd);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-23 03:19:53 +08:00
buf = kmap_atomic(sg_page(sg), KM_IRQ0) + sg->offset;
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
[SCSI] Fix incorrect pointer in megaraid.c MODE_SENSE emulation The SCSI megaraid drive goes to great effort to kmap the scatterlist buffer (if used), but then uses the wrong pointer when copying to it afterward. Signed-off-by: Mark Lord <lkml@rtr.ca> Acked by: Ju, Seokmann <Seokmann.Ju@engenio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-12-08 06:46:57 +08:00
memset(buf, 0, cmd->cmnd[4]);
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
kunmap_atomic(buf - sg->offset, KM_IRQ0);
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
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!
2005-04-17 06:20:36 +08:00
cmd->result = (DID_OK << 16);
cmd->scsi_done(cmd);
return NULL;
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
}
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!
2005-04-17 06:20:36 +08:00
case READ_CAPACITY:
case INQUIRY:
if(!(adapter->flag & (1L << cmd->device->channel))) {
printk(KERN_NOTICE
"scsi%d: scanning scsi channel %d ",
adapter->host->host_no,
cmd->device->channel);
printk("for logical drives.\n");
adapter->flag |= (1L << cmd->device->channel);
}
/* Allocate a SCB and initialize passthru */
if(!(scb = mega_allocate_scb(adapter, cmd))) {
*busy = 1;
return NULL;
}
pthru = scb->pthru;
mbox = (mbox_t *)scb->raw_mbox;
memset(mbox, 0, sizeof(scb->raw_mbox));
memset(pthru, 0, sizeof(mega_passthru));
pthru->timeout = 0;
pthru->ars = 1;
pthru->reqsenselen = 14;
pthru->islogical = 1;
pthru->logdrv = ldrv_num;
pthru->cdblen = cmd->cmd_len;
memcpy(pthru->cdb, cmd->cmnd, cmd->cmd_len);
if( adapter->has_64bit_addr ) {
mbox->m_out.cmd = MEGA_MBOXCMD_PASSTHRU64;
}
else {
mbox->m_out.cmd = MEGA_MBOXCMD_PASSTHRU;
}
scb->dma_direction = PCI_DMA_FROMDEVICE;
pthru->numsgelements = mega_build_sglist(adapter, scb,
&pthru->dataxferaddr, &pthru->dataxferlen);
mbox->m_out.xferaddr = scb->pthru_dma_addr;
return scb;
case READ_6:
case WRITE_6:
case READ_10:
case WRITE_10:
case READ_12:
case WRITE_12:
/* Allocate a SCB and initialize mailbox */
if(!(scb = mega_allocate_scb(adapter, cmd))) {
*busy = 1;
return NULL;
}
mbox = (mbox_t *)scb->raw_mbox;
memset(mbox, 0, sizeof(scb->raw_mbox));
mbox->m_out.logdrv = ldrv_num;
/*
* A little hack: 2nd bit is zero for all scsi read
* commands and is set for all scsi write commands
*/
if( adapter->has_64bit_addr ) {
mbox->m_out.cmd = (*cmd->cmnd & 0x02) ?
MEGA_MBOXCMD_LWRITE64:
MEGA_MBOXCMD_LREAD64 ;
}
else {
mbox->m_out.cmd = (*cmd->cmnd & 0x02) ?
MEGA_MBOXCMD_LWRITE:
MEGA_MBOXCMD_LREAD ;
}
/*
* 6-byte READ(0x08) or WRITE(0x0A) cdb
*/
if( cmd->cmd_len == 6 ) {
mbox->m_out.numsectors = (u32) cmd->cmnd[4];
mbox->m_out.lba =
((u32)cmd->cmnd[1] << 16) |
((u32)cmd->cmnd[2] << 8) |
(u32)cmd->cmnd[3];
mbox->m_out.lba &= 0x1FFFFF;
#if MEGA_HAVE_STATS
/*
* Take modulo 0x80, since the logical drive
* number increases by 0x80 when a logical
* drive was deleted
*/
if (*cmd->cmnd == READ_6) {
adapter->nreads[ldrv_num%0x80]++;
adapter->nreadblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
} else {
adapter->nwrites[ldrv_num%0x80]++;
adapter->nwriteblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
}
#endif
}
/*
* 10-byte READ(0x28) or WRITE(0x2A) cdb
*/
if( cmd->cmd_len == 10 ) {
mbox->m_out.numsectors =
(u32)cmd->cmnd[8] |
((u32)cmd->cmnd[7] << 8);
mbox->m_out.lba =
((u32)cmd->cmnd[2] << 24) |
((u32)cmd->cmnd[3] << 16) |
((u32)cmd->cmnd[4] << 8) |
(u32)cmd->cmnd[5];
#if MEGA_HAVE_STATS
if (*cmd->cmnd == READ_10) {
adapter->nreads[ldrv_num%0x80]++;
adapter->nreadblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
} else {
adapter->nwrites[ldrv_num%0x80]++;
adapter->nwriteblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
}
#endif
}
/*
* 12-byte READ(0xA8) or WRITE(0xAA) cdb
*/
if( cmd->cmd_len == 12 ) {
mbox->m_out.lba =
((u32)cmd->cmnd[2] << 24) |
((u32)cmd->cmnd[3] << 16) |
((u32)cmd->cmnd[4] << 8) |
(u32)cmd->cmnd[5];
mbox->m_out.numsectors =
((u32)cmd->cmnd[6] << 24) |
((u32)cmd->cmnd[7] << 16) |
((u32)cmd->cmnd[8] << 8) |
(u32)cmd->cmnd[9];
#if MEGA_HAVE_STATS
if (*cmd->cmnd == READ_12) {
adapter->nreads[ldrv_num%0x80]++;
adapter->nreadblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
} else {
adapter->nwrites[ldrv_num%0x80]++;
adapter->nwriteblocks[ldrv_num%0x80] +=
mbox->m_out.numsectors;
}
#endif
}
/*
* If it is a read command
*/
if( (*cmd->cmnd & 0x0F) == 0x08 ) {
scb->dma_direction = PCI_DMA_FROMDEVICE;
}
else {
scb->dma_direction = PCI_DMA_TODEVICE;
}
/* Calculate Scatter-Gather info */
mbox->m_out.numsgelements = mega_build_sglist(adapter, scb,
(u32 *)&mbox->m_out.xferaddr, (u32 *)&seg);
return scb;
#if MEGA_HAVE_CLUSTERING
case RESERVE: /* Fall through */
case RELEASE:
/*
* Do we support clustering and is the support enabled
*/
if( ! adapter->has_cluster ) {
cmd->result = (DID_BAD_TARGET << 16);
cmd->scsi_done(cmd);
return NULL;
}
/* Allocate a SCB and initialize mailbox */
if(!(scb = mega_allocate_scb(adapter, cmd))) {
*busy = 1;
return NULL;
}
scb->raw_mbox[0] = MEGA_CLUSTER_CMD;
scb->raw_mbox[2] = ( *cmd->cmnd == RESERVE ) ?
MEGA_RESERVE_LD : MEGA_RELEASE_LD;
scb->raw_mbox[3] = ldrv_num;
scb->dma_direction = PCI_DMA_NONE;
return scb;
#endif
default:
cmd->result = (DID_BAD_TARGET << 16);
cmd->scsi_done(cmd);
return NULL;
}
}
/*
* Passthru drive commands
*/
else {
/* Allocate a SCB and initialize passthru */
if(!(scb = mega_allocate_scb(adapter, cmd))) {
*busy = 1;
return NULL;
}
mbox = (mbox_t *)scb->raw_mbox;
memset(mbox, 0, sizeof(scb->raw_mbox));
if( adapter->support_ext_cdb ) {
epthru = mega_prepare_extpassthru(adapter, scb, cmd,
channel, target);
mbox->m_out.cmd = MEGA_MBOXCMD_EXTPTHRU;
mbox->m_out.xferaddr = scb->epthru_dma_addr;
}
else {
pthru = mega_prepare_passthru(adapter, scb, cmd,
channel, target);
/* Initialize mailbox */
if( adapter->has_64bit_addr ) {
mbox->m_out.cmd = MEGA_MBOXCMD_PASSTHRU64;
}
else {
mbox->m_out.cmd = MEGA_MBOXCMD_PASSTHRU;
}
mbox->m_out.xferaddr = scb->pthru_dma_addr;
}
return scb;
}
return NULL;
}
/**
* mega_prepare_passthru()
* @adapter - pointer to our soft state
* @scb - our scsi control block
* @cmd - scsi command from the mid-layer
* @channel - actual channel on the controller
* @target - actual id on the controller.
*
* prepare a command for the scsi physical devices.
*/
static mega_passthru *
mega_prepare_passthru(adapter_t *adapter, scb_t *scb, Scsi_Cmnd *cmd,
int channel, int target)
{
mega_passthru *pthru;
pthru = scb->pthru;
memset(pthru, 0, sizeof (mega_passthru));
/* 0=6sec/1=60sec/2=10min/3=3hrs */
pthru->timeout = 2;
pthru->ars = 1;
pthru->reqsenselen = 14;
pthru->islogical = 0;
pthru->channel = (adapter->flag & BOARD_40LD) ? 0 : channel;
pthru->target = (adapter->flag & BOARD_40LD) ?
(channel << 4) | target : target;
pthru->cdblen = cmd->cmd_len;
pthru->logdrv = cmd->device->lun;
memcpy(pthru->cdb, cmd->cmnd, cmd->cmd_len);
/* Not sure about the direction */
scb->dma_direction = PCI_DMA_BIDIRECTIONAL;
/* Special Code for Handling READ_CAPA/ INQ using bounce buffers */
switch (cmd->cmnd[0]) {
case INQUIRY:
case READ_CAPACITY:
if(!(adapter->flag & (1L << cmd->device->channel))) {
printk(KERN_NOTICE
"scsi%d: scanning scsi channel %d [P%d] ",
adapter->host->host_no,
cmd->device->channel, channel);
printk("for physical devices.\n");
adapter->flag |= (1L << cmd->device->channel);
}
/* Fall through */
default:
pthru->numsgelements = mega_build_sglist(adapter, scb,
&pthru->dataxferaddr, &pthru->dataxferlen);
break;
}
return pthru;
}
/**
* mega_prepare_extpassthru()
* @adapter - pointer to our soft state
* @scb - our scsi control block
* @cmd - scsi command from the mid-layer
* @channel - actual channel on the controller
* @target - actual id on the controller.
*
* prepare a command for the scsi physical devices. This rountine prepares
* commands for devices which can take extended CDBs (>10 bytes)
*/
static mega_ext_passthru *
mega_prepare_extpassthru(adapter_t *adapter, scb_t *scb, Scsi_Cmnd *cmd,
int channel, int target)
{
mega_ext_passthru *epthru;
epthru = scb->epthru;
memset(epthru, 0, sizeof(mega_ext_passthru));
/* 0=6sec/1=60sec/2=10min/3=3hrs */
epthru->timeout = 2;
epthru->ars = 1;
epthru->reqsenselen = 14;
epthru->islogical = 0;
epthru->channel = (adapter->flag & BOARD_40LD) ? 0 : channel;
epthru->target = (adapter->flag & BOARD_40LD) ?
(channel << 4) | target : target;
epthru->cdblen = cmd->cmd_len;
epthru->logdrv = cmd->device->lun;
memcpy(epthru->cdb, cmd->cmnd, cmd->cmd_len);
/* Not sure about the direction */
scb->dma_direction = PCI_DMA_BIDIRECTIONAL;
switch(cmd->cmnd[0]) {
case INQUIRY:
case READ_CAPACITY:
if(!(adapter->flag & (1L << cmd->device->channel))) {
printk(KERN_NOTICE
"scsi%d: scanning scsi channel %d [P%d] ",
adapter->host->host_no,
cmd->device->channel, channel);
printk("for physical devices.\n");
adapter->flag |= (1L << cmd->device->channel);
}
/* Fall through */
default:
epthru->numsgelements = mega_build_sglist(adapter, scb,
&epthru->dataxferaddr, &epthru->dataxferlen);
break;
}
return epthru;
}
static void
__mega_runpendq(adapter_t *adapter)
{
scb_t *scb;
struct list_head *pos, *next;
/* Issue any pending commands to the card */
list_for_each_safe(pos, next, &adapter->pending_list) {
scb = list_entry(pos, scb_t, list);
if( !(scb->state & SCB_ISSUED) ) {
if( issue_scb(adapter, scb) != 0 )
return;
}
}
return;
}
/**
* issue_scb()
* @adapter - pointer to our soft state
* @scb - scsi control block
*
* Post a command to the card if the mailbox is available, otherwise return
* busy. We also take the scb from the pending list if the mailbox is
* available.
*/
static int
issue_scb(adapter_t *adapter, scb_t *scb)
{
volatile mbox64_t *mbox64 = adapter->mbox64;
volatile mbox_t *mbox = adapter->mbox;
unsigned int i = 0;
if(unlikely(mbox->m_in.busy)) {
do {
udelay(1);
i++;
} while( mbox->m_in.busy && (i < max_mbox_busy_wait) );
if(mbox->m_in.busy) return -1;
}
/* Copy mailbox data into host structure */
memcpy((char *)&mbox->m_out, (char *)scb->raw_mbox,
sizeof(struct mbox_out));
mbox->m_out.cmdid = scb->idx; /* Set cmdid */
mbox->m_in.busy = 1; /* Set busy */
/*
* Increment the pending queue counter
*/
atomic_inc(&adapter->pend_cmds);
switch (mbox->m_out.cmd) {
case MEGA_MBOXCMD_LREAD64:
case MEGA_MBOXCMD_LWRITE64:
case MEGA_MBOXCMD_PASSTHRU64:
case MEGA_MBOXCMD_EXTPTHRU:
mbox64->xfer_segment_lo = mbox->m_out.xferaddr;
mbox64->xfer_segment_hi = 0;
mbox->m_out.xferaddr = 0xFFFFFFFF;
break;
default:
mbox64->xfer_segment_lo = 0;
mbox64->xfer_segment_hi = 0;
}
/*
* post the command
*/
scb->state |= SCB_ISSUED;
if( likely(adapter->flag & BOARD_MEMMAP) ) {
mbox->m_in.poll = 0;
mbox->m_in.ack = 0;
WRINDOOR(adapter, adapter->mbox_dma | 0x1);
}
else {
irq_enable(adapter);
issue_command(adapter);
}
return 0;
}
/*
* Wait until the controller's mailbox is available
*/
static inline int
mega_busywait_mbox (adapter_t *adapter)
{
if (adapter->mbox->m_in.busy)
return __mega_busywait_mbox(adapter);
return 0;
}
/**
* issue_scb_block()
* @adapter - pointer to our soft state
* @raw_mbox - the mailbox
*
* Issue a scb in synchronous and non-interrupt mode
*/
static int
issue_scb_block(adapter_t *adapter, u_char *raw_mbox)
{
volatile mbox64_t *mbox64 = adapter->mbox64;
volatile mbox_t *mbox = adapter->mbox;
u8 byte;
/* Wait until mailbox is free */
if(mega_busywait_mbox (adapter))
goto bug_blocked_mailbox;
/* Copy mailbox data into host structure */
memcpy((char *) mbox, raw_mbox, sizeof(struct mbox_out));
mbox->m_out.cmdid = 0xFE;
mbox->m_in.busy = 1;
switch (raw_mbox[0]) {
case MEGA_MBOXCMD_LREAD64:
case MEGA_MBOXCMD_LWRITE64:
case MEGA_MBOXCMD_PASSTHRU64:
case MEGA_MBOXCMD_EXTPTHRU:
mbox64->xfer_segment_lo = mbox->m_out.xferaddr;
mbox64->xfer_segment_hi = 0;
mbox->m_out.xferaddr = 0xFFFFFFFF;
break;
default:
mbox64->xfer_segment_lo = 0;
mbox64->xfer_segment_hi = 0;
}
if( likely(adapter->flag & BOARD_MEMMAP) ) {
mbox->m_in.poll = 0;
mbox->m_in.ack = 0;
mbox->m_in.numstatus = 0xFF;
mbox->m_in.status = 0xFF;
WRINDOOR(adapter, adapter->mbox_dma | 0x1);
while((volatile u8)mbox->m_in.numstatus == 0xFF)
cpu_relax();
mbox->m_in.numstatus = 0xFF;
while( (volatile u8)mbox->m_in.poll != 0x77 )
cpu_relax();
mbox->m_in.poll = 0;
mbox->m_in.ack = 0x77;
WRINDOOR(adapter, adapter->mbox_dma | 0x2);
while(RDINDOOR(adapter) & 0x2)
cpu_relax();
}
else {
irq_disable(adapter);
issue_command(adapter);
while (!((byte = irq_state(adapter)) & INTR_VALID))
cpu_relax();
set_irq_state(adapter, byte);
irq_enable(adapter);
irq_ack(adapter);
}
return mbox->m_in.status;
bug_blocked_mailbox:
printk(KERN_WARNING "megaraid: Blocked mailbox......!!\n");
udelay (1000);
return -1;
}
/**
* megaraid_isr_iomapped()
* @irq - irq
* @devp - pointer to our soft state
*
* Interrupt service routine for io-mapped controllers.
