ACPI source release updated (20020829)

ACPI source release updated (20020829)

[Grover, Andrew]

Patches against 2.4.20-pre5 and 2.5.32 are up on http://sf.net/projects/acpi
now. iasl and other source releases will be updated by tomorrow on the
developer.intel.com site, as usual.

----------------------------------------
29 August 2002.  Summary of changes for this release.

1) ACPI CA Core Subsystem Version 20020829:

If the target of a Scope() operator already exists, it must
be an object type that actually opens a scope -- such as a
Device, Method, Scope, etc.  This is a fatal runtime error.
Similar error check has been added to the iASL compiler
also.

Tightened up the namespace load to disallow multiple names
in the same scope.  This previously was allowed if both
objects were of the same type.  (i.e., a lookup was the same
as entering a new name).


2) Linux

Ensure that the ACPI interrupt has the proper trigger and
polarity.

local_irq_disable is extraneous. (Matthew Wilcox)

Make "acpi=off" actually do what it says, and not use the
ACPI interpreter *or* the tables.

Added arch-neutral support for parsing SLIT and SRAT tables
(Kochi Takayoshi)


3) iASL Compiler/Disassembler  Version 20020829:

Implemented namepath optimization for name declarations.
For example, a declaration like "Method (\_SB_.ABCD)" would
get optimized to "Method (ABCD)" if the declaration is
within the \_SB_ scope.  This optimization is in addition to
the named reference path optimization first released in the
previous version. This would seem to complete all possible
optimizations for namepaths within the ASL/AML.

If the target of a Scope() operator already exists, it must
be an object type that actually opens a scope -- such as a
Device, Method, Scope, etc.

Implemented a check and warning for unreachable code in the
same block below a Return() statement.

Fixed a problem where the listing file was not generated if
the compiler aborted if the maximum error count was exceeded
(200).

Fixed a problem where the typechecking of method return
values was broken.  This includes the check for a return
value when the method is invoked as a TermArg (a return
value is expected.)

Fixed a reported problem where EOF conditions during a
quoted string or comment caused a fault.

-----------------------------
Andrew Grover
Intel Labs / Mobile Architecture
andrew.grover-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: ACPI source release updated (20020829)

[KOCHI, Takayoshi]

[-- Attachment #1: Type: text/plain, Size: 15663 bytes --]

Hi Andy,

I checked difference today's releases between 2.4 and 2.5.
Except Makefiles, Config.help and toshiba_acpi.c, I found
some important differences.

I made patches to add some missed changes for both 2.4 and 2.5.
If this seems reasonable, please apply.

For 2.4:
 * bus.c
   - blacklist Compaq Presario 711FR was missing
   - removed unused argument from arguments of acpi_device_remove()
     (Should we remove whole #ifdef CONFIG_LDM for 2.4?)
   - fix remove_proc_entry("ACPI", ...) ->
         remove_proc_entry(ACPI_BUS_FILE_ROOT, ...)
   - remove unnecessary memory clear for acpi_fadt
   - remove unnecessary version check
 ... and several whitespace changes

For 2.5:
 * acpi_ksyms.c
   - export acpi_walk_namespace
 * battery.c
   - use acpi_os_free instead of kfree for those allocated
     by ACPI_ALLOCATE_BUFFER
 * system.c
   - remove garbage (#define ACPI_C)
 * include/linux/acpi.h
   - remove #else/#endif comments (Pavel Machek)
 ... and several whitespace changes

With these patches applied and Config.help, Makefiles, toshiba_acpi.c
and $Revision header-only difference excluded,
the difference looks like as follows.
IMO, as to bus.c and pci_irq.c, i386-specific mp* stuff should be
ifdef'ed so that ia64 can use the common sources.

diff -x CVS -Nuar acpi-2.4/drivers/acpi/Config.in acpi-2.5/drivers/acpi/Config.in
--- acpi-2.4/drivers/acpi/Config.in	Thu Aug 29 17:27:11 2002
+++ acpi-2.5/drivers/acpi/Config.in	Thu Aug 29 16:30:14 2002
@@ -23,7 +23,7 @@
       if [ "$CONFIG_PCI" = "y" ]; then
         define_bool CONFIG_ACPI_PCI		y
       fi
-      define_bool CONFIG_ACPI_SLEEP		y
+      define_bool CONFIG_ACPI_SLEEP		$CONFIG_SOFTWARE_SUSPEND
       define_bool CONFIG_ACPI_SYSTEM		y
       tristate     '  AC Adapter'	CONFIG_ACPI_AC
       tristate     '  Battery'		CONFIG_ACPI_BATTERY
@@ -34,6 +34,7 @@
       if [ "$CONFIG_NUMA" = "y" ]; then
         bool       '  NUMA support'     CONFIG_ACPI_NUMA $CONFIG_NUMA
       fi
+      tristate     '  Toshiba Laptop Extras'	CONFIG_ACPI_TOSHIBA
       bool         '  Debug Statements'	CONFIG_ACPI_DEBUG
     fi
   fi
diff -x CVS -Nuar acpi-2.4/drivers/acpi/bus.c acpi-2.5/drivers/acpi/bus.c
--- acpi-2.4/drivers/acpi/bus.c	Thu Aug 29 18:04:04 2002
+++ acpi-2.5/drivers/acpi/bus.c	Thu Aug 29 16:30:14 2002
@@ -1,5 +1,5 @@
 /*
- *  acpi_bus.c - ACPI Bus Driver ($Revision: 77 $)
+ *  acpi_bus.c - ACPI Bus Driver ($Revision: 80 $)
  *
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
  *
@@ -32,6 +32,7 @@
 #include <linux/sched.h>
 #include <linux/pm.h>
 #include <linux/proc_fs.h>
+#include <asm/mpspec.h>
 #include "acpi_bus.h"
 #include "acpi_drivers.h"
 #include "include/acinterp.h"	/* for acpi_ex_eisa_id_to_string() */
@@ -46,6 +47,8 @@
 
