- Kernel Parameters
- ~~~~~~~~~~~~~~~~~
+ Kernel Parameters
+ ~~~~~~~~~~~~~~~~~
The following is a consolidated list of the kernel parameters as implemented
(mostly) by the __setup() macro and sorted into English Dictionary order
reboot= [BUGS=X86-32,BUGS=ARM,BUGS=IA-64] Rebooting mode
Format: <reboot_mode>[,<reboot_mode2>[,...]]
- See arch/*/kernel/reboot.c or arch/*/kernel/process.c
+ See arch/*/kernel/reboot.c or arch/*/kernel/process.c
reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
selinux_compat_net =
[SELINUX] Set initial selinux_compat_net flag value.
- Format: { "0" | "1" }
- 0 -- use new secmark-based packet controls
- 1 -- use legacy packet controls
- Default value is 0 (preferred).
- Value can be changed at runtime via
- /selinux/compat_net.
+ Format: { "0" | "1" }
+ 0 -- use new secmark-based packet controls
+ 1 -- use legacy packet controls
+ Default value is 0 (preferred).
+ Value can be changed at runtime via
+ /selinux/compat_net.
serialnumber [BUGS=X86-32]
norandmaps Don't use address space randomization
Equivalent to echo 0 > /proc/sys/kernel/randomize_va_space
- unwind_debug=N N > 0 will enable dwarf2 unwinder debugging
+ unwind_debug=N N > 0 will enable dwarf2 unwinder debugging
This is useful to get more information why
you got a "dwarf2 unwinder stuck"
static void xen_write_cr4(unsigned long cr4)
{
- /* never allow TSC to be disabled */
- native_write_cr4(cr4 & ~X86_CR4_TSD);
+ /* Just ignore cr4 changes; Xen doesn't allow us to do
+ anything anyway. */
}
static unsigned long xen_read_cr3(void)
struct cpu_purr_data *p = &__get_cpu_var(cpu_purr_data);
local_irq_save(flags);
- p->tb = mftb();
+ p->tb = get_tb_or_rtc();
p->purr = mfspr(SPRN_PURR);
wmb();
p->initialized = 1;
*/
void snapshot_timebase(void)
{
- __get_cpu_var(last_jiffy) = get_tb();
+ __get_cpu_var(last_jiffy) = get_tb_or_rtc();
snapshot_purr();
}
write_seqlock(&xtime_lock);
tb_next_jiffy = tb_last_jiffy + tb_ticks_per_jiffy;
+ if (__USE_RTC() && tb_next_jiffy >= 1000000000)
+ tb_next_jiffy -= 1000000000;
if (per_cpu(last_jiffy, cpu) >= tb_next_jiffy) {
tb_last_jiffy = tb_next_jiffy;
do_timer(1);
tb_to_ns_scale = scale;
tb_to_ns_shift = shift;
/* Save the current timebase to pretty up CONFIG_PRINTK_TIME */
- boot_tb = get_tb();
+ boot_tb = get_tb_or_rtc();
tm = get_boot_time();
CPU_FTR_USE_TB, 0,
"__kernel_gettimeofday", NULL
},
+ {
+ CPU_FTR_USE_TB, 0,
+ "__kernel_clock_gettime", NULL
+ },
+ {
+ CPU_FTR_USE_TB, 0,
+ "__kernel_clock_getres", NULL
+ },
+ {
+ CPU_FTR_USE_TB, 0,
+ "__kernel_get_tbfreq", NULL
+ },
};
/*
list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) {
struct spu_context *tmp = spu->ctx;
- if (tmp->prio > ctx->prio &&
+ if (tmp && tmp->prio > ctx->prio &&
(!victim || tmp->prio > victim->prio))
victim = spu->ctx;
}
mutex_lock(&cbe_spu_info[node].list_mutex);
cbe_spu_info[node].nr_active--;
+ spu_unbind_context(spu, victim);
mutex_unlock(&cbe_spu_info[node].list_mutex);
- spu_unbind_context(spu, victim);
victim->stats.invol_ctx_switch++;
spu->stats.invol_ctx_switch++;
mutex_unlock(&victim->state_mutex);
extern void copy_sc(union uml_pt_regs *regs, void *from);
-unsigned long to_irq_stack(int sig, unsigned long *mask_out);
+extern unsigned long to_irq_stack(unsigned long *mask_out);
unsigned long from_irq_stack(int nested);
#endif
static unsigned long pending_mask;
-unsigned long to_irq_stack(int sig, unsigned long *mask_out)
+unsigned long to_irq_stack(unsigned long *mask_out)
{
struct thread_info *ti;
unsigned long mask, old;
int nested;
- mask = xchg(&pending_mask, 1 << sig);
+ mask = xchg(&pending_mask, *mask_out);
if(mask != 0){
/* If any interrupts come in at this point, we want to
* make sure that their bits aren't lost by our
* and pending_mask contains a bit for each interrupt
* that came in.
*/
- old = 1 << sig;
+ old = *mask_out;
do {
old |= mask;
mask = xchg(&pending_mask, old);
task = cpu_tasks[ti->cpu].task;
tti = task_thread_info(task);
+
*ti = *tti;
ti->real_thread = tti;
task->stack = ti;
}
if(S_ISBLK(buf.ust_mode)){
- int fd, blocks;
+ int fd;
+ long blocks;
fd = os_open_file(file, of_read(OPENFLAGS()), 0);
if(fd < 0){
void handle_signal(int sig, struct sigcontext *sc)
{
- unsigned long pending = 0;
+ unsigned long pending = 1UL << sig;
do {
int nested, bail;
* have to return, and the upper handler will deal
* with this interrupt.
*/
- bail = to_irq_stack(sig, &pending);
+ bail = to_irq_stack(&pending);
if(bail)
return;
if (unlikely(in_atomic() || !mm))
goto bad_area_nosemaphore;
+ /*
+ * User-mode registers count as a user access even for any
+ * potential system fault or CPU buglet.