* Find out if our device is interrupting. If yes, acknowledge the interrupt
* and service the completed commands.
*/
static irqreturn_t
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
megaraid_isr_iomapped(int irq, void *devp)
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!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter = devp;
unsigned long flags;
u8 status;
u8 nstatus;
u8 completed[MAX_FIRMWARE_STATUS];
u8 byte;
int handled = 0;
/*
* loop till F/W has more commands for us to complete.
*/
spin_lock_irqsave(&adapter->lock, flags);
do {
/* Check if a valid interrupt is pending */
byte = irq_state(adapter);
if( (byte & VALID_INTR_BYTE) == 0 ) {
/*
* No more pending commands
*/
goto out_unlock;
}
set_irq_state(adapter, byte);
while((nstatus = (volatile u8)adapter->mbox->m_in.numstatus)
== 0xFF)
cpu_relax();
adapter->mbox->m_in.numstatus = 0xFF;
status = adapter->mbox->m_in.status;
/*
* decrement the pending queue counter
*/
atomic_sub(nstatus, &adapter->pend_cmds);
memcpy(completed, (void *)adapter->mbox->m_in.completed,
nstatus);
/* Acknowledge interrupt */
irq_ack(adapter);
mega_cmd_done(adapter, completed, nstatus, status);
mega_rundoneq(adapter);
handled = 1;
/* Loop through any pending requests */
if(atomic_read(&adapter->quiescent) == 0) {
mega_runpendq(adapter);
}
} while(1);
out_unlock:
spin_unlock_irqrestore(&adapter->lock, flags);
return IRQ_RETVAL(handled);
}
/**
* megaraid_isr_memmapped()
* @irq - irq
* @devp - pointer to our soft state
*
* Interrupt service routine for memory-mapped controllers.
* Find out if our device is interrupting. If yes, acknowledge the interrupt
* and service the completed commands.
*/
static irqreturn_t
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
megaraid_isr_memmapped(int irq, void *devp)
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!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter = devp;
unsigned long flags;
u8 status;
u32 dword = 0;
u8 nstatus;
u8 completed[MAX_FIRMWARE_STATUS];
int handled = 0;
/*
* loop till F/W has more commands for us to complete.
*/
spin_lock_irqsave(&adapter->lock, flags);
do {
/* Check if a valid interrupt is pending */
dword = RDOUTDOOR(adapter);
if(dword != 0x10001234) {
/*
* No more pending commands
*/
goto out_unlock;
}
WROUTDOOR(adapter, 0x10001234);
while((nstatus = (volatile u8)adapter->mbox->m_in.numstatus)
== 0xFF) {
cpu_relax();
}
adapter->mbox->m_in.numstatus = 0xFF;
status = adapter->mbox->m_in.status;
/*
* decrement the pending queue counter
*/
atomic_sub(nstatus, &adapter->pend_cmds);
memcpy(completed, (void *)adapter->mbox->m_in.completed,
nstatus);
/* Acknowledge interrupt */
WRINDOOR(adapter, 0x2);
handled = 1;
[SCSI] megaraid: fix MMIO casts megaraid's MMIO RD*/WR* macros directly call readl() and writel() with an 'unsigned long' argument. This throws a warning, but is otherwise OK because the 'unsigned long' is really the result of ioremap(). This setup is also OK because the variable can hold an ioremap cookie /or/ a PCI I/O port (PIO). However, to fix the warning thrown when readl() and writel() are passed an unsigned long cookie, I introduce 'void __iomem *mmio_base', holding the same value as 'base'. This will silence the warnings, and also cause an oops whenever these MMIO-only functions are ever accidentally passed an I/O address. Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-12-04 09:49:23 +08:00
while( RDINDOOR(adapter) & 0x02 )
cpu_relax();
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!
2005-04-17 06:20:36 +08:00
mega_cmd_done(adapter, completed, nstatus, status);
mega_rundoneq(adapter);
/* Loop through any pending requests */
if(atomic_read(&adapter->quiescent) == 0) {
mega_runpendq(adapter);
}
} while(1);
out_unlock:
spin_unlock_irqrestore(&adapter->lock, flags);
return IRQ_RETVAL(handled);
}
/**
* mega_cmd_done()
* @adapter - pointer to our soft state
* @completed - array of ids of completed commands
* @nstatus - number of completed commands
* @status - status of the last command completed
*
* Complete the comamnds and call the scsi mid-layer callback hooks.
*/
static void
mega_cmd_done(adapter_t *adapter, u8 completed[], int nstatus, int status)
{
mega_ext_passthru *epthru = NULL;
struct scatterlist *sgl;
Scsi_Cmnd *cmd = NULL;
mega_passthru *pthru = NULL;
mbox_t *mbox = NULL;
u8 c;
scb_t *scb;
int islogical;
int cmdid;
int i;
/*
* for all the commands completed, call the mid-layer callback routine
* and free the scb.
*/
for( i = 0; i < nstatus; i++ ) {
cmdid = completed[i];
if( cmdid == CMDID_INT_CMDS ) { /* internal command */
scb = &adapter->int_scb;
cmd = scb->cmd;
mbox = (mbox_t *)scb->raw_mbox;
/*
* Internal command interface do not fire the extended
* passthru or 64-bit passthru
*/
pthru = scb->pthru;
}
else {
scb = &adapter->scb_list[cmdid];
/*
* Make sure f/w has completed a valid command
*/
if( !(scb->state & SCB_ISSUED) || scb->cmd == NULL ) {
printk(KERN_CRIT
"megaraid: invalid command ");
printk("Id %d, scb->state:%x, scsi cmd:%p\n",
cmdid, scb->state, scb->cmd);
continue;
}
/*
* Was a abort issued for this command
*/
if( scb->state & SCB_ABORT ) {
printk(KERN_WARNING
"megaraid: aborted cmd %lx[%x] complete.\n",
scb->cmd->serial_number, scb->idx);
scb->cmd->result = (DID_ABORT << 16);
list_add_tail(SCSI_LIST(scb->cmd),
&adapter->completed_list);
mega_free_scb(adapter, scb);
continue;
}
/*
* Was a reset issued for this command
*/
if( scb->state & SCB_RESET ) {
printk(KERN_WARNING
"megaraid: reset cmd %lx[%x] complete.\n",
scb->cmd->serial_number, scb->idx);
scb->cmd->result = (DID_RESET << 16);
list_add_tail(SCSI_LIST(scb->cmd),
&adapter->completed_list);
mega_free_scb (adapter, scb);
continue;
}
cmd = scb->cmd;
pthru = scb->pthru;
epthru = scb->epthru;
mbox = (mbox_t *)scb->raw_mbox;
#if MEGA_HAVE_STATS
{
int logdrv = mbox->m_out.logdrv;
islogical = adapter->logdrv_chan[cmd->channel];
/*
* Maintain an error counter for the logical drive.
* Some application like SNMP agent need such
* statistics
*/
if( status && islogical && (cmd->cmnd[0] == READ_6 ||
cmd->cmnd[0] == READ_10 ||
cmd->cmnd[0] == READ_12)) {
/*
* Logical drive number increases by 0x80 when
* a logical drive is deleted
*/
adapter->rd_errors[logdrv%0x80]++;
}
if( status && islogical && (cmd->cmnd[0] == WRITE_6 ||
cmd->cmnd[0] == WRITE_10 ||
cmd->cmnd[0] == WRITE_12)) {
/*
* Logical drive number increases by 0x80 when
* a logical drive is deleted
*/
adapter->wr_errors[logdrv%0x80]++;
}
}
#endif
}
/*
* Do not return the presence of hard disk on the channel so,
* inquiry sent, and returned data==hard disk or removable
* hard disk and not logical, request should return failure! -
* PJ
*/
islogical = adapter->logdrv_chan[cmd->device->channel];
if( cmd->cmnd[0] == INQUIRY && !islogical ) {
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
sgl = scsi_sglist(cmd);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-23 03:19:53 +08:00
if( sg_page(sgl) ) {
c = *(unsigned char *) sg_virt(&sgl[0]);
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
} else {
printk(KERN_WARNING
"megaraid: invalid sg.\n");
c = 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!
2005-04-17 06:20:36 +08:00
}
if(IS_RAID_CH(adapter, cmd->device->channel) &&
((c & 0x1F ) == TYPE_DISK)) {
status = 0xF0;
}
}
/* clear result; otherwise, success returns corrupt value */
cmd->result = 0;
/* Convert MegaRAID status to Linux error code */
switch (status) {
case 0x00: /* SUCCESS , i.e. SCSI_STATUS_GOOD */
cmd->result |= (DID_OK << 16);
break;
case 0x02: /* ERROR_ABORTED, i.e.
SCSI_STATUS_CHECK_CONDITION */
/* set sense_buffer and result fields */
if( mbox->m_out.cmd == MEGA_MBOXCMD_PASSTHRU ||
mbox->m_out.cmd == MEGA_MBOXCMD_PASSTHRU64 ) {
memcpy(cmd->sense_buffer, pthru->reqsensearea,
14);
cmd->result = (DRIVER_SENSE << 24) |
(DID_OK << 16) |
(CHECK_CONDITION << 1);
}
else {
if (mbox->m_out.cmd == MEGA_MBOXCMD_EXTPTHRU) {
memcpy(cmd->sense_buffer,
epthru->reqsensearea, 14);
cmd->result = (DRIVER_SENSE << 24) |
(DID_OK << 16) |
(CHECK_CONDITION << 1);
} else {
cmd->sense_buffer[0] = 0x70;
cmd->sense_buffer[2] = ABORTED_COMMAND;
cmd->result |= (CHECK_CONDITION << 1);
}
}
break;
case 0x08: /* ERR_DEST_DRIVE_FAILED, i.e.
SCSI_STATUS_BUSY */
cmd->result |= (DID_BUS_BUSY << 16) | status;
break;
default:
#if MEGA_HAVE_CLUSTERING
/*
* If TEST_UNIT_READY fails, we know
* MEGA_RESERVATION_STATUS failed
*/
if( cmd->cmnd[0] == TEST_UNIT_READY ) {
cmd->result |= (DID_ERROR << 16) |
(RESERVATION_CONFLICT << 1);
}
else
/*
* Error code returned is 1 if Reserve or Release
* failed or the input parameter is invalid
*/
if( status == 1 &&
(cmd->cmnd[0] == RESERVE ||
cmd->cmnd[0] == RELEASE) ) {
cmd->result |= (DID_ERROR << 16) |
(RESERVATION_CONFLICT << 1);
}
else
#endif
cmd->result |= (DID_BAD_TARGET << 16)|status;
}
/*
* Only free SCBs for the commands coming down from the
* mid-layer, not for which were issued internally
*
* For internal command, restore the status returned by the
* firmware so that user can interpret it.
*/
if( cmdid == CMDID_INT_CMDS ) { /* internal command */
cmd->result = status;
/*
* Remove the internal command from the pending list
*/
list_del_init(&scb->list);
scb->state = SCB_FREE;
}
else {
mega_free_scb(adapter, scb);
}
/* Add Scsi_Command to end of completed queue */
list_add_tail(SCSI_LIST(cmd), &adapter->completed_list);
}
}
/*
* mega_runpendq()
*
* Run through the list of completed requests and finish it
*/
static void
mega_rundoneq (adapter_t *adapter)
{
Scsi_Cmnd *cmd;
struct list_head *pos;
list_for_each(pos, &adapter->completed_list) {
[SCSI] remove Scsi_Pointer typedef Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 01:31:56 +08:00
struct scsi_pointer* spos = (struct scsi_pointer *)pos;
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!
2005-04-17 06:20:36 +08:00
cmd = list_entry(spos, Scsi_Cmnd, SCp);
cmd->scsi_done(cmd);
}
INIT_LIST_HEAD(&adapter->completed_list);
}
/*
* Free a SCB structure
* Note: We assume the scsi commands associated with this scb is not free yet.
*/
static void
mega_free_scb(adapter_t *adapter, scb_t *scb)
{
switch( scb->dma_type ) {
case MEGA_DMA_TYPE_NONE:
break;
case MEGA_SGLIST:
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
scsi_dma_unmap(scb->cmd);
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!
2005-04-17 06:20:36 +08:00
break;
default:
break;
}
/*
* Remove from the pending list
*/
list_del_init(&scb->list);
/* Link the scb back into free list */
scb->state = SCB_FREE;
scb->cmd = NULL;
list_add(&scb->list, &adapter->free_list);
}
static int
__mega_busywait_mbox (adapter_t *adapter)
{
volatile mbox_t *mbox = adapter->mbox;
long counter;
for (counter = 0; counter < 10000; counter++) {
if (!mbox->m_in.busy)
return 0;
[SCSI] megaraid: replace yield() with cond_resched() For this driver cond_resched() seems to be a better alternative Signed-off-by: Amol Lad <amol@verismonetworks.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:35:13 +08:00
udelay(100);
cond_resched();
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!
2005-04-17 06:20:36 +08:00
}
return -1; /* give up after 1 second */
}
/*
* Copies data to SGLIST
* Note: For 64 bit cards, we need a minimum of one SG element for read/write
*/
static int
mega_build_sglist(adapter_t *adapter, scb_t *scb, u32 *buf, u32 *len)
{
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
struct scatterlist *sg;
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!
2005-04-17 06:20:36 +08:00
Scsi_Cmnd *cmd;
int sgcnt;
int idx;
cmd = scb->cmd;
/*
* Copy Scatter-Gather list info into controller structure.
*
* The number of sg elements returned must not exceed our limit
*/
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
sgcnt = scsi_dma_map(cmd);
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!
2005-04-17 06:20:36 +08:00
scb->dma_type = MEGA_SGLIST;
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
BUG_ON(sgcnt > adapter->sglen || sgcnt < 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!
2005-04-17 06:20:36 +08:00
[SCSI] Legacy MegaRAID: Fix READ CAPACITY Some Legacy megaraid cards can't actually cope with the scatter/gather version of the READ CAPACITY command (which is what we now send them since altering all SCSI internal I/O to go via the block layer). Fix this (and a few other broken megaraid driver assumptions) by sending the non-sg version of the command if the sg list only has a single element. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-10-01 22:38:05 +08:00
*len = 0;
[SCSI] megaraid_old: fix READ_CAPACITY The bulk transfer mode got eleminated by 3f6270ef76f2ce5c134615a470685d6c2a66c07e. Unfortunately, this mode is required for READ_CAPACITY commands on certain cards, so put it back again. This fixes a boot failure regression reported by Burton Windle. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-10-03 08:00:58 +08:00
if (scsi_sg_count(cmd) == 1 && !adapter->has_64bit_addr) {
sg = scsi_sglist(cmd);
scb->dma_h_bulkdata = sg_dma_address(sg);
*buf = (u32)scb->dma_h_bulkdata;
*len = sg_dma_len(sg);
return 0;
}
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
scsi_for_each_sg(cmd, sg, sgcnt, idx) {
if (adapter->has_64bit_addr) {
scb->sgl64[idx].address = sg_dma_address(sg);
*len += scb->sgl64[idx].length = sg_dma_len(sg);
} else {
scb->sgl[idx].address = sg_dma_address(sg);
*len += scb->sgl[idx].length = sg_dma_len(sg);
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!