 #define	PREFIX			"ACPI: "
 
+extern void eisa_set_level_irq(unsigned int irq);
+
 extern int			acpi_disabled;
 
 FADT_DESCRIPTOR			acpi_fadt;
@@ -1964,6 +1967,12 @@
 		goto error1;
 	}
 
+	/* Ensure the SCI is set to level-triggered, active-low */
+	if (acpi_ioapic)
+		mp_override_legacy_irq(acpi_fadt.sci_int, 3, 3, acpi_fadt.sci_int);
+	else
+		eisa_set_level_irq(acpi_fadt.sci_int);
+
 	status = acpi_enable_subsystem(ACPI_FULL_INITIALIZATION);
 	if (ACPI_FAILURE(status)) {
 		printk(KERN_ERR PREFIX "Unable to start the ACPI Interpreter\n");
@@ -2168,5 +2177,6 @@
 	return 1;
 }
 
+subsys_initcall(acpi_init);
 
 __setup("acpi=", acpi_setup);
diff -x CVS -Nuar acpi-2.4/drivers/acpi/osl.c acpi-2.5/drivers/acpi/osl.c
--- acpi-2.4/drivers/acpi/osl.c	Thu Aug 29 17:44:27 2002
+++ acpi-2.5/drivers/acpi/osl.c	Wed Aug 28 00:13:27 2002
@@ -1,5 +1,5 @@
 /*
- *  acpi_osl.c - OS-dependent functions ($Revision: 80 $)
+ *  acpi_osl.c - OS-dependent functions ($Revision: 83 $)
  *
  *  Copyright (C) 2000       Andrew Henroid
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
@@ -33,6 +33,7 @@
 #include <linux/interrupt.h>
 #include <linux/kmod.h>
 #include <linux/delay.h>
+#include <linux/tqueue.h>
 #include <asm/io.h>
 #include "acpi.h"
 
@@ -79,7 +80,6 @@
 	 * it while walking the namespace (bus 0 and root bridges w/ _BBNs).
 	 */
 #ifdef CONFIG_ACPI_PCI
-	pcibios_config_init();
 	if (!pci_config_read || !pci_config_write) {
 		printk(KERN_ERR PREFIX "Access to PCI configuration space unavailable\n");
 		return AE_NULL_ENTRY;
diff -x CVS -Nuar acpi-2.4/drivers/acpi/pci_irq.c acpi-2.5/drivers/acpi/pci_irq.c
--- acpi-2.4/drivers/acpi/pci_irq.c	Thu Aug 29 16:58:11 2002
+++ acpi-2.5/drivers/acpi/pci_irq.c	Thu Aug 29 17:45:07 2002
@@ -33,6 +33,7 @@
 #include <linux/pm.h>
 #include <linux/pci.h>
 #include <linux/acpi.h>
+#include <asm/mpspec.h>
 #include "acpi_bus.h"
 #include "acpi_drivers.h"
 
diff -x CVS -Nuar acpi-2.4/drivers/acpi/processor.c acpi-2.5/drivers/acpi/processor.c
--- acpi-2.4/drivers/acpi/processor.c	Thu Aug 29 16:58:11 2002
+++ acpi-2.5/drivers/acpi/processor.c	Wed Aug 28 00:13:27 2002
@@ -1,5 +1,5 @@
 /*
- * acpi_processor.c - ACPI Processor Driver ($Revision: 69 $)
+ * acpi_processor.c - ACPI Processor Driver ($Revision: 71 $)
  *
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
@@ -416,7 +416,7 @@
 	 * Interrupts must be disabled during bus mastering calculations and
 	 * for C2/C3 transitions.
 	 */
-	__cli();
+	local_irq_disable();
 
 	cx = &(pr->power.states[pr->power.state]);
 
@@ -461,7 +461,7 @@
 		 *      issues (e.g. floppy DMA transfer overrun/underrun).
 		 */
 		if (pr->power.bm_activity & cx->demotion.threshold.bm) {
-			__sti();
+			local_irq_enable();
 			next_state = cx->demotion.state;
 			goto end;
 		}
@@ -497,7 +497,7 @@
 		/* Get end time (ticks) */
 		t2 = inl(acpi_fadt.Xpm_tmr_blk.address);
 		/* Re-enable interrupts */
-		__sti();
+		local_irq_enable();
 		/* Compute time (ticks) that we were actually asleep */
 		sleep_ticks = ticks_elapsed(t1, t2) - cx->latency_ticks - C2_OVERHEAD;
 		break;
@@ -516,13 +516,13 @@
 		/* Enable bus master arbitration */
 		acpi_set_register(ACPI_BITREG_ARB_DISABLE, 0, ACPI_MTX_DO_NOT_LOCK);
 		/* Re-enable interrupts */
-		__sti();
+		local_irq_enable();
 		/* Compute time (ticks) that we were actually asleep */
 		sleep_ticks = ticks_elapsed(t1, t2) - cx->latency_ticks - C3_OVERHEAD;
 		break;
 
 	default:
-		__sti();
+		local_irq_enable();
 		return;
 	}
 
@@ -1170,7 +1170,7 @@
 
 	pr->throttling.state = 0;
 