+ */
+ if (user_mode_vm(regs))
+ error_code |= PF_USER;
+
again:
/* When running in the kernel we expect faults to occur only to
* addresses in user space. All other faults represent errors in the
goto out_subsys;
}
if (dev->parent) {
- error = sysfs_create_link(&dev->kobj, &dev->parent->kobj,
- "device");
- if (error)
- goto out_busid;
#ifdef CONFIG_SYSFS_DEPRECATED
{
- char * class_name = make_class_name(dev->class->name,
- &dev->kobj);
+ struct device *parent = dev->parent;
+ char *class_name;
+
+ /*
+ * In old sysfs stacked class devices had 'device'
+ * link pointing to real device instead of parent
+ */
+ while (parent->class && !parent->bus && parent->parent)
+ parent = parent->parent;
+
+ error = sysfs_create_link(&dev->kobj,
+ &parent->kobj,
+ "device");
+ if (error)
+ goto out_busid;
+
+ class_name = make_class_name(dev->class->name,
+ &dev->kobj);
if (class_name)
error = sysfs_create_link(&dev->parent->kobj,
&dev->kobj, class_name);
if (error)
goto out_device;
}
+#else
+ error = sysfs_create_link(&dev->kobj, &dev->parent->kobj,
+ "device");
+ if (error)
+ goto out_busid;
#endif
}
return 0;
#include <linux/genhd.h>
#include <linux/hdreg.h>
#include <linux/blkpg.h>
+#include <linux/dma-mapping.h>
#include <linux/interrupt.h>
#include <linux/ioport.h>
#include <linux/mm.h>
#define I830_GMCH_MEM_MASK 0x1
#define I830_GMCH_MEM_64M 0x1
#define I830_GMCH_MEM_128M 0
-#define I830_GMCH_GMS_MASK 0xF0
+#define I830_GMCH_GMS_MASK 0x70
#define I830_GMCH_GMS_DISABLED 0x00
#define I830_GMCH_GMS_LOCAL 0x10
#define I830_GMCH_GMS_STOLEN_512 0x20
#define INTEL_I830_ERRSTS 0x92
/* Intel 855GM/852GM registers */
+#define I855_GMCH_GMS_MASK 0xF0
#define I855_GMCH_GMS_STOLEN_0M 0x0
#define I855_GMCH_GMS_STOLEN_1M (0x1 << 4)
#define I855_GMCH_GMS_STOLEN_4M (0x2 << 4)
*/
if (IS_G33)
size = 0;
- switch (gmch_ctrl & I830_GMCH_GMS_MASK) {
+ switch (gmch_ctrl & I855_GMCH_GMS_MASK) {
case I855_GMCH_GMS_STOLEN_1M:
gtt_entries = MB(1) - KB(size);
break;
static struct pci_device_id ipmi_pci_devices[] = {
{ PCI_DEVICE(PCI_HP_VENDOR_ID, PCI_MMC_DEVICE_ID) },
- { PCI_DEVICE_CLASS(PCI_ERMC_CLASSCODE, PCI_ERMC_CLASSCODE_MASK) }
+ { PCI_DEVICE_CLASS(PCI_ERMC_CLASSCODE, PCI_ERMC_CLASSCODE_MASK) },
+ { 0, }
};
MODULE_DEVICE_TABLE(pci, ipmi_pci_devices);
/*
* Page types allocated by the device.
*/
-enum {
+enum mspec_page_type {
MSPEC_FETCHOP = 1,
MSPEC_CACHED,
MSPEC_UNCACHED
* One of these structures is allocated when an mspec region is mmaped. The
* structure is pointed to by the vma->vm_private_data field in the vma struct.
* This structure is used to record the addresses of the mspec pages.
+ * This structure is shared by all vma's that are split off from the
+ * original vma when split_vma()'s are done.
+ *
+ * The refcnt is incremented atomically because mm->mmap_sem does not
+ * protect in fork case where multiple tasks share the vma_data.
*/
struct vma_data {
atomic_t refcnt; /* Number of vmas sharing the data. */
- spinlock_t lock; /* Serialize access to the vma. */
+ spinlock_t lock; /* Serialize access to this structure. */
int count; /* Number of pages allocated. */
- int type; /* Type of pages allocated. */
+ enum mspec_page_type type; /* Type of pages allocated. */
+ int flags; /* See VMD_xxx below. */
+ unsigned long vm_start; /* Original (unsplit) base. */
+ unsigned long vm_end; /* Original (unsplit) end. */
unsigned long maddr[0]; /* Array of MSPEC addresses. */
};
+#define VMD_VMALLOCED 0x1 /* vmalloc'd rather than kmalloc'd */
+
/* used on shub2 to clear FOP cache in the HUB */
static unsigned long scratch_page[MAX_NUMNODES];
#define SH2_AMO_CACHE_ENTRIES 4
* mspec_open
*
* Called when a device mapping is created by a means other than mmap
- * (via fork, etc.). Increments the reference count on the underlying
- * mspec data so it is not freed prematurely.
+ * (via fork, munmap, etc.). Increments the reference count on the
+ * underlying mspec data so it is not freed prematurely.
*/
static void
mspec_open(struct vm_area_struct *vma)
mspec_close(struct vm_area_struct *vma)
{
struct vma_data *vdata;
- int i, pages, result, vdata_size;
+ int index, last_index, result;
+ unsigned long my_page;
vdata = vma->vm_private_data;
- if (!atomic_dec_and_test(&vdata->refcnt))
- return;
- pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
- vdata_size = sizeof(struct vma_data) + pages * sizeof(long);
- for (i = 0; i < pages; i++) {
- if (vdata->maddr[i] == 0)
+ BUG_ON(vma->vm_start < vdata->vm_start || vma->vm_end > vdata->vm_end);
+
+ spin_lock(&vdata->lock);
+ index = (vma->vm_start - vdata->vm_start) >> PAGE_SHIFT;
+ last_index = (vma->vm_end - vdata->vm_start) >> PAGE_SHIFT;
+ for (; index < last_index; index++) {
+ if (vdata->maddr[index] == 0)
continue;
/*
* Clear the page before sticking it back
* into the pool.
*/
- result = mspec_zero_block(vdata->maddr[i], PAGE_SIZE);
+ my_page = vdata->maddr[index];
+ vdata->maddr[index] = 0;
+ spin_unlock(&vdata->lock);
+ result = mspec_zero_block(my_page, PAGE_SIZE);
if (!result)
- uncached_free_page(vdata->maddr[i]);
+ uncached_free_page(my_page);
else
printk(KERN_WARNING "mspec_close(): "
"failed to zero page %i\n",
result);
+ spin_lock(&vdata->lock);
}
+ spin_unlock(&vdata->lock);
- if (vdata_size <= PAGE_SIZE)
- kfree(vdata);
- else
+ if (!atomic_dec_and_test(&vdata->refcnt))
+ return;
+
+ if (vdata->flags & VMD_VMALLOCED)
vfree(vdata);
+ else
+ kfree(vdata);
}
int index;
struct vma_data *vdata = vma->vm_private_data;
- index = (address - vma->vm_start) >> PAGE_SHIFT;
+ BUG_ON(address < vdata->vm_start || address >= vdata->vm_end);
+ index = (address - vdata->vm_start) >> PAGE_SHIFT;
maddr = (volatile unsigned long) vdata->maddr[index];
if (maddr == 0) {
maddr = uncached_alloc_page(numa_node_id());
* underlying pages.