2005-04-17 06:20:36 +08:00
}
}
/* Reset pointer and length fields */
*buf = scb->sgl_dma_addr;
/* Return count of SG requests */
return sgcnt;
}
/*
* mega_8_to_40ld()
*
* takes all info in AdapterInquiry structure and puts it into ProductInfo and
* Enquiry3 structures for later use
*/
static void
mega_8_to_40ld(mraid_inquiry *inquiry, mega_inquiry3 *enquiry3,
mega_product_info *product_info)
{
int i;
product_info->max_commands = inquiry->adapter_info.max_commands;
enquiry3->rebuild_rate = inquiry->adapter_info.rebuild_rate;
product_info->nchannels = inquiry->adapter_info.nchannels;
for (i = 0; i < 4; i++) {
product_info->fw_version[i] =
inquiry->adapter_info.fw_version[i];
product_info->bios_version[i] =
inquiry->adapter_info.bios_version[i];
}
enquiry3->cache_flush_interval =
inquiry->adapter_info.cache_flush_interval;
product_info->dram_size = inquiry->adapter_info.dram_size;
enquiry3->num_ldrv = inquiry->logdrv_info.num_ldrv;
for (i = 0; i < MAX_LOGICAL_DRIVES_8LD; i++) {
enquiry3->ldrv_size[i] = inquiry->logdrv_info.ldrv_size[i];
enquiry3->ldrv_prop[i] = inquiry->logdrv_info.ldrv_prop[i];
enquiry3->ldrv_state[i] = inquiry->logdrv_info.ldrv_state[i];
}
for (i = 0; i < (MAX_PHYSICAL_DRIVES); i++)
enquiry3->pdrv_state[i] = inquiry->pdrv_info.pdrv_state[i];
}
static inline void
mega_free_sgl(adapter_t *adapter)
{
scb_t *scb;
int i;
for(i = 0; i < adapter->max_cmds; i++) {
scb = &adapter->scb_list[i];
if( scb->sgl64 ) {
pci_free_consistent(adapter->dev,
sizeof(mega_sgl64) * adapter->sglen,
scb->sgl64,
scb->sgl_dma_addr);
scb->sgl64 = NULL;
}
if( scb->pthru ) {
pci_free_consistent(adapter->dev, sizeof(mega_passthru),
scb->pthru, scb->pthru_dma_addr);
scb->pthru = NULL;
}
if( scb->epthru ) {
pci_free_consistent(adapter->dev,
sizeof(mega_ext_passthru),
scb->epthru, scb->epthru_dma_addr);
scb->epthru = NULL;
}
}
}
/*
* Get information about the card/driver
*/
const char *
megaraid_info(struct Scsi_Host *host)
{
static char buffer[512];
adapter_t *adapter;
adapter = (adapter_t *)host->hostdata;
sprintf (buffer,
"LSI Logic MegaRAID %s %d commands %d targs %d chans %d luns",
adapter->fw_version, adapter->product_info.max_commands,
adapter->host->max_id, adapter->host->max_channel,
adapter->host->max_lun);
return buffer;
}
/*
* Abort a previous SCSI request. Only commands on the pending list can be
* aborted. All the commands issued to the F/W must complete.
*/
static int
megaraid_abort(Scsi_Cmnd *cmd)
{
adapter_t *adapter;
int rval;
adapter = (adapter_t *)cmd->device->host->hostdata;
rval = megaraid_abort_and_reset(adapter, cmd, SCB_ABORT);
/*
* This is required here to complete any completed requests
* to be communicated over to the mid layer.
*/
mega_rundoneq(adapter);
return rval;
}
static int
[SCSI] megaraid: fix compilation after eh locking changes From: Christoph Hellwig <hch@lst.de> Patch fixed up and Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-26 21:45:39 +08:00
megaraid_reset(struct scsi_cmnd *cmd)
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!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter;
megacmd_t mc;
int rval;
adapter = (adapter_t *)cmd->device->host->hostdata;
#if MEGA_HAVE_CLUSTERING
mc.cmd = MEGA_CLUSTER_CMD;
mc.opcode = MEGA_RESET_RESERVATIONS;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
if( mega_internal_command(adapter, &mc, NULL) != 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!
2005-04-17 06:20:36 +08:00
printk(KERN_WARNING
"megaraid: reservation reset failed.\n");
}
else {
printk(KERN_INFO "megaraid: reservation reset.\n");
}
#endif
[SCSI] megaraid: fix compilation after eh locking changes From: Christoph Hellwig <hch@lst.de> Patch fixed up and Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-26 21:45:39 +08:00
spin_lock_irq(&adapter->lock);
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!
2005-04-17 06:20:36 +08:00
rval = megaraid_abort_and_reset(adapter, cmd, SCB_RESET);
/*
* This is required here to complete any completed requests
* to be communicated over to the mid layer.
*/
mega_rundoneq(adapter);
[SCSI] allow sleeping in ->eh_device_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 19:55:48 +08:00
spin_unlock_irq(&adapter->lock);
[SCSI] megaraid: fix compilation after eh locking changes From: Christoph Hellwig <hch@lst.de> Patch fixed up and Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-26 21:45:39 +08:00
return rval;
[SCSI] allow sleeping in ->eh_device_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 19:55:48 +08:00
}
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!
2005-04-17 06:20:36 +08:00
/**
* megaraid_abort_and_reset()
* @adapter - megaraid soft state
* @cmd - scsi command to be aborted or reset
* @aor - abort or reset flag
*
* Try to locate the scsi command in the pending queue. If found and is not
* issued to the controller, abort/reset it. Otherwise return failure
*/
static int
megaraid_abort_and_reset(adapter_t *adapter, Scsi_Cmnd *cmd, int aor)
{
struct list_head *pos, *next;
scb_t *scb;
printk(KERN_WARNING "megaraid: %s-%lx cmd=%x <c=%d t=%d l=%d>\n",
(aor == SCB_ABORT)? "ABORTING":"RESET", cmd->serial_number,
cmd->cmnd[0], cmd->device->channel,
cmd->device->id, cmd->device->lun);
if(list_empty(&adapter->pending_list))
return FALSE;
list_for_each_safe(pos, next, &adapter->pending_list) {
scb = list_entry(pos, scb_t, list);
if (scb->cmd == cmd) { /* Found command */
scb->state |= aor;
/*
trivial: Fix misspelling of "firmware" in megaraid.c Fix misspelling of "firmware" in megaraid.c Fixed "firmware", "ownership" and grammar in the same comment. Signed-off-by: Nick Andrew <nick@nick-andrew.net> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2009-01-03 15:55:39 +08:00
* Check if this command has firmware ownership. If
* yes, we cannot reset this command. Whenever f/w
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!
2005-04-17 06:20:36 +08:00
* completes this command, we will return appropriate
* status from ISR.
*/
if( scb->state & SCB_ISSUED ) {
printk(KERN_WARNING
"megaraid: %s-%lx[%x], fw owner.\n",
(aor==SCB_ABORT) ? "ABORTING":"RESET",
cmd->serial_number, scb->idx);
return FALSE;
}
else {
/*
* Not yet issued! Remove from the pending
* list
*/
printk(KERN_WARNING
"megaraid: %s-%lx[%x], driver owner.\n",
(aor==SCB_ABORT) ? "ABORTING":"RESET",
cmd->serial_number, scb->idx);
mega_free_scb(adapter, scb);
if( aor == SCB_ABORT ) {
cmd->result = (DID_ABORT << 16);
}
else {
cmd->result = (DID_RESET << 16);
}
list_add_tail(SCSI_LIST(cmd),
&adapter->completed_list);
return TRUE;
}
}
}
return FALSE;
}
static inline int
make_local_pdev(adapter_t *adapter, struct pci_dev **pdev)
{
PCI: Convert to alloc_pci_dev() Convert code that allocs a struct pci_dev to use alloc_pci_dev(). Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-04-05 15:19:09 +08:00
*pdev = alloc_pci_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!
2005-04-17 06:20:36 +08:00
if( *pdev == NULL ) return -1;
memcpy(*pdev, adapter->dev, sizeof(struct pci_dev));
dma-mapping: replace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32) Replace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32) Signed-off-by: Yang Hongyang<yanghy@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-07 10:01:15 +08:00
if( pci_set_dma_mask(*pdev, DMA_BIT_MASK(32)) != 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!
2005-04-17 06:20:36 +08:00
kfree(*pdev);
return -1;
}
return 0;
}
static inline void
free_local_pdev(struct pci_dev *pdev)
{
kfree(pdev);
}
/**
* mega_allocate_inquiry()
* @dma_handle - handle returned for dma address
* @pdev - handle to pci device
*
* allocates memory for inquiry structure
*/
static inline void *
mega_allocate_inquiry(dma_addr_t *dma_handle, struct pci_dev *pdev)
{
return pci_alloc_consistent(pdev, sizeof(mega_inquiry3), dma_handle);
}
static inline void
mega_free_inquiry(void *inquiry, dma_addr_t dma_handle, struct pci_dev *pdev)
{
pci_free_consistent(pdev, sizeof(mega_inquiry3), inquiry, dma_handle);
}
#ifdef CONFIG_PROC_FS
/* Following code handles /proc fs */
#define CREATE_READ_PROC(string, func) create_proc_read_entry(string, \
S_IRUSR | S_IFREG, \
controller_proc_dir_entry, \
func, adapter)
/**
* mega_create_proc_entry()
* @index - index in soft state array
* @parent - parent node for this /proc entry
*
* Creates /proc entries for our controllers.
*/
static void
mega_create_proc_entry(int index, struct proc_dir_entry *parent)
{
struct proc_dir_entry *controller_proc_dir_entry = NULL;
u8 string[64] = { 0 };
adapter_t *adapter = hba_soft_state[index];
sprintf(string, "hba%d", adapter->host->host_no);
controller_proc_dir_entry =
adapter->controller_proc_dir_entry = proc_mkdir(string, parent);
if(!controller_proc_dir_entry) {
printk(KERN_WARNING "\nmegaraid: proc_mkdir failed\n");
return;
}
adapter->proc_read = CREATE_READ_PROC("config", proc_read_config);
adapter->proc_stat = CREATE_READ_PROC("stat", proc_read_stat);
adapter->proc_mbox = CREATE_READ_PROC("mailbox", proc_read_mbox);
#if MEGA_HAVE_ENH_PROC
adapter->proc_rr = CREATE_READ_PROC("rebuild-rate", proc_rebuild_rate);
adapter->proc_battery = CREATE_READ_PROC("battery-status",
proc_battery);
/*
* Display each physical drive on its channel
*/
adapter->proc_pdrvstat[0] = CREATE_READ_PROC("diskdrives-ch0",
proc_pdrv_ch0);
adapter->proc_pdrvstat[1] = CREATE_READ_PROC("diskdrives-ch1",
proc_pdrv_ch1);
adapter->proc_pdrvstat[2] = CREATE_READ_PROC("diskdrives-ch2",
proc_pdrv_ch2);
adapter->proc_pdrvstat[3] = CREATE_READ_PROC("diskdrives-ch3",
proc_pdrv_ch3);
/*
* Display a set of up to 10 logical drive through each of following
* /proc entries
*/
adapter->proc_rdrvstat[0] = CREATE_READ_PROC("raiddrives-0-9",
proc_rdrv_10);
adapter->proc_rdrvstat[1] = CREATE_READ_PROC("raiddrives-10-19",
proc_rdrv_20);
adapter->proc_rdrvstat[2] = CREATE_READ_PROC("raiddrives-20-29",
proc_rdrv_30);
adapter->proc_rdrvstat[3] = CREATE_READ_PROC("raiddrives-30-39",
proc_rdrv_40);
#endif
}
/**
* proc_read_config()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display configuration information about the controller.
*/
static int
proc_read_config(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
int len = 0;
len += sprintf(page+len, "%s", MEGARAID_VERSION);
if(adapter->product_info.product_name[0])
len += sprintf(page+len, "%s\n",
adapter->product_info.product_name);
len += sprintf(page+len, "Controller Type: ");
if( adapter->flag & BOARD_MEMMAP ) {
len += sprintf(page+len,
"438/466/467/471/493/518/520/531/532\n");
}
else {
len += sprintf(page+len,
"418/428/434\n");
}
if(adapter->flag & BOARD_40LD) {
len += sprintf(page+len,
"Controller Supports 40 Logical Drives\n");
}
if(adapter->flag & BOARD_64BIT) {
len += sprintf(page+len,
"Controller capable of 64-bit memory addressing\n");
}
if( adapter->has_64bit_addr ) {
len += sprintf(page+len,
"Controller using 64-bit memory addressing\n");
}
else {
len += sprintf(page+len,
"Controller is not using 64-bit memory addressing\n");
}
len += sprintf(page+len, "Base = %08lx, Irq = %d, ", adapter->base,
adapter->host->irq);
len += sprintf(page+len, "Logical Drives = %d, Channels = %d\n",
adapter->numldrv, adapter->product_info.nchannels);
len += sprintf(page+len, "Version =%s:%s, DRAM = %dMb\n",
adapter->fw_version, adapter->bios_version,
adapter->product_info.dram_size);
len += sprintf(page+len,
"Controller Queue Depth = %d, Driver Queue Depth = %d\n",
adapter->product_info.max_commands, adapter->max_cmds);
len += sprintf(page+len, "support_ext_cdb = %d\n",
adapter->support_ext_cdb);
len += sprintf(page+len, "support_random_del = %d\n",
adapter->support_random_del);
len += sprintf(page+len, "boot_ldrv_enabled = %d\n",
adapter->boot_ldrv_enabled);
len += sprintf(page+len, "boot_ldrv = %d\n",
adapter->boot_ldrv);
len += sprintf(page+len, "boot_pdrv_enabled = %d\n",
adapter->boot_pdrv_enabled);
len += sprintf(page+len, "boot_pdrv_ch = %d\n",
adapter->boot_pdrv_ch);
len += sprintf(page+len, "boot_pdrv_tgt = %d\n",
adapter->boot_pdrv_tgt);
len += sprintf(page+len, "quiescent = %d\n",
atomic_read(&adapter->quiescent));
len += sprintf(page+len, "has_cluster = %d\n",
adapter->has_cluster);
len += sprintf(page+len, "\nModule Parameters:\n");
len += sprintf(page+len, "max_cmd_per_lun = %d\n",
max_cmd_per_lun);
len += sprintf(page+len, "max_sectors_per_io = %d\n",
max_sectors_per_io);
*eof = 1;
return len;
}
/**
* proc_read_stat()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Diaplay statistical information about the I/O activity.
*/
static int
proc_read_stat(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter;
int len;
int i;
i = 0; /* avoid compilation warnings */
len = 0;
adapter = (adapter_t *)data;
len = sprintf(page, "Statistical Information for this controller\n");
len += sprintf(page+len, "pend_cmds = %d\n",
atomic_read(&adapter->pend_cmds));
#if MEGA_HAVE_STATS
for(i = 0; i < adapter->numldrv; i++) {
len += sprintf(page+len, "Logical Drive %d:\n", i);
len += sprintf(page+len,
"\tReads Issued = %lu, Writes Issued = %lu\n",
adapter->nreads[i], adapter->nwrites[i]);
len += sprintf(page+len,
"\tSectors Read = %lu, Sectors Written = %lu\n",
adapter->nreadblocks[i], adapter->nwriteblocks[i]);
len += sprintf(page+len,
"\tRead errors = %lu, Write errors = %lu\n\n",
adapter->rd_errors[i], adapter->wr_errors[i]);
}
#else
len += sprintf(page+len,
"IO and error counters not compiled in driver.\n");
#endif
*eof = 1;
return len;
}
/**
* proc_read_mbox()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display mailbox information for the last command issued. This information
* is good for debugging.
*/
static int
proc_read_mbox(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
volatile mbox_t *mbox = adapter->mbox;
int len = 0;
len = sprintf(page, "Contents of Mail Box Structure\n");
len += sprintf(page+len, " Fw Command = 0x%02x\n",
mbox->m_out.cmd);
len += sprintf(page+len, " Cmd Sequence = 0x%02x\n",
mbox->m_out.cmdid);
len += sprintf(page+len, " No of Sectors= %04d\n",
mbox->m_out.numsectors);
len += sprintf(page+len, " LBA = 0x%02x\n",
mbox->m_out.lba);
len += sprintf(page+len, " DTA = 0x%08x\n",
mbox->m_out.xferaddr);
len += sprintf(page+len, " Logical Drive= 0x%02x\n",
mbox->m_out.logdrv);
len += sprintf(page+len, " No of SG Elmt= 0x%02x\n",
mbox->m_out.numsgelements);
len += sprintf(page+len, " Busy = %01x\n",
mbox->m_in.busy);
len += sprintf(page+len, " Status = 0x%02x\n",
mbox->m_in.status);
*eof = 1;
return len;
}
/**
* proc_rebuild_rate()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display current rebuild rate
*/
static int
proc_rebuild_rate(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
dma_addr_t dma_handle;
caddr_t inquiry;
struct pci_dev *pdev;
int len = 0;
if( make_local_pdev(adapter, &pdev) != 0 ) {
*eof = 1;
return len;
}
if( (inquiry = mega_allocate_inquiry(&dma_handle, pdev)) == NULL ) {
free_local_pdev(pdev);
*eof = 1;
return len;
}
if( mega_adapinq(adapter, dma_handle) != 0 ) {
len = sprintf(page, "Adapter inquiry failed.\n");
printk(KERN_WARNING "megaraid: inquiry failed.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
*eof = 1;
return len;
}
if( adapter->flag & BOARD_40LD ) {
len = sprintf(page, "Rebuild Rate: [%d%%]\n",
((mega_inquiry3 *)inquiry)->rebuild_rate);
}
else {
len = sprintf(page, "Rebuild Rate: [%d%%]\n",
((mraid_ext_inquiry *)
inquiry)->raid_inq.adapter_info.rebuild_rate);
}
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
*eof = 1;
return len;
}
/**
* proc_battery()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display information about the battery module on the controller.