-	__cli();
+	local_irq_disable();
 
 	duty_mask = pr->throttling.state_count - 1;
 
@@ -1192,7 +1192,7 @@
 
 	pr->throttling.state = state;
 
-	__sti();
+	local_irq_enable();
 
 	ACPI_DEBUG_PRINT((ACPI_DB_INFO, 
 		"Throttling state is T%d (%d%% throttling applied)\n",
@@ -1225,7 +1225,7 @@
 	if (state == pr->throttling.state)
 		return_VALUE(0);
 
-	__cli();
+	local_irq_disable();
 
 	/*
 	 * Calculate the duty_value and duty_mask.
@@ -1267,7 +1267,7 @@
 
 	pr->throttling.state = state;
 
-	__sti();
+	local_irq_enable();
 
 	ACPI_DEBUG_PRINT((ACPI_DB_INFO, 
 		"Throttling state set to T%d (%d%%)\n", state, 
@@ -2066,10 +2066,8 @@
 	if (!pr)
 		return_VALUE(-EINVAL);
 
-#ifdef CONFIG_SMP
-	if (smp_num_cpus > 1)
-		errata.smp = smp_num_cpus;
-#endif
+	if (num_online_cpus() > 1)
+		errata.smp = TRUE;
 
 	acpi_processor_errata(pr);
 
diff -x CVS -Nuar acpi-2.4/drivers/acpi/system.c acpi-2.5/drivers/acpi/system.c
--- acpi-2.4/drivers/acpi/system.c	Thu Aug 29 16:58:11 2002
+++ acpi-2.5/drivers/acpi/system.c	Thu Aug 29 17:50:08 2002
@@ -1,5 +1,5 @@
 /*
- *  acpi_system.c - ACPI System Driver ($Revision: 57 $)
+ *  acpi_system.c - ACPI System Driver ($Revision: 63 $)
  *
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
@@ -32,9 +32,12 @@
 #include <linux/poll.h>
 #include <linux/delay.h>
 #include <linux/sysrq.h>
-#include <linux/pm.h>
 #include <linux/compatmac.h>
 #include <linux/proc_fs.h>
+#include <linux/irq.h>
+#include <linux/pm.h>
+#include <linux/device.h>
+#include <linux/suspend.h>
 #include <asm/uaccess.h>
 #include <asm/acpi.h>
 #include "acpi_bus.h"
@@ -99,6 +102,179 @@
 #ifdef CONFIG_ACPI_SLEEP
 