*/
static int
-mspec_mmap(struct file *file, struct vm_area_struct *vma, int type)
+mspec_mmap(struct file *file, struct vm_area_struct *vma,
+ enum mspec_page_type type)
{
struct vma_data *vdata;
- int pages, vdata_size;
+ int pages, vdata_size, flags = 0;
if (vma->vm_pgoff != 0)
return -EINVAL;
vdata_size = sizeof(struct vma_data) + pages * sizeof(long);
if (vdata_size <= PAGE_SIZE)
vdata = kmalloc(vdata_size, GFP_KERNEL);
- else
+ else {
vdata = vmalloc(vdata_size);
+ flags = VMD_VMALLOCED;
+ }
if (!vdata)
return -ENOMEM;
memset(vdata, 0, vdata_size);
+ vdata->vm_start = vma->vm_start;
+ vdata->vm_end = vma->vm_end;
+ vdata->flags = flags;
vdata->type = type;
spin_lock_init(&vdata->lock);
vdata->refcnt = ATOMIC_INIT(1);
*/
static const struct drive_list_entry hpa_list[] = {
{ "ST340823A", NULL },
+ { "ST320413A", NULL },
{ NULL, NULL }
};
{
struct hd_driveid *id = drive->id;
ide_hwif_t *hwif = HWIF(drive);
- pmac_ide_hwif_t* pmif = (pmac_ide_hwif_t *)hwif->hwif_data;
int enable = 1;
- int map;
drive->using_dma = 0;
if (drive->media == ide_floppy)
{ USB_DEVICE(0x2304, 0x0301), .driver_info=PINNA_LINX_VD_IN_CAB_PAL },
{ USB_DEVICE(0x2304, 0x0419), .driver_info=PINNA_PCTV_BUNGEE_PAL_FM },
{ USB_DEVICE(0x2400, 0x4200), .driver_info=HPG_WINTV },
+ { }, /* terminate list */
};
MODULE_DEVICE_TABLE (usb, usbvision_table);
}
static struct pci_device_id cafe_nand_tbl[] = {
- { 0x11ab, 0x4100, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_MEMORY_FLASH << 8, 0xFFFF0 }
+ { 0x11ab, 0x4100, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_MEMORY_FLASH << 8, 0xFFFF0 },
+ { 0, }
};
MODULE_DEVICE_TABLE(pci, cafe_nand_tbl);
struct rtc_plat_data {
struct rtc_device *rtc;
void __iomem *ioaddr;
- unsigned long baseaddr;
+ resource_size_t baseaddr;
unsigned long last_jiffies;
int irq;
unsigned int irqen;
void __iomem *ioaddr_rtc;
size_t size_nvram;
size_t size;
- unsigned long baseaddr;
+ resource_size_t baseaddr;
unsigned long last_jiffies;
};
unsigned char interrupt_mask1;/* ISR1 masking */
unsigned char pvr_dtr_bit; /* Which PVR bit is DTR */
unsigned char pvr_dsr_bit; /* Which PVR bit is DSR */
+ unsigned int gis_shift;
int type; /* SAB82532 version */
/* Setting configuration bits while the transmitter is active
struct tty_struct *tty;
union sab82532_irq_status status;
unsigned long flags;
+ unsigned char gis;
spin_lock_irqsave(&up->port.lock, flags);
status.stat = 0;
- if (readb(&up->regs->r.gis) & SAB82532_GIS_ISA0)
+ gis = readb(&up->regs->r.gis) >> up->gis_shift;
+ if (gis & 1)
status.sreg.isr0 = readb(&up->regs->r.isr0);
- if (readb(&up->regs->r.gis) & SAB82532_GIS_ISA1)
+ if (gis & 2)
status.sreg.isr1 = readb(&up->regs->r.isr1);
tty = NULL;
transmit_chars(up, &status);
}
- spin_unlock(&up->port.lock);
-
- if (tty)
- tty_flip_buffer_push(tty);
-
- up++;
-
- spin_lock(&up->port.lock);
-
- status.stat = 0;
- if (readb(&up->regs->r.gis) & SAB82532_GIS_ISB0)
- status.sreg.isr0 = readb(&up->regs->r.isr0);
- if (readb(&up->regs->r.gis) & SAB82532_GIS_ISB1)
- status.sreg.isr1 = readb(&up->regs->r.isr1);
-
- tty = NULL;
- if (status.stat) {
- if ((status.sreg.isr0 & (SAB82532_ISR0_TCD | SAB82532_ISR0_TIME |
- SAB82532_ISR0_RFO | SAB82532_ISR0_RPF)) ||
- (status.sreg.isr1 & SAB82532_ISR1_BRK))
-
- tty = receive_chars(up, &status);
- if ((status.sreg.isr0 & SAB82532_ISR0_CDSC) ||
- (status.sreg.isr1 & (SAB82532_ISR1_BRK | SAB82532_ISR1_CSC)))
- check_status(up, &status);
- if (status.sreg.isr1 & (SAB82532_ISR1_ALLS | SAB82532_ISR1_XPR))
- transmit_chars(up, &status);
- }
-
spin_unlock_irqrestore(&up->port.lock, flags);
if (tty)
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
unsigned char tmp;
+ int err = request_irq(up->port.irq, sunsab_interrupt,
+ IRQF_SHARED, "sab", up);
+ if (err)
+ return err;
spin_lock_irqsave(&up->port.lock, flags);
#endif
spin_unlock_irqrestore(&up->port.lock, flags);
+ free_irq(up->port.irq, up);
}
/*
if ((up->port.line & 0x1) == 0) {
up->pvr_dsr_bit = (1 << 0);
up->pvr_dtr_bit = (1 << 1);
+ up->gis_shift = 2;
} else {
up->pvr_dsr_bit = (1 << 3);
up->pvr_dtr_bit = (1 << 2);
+ up->gis_shift = 0;
}
up->cached_pvr = (1 << 1) | (1 << 2) | (1 << 4);
writeb(up->cached_pvr, &up->regs->w.pvr);
up->tec_timeout = SAB82532_MAX_TEC_TIMEOUT;
up->cec_timeout = SAB82532_MAX_CEC_TIMEOUT;
- if (!(up->port.line & 0x01)) {
- int err;
-
- err = request_irq(up->port.irq, sunsab_interrupt,
- IRQF_SHARED, "sab", up);
- if (err) {
- of_iounmap(&op->resource[0],
- up->port.membase,
- sizeof(union sab82532_async_regs));
- return err;
- }
- }
-
return 0;
}
0,
(inst * 2) + 0);
if (err)
- return err;
+ goto out;
err = sunsab_init_one(&up[1], op,
sizeof(union sab82532_async_regs),
(inst * 2) + 1);
- if (err) {
- of_iounmap(&op->resource[0],
- up[0].port.membase,
- sizeof(union sab82532_async_regs));
- free_irq(up[0].port.irq, &up[0]);
- return err;