*/
static int
proc_battery(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
dma_addr_t dma_handle;
caddr_t inquiry;
struct pci_dev *pdev;
u8 battery_status = 0;
char str[256];
int len = 0;
if( make_local_pdev(adapter, &pdev) != 0 ) {
*eof = 1;
return len;
}
if( (inquiry = mega_allocate_inquiry(&dma_handle, pdev)) == NULL ) {
free_local_pdev(pdev);
*eof = 1;
return len;
}
if( mega_adapinq(adapter, dma_handle) != 0 ) {
len = sprintf(page, "Adapter inquiry failed.\n");
printk(KERN_WARNING "megaraid: inquiry failed.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
*eof = 1;
return len;
}
if( adapter->flag & BOARD_40LD ) {
battery_status = ((mega_inquiry3 *)inquiry)->battery_status;
}
else {
battery_status = ((mraid_ext_inquiry *)inquiry)->
raid_inq.adapter_info.battery_status;
}
/*
* Decode the battery status
*/
sprintf(str, "Battery Status:[%d]", battery_status);
if(battery_status == MEGA_BATT_CHARGE_DONE)
strcat(str, " Charge Done");
if(battery_status & MEGA_BATT_MODULE_MISSING)
strcat(str, " Module Missing");
if(battery_status & MEGA_BATT_LOW_VOLTAGE)
strcat(str, " Low Voltage");
if(battery_status & MEGA_BATT_TEMP_HIGH)
strcat(str, " Temperature High");
if(battery_status & MEGA_BATT_PACK_MISSING)
strcat(str, " Pack Missing");
if(battery_status & MEGA_BATT_CHARGE_INPROG)
strcat(str, " Charge In-progress");
if(battery_status & MEGA_BATT_CHARGE_FAIL)
strcat(str, " Charge Fail");
if(battery_status & MEGA_BATT_CYCLES_EXCEEDED)
strcat(str, " Cycles Exceeded");
len = sprintf(page, "%s\n", str);
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
*eof = 1;
return len;
}
/**
* proc_pdrv_ch0()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display information about the physical drives on physical channel 0.
*/
static int
proc_pdrv_ch0(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_pdrv(adapter, page, 0));
}
/**
* proc_pdrv_ch1()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display information about the physical drives on physical channel 1.
*/
static int
proc_pdrv_ch1(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_pdrv(adapter, page, 1));
}
/**
* proc_pdrv_ch2()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display information about the physical drives on physical channel 2.
*/
static int
proc_pdrv_ch2(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_pdrv(adapter, page, 2));
}
/**
* proc_pdrv_ch3()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display information about the physical drives on physical channel 3.
*/
static int
proc_pdrv_ch3(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_pdrv(adapter, page, 3));
}
/**
* proc_pdrv()
* @page - buffer to write the data in
* @adapter - pointer to our soft state
*
* Display information about the physical drives.
*/
static int
proc_pdrv(adapter_t *adapter, char *page, int channel)
{
dma_addr_t dma_handle;
char *scsi_inq;
dma_addr_t scsi_inq_dma_handle;
caddr_t inquiry;
struct pci_dev *pdev;
u8 *pdrv_state;
u8 state;
int tgt;
int max_channels;
int len = 0;
char str[80];
int i;
if( make_local_pdev(adapter, &pdev) != 0 ) {
return len;
}
if( (inquiry = mega_allocate_inquiry(&dma_handle, pdev)) == NULL ) {
goto free_pdev;
}
if( mega_adapinq(adapter, dma_handle) != 0 ) {
len = sprintf(page, "Adapter inquiry failed.\n");
printk(KERN_WARNING "megaraid: inquiry failed.\n");
goto free_inquiry;
}
scsi_inq = pci_alloc_consistent(pdev, 256, &scsi_inq_dma_handle);
if( scsi_inq == NULL ) {
len = sprintf(page, "memory not available for scsi inq.\n");
goto free_inquiry;
}
if( adapter->flag & BOARD_40LD ) {
pdrv_state = ((mega_inquiry3 *)inquiry)->pdrv_state;
}
else {
pdrv_state = ((mraid_ext_inquiry *)inquiry)->
raid_inq.pdrv_info.pdrv_state;
}
max_channels = adapter->product_info.nchannels;
if( channel >= max_channels ) {
goto free_pci;
}
for( tgt = 0; tgt <= MAX_TARGET; tgt++ ) {
i = channel*16 + tgt;
state = *(pdrv_state + i);
switch( state & 0x0F ) {
case PDRV_ONLINE:
sprintf(str,
"Channel:%2d Id:%2d State: Online",
channel, tgt);
break;
case PDRV_FAILED:
sprintf(str,
"Channel:%2d Id:%2d State: Failed",
channel, tgt);
break;
case PDRV_RBLD:
sprintf(str,
"Channel:%2d Id:%2d State: Rebuild",
channel, tgt);
break;
case PDRV_HOTSPARE:
sprintf(str,
"Channel:%2d Id:%2d State: Hot spare",
channel, tgt);
break;
default:
sprintf(str,
"Channel:%2d Id:%2d State: Un-configured",
channel, tgt);
break;
}
/*
* This interface displays inquiries for disk drives
* only. Inquries for logical drives and non-disk
* devices are available through /proc/scsi/scsi
*/
memset(scsi_inq, 0, 256);
if( mega_internal_dev_inquiry(adapter, channel, tgt,
scsi_inq_dma_handle) ||
(scsi_inq[0] & 0x1F) != TYPE_DISK ) {
continue;
}
/*
* Check for overflow. We print less than 240
* characters for inquiry
*/
if( (len + 240) >= PAGE_SIZE ) break;
len += sprintf(page+len, "%s.\n", str);
len += mega_print_inquiry(page+len, scsi_inq);
}
free_pci:
pci_free_consistent(pdev, 256, scsi_inq, scsi_inq_dma_handle);
free_inquiry:
mega_free_inquiry(inquiry, dma_handle, pdev);
free_pdev:
free_local_pdev(pdev);
return len;
}
/*
* Display scsi inquiry
*/
static int
mega_print_inquiry(char *page, char *scsi_inq)
{
int len = 0;
int i;
len = sprintf(page, " Vendor: ");
for( i = 8; i < 16; i++ ) {
len += sprintf(page+len, "%c", scsi_inq[i]);
}
len += sprintf(page+len, " Model: ");
for( i = 16; i < 32; i++ ) {
len += sprintf(page+len, "%c", scsi_inq[i]);
}
len += sprintf(page+len, " Rev: ");
for( i = 32; i < 36; i++ ) {
len += sprintf(page+len, "%c", scsi_inq[i]);
}
len += sprintf(page+len, "\n");
i = scsi_inq[0] & 0x1f;
[SCSI] Improve inquiry printing - Replace scsi_device_types array API with scsi_device_type function API. Gets rid of a lot of common code, as well as being easier to use. - Add the new device types in SPC4 r05a, and rename some of the older ones. - Reformat the printing of inquiry data; now fits on one line and includes PQ. I think I've addressed all the feedback from the previous versions. My current test box prints: scsi 2:0:1:0: Direct access HP 18.2G ATLAS10K3_18_SCA HP05 PQ: 0 ANSI: 2 Signed-off-by: Matthew Wilcox <matthew@wil.cx> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-05 02:15:20 +08:00
len += sprintf(page+len, " Type: %s ", scsi_device_type(i));
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!
2005-04-17 06:20:36 +08:00
len += sprintf(page+len,
" ANSI SCSI revision: %02x", scsi_inq[2] & 0x07);
if( (scsi_inq[2] & 0x07) == 1 && (scsi_inq[3] & 0x0f) == 1 )
len += sprintf(page+len, " CCS\n");
else
len += sprintf(page+len, "\n");
return len;
}
/**
* proc_rdrv_10()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display real time information about the logical drives 0 through 9.
*/
static int
proc_rdrv_10(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_rdrv(adapter, page, 0, 9));
}
/**
* proc_rdrv_20()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display real time information about the logical drives 0 through 9.
*/
static int
proc_rdrv_20(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_rdrv(adapter, page, 10, 19));
}
/**
* proc_rdrv_30()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display real time information about the logical drives 0 through 9.
*/
static int
proc_rdrv_30(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_rdrv(adapter, page, 20, 29));
}
/**
* proc_rdrv_40()
* @page - buffer to write the data in
* @start - where the actual data has been written in page
* @offset - same meaning as the read system call
* @count - same meaning as the read system call
* @eof - set if no more data needs to be returned
* @data - pointer to our soft state
*
* Display real time information about the logical drives 0 through 9.
*/
static int
proc_rdrv_40(char *page, char **start, off_t offset, int count, int *eof,
void *data)
{
adapter_t *adapter = (adapter_t *)data;
*eof = 1;
return (proc_rdrv(adapter, page, 30, 39));
}
/**
* proc_rdrv()
* @page - buffer to write the data in
* @adapter - pointer to our soft state
* @start - starting logical drive to display
* @end - ending logical drive to display
*
* We do not print the inquiry information since its already available through
* /proc/scsi/scsi interface
*/
static int
proc_rdrv(adapter_t *adapter, char *page, int start, int end )
{
dma_addr_t dma_handle;
logdrv_param *lparam;
megacmd_t mc;
char *disk_array;
dma_addr_t disk_array_dma_handle;
caddr_t inquiry;
struct pci_dev *pdev;
u8 *rdrv_state;
int num_ldrv;
u32 array_sz;
int len = 0;
int i;
if( make_local_pdev(adapter, &pdev) != 0 ) {
return len;
}
if( (inquiry = mega_allocate_inquiry(&dma_handle, pdev)) == NULL ) {
free_local_pdev(pdev);
return len;
}
if( mega_adapinq(adapter, dma_handle) != 0 ) {
len = sprintf(page, "Adapter inquiry failed.\n");
printk(KERN_WARNING "megaraid: inquiry failed.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
return len;
}
memset(&mc, 0, sizeof(megacmd_t));
if( adapter->flag & BOARD_40LD ) {
array_sz = sizeof(disk_array_40ld);
rdrv_state = ((mega_inquiry3 *)inquiry)->ldrv_state;
num_ldrv = ((mega_inquiry3 *)inquiry)->num_ldrv;
}
else {
array_sz = sizeof(disk_array_8ld);
rdrv_state = ((mraid_ext_inquiry *)inquiry)->
raid_inq.logdrv_info.ldrv_state;
num_ldrv = ((mraid_ext_inquiry *)inquiry)->
raid_inq.logdrv_info.num_ldrv;
}
disk_array = pci_alloc_consistent(pdev, array_sz,
&disk_array_dma_handle);
if( disk_array == NULL ) {
len = sprintf(page, "memory not available.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
free_local_pdev(pdev);
return len;
}
mc.xferaddr = (u32)disk_array_dma_handle;
if( adapter->flag & BOARD_40LD ) {
mc.cmd = FC_NEW_CONFIG;
mc.opcode = OP_DCMD_READ_CONFIG;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
if( mega_internal_command(adapter, &mc, NULL) ) {
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!
2005-04-17 06:20:36 +08:00
len = sprintf(page, "40LD read config failed.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
pci_free_consistent(pdev, array_sz, disk_array,
disk_array_dma_handle);
free_local_pdev(pdev);
return len;
}
}
else {
mc.cmd = NEW_READ_CONFIG_8LD;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
if( mega_internal_command(adapter, &mc, NULL) ) {
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!
2005-04-17 06:20:36 +08:00
mc.cmd = READ_CONFIG_8LD;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
if( mega_internal_command(adapter, &mc,
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!
2005-04-17 06:20:36 +08:00
NULL) ){
len = sprintf(page,
"8LD read config failed.\n");
mega_free_inquiry(inquiry, dma_handle, pdev);
pci_free_consistent(pdev, array_sz,
disk_array,
disk_array_dma_handle);
free_local_pdev(pdev);
return len;
}
}
}
for( i = start; i < ( (end+1 < num_ldrv) ? end+1 : num_ldrv ); i++ ) {
if( adapter->flag & BOARD_40LD ) {
lparam =
&((disk_array_40ld *)disk_array)->ldrv[i].lparam;
}
else {
lparam =
&((disk_array_8ld *)disk_array)->ldrv[i].lparam;
}
/*
* Check for overflow. We print less than 240 characters for
* information about each logical drive.
*/
if( (len + 240) >= PAGE_SIZE ) break;
len += sprintf(page+len, "Logical drive:%2d:, ", i);
switch( rdrv_state[i] & 0x0F ) {
case RDRV_OFFLINE:
len += sprintf(page+len, "state: offline");
break;
case RDRV_DEGRADED:
len += sprintf(page+len, "state: degraded");
break;
case RDRV_OPTIMAL:
len += sprintf(page+len, "state: optimal");
break;
case RDRV_DELETED:
len += sprintf(page+len, "state: deleted");
break;
default:
len += sprintf(page+len, "state: unknown");
break;
}
/*
* Check if check consistency or initialization is going on
* for this logical drive.
*/
if( (rdrv_state[i] & 0xF0) == 0x20 ) {
len += sprintf(page+len,
", check-consistency in progress");
}
else if( (rdrv_state[i] & 0xF0) == 0x10 ) {
len += sprintf(page+len,
", initialization in progress");
}
len += sprintf(page+len, "\n");
len += sprintf(page+len, "Span depth:%3d, ",
lparam->span_depth);
len += sprintf(page+len, "RAID level:%3d, ",
lparam->level);
len += sprintf(page+len, "Stripe size:%3d, ",
lparam->stripe_sz ? lparam->stripe_sz/2: 128);
len += sprintf(page+len, "Row size:%3d\n",
lparam->row_size);
len += sprintf(page+len, "Read Policy: ");
switch(lparam->read_ahead) {
case NO_READ_AHEAD:
len += sprintf(page+len, "No read ahead, ");
break;
case READ_AHEAD:
len += sprintf(page+len, "Read ahead, ");
break;
case ADAP_READ_AHEAD:
len += sprintf(page+len, "Adaptive, ");
break;
}
len += sprintf(page+len, "Write Policy: ");
switch(lparam->write_mode) {
case WRMODE_WRITE_THRU:
len += sprintf(page+len, "Write thru, ");
break;
case WRMODE_WRITE_BACK:
len += sprintf(page+len, "Write back, ");
break;
}
len += sprintf(page+len, "Cache Policy: ");
switch(lparam->direct_io) {
case CACHED_IO:
len += sprintf(page+len, "Cached IO\n\n");
break;
case DIRECT_IO:
len += sprintf(page+len, "Direct IO\n\n");
break;
}
}
mega_free_inquiry(inquiry, dma_handle, pdev);
pci_free_consistent(pdev, array_sz, disk_array,
disk_array_dma_handle);
free_local_pdev(pdev);
return len;
}
[SCSI] megaraid: fix warnings when CONFIG_PROC_FS=n drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:4893: warning: implicit declaration of function 'mega_create_proc_entry' drivers/scsi/megaraid.c: In function 'megaraid_remove_one': drivers/scsi/megaraid.c:4968: warning: unused variable 'buf' Fix by adding #defines Signed-off-by: walter harms <wharms@bfs.de> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:35:09 +08:00
#else
static inline void mega_create_proc_entry(int index, struct proc_dir_entry *parent)
{
}
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!
2005-04-17 06:20:36 +08:00
#endif
/**
* megaraid_biosparam()
*
* Return the disk geometry for a particular disk
*/
static int
megaraid_biosparam(struct scsi_device *sdev, struct block_device *bdev,
sector_t capacity, int geom[])
{
adapter_t *adapter;
unsigned char *bh;
int heads;
int sectors;
int cylinders;
int rval;
/* Get pointer to host config structure */
adapter = (adapter_t *)sdev->host->hostdata;
if (IS_RAID_CH(adapter, sdev->channel)) {
/* Default heads (64) & sectors (32) */
heads = 64;
sectors = 32;
cylinders = (ulong)capacity / (heads * sectors);
/*
* Handle extended translation size for logical drives
* > 1Gb
*/
if ((ulong)capacity >= 0x200000) {
heads = 255;
sectors = 63;
cylinders = (ulong)capacity / (heads * sectors);
}
/* return result */
geom[0] = heads;
geom[1] = sectors;
geom[2] = cylinders;
}
else {
bh = scsi_bios_ptable(bdev);
if( bh ) {
rval = scsi_partsize(bh, capacity,
&geom[2], &geom[0], &geom[1]);
kfree(bh);
if( rval != -1 )
return rval;
}
printk(KERN_INFO
"megaraid: invalid partition on this disk on channel %d\n",
sdev->channel);
/* Default heads (64) & sectors (32) */
heads = 64;
sectors = 32;
cylinders = (ulong)capacity / (heads * sectors);
/* Handle extended translation size for logical drives > 1Gb */
if ((ulong)capacity >= 0x200000) {
heads = 255;
sectors = 63;
cylinders = (ulong)capacity / (heads * sectors);
}
/* return result */
geom[0] = heads;
geom[1] = sectors;
geom[2] = cylinders;
}
return 0;
}
/**
* mega_init_scb()
* @adapter - pointer to our soft state
*
* Allocate memory for the various pointers in the scb structures:
* scatter-gather list pointer, passthru and extended passthru structure
* pointers.