 /**
+ * acpi_system_restore_state - OS-specific restoration of state
+ * @state:	sleep state we're exiting
+ *
+ * Note that if we're coming back from S4, the memory image should have already
+ * been loaded from the disk and is already in place. (Otherwise how else would we
+ * be here?).
+ */
+acpi_status
+acpi_system_restore_state (
+	u32			state)
+{
+	/* restore processor state
+	 * We should only be here if we're coming back from STR or STD.
+	 * And, in the case of the latter, the memory image should have already
+	 * been loaded from disk.
+	 */
+	if (state > ACPI_STATE_S1)
+		acpi_restore_state_mem();
+
+	/* wait for power to come back */
+	mdelay(10);
+
+	/* turn all the devices back on */
+	device_resume(RESUME_POWER_ON);
+
+	/* enable interrupts once again */
+	ACPI_ENABLE_IRQS();
+
+	/* restore device context */
+	device_resume(RESUME_RESTORE_STATE);
+
+	if (dmi_broken & BROKEN_INIT_AFTER_S1) {
+		printk("Broken toshiba laptop -> kicking interrupts\n");
+		init_8259A(0);
+	}
+
+	return AE_OK;
+}
+
+/**
+ * acpi_system_save_state - save OS specific state and power down devices
+ * @state:	sleep state we're entering.
+ *
+ * This handles saving all context to memory, and possibly disk.
+ * First, we call to the device driver layer to save device state.
+ * Once we have that, we save whatevery processor and kernel state we
+ * need to memory.
+ * If we're entering S4, we then write the memory image to disk.
+ *
+ * Only then is it safe for us to power down devices, since we may need
+ * the disks and upstream buses to write to.
+ */
+acpi_status
+acpi_system_save_state(
+	u32			state)
+{
+	int			error = 0;
+
+	/* Send notification to devices that they will be suspended.
+	 * If any device or driver cannot make the transition, either up
+	 * or down, we'll get an error back.
+	 */
+	error = device_suspend(state, SUSPEND_NOTIFY);
+	if (error)
+		return AE_ERROR;
+
+	if (state < ACPI_STATE_S5) {
+
+		/* Tell devices to stop I/O and actually save their state.
+		 * It is theoretically possible that something could fail,
+		 * so handle that gracefully..
+		 */
+		error = device_suspend(state, SUSPEND_SAVE_STATE);
+		if (error) {
+			/* tell devices to restore state if they have
+			 * it saved and to start taking I/O requests.
+			 */
+			device_resume(RESUME_RESTORE_STATE);
+			return error;
+		}
+
+		/* flush caches */
+		ACPI_FLUSH_CPU_CACHE();
+
+		/* Do arch specific saving of state. */
+		if (state > ACPI_STATE_S1) {
+			error = acpi_save_state_mem();
+
+			if (!error && (state == ACPI_STATE_S4))
+				error = acpi_save_state_disk();
+
+			if (error) {
+				device_resume(RESUME_RESTORE_STATE);
+				return error;
+			}
+		}
+	}
+
+	/* disable interrupts
+	 * Note that acpi_suspend -- our caller -- will do this once we return.
+	 * But, we want it done early, so we don't get any suprises during
+	 * the device suspend sequence.
+	 */
+	ACPI_DISABLE_IRQS();
+
+	/* Unconditionally turn off devices.
+	 * Obvious if we enter a sleep state.
+	 * If entering S5 (soft off), this should put devices in a
+	 * quiescent state.
+	 */
+	error = device_suspend(state, SUSPEND_POWER_DOWN);
+
+	/* We're pretty screwed if we got an error from this.
+	 * We try to recover by simply calling our own restore_state
+	 * function; see above for definition.
+	 *
+	 * If it's S5 though, go through with it anyway..
+	 */
+	if (error && state != ACPI_STATE_S5)
+		acpi_system_restore_state(state);
+
+	return error ? AE_ERROR : AE_OK;
+}
+
+
+/****************************************************************************
+ *
+ * FUNCTION:    acpi_system_suspend
+ *
+ * PARAMETERS:  %state: Sleep state to enter.
+ *
+ * RETURN:      acpi_status, whether or not we successfully entered and
+ *              exited sleep.
+ *
+ * DESCRIPTION: Perform OS-specific action to enter sleep state.
+ *              This is the final step in going to sleep, per spec.  If we
+ *              know we're coming back (i.e. not entering S5), we save the
+ *              processor flags. [ We'll have to save and restore them anyway,
+ *              so we use the arch-agnostic save_flags and restore_flags
+ *              here.]  We then set the place to return to in arch-specific
+ *              globals using arch_set_return_point. Finally, we call the
+ *              ACPI function to write the proper values to I/O ports.
+ *
+ ****************************************************************************/
+
+acpi_status
+acpi_system_suspend(
+	u32			state)
+{
+	acpi_status		status = AE_ERROR;
+	unsigned long		flags = 0;
+
+	local_irq_save(flags);
+	
+	switch (state)
+	{
+	case ACPI_STATE_S1:
+		barrier();
+		status = acpi_enter_sleep_state(state);
+		break;
+
+	case ACPI_STATE_S2:
+	case ACPI_STATE_S3:
+		do_suspend_lowlevel(0);
+		break;
+	}
+	local_irq_restore(flags);
+
+	return status;
+}
+
+
+/**
  * acpi_suspend - OS-agnostic system suspend/resume support (S? states)
  * @state:	state we're entering
  *
@@ -109,28 +285,46 @@
 {
 	acpi_status status;
 
-	/* only support S1 and S5 on kernel 2.4 */
-	if (state != ACPI_STATE_S1 && state != ACPI_STATE_S5)
+	/* get out if state is invalid */
+	if (state < ACPI_STATE_S1 || state > ACPI_STATE_S5)
 		return AE_ERROR;
 
+	freeze_processes();		/* device_suspend needs processes to be stopped */
+
+	/* do we have a wakeup address for S2 and S3? */
+	if (state == ACPI_STATE_S2 || state == ACPI_STATE_S3) {
+		if (!acpi_wakeup_address)
+			return AE_ERROR;
+		acpi_set_firmware_waking_vector((ACPI_PHYSICAL_ADDRESS) acpi_wakeup_address);
+	}
+
 	acpi_enter_sleep_state_prep(state);
 
+	status = acpi_system_save_state(state);
+	if (!ACPI_SUCCESS(status))
+		return status;
+
 	/* disable interrupts and flush caches */
 	ACPI_DISABLE_IRQS();
 	ACPI_FLUSH_CPU_CACHE();
 
 	/* perform OS-specific sleep actions */
-	status = acpi_enter_sleep_state(state);
+	status = acpi_system_suspend(state);
 
 	/* Even if we failed to go to sleep, all of the devices are in an suspended
 	 * mode. So, we run these unconditionaly to make sure we have a usable system
 	 * no matter what.
 	 */
+	acpi_system_restore_state(state);
 	acpi_leave_sleep_state(state);
 
 	/* make sure interrupts are enabled */
 	ACPI_ENABLE_IRQS();
 
+	/* reset firmware waking vector */
+	acpi_set_firmware_waking_vector((ACPI_PHYSICAL_ADDRESS) 0);
+	thaw_processes();
+
 	return status;
 }
 
@@ -509,8 +703,16 @@
 	
 	if (!system->states[state])
 		return_VALUE(-ENODEV);
+
 	
+#ifdef CONFIG_SOFTWARE_SUSPEND
+	if (state == 4) {
+		software_suspend();
+		return_VALUE(count);
+	}
+#endif
 	status = acpi_suspend(state);
+
 	if (ACPI_FAILURE(status))
 		return_VALUE(-ENODEV);
 	