- }
+ if (err)
+ goto out1;
sunserial_console_match(SUNSAB_CONSOLE(), op->node,
&sunsab_reg, up[0].port.line);
- uart_add_one_port(&sunsab_reg, &up[0].port);
sunserial_console_match(SUNSAB_CONSOLE(), op->node,
&sunsab_reg, up[1].port.line);
- uart_add_one_port(&sunsab_reg, &up[1].port);
+
+ err = uart_add_one_port(&sunsab_reg, &up[0].port);
+ if (err)
+ goto out2;
+
+ err = uart_add_one_port(&sunsab_reg, &up[1].port);
+ if (err)
+ goto out3;
dev_set_drvdata(&op->dev, &up[0]);
inst++;
return 0;
-}
-
-static void __devexit sab_remove_one(struct uart_sunsab_port *up)
-{
- struct of_device *op = to_of_device(up->port.dev);
- uart_remove_one_port(&sunsab_reg, &up->port);
- if (!(up->port.line & 1))
- free_irq(up->port.irq, up);
+out3:
+ uart_remove_one_port(&sunsab_reg, &up[0].port);
+out2:
of_iounmap(&op->resource[0],
- up->port.membase,
+ up[1].port.membase,
sizeof(union sab82532_async_regs));
+out1:
+ of_iounmap(&op->resource[0],
+ up[0].port.membase,
+ sizeof(union sab82532_async_regs));
+out:
+ return err;
}
static int __devexit sab_remove(struct of_device *op)
{
struct uart_sunsab_port *up = dev_get_drvdata(&op->dev);
- sab_remove_one(&up[0]);
- sab_remove_one(&up[1]);
+ uart_remove_one_port(&sunsab_reg, &up[1].port);
+ uart_remove_one_port(&sunsab_reg, &up[0].port);
+ of_iounmap(&op->resource[0],
+ up[1].port.membase,
+ sizeof(union sab82532_async_regs));
+ of_iounmap(&op->resource[0],
+ up[0].port.membase,
+ sizeof(union sab82532_async_regs));
dev_set_drvdata(&op->dev, NULL);
sunsab_reg.minor = sunserial_current_minor;
sunsab_reg.nr = num_channels;
+ sunsab_reg.cons = SUNSAB_CONSOLE();
err = uart_register_driver(&sunsab_reg);
if (err) {
/* turn off PLL */
tmp = INREG(dpll_reg);
- dpll_reg &= ~DPLL_VCO_ENABLE;
+ tmp &= ~DPLL_VCO_ENABLE;
OUTREG(dpll_reg, tmp);
/* Set PLL parameters */
struct dx_map_entry
{
u32 hash;
- u32 offs;
+ u16 offs;
+ u16 size;
};
#ifdef CONFIG_EXT3_INDEX
entries = (struct dx_entry *) (((char *)&root->info) +
root->info.info_length);
- assert(dx_get_limit(entries) == dx_root_limit(dir,
- root->info.info_length));
+
+ if (dx_get_limit(entries) != dx_root_limit(dir,
+ root->info.info_length)) {
+ ext3_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: limit != root limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail;
+ }
+
dxtrace (printk("Look up %x", hash));
while (1)
{
count = dx_get_count(entries);
- assert (count && count <= dx_get_limit(entries));
+ if (!count || count > dx_get_limit(entries)) {
+ ext3_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: no count or count > limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail2;
+ }
+
p = entries + 1;
q = entries + count - 1;
while (p <= q)
if (!(bh = ext3_bread (NULL,dir, dx_get_block(at), 0, err)))
goto fail2;
at = entries = ((struct dx_node *) bh->b_data)->entries;
- assert (dx_get_limit(entries) == dx_node_limit (dir));
+ if (dx_get_limit(entries) != dx_node_limit (dir)) {
+ ext3_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: limit != node limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail2;
+ }
frame++;
+ frame->bh = NULL;
}
fail2:
while (frame >= frame_in) {
frame--;
}
fail:
+ if (*err == ERR_BAD_DX_DIR)
+ ext3_warning(dir->i_sb, __FUNCTION__,
+ "Corrupt dir inode %ld, running e2fsck is "
+ "recommended.", dir->i_ino);
return NULL;
}
* Directory block splitting, compacting
*/
+/*
+ * Create map of hash values, offsets, and sizes, stored at end of block.
+ * Returns number of entries mapped.
+ */
static int dx_make_map (struct ext3_dir_entry_2 *de, int size,
struct dx_hash_info *hinfo, struct dx_map_entry *map_tail)
{
ext3fs_dirhash(de->name, de->name_len, &h);
map_tail--;
map_tail->hash = h.hash;
- map_tail->offs = (u32) ((char *) de - base);
+ map_tail->offs = (u16) ((char *) de - base);
+ map_tail->size = le16_to_cpu(de->rec_len);
count++;
cond_resched();
}
return count;
}
+/* Sort map by hash value */
static void dx_sort_map (struct dx_map_entry *map, unsigned count)
{
struct dx_map_entry *p, *q, *top = map + count - 1;
}
#ifdef CONFIG_EXT3_INDEX
+/*
+ * Move count entries from end of map between two memory locations.
+ * Returns pointer to last entry moved.
+ */
static struct ext3_dir_entry_2 *
dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count)
{
return (struct ext3_dir_entry_2 *) (to - rec_len);
}
+/*
+ * Compact each dir entry in the range to the minimal rec_len.
+ * Returns pointer to last entry in range.
+ */
static struct ext3_dir_entry_2* dx_pack_dirents(char *base, int size)
{
struct ext3_dir_entry_2 *next, *to, *prev, *de = (struct ext3_dir_entry_2 *) base;
return prev;
}
+/*
+ * Split a full leaf block to make room for a new dir entry.
+ * Allocate a new block, and move entries so that they are approx. equally full.
+ * Returns pointer to de in block into which the new entry will be inserted.