*/
static int
mega_init_scb(adapter_t *adapter)
{
scb_t *scb;
int i;
for( i = 0; i < adapter->max_cmds; i++ ) {
scb = &adapter->scb_list[i];
scb->sgl64 = NULL;
scb->sgl = NULL;
scb->pthru = NULL;
scb->epthru = NULL;
}
for( i = 0; i < adapter->max_cmds; i++ ) {
scb = &adapter->scb_list[i];
scb->idx = i;
scb->sgl64 = pci_alloc_consistent(adapter->dev,
sizeof(mega_sgl64) * adapter->sglen,
&scb->sgl_dma_addr);
scb->sgl = (mega_sglist *)scb->sgl64;
if( !scb->sgl ) {
printk(KERN_WARNING "RAID: Can't allocate sglist.\n");
mega_free_sgl(adapter);
return -1;
}
scb->pthru = pci_alloc_consistent(adapter->dev,
sizeof(mega_passthru),
&scb->pthru_dma_addr);
if( !scb->pthru ) {
printk(KERN_WARNING "RAID: Can't allocate passthru.\n");
mega_free_sgl(adapter);
return -1;
}
scb->epthru = pci_alloc_consistent(adapter->dev,
sizeof(mega_ext_passthru),
&scb->epthru_dma_addr);
if( !scb->epthru ) {
printk(KERN_WARNING
"Can't allocate extended passthru.\n");
mega_free_sgl(adapter);
return -1;
}
scb->dma_type = MEGA_DMA_TYPE_NONE;
/*
* Link to free list
* lock not required since we are loading the driver, so no
* commands possible right now.
*/
scb->state = SCB_FREE;
scb->cmd = NULL;
list_add(&scb->list, &adapter->free_list);
}
return 0;
}
/**
* megadev_open()
* @inode - unused
* @filep - unused
*
* Routines for the character/ioctl interface to the driver. Find out if this
Add "no BKL needed" comments to several drivers This documents the fact that somebody looked at the relevant open() functions and concluded that, due to their trivial nature, no locking was needed. Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-05-17 03:40:30 +08:00
* is a valid open.
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!
2005-04-17 06:20:36 +08:00
*/
static int
megadev_open (struct inode *inode, struct file *filep)
{
/*
* Only allow superuser to access private ioctl interface
*/
if( !capable(CAP_SYS_ADMIN) ) return -EACCES;
return 0;
}
/**
* megadev_ioctl()
* @inode - Our device inode
* @filep - unused
* @cmd - ioctl command
* @arg - user buffer
*
* ioctl entry point for our private ioctl interface. We move the data in from
* the user space, prepare the command (if necessary, convert the old MIMD
* ioctl to new ioctl command), and issue a synchronous command to the
* controller.
*/
static int
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
megadev_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
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!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter;
nitioctl_t uioc;
int adapno;
int rval;
mega_passthru __user *upthru; /* user address for passthru */
mega_passthru *pthru; /* copy user passthru here */
dma_addr_t pthru_dma_hndl;
void *data = NULL; /* data to be transferred */
dma_addr_t data_dma_hndl; /* dma handle for data xfer area */
megacmd_t mc;
megastat_t __user *ustats;
int num_ldrv;
u32 uxferaddr = 0;
struct pci_dev *pdev;
ustats = NULL; /* avoid compilation warnings */
num_ldrv = 0;
/*
* Make sure only USCSICMD are issued through this interface.
* MIMD application would still fire different command.
*/
if( (_IOC_TYPE(cmd) != MEGAIOC_MAGIC) && (cmd != USCSICMD) ) {
return -EINVAL;
}
/*
* Check and convert a possible MIMD command to NIT command.
* mega_m_to_n() copies the data from the user space, so we do not
* have to do it here.
* NOTE: We will need some user address to copyout the data, therefore
* the inteface layer will also provide us with the required user
* addresses.
*/
memset(&uioc, 0, sizeof(nitioctl_t));
if( (rval = mega_m_to_n( (void __user *)arg, &uioc)) != 0 )
return rval;
switch( uioc.opcode ) {
case GET_DRIVER_VER:
if( put_user(driver_ver, (u32 __user *)uioc.uioc_uaddr) )
return (-EFAULT);
break;
case GET_N_ADAP:
if( put_user(hba_count, (u32 __user *)uioc.uioc_uaddr) )
return (-EFAULT);
/*
* Shucks. MIMD interface returns a positive value for number
* of adapters. TODO: Change it to return 0 when there is no
* applicatio using mimd interface.
*/
return hba_count;
case GET_ADAP_INFO:
/*
* Which adapter
*/
if( (adapno = GETADAP(uioc.adapno)) >= hba_count )
return (-ENODEV);
if( copy_to_user(uioc.uioc_uaddr, mcontroller+adapno,
sizeof(struct mcontroller)) )
return (-EFAULT);
break;
#if MEGA_HAVE_STATS
case GET_STATS:
/*
* Which adapter
*/
if( (adapno = GETADAP(uioc.adapno)) >= hba_count )
return (-ENODEV);
adapter = hba_soft_state[adapno];
ustats = uioc.uioc_uaddr;
if( copy_from_user(&num_ldrv, &ustats->num_ldrv, sizeof(int)) )
return (-EFAULT);
/*
* Check for the validity of the logical drive number
*/
if( num_ldrv >= MAX_LOGICAL_DRIVES_40LD ) return -EINVAL;
if( copy_to_user(ustats->nreads, adapter->nreads,
num_ldrv*sizeof(u32)) )
return -EFAULT;
if( copy_to_user(ustats->nreadblocks, adapter->nreadblocks,
num_ldrv*sizeof(u32)) )
return -EFAULT;
if( copy_to_user(ustats->nwrites, adapter->nwrites,
num_ldrv*sizeof(u32)) )
return -EFAULT;
if( copy_to_user(ustats->nwriteblocks, adapter->nwriteblocks,
num_ldrv*sizeof(u32)) )
return -EFAULT;
if( copy_to_user(ustats->rd_errors, adapter->rd_errors,
num_ldrv*sizeof(u32)) )
return -EFAULT;
if( copy_to_user(ustats->wr_errors, adapter->wr_errors,
num_ldrv*sizeof(u32)) )
return -EFAULT;
return 0;
#endif
case MBOX_CMD:
/*
* Which adapter
*/
if( (adapno = GETADAP(uioc.adapno)) >= hba_count )
return (-ENODEV);
adapter = hba_soft_state[adapno];
/*
* Deletion of logical drive is a special case. The adapter
* should be quiescent before this command is issued.
*/
if( uioc.uioc_rmbox[0] == FC_DEL_LOGDRV &&
uioc.uioc_rmbox[2] == OP_DEL_LOGDRV ) {
/*
* Do we support this feature
*/
if( !adapter->support_random_del ) {
printk(KERN_WARNING "megaraid: logdrv ");
printk("delete on non-supporting F/W.\n");
return (-EINVAL);
}
rval = mega_del_logdrv( adapter, uioc.uioc_rmbox[3] );
if( rval == 0 ) {
memset(&mc, 0, sizeof(megacmd_t));
mc.status = rval;
rval = mega_n_to_m((void __user *)arg, &mc);
}
return rval;
}
/*
* This interface only support the regular passthru commands.
* Reject extended passthru and 64-bit passthru
*/
if( uioc.uioc_rmbox[0] == MEGA_MBOXCMD_PASSTHRU64 ||
uioc.uioc_rmbox[0] == MEGA_MBOXCMD_EXTPTHRU ) {
printk(KERN_WARNING "megaraid: rejected passthru.\n");
return (-EINVAL);
}
/*
* For all internal commands, the buffer must be allocated in
* <4GB address range
*/
if( make_local_pdev(adapter, &pdev) != 0 )
return -EIO;
/* Is it a passthru command or a DCMD */
if( uioc.uioc_rmbox[0] == MEGA_MBOXCMD_PASSTHRU ) {
/* Passthru commands */
pthru = pci_alloc_consistent(pdev,
sizeof(mega_passthru),
&pthru_dma_hndl);
if( pthru == NULL ) {
free_local_pdev(pdev);
return (-ENOMEM);
}
/*
* The user passthru structure
*/
[SCSI] megaraid: fix compiler warnings The user ioctl mailbox can only support a 32 bit address for the commands structure. This is fine, since the area it's pointing to is allocated with pci_alloc_consistent(), so it should be physically < 4GB. Thus kill the ptr to u32 conversion warnings on 64 bit. Signed-off-by: Martin J. Bligh <mbligh@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-24 07:11:46 +08:00
upthru = (mega_passthru __user *)(unsigned long)MBOX(uioc)->xferaddr;
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!
2005-04-17 06:20:36 +08:00
/*
* Copy in the user passthru here.
*/
if( copy_from_user(pthru, upthru,
sizeof(mega_passthru)) ) {
pci_free_consistent(pdev,
sizeof(mega_passthru), pthru,
pthru_dma_hndl);
free_local_pdev(pdev);
return (-EFAULT);
}
/*
* Is there a data transfer
*/
if( pthru->dataxferlen ) {
data = pci_alloc_consistent(pdev,
pthru->dataxferlen,
&data_dma_hndl);
if( data == NULL ) {
pci_free_consistent(pdev,
sizeof(mega_passthru),
pthru,
pthru_dma_hndl);
free_local_pdev(pdev);
return (-ENOMEM);
}
/*
* Save the user address and point the kernel
* address at just allocated memory
*/
uxferaddr = pthru->dataxferaddr;
pthru->dataxferaddr = data_dma_hndl;
}
/*
* Is data coming down-stream
*/
if( pthru->dataxferlen && (uioc.flags & UIOC_WR) ) {
/*
* Get the user data
*/
[SCSI] megaraid: fix compiler warnings The user ioctl mailbox can only support a 32 bit address for the commands structure. This is fine, since the area it's pointing to is allocated with pci_alloc_consistent(), so it should be physically < 4GB. Thus kill the ptr to u32 conversion warnings on 64 bit. Signed-off-by: Martin J. Bligh <mbligh@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-24 07:11:46 +08:00
if( copy_from_user(data, (char __user *)(unsigned long) uxferaddr,
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!
2005-04-17 06:20:36 +08:00
pthru->dataxferlen) ) {
rval = (-EFAULT);
goto freemem_and_return;
}
}
memset(&mc, 0, sizeof(megacmd_t));
mc.cmd = MEGA_MBOXCMD_PASSTHRU;
mc.xferaddr = (u32)pthru_dma_hndl;
/*
* Issue the command
*/
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
mega_internal_command(adapter, &mc, pthru);
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!
2005-04-17 06:20:36 +08:00
rval = mega_n_to_m((void __user *)arg, &mc);
if( rval ) goto freemem_and_return;
/*
* Is data going up-stream
*/
if( pthru->dataxferlen && (uioc.flags & UIOC_RD) ) {
[SCSI] megaraid: fix compiler warnings The user ioctl mailbox can only support a 32 bit address for the commands structure. This is fine, since the area it's pointing to is allocated with pci_alloc_consistent(), so it should be physically < 4GB. Thus kill the ptr to u32 conversion warnings on 64 bit. Signed-off-by: Martin J. Bligh <mbligh@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-24 07:11:46 +08:00
if( copy_to_user((char __user *)(unsigned long) uxferaddr, data,
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!
2005-04-17 06:20:36 +08:00
pthru->dataxferlen) ) {
rval = (-EFAULT);
}
}
/*
* Send the request sense data also, irrespective of
* whether the user has asked for it or not.
*/
[SCSI] megaraid: Make megaraid_ioctl() check copy_to_user() return value Check copy_to_user() return value in drivers/scsi/megaraid.c::megadev_ioctl() This gets rid of this little warning: drivers/scsi/megaraid.c:3661: warning: ignoring return value of 'copy_to_user', declared with attribute warn_unused_result Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> Acked-by: "Ju, Seokmann" <Seokmann.Ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-15 20:43:11 +08:00
if (copy_to_user(upthru->reqsensearea,
pthru->reqsensearea, 14))
rval = -EFAULT;
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!
2005-04-17 06:20:36 +08:00
freemem_and_return:
if( pthru->dataxferlen ) {
pci_free_consistent(pdev,
pthru->dataxferlen, data,
data_dma_hndl);
}
pci_free_consistent(pdev, sizeof(mega_passthru),
pthru, pthru_dma_hndl);
free_local_pdev(pdev);
return rval;
}
else {
/* DCMD commands */
/*
* Is there a data transfer
*/
if( uioc.xferlen ) {
data = pci_alloc_consistent(pdev,
uioc.xferlen, &data_dma_hndl);
if( data == NULL ) {
free_local_pdev(pdev);
return (-ENOMEM);
}
uxferaddr = MBOX(uioc)->xferaddr;
}
/*
* Is data coming down-stream
*/
if( uioc.xferlen && (uioc.flags & UIOC_WR) ) {
/*
* Get the user data
*/
[SCSI] megaraid: fix compiler warnings The user ioctl mailbox can only support a 32 bit address for the commands structure. This is fine, since the area it's pointing to is allocated with pci_alloc_consistent(), so it should be physically < 4GB. Thus kill the ptr to u32 conversion warnings on 64 bit. Signed-off-by: Martin J. Bligh <mbligh@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-24 07:11:46 +08:00
if( copy_from_user(data, (char __user *)(unsigned long) uxferaddr,
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!
2005-04-17 06:20:36 +08:00
uioc.xferlen) ) {
pci_free_consistent(pdev,
uioc.xferlen,
data, data_dma_hndl);
free_local_pdev(pdev);
return (-EFAULT);
}
}
memcpy(&mc, MBOX(uioc), sizeof(megacmd_t));
mc.xferaddr = (u32)data_dma_hndl;
/*
* Issue the command
*/
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
mega_internal_command(adapter, &mc, NULL);
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!
2005-04-17 06:20:36 +08:00
rval = mega_n_to_m((void __user *)arg, &mc);
if( rval ) {
if( uioc.xferlen ) {
pci_free_consistent(pdev,
uioc.xferlen, data,
data_dma_hndl);
}
free_local_pdev(pdev);
return rval;
}
/*
* Is data going up-stream
*/
if( uioc.xferlen && (uioc.flags & UIOC_RD) ) {
[SCSI] megaraid: fix compiler warnings The user ioctl mailbox can only support a 32 bit address for the commands structure. This is fine, since the area it's pointing to is allocated with pci_alloc_consistent(), so it should be physically < 4GB. Thus kill the ptr to u32 conversion warnings on 64 bit. Signed-off-by: Martin J. Bligh <mbligh@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-24 07:11:46 +08:00
if( copy_to_user((char __user *)(unsigned long) uxferaddr, data,
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!
2005-04-17 06:20:36 +08:00
uioc.xferlen) ) {
rval = (-EFAULT);
}
}
if( uioc.xferlen ) {
pci_free_consistent(pdev,
uioc.xferlen, data,
data_dma_hndl);
}
free_local_pdev(pdev);
return rval;
}
default:
return (-EINVAL);
}
return 0;
}
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
static long
megadev_unlocked_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
{
int ret;
scsi: autoconvert trivial BKL users to private mutex All these files use the big kernel lock in a trivial way to serialize their private file operations, typically resulting from an earlier semi-automatic pushdown from VFS. None of these drivers appears to want to lock against other code, and they all use the BKL as the top-level lock in their file operations, meaning that there is no lock-order inversion problem. Consequently, we can remove the BKL completely, replacing it with a per-file mutex in every case. Using a scripted approach means we can avoid typos. file=$1 name=$2 if grep -q lock_kernel ${file} ; then if grep -q 'include.*linux.mutex.h' ${file} ; then sed -i '/include.*<linux\/smp_lock.h>/d' ${file} else sed -i 's/include.*<linux\/smp_lock.h>.*$/include <linux\/mutex.h>/g' ${file} fi sed -i ${file} \ -e "/^#include.*linux.mutex.h/,$ { 1,/^\(static\|int\|long\)/ { /^\(static\|int\|long\)/istatic DEFINE_MUTEX(${name}_mutex); } }" \ -e "s/\(un\)*lock_kernel\>[ ]*()/mutex_\1lock(\&${name}_mutex)/g" \ -e '/[ ]*cycle_kernel_lock();/d' else sed -i -e '/include.*\<smp_lock.h\>/d' ${file} \ -e '/cycle_kernel_lock()/d' fi Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: linux-scsi@vger.kernel.org Cc: "James E.J. Bottomley" <James.Bottomley@suse.de>
2010-06-02 20:28:52 +08:00
mutex_lock(&megadev_mutex);
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
ret = megadev_ioctl(filep, cmd, arg);
scsi: autoconvert trivial BKL users to private mutex All these files use the big kernel lock in a trivial way to serialize their private file operations, typically resulting from an earlier semi-automatic pushdown from VFS. None of these drivers appears to want to lock against other code, and they all use the BKL as the top-level lock in their file operations, meaning that there is no lock-order inversion problem. Consequently, we can remove the BKL completely, replacing it with a per-file mutex in every case. Using a scripted approach means we can avoid typos. file=$1 name=$2 if grep -q lock_kernel ${file} ; then if grep -q 'include.*linux.mutex.h' ${file} ; then sed -i '/include.*<linux\/smp_lock.h>/d' ${file} else sed -i 's/include.*<linux\/smp_lock.h>.*$/include <linux\/mutex.h>/g' ${file} fi sed -i ${file} \ -e "/^#include.*linux.mutex.h/,$ { 1,/^\(static\|int\|long\)/ { /^\(static\|int\|long\)/istatic DEFINE_MUTEX(${name}_mutex); } }" \ -e "s/\(un\)*lock_kernel\>[ ]*()/mutex_\1lock(\&${name}_mutex)/g" \ -e '/[ ]*cycle_kernel_lock();/d' else sed -i -e '/include.*\<smp_lock.h\>/d' ${file} \ -e '/cycle_kernel_lock()/d' fi Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: linux-scsi@vger.kernel.org Cc: "James E.J. Bottomley" <James.Bottomley@suse.de>
2010-06-02 20:28:52 +08:00
mutex_unlock(&megadev_mutex);
scsi: Push down BKL into ioctl functions Push down the bkl into ioctl functions on the scsi layer. [jkacur: Forward declaration missing ';'. Conflicting declaraction in megaraid.h changed Fixed missing inodes declarations] Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: John Kacur <jkacur@redhat.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2010-04-27 06:24:01 +08:00
return ret;
}
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!