@@ -961,7 +1163,7 @@
 
 /* Simple wrapper calling power down function. */
 static void acpi_sysrq_power_off(int key, struct pt_regs *pt_regs,
-	struct kbd_struct *kbd, struct tty_struct *tty)
+	struct tty_struct *tty)
 {
 	acpi_power_off();
 }
@@ -1019,6 +1221,11 @@
 	if (system->states[ACPI_STATE_S5]) {
 		pm_power_off = acpi_power_off;
 		register_sysrq_key('o', &sysrq_acpi_poweroff_op);
+
+		/* workaround: some systems don't claim S4 support, but they
+                   do support S5 (power-down). That is all we need, so
+		   indicate support. */
+		system->states[ACPI_STATE_S4] = 1;
 	}
 #endif
 


Thanks,
-- 
KOCHI, Takayoshi <t-kouchi-f7IHDacdhdx8UrSeD/g0lQ@public.gmane.org/t-kouchi>

[-- Attachment #2: for-2.4-20020829.diff --]
[-- Type: application/octet-stream, Size: 3048 bytes --]

Index: acpi-2.4/drivers/acpi/Config.in
diff -u acpi-2.4/drivers/acpi/Config.in:1.1.1.1 acpi-2.4/drivers/acpi/Config.in:1.1.1.1.2.1
--- acpi-2.4/drivers/acpi/Config.in:1.1.1.1	Thu Aug 29 16:58:11 2002
+++ acpi-2.4/drivers/acpi/Config.in	Thu Aug 29 18:01:49 2002
@@ -30,10 +30,10 @@
       tristate     '  Button'		CONFIG_ACPI_BUTTON
       tristate     '  Fan'		CONFIG_ACPI_FAN
       tristate     '  Processor'	CONFIG_ACPI_PROCESSOR
+      dep_tristate '  Thermal Zone' 	CONFIG_ACPI_THERMAL $CONFIG_ACPI_PROCESSOR
       if [ "$CONFIG_NUMA" = "y" ]; then
         bool       '  NUMA support'     CONFIG_ACPI_NUMA $CONFIG_NUMA
       fi
-      dep_tristate '  Thermal Zone' 	CONFIG_ACPI_THERMAL $CONFIG_ACPI_PROCESSOR
       bool         '  Debug Statements'	CONFIG_ACPI_DEBUG
     fi
   fi
Index: acpi-2.4/drivers/acpi/bus.c
diff -u acpi-2.4/drivers/acpi/bus.c:1.1.1.1 acpi-2.4/drivers/acpi/bus.c:1.1.1.1.2.2
--- acpi-2.4/drivers/acpi/bus.c:1.1.1.1	Thu Aug 29 16:58:11 2002
+++ acpi-2.4/drivers/acpi/bus.c	Thu Aug 29 18:04:21 2002
@@ -72,6 +72,8 @@
 	{"ASUS  ", "K7M     ", 0x00001000, ACPI_TABLE_DSDT, less_than_or_equal, "Field beyond end of region", 0},
 	/* Intel 810 Motherboard? */
 	{"MNTRAL", "MO81010A", 0x00000012, ACPI_TABLE_DSDT, less_than_or_equal, "Field beyond end of region", 0},
+	/* Compaq Presario 711FR */
+	{"COMAPQ", "EAGLES", 0x06040000, ACPI_TABLE_DSDT, less_than_or_equal, "SCI issues (C2 disabled)", 0},
 	/* Compaq Presario 1700 */
 	{"PTLTD ", "  DSDT  ", 0x06040000, ACPI_TABLE_DSDT, less_than_or_equal, "Multiple problems", 1},
 	/* Sony FX120, FX140, FX150? */
@@ -91,7 +93,7 @@
 #ifdef CONFIG_LDM
 
 static int acpi_device_probe(struct device *dev);
-static int acpi_device_remove(struct device *dev, u32 flags);
+static int acpi_device_remove(struct device *dev);
 static int acpi_device_suspend(struct device *dev, u32 state, u32 stage);
 static int acpi_device_resume(struct device *dev, u32 stage);
 
@@ -120,8 +122,7 @@
 
 static int
 acpi_device_remove (
-	struct device		*dev,
-	u32			flags)
+	struct device		*dev)
 {
 	ACPI_FUNCTION_TRACE("acpi_device_remove");
 
@@ -2052,7 +2053,7 @@
 
 	/* Mimic structured exception handling */
 error4:
-	remove_proc_entry("ACPI", NULL);
+	remove_proc_entry(ACPI_BUS_FILE_ROOT, NULL);
 error3:
 	acpi_bus_remove(acpi_root, ACPI_BUS_REMOVAL_NORMAL);
 error2:
@@ -2109,8 +2110,6 @@
 
 	ACPI_FUNCTION_TRACE("acpi_init");
 
-	memset(&acpi_fadt, 0, sizeof(FADT_DESCRIPTOR));
-
 	printk(KERN_INFO PREFIX "Subsystem revision %08x\n",
 		ACPI_CA_VERSION);
 