+ */
static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
struct buffer_head **bh,struct dx_frame *frame,
struct dx_hash_info *hinfo, int *error)
u32 hash2;
struct dx_map_entry *map;
char *data1 = (*bh)->b_data, *data2;
- unsigned split;
+ unsigned split, move, size, i;
struct ext3_dir_entry_2 *de = NULL, *de2;
int err = 0;
count = dx_make_map ((struct ext3_dir_entry_2 *) data1,
blocksize, hinfo, map);
map -= count;
- split = count/2; // need to adjust to actual middle
dx_sort_map (map, count);
+ /* Split the existing block in the middle, size-wise */
+ size = 0;
+ move = 0;
+ for (i = count-1; i >= 0; i--) {
+ /* is more than half of this entry in 2nd half of the block? */
+ if (size + map[i].size/2 > blocksize/2)
+ break;
+ size += map[i].size;
+ move++;
+ }
+ /* map index at which we will split */
+ split = count - move;
hash2 = map[split].hash;
continued = hash2 == map[split - 1].hash;
dxtrace(printk("Split block %i at %x, %i/%i\n",
struct dx_map_entry
{
u32 hash;
- u32 offs;
+ u16 offs;
+ u16 size;
};
#ifdef CONFIG_EXT4_INDEX
entries = (struct dx_entry *) (((char *)&root->info) +
root->info.info_length);
- assert(dx_get_limit(entries) == dx_root_limit(dir,
- root->info.info_length));
+
+ if (dx_get_limit(entries) != dx_root_limit(dir,
+ root->info.info_length)) {
+ ext4_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: limit != root limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail;
+ }
+
dxtrace (printk("Look up %x", hash));
while (1)
{
count = dx_get_count(entries);
- assert (count && count <= dx_get_limit(entries));
+ if (!count || count > dx_get_limit(entries)) {
+ ext4_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: no count or count > limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail2;
+ }
+
p = entries + 1;
q = entries + count - 1;
while (p <= q)
if (!(bh = ext4_bread (NULL,dir, dx_get_block(at), 0, err)))
goto fail2;
at = entries = ((struct dx_node *) bh->b_data)->entries;
- assert (dx_get_limit(entries) == dx_node_limit (dir));
+ if (dx_get_limit(entries) != dx_node_limit (dir)) {
+ ext4_warning(dir->i_sb, __FUNCTION__,
+ "dx entry: limit != node limit");
+ brelse(bh);
+ *err = ERR_BAD_DX_DIR;
+ goto fail2;
+ }
frame++;
+ frame->bh = NULL;
}
fail2:
while (frame >= frame_in) {
frame--;
}
fail:
+ if (*err == ERR_BAD_DX_DIR)
+ ext4_warning(dir->i_sb, __FUNCTION__,
+ "Corrupt dir inode %ld, running e2fsck is "
+ "recommended.", dir->i_ino);
return NULL;
}
* Directory block splitting, compacting
*/
+/*
+ * Create map of hash values, offsets, and sizes, stored at end of block.
+ * Returns number of entries mapped.
+ */
static int dx_make_map (struct ext4_dir_entry_2 *de, int size,
struct dx_hash_info *hinfo, struct dx_map_entry *map_tail)
{
ext4fs_dirhash(de->name, de->name_len, &h);
map_tail--;
map_tail->hash = h.hash;
- map_tail->offs = (u32) ((char *) de - base);
+ map_tail->offs = (u16) ((char *) de - base);
+ map_tail->size = le16_to_cpu(de->rec_len);
count++;
cond_resched();
}
return count;
}
+/* Sort map by hash value */
static void dx_sort_map (struct dx_map_entry *map, unsigned count)
{
struct dx_map_entry *p, *q, *top = map + count - 1;
}
#ifdef CONFIG_EXT4_INDEX
+/*
+ * Move count entries from end of map between two memory locations.
+ * Returns pointer to last entry moved.
+ */
static struct ext4_dir_entry_2 *
dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count)
{
return (struct ext4_dir_entry_2 *) (to - rec_len);
}
+/*
+ * Compact each dir entry in the range to the minimal rec_len.
+ * Returns pointer to last entry in range.
+ */
static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size)
{
struct ext4_dir_entry_2 *next, *to, *prev, *de = (struct ext4_dir_entry_2 *) base;
return prev;
}
+/*
+ * Split a full leaf block to make room for a new dir entry.
+ * Allocate a new block, and move entries so that they are approx. equally full.
+ * Returns pointer to de in block into which the new entry will be inserted.
+ */
static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
struct buffer_head **bh,struct dx_frame *frame,
struct dx_hash_info *hinfo, int *error)
u32 hash2;
struct dx_map_entry *map;
char *data1 = (*bh)->b_data, *data2;
- unsigned split;
+ unsigned split, move, size, i;
struct ext4_dir_entry_2 *de = NULL, *de2;
int err = 0;
count = dx_make_map ((struct ext4_dir_entry_2 *) data1,
blocksize, hinfo, map);
map -= count;
- split = count/2; // need to adjust to actual middle
dx_sort_map (map, count);
+ /* Split the existing block in the middle, size-wise */
+ size = 0;
+ move = 0;
+ for (i = count-1; i >= 0; i--) {
+ /* is more than half of this entry in 2nd half of the block? */
+ if (size + map[i].size/2 > blocksize/2)
+ break;
+ size += map[i].size;
+ move++;
+ }
+ /* map index at which we will split */
+ split = count - move;
hash2 = map[split].hash;
continued = hash2 == map[split - 1].hash;
dxtrace(printk("Split block %i at %x, %i/%i\n",
unregister_shrinker(&acl_shrinker);
#ifdef CONFIG_NFS_V4
unregister_filesystem(&nfs4_fs_type);
- nfs_unregister_sysctl();
#endif
+ nfs_unregister_sysctl();
unregister_filesystem(&nfs_fs_type);
}
ip->i_d.di_size = isize;
ip->i_update_core = 1;
ip->i_update_size = 1;
+ mark_inode_dirty_sync(vn_to_inode(ioend->io_vnode));
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (vp) {
vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address);
- if (sync)
+ if (sync) {
+ filemap_fdatawait(inode->i_mapping);
flags |= FLUSH_SYNC;
+ }
error = bhv_vop_iflush(vp, flags);
if (error == EAGAIN)
error = sync? bhv_vop_iflush(vp, flags | FLUSH_LOG) : 0;
#define XFS_BLI_UDQUOT_BUF 0x4
#define XFS_BLI_PDQUOT_BUF 0x8
#define XFS_BLI_GDQUOT_BUF 0x10
+/*
+ * This flag indicates that the buffer contains newly allocated
+ * inodes.
+ */
+#define XFS_BLI_INODE_NEW_BUF 0x20
#define XFS_BLI_CHUNK 128
#define XFS_BLI_SHIFT 7
xfs_filestream_flush(
xfs_mount_t *mp)
{
- /* point in time flush, so keep the reaper running */
- xfs_mru_cache_flush(mp->m_filestream, 1);
+ xfs_mru_cache_flush(mp->m_filestream);
}
/*
/*ARGSUSED*/
STATIC void
xlog_recover_do_reg_buffer(
+ xfs_mount_t *mp,
xlog_recover_item_t *item,
xfs_buf_t *bp,
xfs_buf_log_format_t *buf_f)
unsigned int *data_map = NULL;
unsigned int map_size = 0;
int error;
+ int stale_buf = 1;
+
+ /*
+ * Scan through the on-disk inode buffer and attempt to
+ * determine if it has been written to since it was logged.