2005-04-17 06:20:36 +08:00
/**
* mega_m_to_n()
* @arg - user address
* @uioc - new ioctl structure
*
* A thin layer to convert older mimd interface ioctl structure to NIT ioctl
* structure
*
* Converts the older mimd ioctl structure to newer NIT structure
*/
static int
mega_m_to_n(void __user *arg, nitioctl_t *uioc)
{
struct uioctl_t uioc_mimd;
char signature[8] = {0};
u8 opcode;
u8 subopcode;
/*
* check is the application conforms to NIT. We do not have to do much
* in that case.
* We exploit the fact that the signature is stored in the very
* begining of the structure.
*/
if( copy_from_user(signature, arg, 7) )
return (-EFAULT);
if( memcmp(signature, "MEGANIT", 7) == 0 ) {
/*
* NOTE NOTE: The nit ioctl is still under flux because of
* change of mailbox definition, in HPE. No applications yet
* use this interface and let's not have applications use this
* interface till the new specifitions are in place.
*/
return -EINVAL;
#if 0
if( copy_from_user(uioc, arg, sizeof(nitioctl_t)) )
return (-EFAULT);
return 0;
#endif
}
/*
* Else assume we have mimd uioctl_t as arg. Convert to nitioctl_t
*
* Get the user ioctl structure
*/
if( copy_from_user(&uioc_mimd, arg, sizeof(struct uioctl_t)) )
return (-EFAULT);
/*
* Get the opcode and subopcode for the commands
*/
opcode = uioc_mimd.ui.fcs.opcode;
subopcode = uioc_mimd.ui.fcs.subopcode;
switch (opcode) {
case 0x82:
switch (subopcode) {
case MEGAIOC_QDRVRVER: /* Query driver version */
uioc->opcode = GET_DRIVER_VER;
uioc->uioc_uaddr = uioc_mimd.data;
break;
case MEGAIOC_QNADAP: /* Get # of adapters */
uioc->opcode = GET_N_ADAP;
uioc->uioc_uaddr = uioc_mimd.data;
break;
case MEGAIOC_QADAPINFO: /* Get adapter information */
uioc->opcode = GET_ADAP_INFO;
uioc->adapno = uioc_mimd.ui.fcs.adapno;
uioc->uioc_uaddr = uioc_mimd.data;
break;
default:
return(-EINVAL);
}
break;
case 0x81:
uioc->opcode = MBOX_CMD;
uioc->adapno = uioc_mimd.ui.fcs.adapno;
memcpy(uioc->uioc_rmbox, uioc_mimd.mbox, 18);
uioc->xferlen = uioc_mimd.ui.fcs.length;
if( uioc_mimd.outlen ) uioc->flags = UIOC_RD;
if( uioc_mimd.inlen ) uioc->flags |= UIOC_WR;
break;
case 0x80:
uioc->opcode = MBOX_CMD;
uioc->adapno = uioc_mimd.ui.fcs.adapno;
memcpy(uioc->uioc_rmbox, uioc_mimd.mbox, 18);
/*
* Choose the xferlen bigger of input and output data
*/
uioc->xferlen = uioc_mimd.outlen > uioc_mimd.inlen ?
uioc_mimd.outlen : uioc_mimd.inlen;
if( uioc_mimd.outlen ) uioc->flags = UIOC_RD;
if( uioc_mimd.inlen ) uioc->flags |= UIOC_WR;
break;
default:
return (-EINVAL);
}
return 0;
}
/*
* mega_n_to_m()
* @arg - user address
* @mc - mailbox command
*
* Updates the status information to the application, depending on application
* conforms to older mimd ioctl interface or newer NIT ioctl interface
*/
static int
mega_n_to_m(void __user *arg, megacmd_t *mc)
{
nitioctl_t __user *uiocp;
megacmd_t __user *umc;
mega_passthru __user *upthru;
struct uioctl_t __user *uioc_mimd;
char signature[8] = {0};
/*
* check is the application conforms to NIT.
*/
if( copy_from_user(signature, arg, 7) )
return -EFAULT;
if( memcmp(signature, "MEGANIT", 7) == 0 ) {
uiocp = arg;
if( put_user(mc->status, (u8 __user *)&MBOX_P(uiocp)->status) )
return (-EFAULT);
if( mc->cmd == MEGA_MBOXCMD_PASSTHRU ) {
umc = MBOX_P(uiocp);
if (get_user(upthru, (mega_passthru __user * __user *)&umc->xferaddr))
return -EFAULT;
if( put_user(mc->status, (u8 __user *)&upthru->scsistatus))
return (-EFAULT);
}
}
else {
uioc_mimd = arg;
if( put_user(mc->status, (u8 __user *)&uioc_mimd->mbox[17]) )
return (-EFAULT);
if( mc->cmd == MEGA_MBOXCMD_PASSTHRU ) {
umc = (megacmd_t __user *)uioc_mimd->mbox;
if (get_user(upthru, (mega_passthru __user * __user *)&umc->xferaddr))
return (-EFAULT);
if( put_user(mc->status, (u8 __user *)&upthru->scsistatus) )
return (-EFAULT);
}
}
return 0;
}
/*
* MEGARAID 'FW' commands.
*/
/**
* mega_is_bios_enabled()
* @adapter - pointer to our soft state
*
* issue command to find out if the BIOS is enabled for this controller
*/
static int
mega_is_bios_enabled(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
int ret;
mbox = (mbox_t *)raw_mbox;
memset(&mbox->m_out, 0, sizeof(raw_mbox));
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
raw_mbox[0] = IS_BIOS_ENABLED;
raw_mbox[2] = GET_BIOS;
ret = issue_scb_block(adapter, raw_mbox);
return *(char *)adapter->mega_buffer;
}
/**
* mega_enum_raid_scsi()
* @adapter - pointer to our soft state
*
* Find out what channels are RAID/SCSI. This information is used to
* differentiate the virtual channels and physical channels and to support
* ROMB feature and non-disk devices.
*/
static void
mega_enum_raid_scsi(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
int i;
mbox = (mbox_t *)raw_mbox;
memset(&mbox->m_out, 0, sizeof(raw_mbox));
/*
* issue command to find out what channels are raid/scsi
*/
raw_mbox[0] = CHNL_CLASS;
raw_mbox[2] = GET_CHNL_CLASS;
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
/*
* Non-ROMB firmware fail this command, so all channels
* must be shown RAID
*/
adapter->mega_ch_class = 0xFF;
if(!issue_scb_block(adapter, raw_mbox)) {
adapter->mega_ch_class = *((char *)adapter->mega_buffer);
}
for( i = 0; i < adapter->product_info.nchannels; i++ ) {
if( (adapter->mega_ch_class >> i) & 0x01 ) {
printk(KERN_INFO "megaraid: channel[%d] is raid.\n",
i);
}
else {
printk(KERN_INFO "megaraid: channel[%d] is scsi.\n",
i);
}
}
return;
}
/**
* mega_get_boot_drv()
* @adapter - pointer to our soft state
*
* Find out which device is the boot device. Note, any logical drive or any
* phyical device (e.g., a CDROM) can be designated as a boot device.
*/
static void
mega_get_boot_drv(adapter_t *adapter)
{
struct private_bios_data *prv_bios_data;
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
u16 cksum = 0;
u8 *cksum_p;
u8 boot_pdrv;
int i;
mbox = (mbox_t *)raw_mbox;
memset(&mbox->m_out, 0, sizeof(raw_mbox));
raw_mbox[0] = BIOS_PVT_DATA;
raw_mbox[2] = GET_BIOS_PVT_DATA;
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
adapter->boot_ldrv_enabled = 0;
adapter->boot_ldrv = 0;
adapter->boot_pdrv_enabled = 0;
adapter->boot_pdrv_ch = 0;
adapter->boot_pdrv_tgt = 0;
if(issue_scb_block(adapter, raw_mbox) == 0) {
prv_bios_data =
(struct private_bios_data *)adapter->mega_buffer;
cksum = 0;
cksum_p = (char *)prv_bios_data;
for (i = 0; i < 14; i++ ) {
cksum += (u16)(*cksum_p++);
}
if (prv_bios_data->cksum == (u16)(0-cksum) ) {
/*
* If MSB is set, a physical drive is set as boot
* device
*/
if( prv_bios_data->boot_drv & 0x80 ) {
adapter->boot_pdrv_enabled = 1;
boot_pdrv = prv_bios_data->boot_drv & 0x7F;
adapter->boot_pdrv_ch = boot_pdrv / 16;
adapter->boot_pdrv_tgt = boot_pdrv % 16;
}
else {
adapter->boot_ldrv_enabled = 1;
adapter->boot_ldrv = prv_bios_data->boot_drv;
}
}
}
}
/**
* mega_support_random_del()
* @adapter - pointer to our soft state
*
* Find out if this controller supports random deletion and addition of
* logical drives
*/
static int
mega_support_random_del(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
int rval;
mbox = (mbox_t *)raw_mbox;
memset(&mbox->m_out, 0, sizeof(raw_mbox));
/*
* issue command
*/
raw_mbox[0] = FC_DEL_LOGDRV;
raw_mbox[2] = OP_SUP_DEL_LOGDRV;
rval = issue_scb_block(adapter, raw_mbox);
return !rval;
}
/**
* mega_support_ext_cdb()
* @adapter - pointer to our soft state
*
* Find out if this firmware support cdblen > 10
*/
static int
mega_support_ext_cdb(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
int rval;
mbox = (mbox_t *)raw_mbox;
memset(&mbox->m_out, 0, sizeof(raw_mbox));
/*
* issue command to find out if controller supports extended CDBs.
*/
raw_mbox[0] = 0xA4;
raw_mbox[2] = 0x16;
rval = issue_scb_block(adapter, raw_mbox);
return !rval;
}
/**
* mega_del_logdrv()
* @adapter - pointer to our soft state
* @logdrv - logical drive to be deleted
*
* Delete the specified logical drive. It is the responsibility of the user
* app to let the OS know about this operation.
*/
static int
mega_del_logdrv(adapter_t *adapter, int logdrv)
{
unsigned long flags;
scb_t *scb;
int rval;
/*
* Stop sending commands to the controller, queue them internally.
* When deletion is complete, ISR will flush the queue.
*/
atomic_set(&adapter->quiescent, 1);
/*
* Wait till all the issued commands are complete and there are no
* commands in the pending queue
*/
while (atomic_read(&adapter->pend_cmds) > 0 ||
!list_empty(&adapter->pending_list))
msleep(1000); /* sleep for 1s */
rval = mega_do_del_logdrv(adapter, logdrv);
spin_lock_irqsave(&adapter->lock, flags);
/*
* If delete operation was successful, add 0x80 to the logical drive
* ids for commands in the pending queue.
*/
if (adapter->read_ldidmap) {
struct list_head *pos;
list_for_each(pos, &adapter->pending_list) {
scb = list_entry(pos, scb_t, list);
if (scb->pthru->logdrv < 0x80 )
scb->pthru->logdrv += 0x80;
}
}
atomic_set(&adapter->quiescent, 0);
mega_runpendq(adapter);
spin_unlock_irqrestore(&adapter->lock, flags);
return rval;
}
static int
mega_do_del_logdrv(adapter_t *adapter, int logdrv)
{
megacmd_t mc;
int rval;
memset( &mc, 0, sizeof(megacmd_t));
mc.cmd = FC_DEL_LOGDRV;
mc.opcode = OP_DEL_LOGDRV;
mc.subopcode = logdrv;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
rval = mega_internal_command(adapter, &mc, NULL);
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!
2005-04-17 06:20:36 +08:00
/* log this event */
if(rval) {
printk(KERN_WARNING "megaraid: Delete LD-%d failed.", logdrv);
return rval;
}
/*
* After deleting first logical drive, the logical drives must be
* addressed by adding 0x80 to the logical drive id.
*/
adapter->read_ldidmap = 1;
return rval;
}
/**
* mega_get_max_sgl()
* @adapter - pointer to our soft state
*
* Find out the maximum number of scatter-gather elements supported by this
* version of the firmware
*/
static void
mega_get_max_sgl(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
mbox = (mbox_t *)raw_mbox;
memset(mbox, 0, sizeof(raw_mbox));
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
raw_mbox[0] = MAIN_MISC_OPCODE;
raw_mbox[2] = GET_MAX_SG_SUPPORT;
if( issue_scb_block(adapter, raw_mbox) ) {
/*
* f/w does not support this command. Choose the default value
*/
adapter->sglen = MIN_SGLIST;
}
else {
adapter->sglen = *((char *)adapter->mega_buffer);
/*
* Make sure this is not more than the resources we are
* planning to allocate
*/
if ( adapter->sglen > MAX_SGLIST )
adapter->sglen = MAX_SGLIST;
}
return;
}
/**
* mega_support_cluster()
* @adapter - pointer to our soft state
*
* Find out if this firmware support cluster calls.
*/
static int
mega_support_cluster(adapter_t *adapter)
{
unsigned char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox;
mbox = (mbox_t *)raw_mbox;
memset(mbox, 0, sizeof(raw_mbox));
memset((void *)adapter->mega_buffer, 0, MEGA_BUFFER_SIZE);
mbox->m_out.xferaddr = (u32)adapter->buf_dma_handle;
/*
* Try to get the initiator id. This command will succeed iff the
* clustering is available on this HBA.
*/
raw_mbox[0] = MEGA_GET_TARGET_ID;
if( issue_scb_block(adapter, raw_mbox) == 0 ) {
/*
* Cluster support available. Get the initiator target id.
* Tell our id to mid-layer too.
*/
adapter->this_id = *(u32 *)adapter->mega_buffer;
adapter->host->this_id = adapter->this_id;
return 1;
}
return 0;
}
[SCSI] megaraid: fix warnings when CONFIG_PROC_FS=n drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:4893: warning: implicit declaration of function 'mega_create_proc_entry' drivers/scsi/megaraid.c: In function 'megaraid_remove_one': drivers/scsi/megaraid.c:4968: warning: unused variable 'buf' Fix by adding #defines Signed-off-by: walter harms <wharms@bfs.de> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:35:09 +08:00
#ifdef CONFIG_PROC_FS
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!
2005-04-17 06:20:36 +08:00
/**
* mega_adapinq()
* @adapter - pointer to our soft state
* @dma_handle - DMA address of the buffer
*
* Issue internal comamnds while interrupts are available.
* We only issue direct mailbox commands from within the driver. ioctl()
* interface using these routines can issue passthru commands.
*/
static int
mega_adapinq(adapter_t *adapter, dma_addr_t dma_handle)
{
megacmd_t mc;
memset(&mc, 0, sizeof(megacmd_t));
if( adapter->flag & BOARD_40LD ) {
mc.cmd = FC_NEW_CONFIG;
mc.opcode = NC_SUBOP_ENQUIRY3;
mc.subopcode = ENQ3_GET_SOLICITED_FULL;
}
else {
mc.cmd = MEGA_MBOXCMD_ADPEXTINQ;
}
mc.xferaddr = (u32)dma_handle;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
if ( mega_internal_command(adapter, &mc, NULL) != 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!