@@ -2170,9 +2169,4 @@
 }
 
 
-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,5,0)
-subsys_initcall(acpi_init);
-#endif
-
 __setup("acpi=", acpi_setup);
-
Index: acpi-2.4/drivers/acpi/osl.c
diff -u acpi-2.4/drivers/acpi/osl.c:1.1.1.1 acpi-2.4/drivers/acpi/osl.c:1.1.1.1.2.1
--- acpi-2.4/drivers/acpi/osl.c:1.1.1.1	Thu Aug 29 16:58:11 2002
+++ acpi-2.4/drivers/acpi/osl.c	Thu Aug 29 18:01:49 2002
@@ -658,7 +658,6 @@
 		}
 	}
 	break;
-
 	default:
 		/*
 		 * Allocate/initialize DPC structure.  Note that this memory will be

[-- Attachment #3: for-2.5-20020829.diff --]
[-- Type: application/octet-stream, Size: 4750 bytes --]

Index: acpi-2.5/drivers/acpi/ac.c
diff -u acpi-2.5/drivers/acpi/ac.c:1.1.1.1 acpi-2.5/drivers/acpi/ac.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/ac.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/ac.c	Thu Aug 29 18:01:23 2002
@@ -51,7 +51,7 @@
 	.class =	ACPI_AC_CLASS,
 	.ids =		ACPI_AC_HID,
 	.ops =		{
-				.add =	acpi_ac_add,
+				.add =		acpi_ac_add,
 				.remove =	acpi_ac_remove,
 			},
 };
Index: acpi-2.5/drivers/acpi/acpi_ksyms.c
diff -u acpi-2.5/drivers/acpi/acpi_ksyms.c:1.1.1.1 acpi-2.5/drivers/acpi/acpi_ksyms.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/acpi_ksyms.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/acpi_ksyms.c	Thu Aug 29 18:01:23 2002
@@ -114,6 +114,7 @@
 #ifdef CONFIG_ACPI_BUS
 
 EXPORT_SYMBOL(acpi_fadt);
+EXPORT_SYMBOL(acpi_walk_namespace);
 EXPORT_SYMBOL(acpi_root_dir);
 EXPORT_SYMBOL(acpi_bus_get_device);
 EXPORT_SYMBOL(acpi_bus_get_status);
@@ -127,4 +128,3 @@
 EXPORT_SYMBOL(acpi_init);
 
 #endif /*CONFIG_ACPI_BUS*/
-
Index: acpi-2.5/drivers/acpi/battery.c
diff -u acpi-2.5/drivers/acpi/battery.c:1.1.1.1 acpi-2.5/drivers/acpi/battery.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/battery.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/battery.c	Thu Aug 29 18:01:23 2002
@@ -162,7 +162,7 @@
 	}
 
 end:
-	kfree(buffer.pointer);
+	acpi_os_free(buffer.pointer);
 
 	if (!result)
 		(*bif) = (struct acpi_battery_info *) data.pointer;
@@ -223,7 +223,7 @@
 	}
 
 end:
-	kfree(buffer.pointer);
+	acpi_os_free(buffer.pointer);
 
 	if (!result)
 		(*bst) = (struct acpi_battery_status *) data.pointer;
Index: acpi-2.5/drivers/acpi/button.c
diff -u acpi-2.5/drivers/acpi/button.c:1.1.1.1 acpi-2.5/drivers/acpi/button.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/button.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/button.c	Thu Aug 29 18:01:23 2002
@@ -70,7 +70,6 @@
 
 static struct proc_dir_entry	*acpi_button_dir = NULL;
 
-
 static int
 acpi_button_read_info (
 	char			*page,
Index: acpi-2.5/drivers/acpi/pci_irq.c
diff -u acpi-2.5/drivers/acpi/pci_irq.c:1.1.1.1 acpi-2.5/drivers/acpi/pci_irq.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/pci_irq.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/pci_irq.c	Thu Aug 29 18:01:23 2002
@@ -288,7 +288,7 @@
 	while (!irq && (bridge = bridge->bus->self)) {
 		pin = (pin + PCI_SLOT(bridge->devfn)) % 4;
 		irq = acpi_pci_irq_lookup(0, bridge->bus->number, PCI_SLOT(bridge->devfn), pin);
-	};
+	}
 
 	if (!irq) {
 		ACPI_DEBUG_PRINT((ACPI_DB_WARN, "Unable to derive IRQ for device %s\n", dev->slot_name));
Index: acpi-2.5/drivers/acpi/pci_link.c
diff -u acpi-2.5/drivers/acpi/pci_link.c:1.1.1.1 acpi-2.5/drivers/acpi/pci_link.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/pci_link.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/pci_link.c	Thu Aug 29 18:01:23 2002
@@ -54,12 +54,12 @@
 static int acpi_pci_link_remove (struct acpi_device *device, int type);
 
 static struct acpi_driver acpi_pci_link_driver = {
-        .name =		ACPI_PCI_LINK_DRIVER_NAME,
-        .class =	ACPI_PCI_LINK_CLASS,
-        .ids =		ACPI_PCI_LINK_HID,
-        .ops =		{
-				.add =		acpi_pci_link_add,
-				.remove =	acpi_pci_link_remove,
+	.name =		ACPI_PCI_LINK_DRIVER_NAME,
+	.class =	ACPI_PCI_LINK_CLASS,
+	.ids =		ACPI_PCI_LINK_HID,
+	.ops =		{
+				.add =    acpi_pci_link_add,
+				.remove = acpi_pci_link_remove,
 			},
 };
 