+ *
+ * - If any of the magic numbers are incorrect then the buffer is stale
+ * - If any of the modes are non-zero then the buffer is not stale
+ * - If all of the modes are zero and at least one of the generation
+ * counts is non-zero then the buffer is stale
+ *
+ * If the end result is a stale buffer then the log buffer is replayed
+ * otherwise it is skipped.
+ *
+ * This heuristic is not perfect. It can be improved by scanning the
+ * entire inode chunk for evidence that any of the inode clusters have
+ * been updated. To fix this problem completely we will need a major
+ * architectural change to the logging system.
+ */
+ if (buf_f->blf_flags & XFS_BLI_INODE_NEW_BUF) {
+ xfs_dinode_t *dip;
+ int inodes_per_buf;
+ int mode_count = 0;
+ int gen_count = 0;
+
+ stale_buf = 0;
+ inodes_per_buf = XFS_BUF_COUNT(bp) >> mp->m_sb.sb_inodelog;
+ for (i = 0; i < inodes_per_buf; i++) {
+ dip = (xfs_dinode_t *)xfs_buf_offset(bp,
+ i * mp->m_sb.sb_inodesize);
+ if (be16_to_cpu(dip->di_core.di_magic) !=
+ XFS_DINODE_MAGIC) {
+ stale_buf = 1;
+ break;
+ }
+ if (be16_to_cpu(dip->di_core.di_mode))
+ mode_count++;
+ if (be16_to_cpu(dip->di_core.di_gen))
+ gen_count++;
+ }
+
+ if (!mode_count && gen_count)
+ stale_buf = 1;
+ }
switch (buf_f->blf_type) {
case XFS_LI_BUF:
-1, 0, XFS_QMOPT_DOWARN,
"dquot_buf_recover");
}
- if (!error)
+ if (!error && stale_buf)
memcpy(xfs_buf_offset(bp,
(uint)bit << XFS_BLI_SHIFT), /* dest */
item->ri_buf[i].i_addr, /* source */
if (log->l_quotaoffs_flag & type)
return;
- xlog_recover_do_reg_buffer(item, bp, buf_f);
+ xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
}
/*
(XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) {
xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
} else {
- xlog_recover_do_reg_buffer(item, bp, buf_f);
+ xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
}
if (error)
return XFS_ERROR(error);
*/
if (!_xfs_mru_cache_migrate(mru, now)) {
mru->time_zero = now;
- if (!mru->next_reap)
- mru->next_reap = mru->grp_count * mru->grp_time;
+ if (!mru->queued) {
+ mru->queued = 1;
+ queue_delayed_work(xfs_mru_reap_wq, &mru->work,
+ mru->grp_count * mru->grp_time);
+ }
} else {
grp = (now - mru->time_zero) / mru->grp_time;
grp = (mru->lru_grp + grp) % mru->grp_count;
struct work_struct *work)
{
xfs_mru_cache_t *mru = container_of(work, xfs_mru_cache_t, work.work);
- unsigned long now;
+ unsigned long now, next;
ASSERT(mru && mru->lists);
if (!mru || !mru->lists)
return;
mutex_spinlock(&mru->lock);
- now = jiffies;
- if (mru->reap_all ||
- (mru->next_reap && time_after(now, mru->next_reap))) {
- if (mru->reap_all)
- now += mru->grp_count * mru->grp_time * 2;
- mru->next_reap = _xfs_mru_cache_migrate(mru, now);
- _xfs_mru_cache_clear_reap_list(mru);
+ next = _xfs_mru_cache_migrate(mru, jiffies);
+ _xfs_mru_cache_clear_reap_list(mru);
+
+ mru->queued = next;
+ if ((mru->queued > 0)) {
+ now = jiffies;
+ if (next <= now)
+ next = 0;
+ else
+ next -= now;
+ queue_delayed_work(xfs_mru_reap_wq, &mru->work, next);
}
- /*
- * the process that triggered the reap_all is responsible
- * for restating the periodic reap if it is required.
- */
- if (!mru->reap_all)
- queue_delayed_work(xfs_mru_reap_wq, &mru->work, mru->grp_time);
- mru->reap_all = 0;
mutex_spinunlock(&mru->lock, 0);
}
/* An extra list is needed to avoid reaping up to a grp_time early. */
mru->grp_count = grp_count + 1;
- mru->lists = kmem_alloc(mru->grp_count * sizeof(*mru->lists), KM_SLEEP);
+ mru->lists = kmem_zalloc(mru->grp_count * sizeof(*mru->lists), KM_SLEEP);
if (!mru->lists) {
err = ENOMEM;
mru->grp_time = grp_time;
mru->free_func = free_func;
- /* start up the reaper event */
- mru->next_reap = 0;
- mru->reap_all = 0;
- queue_delayed_work(xfs_mru_reap_wq, &mru->work, mru->grp_time);
-
*mrup = mru;
exit:
* Call xfs_mru_cache_flush() to flush out all cached entries, calling their
* free functions as they're deleted. When this function returns, the caller is
* guaranteed that all the free functions for all the elements have finished
- * executing.
- *
- * While we are flushing, we stop the periodic reaper event from triggering.
- * Normally, we want to restart this periodic event, but if we are shutting
- * down the cache we do not want it restarted. hence the restart parameter
- * where 0 = do not restart reaper and 1 = restart reaper.
+ * executing and the reaper is not running.