2005-04-17 06:20:36 +08:00
return -1;
}
return 0;
}
/** mega_internal_dev_inquiry()
* @adapter - pointer to our soft state
* @ch - channel for this device
* @tgt - ID of this device
* @buf_dma_handle - DMA address of the buffer
*
* Issue the scsi inquiry for the specified device.
*/
static int
mega_internal_dev_inquiry(adapter_t *adapter, u8 ch, u8 tgt,
dma_addr_t buf_dma_handle)
{
mega_passthru *pthru;
dma_addr_t pthru_dma_handle;
megacmd_t mc;
int rval;
struct pci_dev *pdev;
/*
* For all internal commands, the buffer must be allocated in <4GB
* address range
*/
if( make_local_pdev(adapter, &pdev) != 0 ) return -1;
pthru = pci_alloc_consistent(pdev, sizeof(mega_passthru),
&pthru_dma_handle);
if( pthru == NULL ) {
free_local_pdev(pdev);
return -1;
}
pthru->timeout = 2;
pthru->ars = 1;
pthru->reqsenselen = 14;
pthru->islogical = 0;
pthru->channel = (adapter->flag & BOARD_40LD) ? 0 : ch;
pthru->target = (adapter->flag & BOARD_40LD) ? (ch << 4)|tgt : tgt;
pthru->cdblen = 6;
pthru->cdb[0] = INQUIRY;
pthru->cdb[1] = 0;
pthru->cdb[2] = 0;
pthru->cdb[3] = 0;
pthru->cdb[4] = 255;
pthru->cdb[5] = 0;
pthru->dataxferaddr = (u32)buf_dma_handle;
pthru->dataxferlen = 256;
memset(&mc, 0, sizeof(megacmd_t));
mc.cmd = MEGA_MBOXCMD_PASSTHRU;
mc.xferaddr = (u32)pthru_dma_handle;
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
rval = mega_internal_command(adapter, &mc, pthru);
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!
2005-04-17 06:20:36 +08:00
pci_free_consistent(pdev, sizeof(mega_passthru), pthru,
pthru_dma_handle);
free_local_pdev(pdev);
return rval;
}
[SCSI] megaraid: fix warnings when CONFIG_PROC_FS=n drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:4893: warning: implicit declaration of function 'mega_create_proc_entry' drivers/scsi/megaraid.c: In function 'megaraid_remove_one': drivers/scsi/megaraid.c:4968: warning: unused variable 'buf' Fix by adding #defines Signed-off-by: walter harms <wharms@bfs.de> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:35:09 +08:00
#endif
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!
2005-04-17 06:20:36 +08:00
/**
* mega_internal_command()
* @adapter - pointer to our soft state
* @mc - the mailbox command
* @pthru - Passthru structure for DCDB commands
*
* Issue the internal commands in interrupt mode.
* The last argument is the address of the passthru structure if the command
* to be fired is a passthru command
*
* lockscope specifies whether the caller has already acquired the lock. Of
* course, the caller must know which lock we are talking about.
*
* Note: parameter 'pthru' is null for non-passthru commands.
*/
static int
[SCSI] megaraid (legacy): remove scsi_assign_lock usage just take the adapter lock in megaraid_queue. Additional benefit is that we can get rid of the awkward conditional locking in mega_internal_command. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 03:12:07 +08:00
mega_internal_command(adapter_t *adapter, megacmd_t *mc, mega_passthru *pthru)
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!
2005-04-17 06:20:36 +08:00
{
Scsi_Cmnd *scmd;
struct scsi_device *sdev;
scb_t *scb;
int rval;
[SCSI] megaraid: fix mega_internal_command oops scsi_cmnd->cmnd was changed from a static array to a pointer post 2.6.25. It breaks mega_internal_command(): static int mega_internal_command(adapter_t *adapter, megacmd_t *mc, mega_passthru *pthru) { ... scb = &adapter->int_scb; memset(scb, 0, sizeof(scb_t)); scmd = &adapter->int_scmd; memset(scmd, 0, sizeof(Scsi_Cmnd)); sdev = kzalloc(sizeof(struct scsi_device), GFP_KERNEL); scmd->device = sdev; scmd->device->host = adapter->host; scmd->host_scribble = (void *)scb; scmd->cmnd[0] = MEGA_INTERNAL_CMD; mega_internal_command() uses scsi_cmnd allocated internally so scmd->cmnd is NULL here. This patch adds a static array for cdb to adapter_t and uses it here. This also uses scsi_allocate_command/scsi_free_command, the recommended way to allocate struct scsi_cmnd since the driver might use sense_buffer in struct scsi_cmnd. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Reviewed-by: Boaz Harrosh <bharrosh@panasas.com> Tested-by: Pascal Terjan <pterjan@gmail.com> Reported-by: Pascal Terjan <pterjan@gmail.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-10-24 08:21:05 +08:00
scmd = scsi_allocate_command(GFP_KERNEL);
if (!scmd)
return -ENOMEM;
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!
2005-04-17 06:20:36 +08:00
/*
* The internal commands share one command id and hence are
* serialized. This is so because we want to reserve maximum number of
* available command ids for the I/O commands.
*/
[SCSI] turn most scsi semaphores into mutexes the scsi layer is using semaphores in a mutex way, this patch converts these into using mutexes instead Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-01-11 20:16:10 +08:00
mutex_lock(&adapter->int_mtx);
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!
2005-04-17 06:20:36 +08:00
scb = &adapter->int_scb;
memset(scb, 0, sizeof(scb_t));
[SCSI] kmalloc + memset conversion to kzalloc In NCR_D700, a4000t, aic7xxx_old, bvme6000, dpt_i2o, gdth, lpfc, megaraid, mvme16x osst, pluto, qla2xxx, zorro7xx Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-08-11 16:13:24 +08:00
sdev = kzalloc(sizeof(struct scsi_device), 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!
2005-04-17 06:20:36 +08:00
scmd->device = sdev;
[SCSI] megaraid: fix mega_internal_command oops scsi_cmnd->cmnd was changed from a static array to a pointer post 2.6.25. It breaks mega_internal_command(): static int mega_internal_command(adapter_t *adapter, megacmd_t *mc, mega_passthru *pthru) { ... scb = &adapter->int_scb; memset(scb, 0, sizeof(scb_t)); scmd = &adapter->int_scmd; memset(scmd, 0, sizeof(Scsi_Cmnd)); sdev = kzalloc(sizeof(struct scsi_device), GFP_KERNEL); scmd->device = sdev; scmd->device->host = adapter->host; scmd->host_scribble = (void *)scb; scmd->cmnd[0] = MEGA_INTERNAL_CMD; mega_internal_command() uses scsi_cmnd allocated internally so scmd->cmnd is NULL here. This patch adds a static array for cdb to adapter_t and uses it here. This also uses scsi_allocate_command/scsi_free_command, the recommended way to allocate struct scsi_cmnd since the driver might use sense_buffer in struct scsi_cmnd. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Reviewed-by: Boaz Harrosh <bharrosh@panasas.com> Tested-by: Pascal Terjan <pterjan@gmail.com> Reported-by: Pascal Terjan <pterjan@gmail.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-10-24 08:21:05 +08:00
memset(adapter->int_cdb, 0, sizeof(adapter->int_cdb));
scmd->cmnd = adapter->int_cdb;
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!
2005-04-17 06:20:36 +08:00
scmd->device->host = adapter->host;
[SCSI] megaraid_old: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Jens Axboe <jens.axboe@oracle.com> did the for_each_sg cleanup. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:17:27 +08:00
scmd->host_scribble = (void *)scb;
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!
2005-04-17 06:20:36 +08:00
scmd->cmnd[0] = MEGA_INTERNAL_CMD;
scb->state |= SCB_ACTIVE;
scb->cmd = scmd;
memcpy(scb->raw_mbox, mc, sizeof(megacmd_t));
/*
* Is it a passthru command
*/
if( mc->cmd == MEGA_MBOXCMD_PASSTHRU ) {
scb->pthru = pthru;
}
scb->idx = CMDID_INT_CMDS;
megaraid_queue(scmd, mega_internal_done);
[SCSI] remove scsi_cmnd->state We never look at it except for the old megaraid driver that abuses it for sending internal commands. That usage can be fixed easily because those internal commands are single-threaded by a mutex and we can easily use a completion there. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-19 19:42:05 +08:00
wait_for_completion(&adapter->int_waitq);
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!
2005-04-17 06:20:36 +08:00
rval = scmd->result;
mc->status = scmd->result;
kfree(sdev);
/*
* Print a debug message for all failed commands. Applications can use
* this information.
*/
if( scmd->result && trace_level ) {
printk("megaraid: cmd [%x, %x, %x] status:[%x]\n",
mc->cmd, mc->opcode, mc->subopcode, scmd->result);
}
[SCSI] turn most scsi semaphores into mutexes the scsi layer is using semaphores in a mutex way, this patch converts these into using mutexes instead Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-01-11 20:16:10 +08:00
mutex_unlock(&adapter->int_mtx);
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!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid: fix mega_internal_command oops scsi_cmnd->cmnd was changed from a static array to a pointer post 2.6.25. It breaks mega_internal_command(): static int mega_internal_command(adapter_t *adapter, megacmd_t *mc, mega_passthru *pthru) { ... scb = &adapter->int_scb; memset(scb, 0, sizeof(scb_t)); scmd = &adapter->int_scmd; memset(scmd, 0, sizeof(Scsi_Cmnd)); sdev = kzalloc(sizeof(struct scsi_device), GFP_KERNEL); scmd->device = sdev; scmd->device->host = adapter->host; scmd->host_scribble = (void *)scb; scmd->cmnd[0] = MEGA_INTERNAL_CMD; mega_internal_command() uses scsi_cmnd allocated internally so scmd->cmnd is NULL here. This patch adds a static array for cdb to adapter_t and uses it here. This also uses scsi_allocate_command/scsi_free_command, the recommended way to allocate struct scsi_cmnd since the driver might use sense_buffer in struct scsi_cmnd. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Reviewed-by: Boaz Harrosh <bharrosh@panasas.com> Tested-by: Pascal Terjan <pterjan@gmail.com> Reported-by: Pascal Terjan <pterjan@gmail.com> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-10-24 08:21:05 +08:00
scsi_free_command(GFP_KERNEL, scmd);
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!
2005-04-17 06:20:36 +08:00
return rval;
}
/**
* mega_internal_done()
* @scmd - internal scsi command
*
* Callback routine for internal commands.
*/
static void
mega_internal_done(Scsi_Cmnd *scmd)
{
adapter_t *adapter;
adapter = (adapter_t *)scmd->device->host->hostdata;
[SCSI] remove scsi_cmnd->state We never look at it except for the old megaraid driver that abuses it for sending internal commands. That usage can be fixed easily because those internal commands are single-threaded by a mutex and we can easily use a completion there. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-19 19:42:05 +08:00
complete(&adapter->int_waitq);
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!
2005-04-17 06:20:36 +08:00
}
static struct scsi_host_template megaraid_template = {
.module = THIS_MODULE,
.name = "MegaRAID",
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
.proc_name = "megaraid_legacy",
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!
2005-04-17 06:20:36 +08:00
.info = megaraid_info,
.queuecommand = megaraid_queue,
.bios_param = megaraid_biosparam,
.max_sectors = MAX_SECTORS_PER_IO,
.can_queue = MAX_COMMANDS,
.this_id = DEFAULT_INITIATOR_ID,
.sg_tablesize = MAX_SGLIST,
.cmd_per_lun = DEF_CMD_PER_LUN,
.use_clustering = ENABLE_CLUSTERING,
.eh_abort_handler = megaraid_abort,
.eh_device_reset_handler = megaraid_reset,
.eh_bus_reset_handler = megaraid_reset,
.eh_host_reset_handler = megaraid_reset,
};
static int __devinit
megaraid_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
{
struct Scsi_Host *host;
adapter_t *adapter;
unsigned long mega_baseport, tbase, flag = 0;
u16 subsysid, subsysvid;
u8 pci_bus, pci_dev_func;
int irq, i, j;
int error = -ENODEV;
if (pci_enable_device(pdev))
goto out;
pci_set_master(pdev);
pci_bus = pdev->bus->number;
pci_dev_func = pdev->devfn;
/*
* The megaraid3 stuff reports the ID of the Intel part which is not
* remotely specific to the megaraid
*/
if (pdev->vendor == PCI_VENDOR_ID_INTEL) {
u16 magic;
/*
* Don't fall over the Compaq management cards using the same
* PCI identifier
*/
if (pdev->subsystem_vendor == PCI_VENDOR_ID_COMPAQ &&
pdev->subsystem_device == 0xC000)
return -ENODEV;
/* Now check the magic signature byte */
pci_read_config_word(pdev, PCI_CONF_AMISIG, &magic);
if (magic != HBA_SIGNATURE_471 && magic != HBA_SIGNATURE)
return -ENODEV;
/* Ok it is probably a megaraid */
}
/*
* For these vendor and device ids, signature offsets are not
* valid and 64 bit is implicit
*/
if (id->driver_data & BOARD_64BIT)
flag |= BOARD_64BIT;
else {
u32 magic64;
pci_read_config_dword(pdev, PCI_CONF_AMISIG64, &magic64);
if (magic64 == HBA_SIGNATURE_64BIT)
flag |= BOARD_64BIT;
}
subsysvid = pdev->subsystem_vendor;
subsysid = pdev->subsystem_device;
printk(KERN_NOTICE "megaraid: found 0x%4.04x:0x%4.04x:bus %d:",
id->vendor, id->device, pci_bus);
printk("slot %d:func %d\n",
PCI_SLOT(pci_dev_func), PCI_FUNC(pci_dev_func));
/* Read the base port and IRQ from PCI */
mega_baseport = pci_resource_start(pdev, 0);
irq = pdev->irq;
tbase = mega_baseport;
if (pci_resource_flags(pdev, 0) & IORESOURCE_MEM) {
flag |= BOARD_MEMMAP;
if (!request_mem_region(mega_baseport, 128, "megaraid")) {
printk(KERN_WARNING "megaraid: mem region busy!\n");
goto out_disable_device;
}
mega_baseport = (unsigned long)ioremap(mega_baseport, 128);
if (!mega_baseport) {
printk(KERN_WARNING
"megaraid: could not map hba memory\n");
goto out_release_region;
}
} else {
flag |= BOARD_IOMAP;
mega_baseport += 0x10;
if (!request_region(mega_baseport, 16, "megaraid"))
goto out_disable_device;
}
/* Initialize SCSI Host structure */
host = scsi_host_alloc(&megaraid_template, sizeof(adapter_t));
if (!host)
goto out_iounmap;
adapter = (adapter_t *)host->hostdata;
memset(adapter, 0, sizeof(adapter_t));
printk(KERN_NOTICE
"scsi%d:Found MegaRAID controller at 0x%lx, IRQ:%d\n",
host->host_no, mega_baseport, irq);
adapter->base = mega_baseport;
[SCSI] megaraid: fix MMIO casts megaraid's MMIO RD*/WR* macros directly call readl() and writel() with an 'unsigned long' argument. This throws a warning, but is otherwise OK because the 'unsigned long' is really the result of ioremap(). This setup is also OK because the variable can hold an ioremap cookie /or/ a PCI I/O port (PIO). However, to fix the warning thrown when readl() and writel() are passed an unsigned long cookie, I introduce 'void __iomem *mmio_base', holding the same value as 'base'. This will silence the warnings, and also cause an oops whenever these MMIO-only functions are ever accidentally passed an I/O address. Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-12-04 09:49:23 +08:00
if (flag & BOARD_MEMMAP)
adapter->mmio_base = (void __iomem *) mega_baseport;
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!
2005-04-17 06:20:36 +08:00
INIT_LIST_HEAD(&adapter->free_list);
INIT_LIST_HEAD(&adapter->pending_list);
INIT_LIST_HEAD(&adapter->completed_list);
adapter->flag = flag;
spin_lock_init(&adapter->lock);
host->cmd_per_lun = max_cmd_per_lun;
host->max_sectors = max_sectors_per_io;
adapter->dev = pdev;
adapter->host = host;
adapter->host->irq = irq;
if (flag & BOARD_MEMMAP)
adapter->host->base = tbase;
else {
adapter->host->io_port = tbase;
adapter->host->n_io_port = 16;
}
adapter->host->unique_id = (pci_bus << 8) | pci_dev_func;
/*
* Allocate buffer to issue internal commands.
*/
adapter->mega_buffer = pci_alloc_consistent(adapter->dev,
MEGA_BUFFER_SIZE, &adapter->buf_dma_handle);
if (!adapter->mega_buffer) {
printk(KERN_WARNING "megaraid: out of RAM.\n");
goto out_host_put;
}
adapter->scb_list = kmalloc(sizeof(scb_t) * MAX_COMMANDS, GFP_KERNEL);
if (!adapter->scb_list) {
printk(KERN_WARNING "megaraid: out of RAM.\n");
goto out_free_cmd_buffer;
}
if (request_irq(irq, (adapter->flag & BOARD_MEMMAP) ?
megaraid_isr_memmapped : megaraid_isr_iomapped,
[PATCH] irq-flags: scsi: Use the new IRQF_ constants Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: "David S. Miller" <davem@davemloft.net> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: James Bottomley <James.Bottomley@steeleye.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-02 10:29:42 +08:00
IRQF_SHARED, "megaraid", adapter)) {
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!