Index: acpi-2.5/drivers/acpi/pci_root.c
diff -u acpi-2.5/drivers/acpi/pci_root.c:1.1.1.1 acpi-2.5/drivers/acpi/pci_root.c:1.1.1.1.4.1
--- acpi-2.5/drivers/acpi/pci_root.c:1.1.1.1	Wed Aug 28 00:13:27 2002
+++ acpi-2.5/drivers/acpi/pci_root.c	Thu Aug 29 18:01:23 2002
@@ -51,8 +51,8 @@
 	.class =	ACPI_PCI_ROOT_CLASS,
 	.ids =		ACPI_PCI_ROOT_HID,
 	.ops =		{
-				.add =		acpi_pci_root_add,
-				.remove =	acpi_pci_root_remove,
+				.add =    acpi_pci_root_add,
+				.remove = acpi_pci_root_remove,
 			},
 };
 
Index: acpi-2.5/drivers/acpi/system.c
diff -u acpi-2.5/drivers/acpi/system.c:1.1.1.2 acpi-2.5/drivers/acpi/system.c:1.1.1.2.2.1
--- acpi-2.5/drivers/acpi/system.c:1.1.1.2	Thu Aug 29 16:30:14 2002
+++ acpi-2.5/drivers/acpi/system.c	Thu Aug 29 18:01:23 2002
@@ -23,8 +23,6 @@
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  */
 
-#define ACPI_C
-
 #include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
Index: acpi-2.5/include/linux/acpi.h
diff -u acpi-2.5/include/linux/acpi.h:1.1.1.2 acpi-2.5/include/linux/acpi.h:1.1.1.2.2.1
--- acpi-2.5/include/linux/acpi.h:1.1.1.2	Thu Aug 29 16:32:30 2002
+++ acpi-2.5/include/linux/acpi.h	Thu Aug 29 18:01:23 2002
@@ -354,11 +354,11 @@
 
 extern int acpi_mp_config;
 
-#else /*!CONFIG_ACPI_BOOT*/
+#else
 
 #define acpi_mp_config	0
 
-#endif /*CONFIG_ACPI_BOOT*/
+#endif
 
 
 #ifdef CONFIG_ACPI_PCI

Re: ACPI source release updated (20020829)

[Alberto Ornaghi]

At 18.43 29-08-2002, you wrote:

>For 2.4:
>  * bus.c
>    - blacklist Compaq Presario 711FR was missing

the problem isn't specific to 711FR but all the 7xx family is affected... 
for example I have a 705EA and the problem is the same....  lost acpi 
interrupt.

are there any chance that the workarounds patches for the blacklisted 
models will be merged with the official release ?

bye

    --==> ALoR <==---------------------- -  -   -

There are only 10 types of people in this world...
those who understand binary, and those who don't.



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI source release updated (20020829)

[Michael Notforyou]

On my 730US, "ACPI lost interrupt detected" seems to occur, but cause no 
problems.


At 10:14 AM 8/30/2002 +0200, you wrote:
>At 18.43 29-08-2002, you wrote:
>
>>For 2.4:
>>  * bus.c
>>    - blacklist Compaq Presario 711FR was missing
>
>the problem isn't specific to 711FR but all the 7xx family is affected... 
>for example I have a 705EA and the problem is the same....  lost acpi 
>interrupt.
>
>are there any chance that the workarounds patches for the blacklisted 
>models will be merged with the official release ?
>
>bye
>
>    --==> ALoR <==---------------------- -  -   -
>
>There are only 10 types of people in this world...
>those who understand binary, and those who don't.
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by: OSDN - Tired of that same old
>cell phone?  Get a new here for FREE!
>https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel

*Michael Notforyou*

"My God, I'm full of stars!"
Darmok and Jalad at Tanagra!



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI source release updated (20020829)

[Alberto Ornaghi]

At 07.34 30-08-2002, you wrote:
>On my 730US, "ACPI lost interrupt detected" seems to occur, but cause no 
>problems.

you are running the kacpid patch, isn't it ?

I mean with vanilla acpi patch you won't get acpi working properly...

and what I'm asking for is: will never kacipd patch merged in the main tree 
(as well as other patched for other blacklisted machines) ?

bye

    --==> ALoR <==---------------------- -  -   -

There are only 10 types of people in this world...
those who understand binary, and those who don't.



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

RE: ACPI source release updated (20020829)

[Grover, Andrew]

> From: Alberto Ornaghi [mailto:alor-G/4Acn03FKk@public.gmane.org] 
> At 07.34 30-08-2002, you wrote:
> >On my 730US, "ACPI lost interrupt detected" seems to occur, 
> but cause no 
> >problems.
> 
> you are running the kacpid patch, isn't it ?
> 
> I mean with vanilla acpi patch you won't get acpi working properly...
> 
> and what I'm asking for is: will never kacipd patch merged in 
> the main tree 
> (as well as other patched for other blacklisted machines) ?

Kacpid is a workaround, it's not the solution.

If we are not receiving SCIs, that is a problem that should be solved.
Polling is...not preferable.

The 0829 patch includes a fix to ensure the SCI is properly configured. Does
that help these systems?