*/
void
xfs_mru_cache_flush(
- xfs_mru_cache_t *mru,
- int restart)
+ xfs_mru_cache_t *mru)
{
if (!mru || !mru->lists)
return;
- cancel_rearming_delayed_workqueue(xfs_mru_reap_wq, &mru->work);
-
mutex_spinlock(&mru->lock);
- mru->reap_all = 1;
- mutex_spinunlock(&mru->lock, 0);
+ if (mru->queued) {
+ mutex_spinunlock(&mru->lock, 0);
+ cancel_rearming_delayed_workqueue(xfs_mru_reap_wq, &mru->work);
+ mutex_spinlock(&mru->lock);
+ }
- queue_work(xfs_mru_reap_wq, &mru->work.work);
- flush_workqueue(xfs_mru_reap_wq);
+ _xfs_mru_cache_migrate(mru, jiffies + mru->grp_count * mru->grp_time);
+ _xfs_mru_cache_clear_reap_list(mru);
- mutex_spinlock(&mru->lock);
- WARN_ON_ONCE(mru->reap_all != 0);
- mru->reap_all = 0;
- if (restart)
- queue_delayed_work(xfs_mru_reap_wq, &mru->work, mru->grp_time);
mutex_spinunlock(&mru->lock, 0);
}
if (!mru || !mru->lists)
return;
- /* we don't want the reaper to restart here */
- xfs_mru_cache_flush(mru, 0);
+ xfs_mru_cache_flush(mru);
kmem_free(mru->lists, mru->grp_count * sizeof(*mru->lists));
kmem_free(mru, sizeof(*mru));
unsigned int grp_time; /* Time period spanned by grps. */
unsigned int lru_grp; /* Group containing time zero. */
unsigned long time_zero; /* Time first element was added. */
- unsigned long next_reap; /* Time that the reaper should
- next do something. */
- unsigned int reap_all; /* if set, reap all lists */
xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */
struct delayed_work work; /* Workqueue data for reaping. */
+ unsigned int queued; /* work has been queued */
} xfs_mru_cache_t;
int xfs_mru_cache_init(void);
int xfs_mru_cache_create(struct xfs_mru_cache **mrup, unsigned int lifetime_ms,
unsigned int grp_count,
xfs_mru_cache_free_func_t free_func);
-void xfs_mru_cache_flush(xfs_mru_cache_t *mru, int restart);
+void xfs_mru_cache_flush(xfs_mru_cache_t *mru);
void xfs_mru_cache_destroy(struct xfs_mru_cache *mru);
int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key,
void *value);
ASSERT(atomic_read(&bip->bli_refcount) > 0);
bip->bli_flags |= XFS_BLI_INODE_ALLOC_BUF;
+ bip->bli_format.blf_flags |= XFS_BLI_INODE_NEW_BUF;
}
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return XFS_ERROR(EIO);
+ if (flag & FSYNC_DATA)
+ filemap_fdatawait(vn_to_inode(XFS_ITOV(ip))->i_mapping);
+
/*
* We always need to make sure that the required inode state
* is safe on disk. The vnode might be clean but because
sync_lsn = log->l_last_sync_lsn;
GRANT_UNLOCK(log, s);
- if ((XFS_LSN_CMP(iip->ili_last_lsn, sync_lsn) <= 0))
- return 0;
+ if ((XFS_LSN_CMP(iip->ili_last_lsn, sync_lsn) > 0)) {
+ if (flags & FLUSH_SYNC)
+ log_flags |= XFS_LOG_SYNC;
+ error = xfs_log_force(mp, iip->ili_last_lsn, log_flags);
+ if (error)
+ return error;
+ }
- if (flags & FLUSH_SYNC)
- log_flags |= XFS_LOG_SYNC;
- return xfs_log_force(mp, iip->ili_last_lsn, log_flags);
+ if (ip->i_update_core == 0)
+ return 0;
}
}
if (flags & FLUSH_INODE) {
int flush_flags;
- if (xfs_ipincount(ip))
- return EAGAIN;
-
if (flags & FLUSH_SYNC) {
xfs_ilock(ip, XFS_ILOCK_SHARED);
xfs_iflock(ip);
}
#endif /* !CONFIG_PPC64 */
+static inline u64 get_tb_or_rtc(void)
+{
+ return __USE_RTC() ? get_rtc() : get_tb();
+}
+
static inline void set_tb(unsigned int upper, unsigned int lower)
{
mtspr(SPRN_TBWL, 0);
extern struct mempolicy default_policy;
extern struct zonelist *huge_zonelist(struct vm_area_struct *vma,
- unsigned long addr, gfp_t gfp_flags);
+ unsigned long addr, gfp_t gfp_flags, struct mempolicy **mpol);
extern unsigned slab_node(struct mempolicy *policy);
extern enum zone_type policy_zone;
#define set_cpuset_being_rebound(x) do {} while (0)
static inline struct zonelist *huge_zonelist(struct vm_area_struct *vma,
- unsigned long addr, gfp_t gfp_flags)
+ unsigned long addr, gfp_t gfp_flags, struct mempolicy **mpol)
{
return NODE_DATA(0)->node_zonelists + gfp_zone(gfp_flags);
}
#endif
/* Hash table maintenance information */
- struct list_head uidhash_list;
+ struct hlist_node uidhash_node;
uid_t uid;
};
}
extern void free_uid(struct user_struct *);
extern void switch_uid(struct user_struct *);
+extern void release_uids(struct user_namespace *ns);
#include <asm/current.h>
struct user_namespace {
struct kref kref;
- struct list_head uidhash_table[UIDHASH_SZ];
+ struct hlist_head uidhash_table[UIDHASH_SZ];
struct user_struct *root_user;
};
config TIMERFD
bool "Enable timerfd() system call" if EMBEDDED
select ANON_INODES
+ depends on BROKEN
default y
help
Enable the timerfd() system call that allows to receive timer
pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD);
if (pid > 0)
- while (pid != sys_wait4(-1, NULL, 0, NULL))
+ while (pid != sys_wait4(-1, NULL, 0, NULL)) {
+ try_to_freeze();
yield();
+ }
/* move initrd to rootfs' /old */
sys_fchdir(old_fd);
/*
* These routines must be called with the uidhash spinlock held!
*/
-static inline void uid_hash_insert(struct user_struct *up, struct list_head *hashent)
+static inline void uid_hash_insert(struct user_struct *up, struct hlist_head *hashent)
{
- list_add(&up->uidhash_list, hashent);
+ hlist_add_head(&up->uidhash_node, hashent);
}
static inline void uid_hash_remove(struct user_struct *up)
{
- list_del(&up->uidhash_list);
+ hlist_del_init(&up->uidhash_node);
}
-static inline struct user_struct *uid_hash_find(uid_t uid, struct list_head *hashent)
+static inline struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
{
- struct list_head *up;
-
- list_for_each(up, hashent) {
- struct user_struct *user;
-
- user = list_entry(up, struct user_struct, uidhash_list);
+ struct user_struct *user;
+ struct hlist_node *h;
+ hlist_for_each_entry(user, h, hashent, uidhash_node) {
if(user->uid == uid) {
atomic_inc(&user->__count);
return user;
struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
{
- struct list_head *hashent = uidhashentry(ns, uid);
+ struct hlist_head *hashent = uidhashentry(ns, uid);
struct user_struct *up;
spin_lock_irq(&uidhash_lock);
suid_keys(current);
}
+void release_uids(struct user_namespace *ns)
+{
+ int i;
+ unsigned long flags;
+ struct hlist_head *head;
+ struct hlist_node *nd;
+
+ spin_lock_irqsave(&uidhash_lock, flags);
+ /*
+ * collapse the chains so that the user_struct-s will
+ * be still alive, but not in hashes. subsequent free_uid()
+ * will free them.