2005-04-17 06:20:36 +08:00
printk(KERN_WARNING
"megaraid: Couldn't register IRQ %d!\n", irq);
goto out_free_scb_list;
}
if (mega_setup_mailbox(adapter))
goto out_free_irq;
if (mega_query_adapter(adapter))
goto out_free_mbox;
/*
* Have checks for some buggy f/w
*/
if ((subsysid == 0x1111) && (subsysvid == 0x1111)) {
/*
* Which firmware
*/
if (!strcmp(adapter->fw_version, "3.00") ||
!strcmp(adapter->fw_version, "3.01")) {
printk( KERN_WARNING
"megaraid: Your card is a Dell PERC "
"2/SC RAID controller with "
"firmware\nmegaraid: 3.00 or 3.01. "
"This driver is known to have "
"corruption issues\nmegaraid: with "
"those firmware versions on this "
"specific card. In order\nmegaraid: "
"to protect your data, please upgrade "
"your firmware to version\nmegaraid: "
"3.10 or later, available from the "
"Dell Technical Support web\n"
"megaraid: site at\nhttp://support."
"dell.com/us/en/filelib/download/"
"index.asp?fileid=2940\n"
);
}
}
/*
* If we have a HP 1M(0x60E7)/2M(0x60E8) controller with
* firmware H.01.07, H.01.08, and H.01.09 disable 64 bit
* support, since this firmware cannot handle 64 bit
* addressing
*/
if ((subsysvid == HP_SUBSYS_VID) &&
((subsysid == 0x60E7) || (subsysid == 0x60E8))) {
/*
* which firmware
*/
if (!strcmp(adapter->fw_version, "H01.07") ||
!strcmp(adapter->fw_version, "H01.08") ||
!strcmp(adapter->fw_version, "H01.09") ) {
printk(KERN_WARNING
"megaraid: Firmware H.01.07, "
"H.01.08, and H.01.09 on 1M/2M "
"controllers\n"
"megaraid: do not support 64 bit "
"addressing.\nmegaraid: DISABLING "
"64 bit support.\n");
adapter->flag &= ~BOARD_64BIT;
}
}
if (mega_is_bios_enabled(adapter))
mega_hbas[hba_count].is_bios_enabled = 1;
mega_hbas[hba_count].hostdata_addr = adapter;
/*
* Find out which channel is raid and which is scsi. This is
* for ROMB support.
*/
mega_enum_raid_scsi(adapter);
/*
* Find out if a logical drive is set as the boot drive. If
* there is one, will make that as the first logical drive.
* ROMB: Do we have to boot from a physical drive. Then all
* the physical drives would appear before the logical disks.
* Else, all the physical drives would be exported to the mid
* layer after logical drives.
*/
mega_get_boot_drv(adapter);
if (adapter->boot_pdrv_enabled) {
j = adapter->product_info.nchannels;
for( i = 0; i < j; i++ )
adapter->logdrv_chan[i] = 0;
for( i = j; i < NVIRT_CHAN + j; i++ )
adapter->logdrv_chan[i] = 1;
} else {
for (i = 0; i < NVIRT_CHAN; i++)
adapter->logdrv_chan[i] = 1;
for (i = NVIRT_CHAN; i < MAX_CHANNELS+NVIRT_CHAN; i++)
adapter->logdrv_chan[i] = 0;
adapter->mega_ch_class <<= NVIRT_CHAN;
}
/*
* Do we support random deletion and addition of logical
* drives
*/
adapter->read_ldidmap = 0; /* set it after first logdrv
delete cmd */
adapter->support_random_del = mega_support_random_del(adapter);
/* Initialize SCBs */
if (mega_init_scb(adapter))
goto out_free_mbox;
/*
* Reset the pending commands counter
*/
atomic_set(&adapter->pend_cmds, 0);
/*
* Reset the adapter quiescent flag
*/
atomic_set(&adapter->quiescent, 0);
hba_soft_state[hba_count] = adapter;
/*
* Fill in the structure which needs to be passed back to the
* application when it does an ioctl() for controller related
* information.
*/
i = hba_count;
mcontroller[i].base = mega_baseport;
mcontroller[i].irq = irq;
mcontroller[i].numldrv = adapter->numldrv;
mcontroller[i].pcibus = pci_bus;
mcontroller[i].pcidev = id->device;
mcontroller[i].pcifun = PCI_FUNC (pci_dev_func);
mcontroller[i].pciid = -1;
mcontroller[i].pcivendor = id->vendor;
mcontroller[i].pcislot = PCI_SLOT(pci_dev_func);
mcontroller[i].uid = (pci_bus << 8) | pci_dev_func;
/* Set the Mode of addressing to 64 bit if we can */
if ((adapter->flag & BOARD_64BIT) && (sizeof(dma_addr_t) == 8)) {
dma-mapping: replace all DMA_64BIT_MASK macro with DMA_BIT_MASK(64) Replace all DMA_64BIT_MASK macro with DMA_BIT_MASK(64) Signed-off-by: Yang Hongyang<yanghy@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-07 10:01:13 +08:00
pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
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!
2005-04-17 06:20:36 +08:00
adapter->has_64bit_addr = 1;
} else {
dma-mapping: replace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32) Replace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32) Signed-off-by: Yang Hongyang<yanghy@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-07 10:01:15 +08:00
pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
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!
2005-04-17 06:20:36 +08:00
adapter->has_64bit_addr = 0;
}
[SCSI] turn most scsi semaphores into mutexes the scsi layer is using semaphores in a mutex way, this patch converts these into using mutexes instead Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-01-11 20:16:10 +08:00
mutex_init(&adapter->int_mtx);
[SCSI] remove scsi_cmnd->state We never look at it except for the old megaraid driver that abuses it for sending internal commands. That usage can be fixed easily because those internal commands are single-threaded by a mutex and we can easily use a completion there. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-06-19 19:42:05 +08:00
init_completion(&adapter->int_waitq);
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!
2005-04-17 06:20:36 +08:00
adapter->this_id = DEFAULT_INITIATOR_ID;
adapter->host->this_id = DEFAULT_INITIATOR_ID;
#if MEGA_HAVE_CLUSTERING
/*
* Is cluster support enabled on this controller
* Note: In a cluster the HBAs ( the initiators ) will have
* different target IDs and we cannot assume it to be 7. Call
* to mega_support_cluster() will get the target ids also if
* the cluster support is available
*/
adapter->has_cluster = mega_support_cluster(adapter);
if (adapter->has_cluster) {
printk(KERN_NOTICE
"megaraid: Cluster driver, initiator id:%d\n",
adapter->this_id);
}
#endif
pci_set_drvdata(pdev, host);
mega_create_proc_entry(hba_count, mega_proc_dir_entry);
error = scsi_add_host(host, &pdev->dev);
if (error)
goto out_free_mbox;
scsi_scan_host(host);
hba_count++;
return 0;
out_free_mbox:
pci_free_consistent(adapter->dev, sizeof(mbox64_t),
adapter->una_mbox64, adapter->una_mbox64_dma);
out_free_irq:
free_irq(adapter->host->irq, adapter);
out_free_scb_list:
kfree(adapter->scb_list);
out_free_cmd_buffer:
pci_free_consistent(adapter->dev, MEGA_BUFFER_SIZE,
adapter->mega_buffer, adapter->buf_dma_handle);
out_host_put:
scsi_host_put(host);
out_iounmap:
if (flag & BOARD_MEMMAP)
iounmap((void *)mega_baseport);
out_release_region:
if (flag & BOARD_MEMMAP)
release_mem_region(tbase, 128);
else
release_region(mega_baseport, 16);
out_disable_device:
pci_disable_device(pdev);
out:
return error;
}
static void
__megaraid_shutdown(adapter_t *adapter)
{
u_char raw_mbox[sizeof(struct mbox_out)];
mbox_t *mbox = (mbox_t *)raw_mbox;
int i;
/* Flush adapter cache */
memset(&mbox->m_out, 0, sizeof(raw_mbox));
raw_mbox[0] = FLUSH_ADAPTER;
free_irq(adapter->host->irq, adapter);
/* Issue a blocking (interrupts disabled) command to the card */
issue_scb_block(adapter, raw_mbox);
/* Flush disks cache */
memset(&mbox->m_out, 0, sizeof(raw_mbox));
raw_mbox[0] = FLUSH_SYSTEM;
/* Issue a blocking (interrupts disabled) command to the card */
issue_scb_block(adapter, raw_mbox);
if (atomic_read(&adapter->pend_cmds) > 0)
printk(KERN_WARNING "megaraid: pending commands!!\n");
/*
* Have a delibrate delay to make sure all the caches are
* actually flushed.
*/
for (i = 0; i <= 10; i++)
mdelay(1000);
}
[SCSI] megaraid: add __devexit annotation megaraid_remove_one() can become __devexit. Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2007-12-12 06:23:31 +08:00
static void __devexit
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!
2005-04-17 06:20:36 +08:00
megaraid_remove_one(struct pci_dev *pdev)
{
struct Scsi_Host *host = pci_get_drvdata(pdev);
adapter_t *adapter = (adapter_t *)host->hostdata;
scsi_remove_host(host);
__megaraid_shutdown(adapter);
/* Free our resources */
if (adapter->flag & BOARD_MEMMAP) {
iounmap((void *)adapter->base);
release_mem_region(adapter->host->base, 128);
} else
release_region(adapter->base, 16);
mega_free_sgl(adapter);
#ifdef CONFIG_PROC_FS
if (adapter->controller_proc_dir_entry) {
remove_proc_entry("stat", adapter->controller_proc_dir_entry);
remove_proc_entry("config",
adapter->controller_proc_dir_entry);
remove_proc_entry("mailbox",
adapter->controller_proc_dir_entry);
#if MEGA_HAVE_ENH_PROC
remove_proc_entry("rebuild-rate",
adapter->controller_proc_dir_entry);
remove_proc_entry("battery-status",
adapter->controller_proc_dir_entry);
remove_proc_entry("diskdrives-ch0",
adapter->controller_proc_dir_entry);
remove_proc_entry("diskdrives-ch1",
adapter->controller_proc_dir_entry);
remove_proc_entry("diskdrives-ch2",
adapter->controller_proc_dir_entry);
remove_proc_entry("diskdrives-ch3",
adapter->controller_proc_dir_entry);
remove_proc_entry("raiddrives-0-9",
adapter->controller_proc_dir_entry);
remove_proc_entry("raiddrives-10-19",
adapter->controller_proc_dir_entry);
remove_proc_entry("raiddrives-20-29",
adapter->controller_proc_dir_entry);
remove_proc_entry("raiddrives-30-39",
adapter->controller_proc_dir_entry);
#endif
[SCSI] megaraid: fix warnings when CONFIG_PROC_FS=n drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:4893: warning: implicit declaration of function 'mega_create_proc_entry' drivers/scsi/megaraid.c: In function 'megaraid_remove_one': drivers/scsi/megaraid.c:4968: warning: unused variable 'buf' Fix by adding #defines Signed-off-by: walter harms <wharms@bfs.de> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:35:09 +08:00
{
char buf[12] = { 0 };
sprintf(buf, "hba%d", adapter->host->host_no);
remove_proc_entry(buf, mega_proc_dir_entry);
}
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!
2005-04-17 06:20:36 +08:00
}
#endif
pci_free_consistent(adapter->dev, MEGA_BUFFER_SIZE,
adapter->mega_buffer, adapter->buf_dma_handle);
kfree(adapter->scb_list);
pci_free_consistent(adapter->dev, sizeof(mbox64_t),
adapter->una_mbox64, adapter->una_mbox64_dma);
scsi_host_put(host);
pci_disable_device(pdev);
hba_count--;
}
static void
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-24 08:35:56 +08:00
megaraid_shutdown(struct pci_dev *pdev)
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!
2005-04-17 06:20:36 +08:00
{
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-24 08:35:56 +08:00
struct Scsi_Host *host = pci_get_drvdata(pdev);
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!
2005-04-17 06:20:36 +08:00
adapter_t *adapter = (adapter_t *)host->hostdata;
__megaraid_shutdown(adapter);
}
static struct pci_device_id megaraid_pci_tbl[] = {
{PCI_VENDOR_ID_AMI, PCI_DEVICE_ID_AMI_MEGARAID,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{PCI_VENDOR_ID_AMI, PCI_DEVICE_ID_AMI_MEGARAID2,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_AMI_MEGARAID3,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{0,}
};
MODULE_DEVICE_TABLE(pci, megaraid_pci_tbl);
static struct pci_driver megaraid_pci_driver = {
[SCSI] megaraid_legacy: kobject_register failure Attached patch fixes problem that cause kobject_register failure during loading. Kobject_register would fail when there are more than 1 module with same module name. This patch will change module name of megaraid_legacy from 'megaraid' to 'megaraid_legacy'. Signed-Off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-10 03:32:59 +08:00
.name = "megaraid_legacy",
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!
2005-04-17 06:20:36 +08:00
.id_table = megaraid_pci_tbl,
.probe = megaraid_probe_one,
.remove = __devexit_p(megaraid_remove_one),
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-24 08:35:56 +08:00
.shutdown = megaraid_shutdown,
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!
2005-04-17 06:20:36 +08:00
};
static int __init megaraid_init(void)
{
int error;
if ((max_cmd_per_lun <= 0) || (max_cmd_per_lun > MAX_CMD_PER_LUN))
max_cmd_per_lun = MAX_CMD_PER_LUN;
if (max_mbox_busy_wait > MBOX_BUSY_WAIT)
max_mbox_busy_wait = MBOX_BUSY_WAIT;
#ifdef CONFIG_PROC_FS
proc: remove proc_root from drivers Remove proc_root export. Creation and removal works well if parent PDE is supplied as NULL -- it worked always that way. So, one useless export removed and consistency added, some drivers created PDEs with &proc_root as parent but removed them as NULL and so on. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 16:01:44 +08:00
mega_proc_dir_entry = proc_mkdir("megaraid", NULL);
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!
2005-04-17 06:20:36 +08:00
if (!mega_proc_dir_entry) {
printk(KERN_WARNING
"megaraid: failed to create megaraid root\n");
}
#endif
[SCSI] megaraid: pci_module_init to pci_register_driver Convert pci_module_init() to pci_register_driver(). Signed-off-by: Richard Knutsson <ricknu-0@student.ltu.se> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-02-14 08:40:39 +08:00
error = pci_register_driver(&megaraid_pci_driver);
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!
2005-04-17 06:20:36 +08:00
if (error) {
#ifdef CONFIG_PROC_FS
proc: remove proc_root from drivers Remove proc_root export. Creation and removal works well if parent PDE is supplied as NULL -- it worked always that way. So, one useless export removed and consistency added, some drivers created PDEs with &proc_root as parent but removed them as NULL and so on. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 16:01:44 +08:00
remove_proc_entry("megaraid", NULL);
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!
2005-04-17 06:20:36 +08:00
#endif
return error;
}
/*
* Register the driver as a character device, for applications
* to access it for ioctls.
* First argument (major) to register_chrdev implies a dynamic
* major number allocation.
*/
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
major = register_chrdev(0, "megadev_legacy", &megadev_fops);
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!
2005-04-17 06:20:36 +08:00
if (!major) {
printk(KERN_WARNING
"megaraid: failed to register char device\n");
}
return 0;
}
static void __exit megaraid_exit(void)
{
/*
* Unregister the character device interface to the driver.
*/
[SCSI] megaraid_legacy: removed PCI ID overlap from the driv er This patch fixes - PCI ID overlap issue - node name changed to 'megaraid_legacy' I hope this patch addresses concerns brought by Daniel Drake. Signed-off by: Seokmann Ju <seokmann.ju@enginio.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:13:31 +08:00
unregister_chrdev(major, "megadev_legacy");
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!
2005-04-17 06:20:36 +08:00
pci_unregister_driver(&megaraid_pci_driver);
#ifdef CONFIG_PROC_FS
proc: remove proc_root from drivers Remove proc_root export. Creation and removal works well if parent PDE is supplied as NULL -- it worked always that way. So, one useless export removed and consistency added, some drivers created PDEs with &proc_root as parent but removed them as NULL and so on. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 16:01:44 +08:00
remove_proc_entry("megaraid", NULL);
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!
2005-04-17 06:20:36 +08:00
#endif
}
module_init(megaraid_init);
module_exit(megaraid_exit);
/* vi: set ts=8 sw=8 tw=78: */