We have the blacklist so that we do not have to fill the ACPI code with
platform-specific workarounds. Although Linux development has traditionally
been very pragmatic about taking workarounds for all possible
configurations, this just isn't feasible for ACPI. We are even worse than
the Tulip driver in terms of the number of different configurations one
driver has to support.

Regards -- Andy


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

RE: ACPI source release updated (20020829)

[Grover, Andrew]

> From: KOCHI, Takayoshi [mailto:t-kouchi-dPjYVeZdYcz+G+EEi5ephHgSJqDPrsil@public.gmane.org] 
> I checked difference today's releases between 2.4 and 2.5.
> Except Makefiles, Config.help and toshiba_acpi.c, I found
> some important differences.
> 
> I made patches to add some missed changes for both 2.4 and 2.5.
> If this seems reasonable, please apply.

Awesome. Thanks.

Regards -- Andy


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

RE: ACPI source release updated (20020829)

[Alberto Ornaghi]

At 13.29 30-08-2002, Grover, Andrew wrote:

>The 0829 patch includes a fix to ensure the SCI is properly configured. Does
>that help these systems?

is it possible to have a patch against 2.4.18 ?
or I have to download 2.4.20-pre5 ?

bye

    --==> ALoR <==---------------------- -  -   -

There are only 10 types of people in this world...
those who understand binary, and those who don't.



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

RE: ACPI source release updated (20020829)

[Grover, Andrew]

> From: Alberto Ornaghi [mailto:alor-G/4Acn03FKk@public.gmane.org] 
> >The 0829 patch includes a fix to ensure the SCI is properly 
> configured. Does
> >that help these systems?
> 
> is it possible to have a patch against 2.4.18 ?
> or I have to download 2.4.20-pre5 ?

Just get pre5. Maybe someone (not me) will backport to 2.4.19, but 2.4.18?
Doubtful.

-- Andy


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI source release updated (20020829)

[Arndt Schoenewald]

Hi Andy!

On Fri, Aug 30, 2002 at 01:29:46PM -0700, Grover, Andrew wrote:
> Kacpid is a workaround, it's not the solution.
> 
> If we are not receiving SCIs, that is a problem that should be solved.
> Polling is...not preferable.
> 
> The 0829 patch includes a fix to ensure the SCI is properly configured. Does
> that help these systems?

Unfortunately, the code you mention does not seem to have made it into
acpi-20020829-2.4.20-pre5.diff.gz.

Patch acpi-20020829-2.5.32.diff.gz contains the following change for
drivers/acpi/bus.c:

+       /* Ensure the SCI is set to level-triggered, active-low */
+       if (acpi_ioapic)
+               mp_override_legacy_irq(acpi_fadt.sci_int, 3, 3, acpi_fadt.sci_int);
+       else
+               eisa_set_level_irq(acpi_fadt.sci_int);

However, this change is missing in acpi-20020829-2.4.20-pre5.diff.gz,
and in fact ACPI events do not get activated for my machine (Gericom
1st Supersonic M6-T) when running 2.4.20-pre5 with acpi-20020829.

(They get enabled when I manually include "eisa_set_level_irq(9);" in
acpi_pci_irq_init() in drivers/acpi/pci_irq.c.)

Otherwise, 2.4.20-pre5 with acpi-20020829 works just fine for me, i.e.
there are no regressions. I haven't tried 2.5.32 yet, though.

Best regards,
Arndt


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI source release updated (20020829)

[Alexander Hoogerhuis]

"Grover, Andrew" <andrew.grover-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> writes:

> > From: Alberto Ornaghi [mailto:alor-G/4Acn03FKk@public.gmane.org] 
> > >The 0829 patch includes a fix to ensure the SCI is properly 
> > configured. Does
> > >that help these systems?
> > 
> > is it possible to have a patch against 2.4.18 ?
> > or I have to download 2.4.20-pre5 ?
> 
> Just get pre5. Maybe someone (not me) will backport to 2.4.19, but 2.4.18?
> Doubtful.
> 

Any chance of seeing diffs against -ac? :)

mvh,
A

-- 
Alexander Hoogerhuis                               | alexh-rpjHciJLgqZBDgjK7y7TUQ@public.gmane.org
CCNP - CCDP - MCNE - CCSE                          | +47 908 21 485
"You have zero privacy anyway. Get over it."  --Scott McNealy


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

RE: ACPI source release updated (20020829)

[Grover, Andrew]

> From: Arndt Schoenewald [mailto:abs-SA7OhAOe25xnNxvc45mVi0K323yFvGpRdefyYXQ/eNw@public.gmane.org] 
> > The 0829 patch includes a fix to ensure the SCI is properly 
> configured. Does
> > that help these systems?
> 
> Unfortunately, the code you mention does not seem to have made it into
> acpi-20020829-2.4.20-pre5.diff.gz.
> 
> Patch acpi-20020829-2.5.32.diff.gz contains the following change for
> drivers/acpi/bus.c:
> 
> +       /* Ensure the SCI is set to level-triggered, active-low */
> +       if (acpi_ioapic)
> +               mp_override_legacy_irq(acpi_fadt.sci_int, 3, 
> 3, acpi_fadt.sci_int);
> +       else
> +               eisa_set_level_irq(acpi_fadt.sci_int);

Hmm, OK, I'll fix that.

-- Andy


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390