+ */
+ for (i = 0; i < UIDHASH_SZ; i++) {
+ head = ns->uidhash_table + i;
+ while (!hlist_empty(head)) {
+ nd = head->first;
+ hlist_del_init(nd);
+ }
+ }
+ spin_unlock_irqrestore(&uidhash_lock, flags);
+
+ free_uid(ns->root_user);
+}
static int __init uid_cache_init(void)
{
0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
for(n = 0; n < UIDHASH_SZ; ++n)
- INIT_LIST_HEAD(init_user_ns.uidhash_table + n);
+ INIT_HLIST_HEAD(init_user_ns.uidhash_table + n);
/* Insert the root user immediately (init already runs as root) */
spin_lock_irq(&uidhash_lock);
kref_init(&ns->kref);
for (n = 0; n < UIDHASH_SZ; ++n)
- INIT_LIST_HEAD(ns->uidhash_table + n);
+ INIT_HLIST_HEAD(ns->uidhash_table + n);
/* Insert new root user. */
ns->root_user = alloc_uid(ns, 0);
struct user_namespace *ns;
ns = container_of(kref, struct user_namespace, kref);
- free_uid(ns->root_user);
+ release_uids(ns);
kfree(ns);
}
if (!ns)
return ERR_PTR(-ENOMEM);
+ down_read(&uts_sem);
memcpy(&ns->name, &old_ns->name, sizeof(ns->name));
+ up_read(&uts_sem);
kref_init(&ns->kref);
return ns;
}
{
int nid;
struct page *page = NULL;
+ struct mempolicy *mpol;
struct zonelist *zonelist = huge_zonelist(vma, address,
- htlb_alloc_mask);
+ htlb_alloc_mask, &mpol);
struct zone **z;
for (z = zonelist->zones; *z; z++) {
break;
}
}
+ mpol_free(mpol); /* unref if mpol !NULL */
return page;
}
#endif
-/* Return effective policy for a VMA */
+/*
+ * get_vma_policy(@task, @vma, @addr)
+ * @task - task for fallback if vma policy == default
+ * @vma - virtual memory area whose policy is sought
+ * @addr - address in @vma for shared policy lookup
+ *
+ * Returns effective policy for a VMA at specified address.
+ * Falls back to @task or system default policy, as necessary.
+ * Returned policy has extra reference count if shared, vma,
+ * or some other task's policy [show_numa_maps() can pass
+ * @task != current]. It is the caller's responsibility to
+ * free the reference in these cases.
+ */
static struct mempolicy * get_vma_policy(struct task_struct *task,
struct vm_area_struct *vma, unsigned long addr)
{
struct mempolicy *pol = task->mempolicy;
+ int shared_pol = 0;
if (vma) {
- if (vma->vm_ops && vma->vm_ops->get_policy)
+ if (vma->vm_ops && vma->vm_ops->get_policy) {
pol = vma->vm_ops->get_policy(vma, addr);
- else if (vma->vm_policy &&
+ shared_pol = 1; /* if pol non-NULL, add ref below */
+ } else if (vma->vm_policy &&
vma->vm_policy->policy != MPOL_DEFAULT)
pol = vma->vm_policy;
}
if (!pol)
pol = &default_policy;
+ else if (!shared_pol && pol != current->mempolicy)
+ mpol_get(pol); /* vma or other task's policy */
return pol;
}
}
#ifdef CONFIG_HUGETLBFS
-/* Return a zonelist suitable for a huge page allocation. */
+/*
+ * huge_zonelist(@vma, @addr, @gfp_flags, @mpol)
+ * @vma = virtual memory area whose policy is sought
+ * @addr = address in @vma for shared policy lookup and interleave policy
+ * @gfp_flags = for requested zone
+ * @mpol = pointer to mempolicy pointer for reference counted 'BIND policy
+ *
+ * Returns a zonelist suitable for a huge page allocation.
+ * If the effective policy is 'BIND, returns pointer to policy's zonelist.
+ * If it is also a policy for which get_vma_policy() returns an extra
+ * reference, we must hold that reference until after allocation.
+ * In that case, return policy via @mpol so hugetlb allocation can drop
+ * the reference. For non-'BIND referenced policies, we can/do drop the
+ * reference here, so the caller doesn't need to know about the special case
+ * for default and current task policy.
+ */
struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr,
- gfp_t gfp_flags)
+ gfp_t gfp_flags, struct mempolicy **mpol)
{
struct mempolicy *pol = get_vma_policy(current, vma, addr);
+ struct zonelist *zl;
+ *mpol = NULL; /* probably no unref needed */
if (pol->policy == MPOL_INTERLEAVE) {
unsigned nid;
nid = interleave_nid(pol, vma, addr, HPAGE_SHIFT);
+ __mpol_free(pol); /* finished with pol */
return NODE_DATA(nid)->node_zonelists + gfp_zone(gfp_flags);
}
- return zonelist_policy(GFP_HIGHUSER, pol);
+
+ zl = zonelist_policy(GFP_HIGHUSER, pol);
+ if (unlikely(pol != &default_policy && pol != current->mempolicy)) {
+ if (pol->policy != MPOL_BIND)
+ __mpol_free(pol); /* finished with pol */
+ else
+ *mpol = pol; /* unref needed after allocation */
+ }
+ return zl;
}
#endif
alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr)
{
struct mempolicy *pol = get_vma_policy(current, vma, addr);
+ struct zonelist *zl;
cpuset_update_task_memory_state();
nid = interleave_nid(pol, vma, addr, PAGE_SHIFT);
return alloc_page_interleave(gfp, 0, nid);
}
- return __alloc_pages(gfp, 0, zonelist_policy(gfp, pol));
+ zl = zonelist_policy(gfp, pol);
+ if (pol != &default_policy && pol != current->mempolicy) {
+ /*
+ * slow path: ref counted policy -- shared or vma
+ */
+ struct page *page = __alloc_pages(gfp, 0, zl);
+ __mpol_free(pol);
+ return page;
+ }
+ /*
+ * fast path: default or task policy
+ */
+ return __alloc_pages(gfp, 0, zl);
}
/**
struct numa_maps *md;
struct file *file = vma->vm_file;
struct mm_struct *mm = vma->vm_mm;
+ struct mempolicy *pol;
int n;
char buffer[50];
if (!md)
return 0;
- mpol_to_str(buffer, sizeof(buffer),
- get_vma_policy(priv->task, vma, vma->vm_start));
+ pol = get_vma_policy(priv->task, vma, vma->vm_start);
+ mpol_to_str(buffer, sizeof(buffer), pol);
+ /*
+ * unref shared or other task's mempolicy
+ */
+ if (pol != &default_policy && pol != current->mempolicy)
+ __mpol_free(pol);
seq_printf(m, "%08lx %s", vma->vm_start, buffer);
projects / karo-tx-linux.git / commitdiff