GIT 3f4894470810fd3821886d3068fd676b23e33415 git+ssh://master.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq.git commit 3f4894470810fd3821886d3068fd676b23e33415 Author: Dave Jones Date: Tue Jul 25 20:22:29 2006 -0400 [CPUFREQ] Mark longhaul as experimental. It's still undergoing surgery, and isn't ready for mass consumption just yet. Signed-off-by: Dave Jones commit 890b23f1f187186e5335c3ce296b2868fe29e268 Author: Jeremy Fitzhardinge Date: Thu Jul 6 12:32:01 2006 -0700 [CPUFREQ] [2/2] demand load governor modules. Demand-load cpufreq governor modules if needed. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Dave Jones commit b4462ff056c65d6122baf283417c3c1404ca21d7 Author: Jeremy Fitzhardinge Date: Thu Jul 6 12:30:26 2006 -0700 [CPUFREQ] [1/2] add __find_governor helper and clean up some error handling. Adds a __find_governor() helper function to look up a governor by name. Also restructures some error handling to conform to the "single-exit" model which is generally preferred for kernel code. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Dave Jones commit 93dbd766e448905c52ebaaadf7199b3981b6ba0b Author: Rafa³ Bilski Date: Sat Jul 15 19:31:30 2006 +0200 [CPUFREQ] Longhaul - Rename & fix multipliers table This table is only used by Ezra-T CPUs currently, and has values for some other CPU. Fix them to match the values used by that CPU, and for now make it clearer by renaming the variable. Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit db1c1cce27347a5fac11f7396cc55713414b7316 Author: Rafa³ Bilski Date: Thu Jul 13 19:26:10 2006 +0200 [CPUFREQ] Longhaul - Fix power state test to do something more useful This is changing "always true" test to something usefull. Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit efabf0b1d6080e30b30b99e795dd48dc5223cd0a Author: Rafa³ Bilski Date: Sun Jul 9 21:47:04 2006 +0200 [CPUFREQ] Longhaul - Readd accidentally dropped line I lost very important line in do_powersaver Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit 72d7f862c901a315e545538c2c3e9e0e35566d98 Author: Adrian Bunk Date: Sun Jul 9 19:53:08 2006 +0200 [CPUFREQ] Make longhaul_walk_callback() static This patch makes the needlessly global longhaul_walk_callback() static. Signed-off-by: Adrian Bunk Signed-off-by: Dave Jones commit f57df63a0c735f7c8b83a511b5e662194601e9c5 Author: Adrian Bunk Date: Sat Jul 8 22:20:26 2006 +0200 [CPUFREQ] X86_GX_SUSPMOD must depend on PCI It seems commit 32ee8c3e470d86588b51dc42ed01e85c5fa0f180 accidentially reverted cdc9cc1d740ffc3d8d8207fbf5df9bf05fcc9955, IOW, it reintroduced the following compile error with CONFIG_PCI=n: <-- snip --> ... CC arch/i386/kernel/cpu/cpufreq/gx-suspmod.o arch/i386/kernel/cpu/cpufreq/gx-suspmod.c: In function ‘gx_detect_chipset’: arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: error: implicit declaration of function ‘pci_match_id’ arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: warning: comparison between pointer and integer make[3]: *** [arch/i386/kernel/cpu/cpufreq/gx-suspmod.o] Error 1 <-- snip --> This patch therefore re-adds the dependency of X86_GX_SUSPMOD on PCI. Signed-off-by: Adrian Bunk Signed-off-by: Dave Jones commit c25cf6b8430d81117c4d6550ac774dfa2f516b6a Author: Rafa³ Bilski Date: Fri Jul 7 08:48:26 2006 +0200 [CPUFREQ] Longhaul - Initialise later. Without this longhaul will always fail when compiled into kernel, as it needs to initialise after the ACPI processor module. I lost this when I was splitting patches. Sorry. Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit 86a547285802aa8adc08971b76b668dd5fa4a368 Author: Rafa³ Bilski Date: Tue Jul 4 17:50:57 2006 +0200 [CPUFREQ] Longhaul - Workaround issues with APIC. There is no need to worry about local APIC. There is need to worry about I/O APIC, because I/O APIC is replacing good old 8259. According to Nehemiah datasheet VIA is using 3-wire bus to connect local APIC to I/O APIC. "[...] When IA32_APIC_BASE[11] is set to 0, processor APICs based on the 3-wire APIC bus cannot be generally re-enabled until a system hardware reset. The 3-wire bus looses track of arbitration that would be necessary for complete re-enabling. Certain (local) APIC functionality can be enabled. [...]" So we must set disable bit for each interrupt in I/O APIC registers. Same situation as for PIC - we must poke registers direcly. How to do this? I don't know. So at the moment it is better to fail. Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit 9b26436353a88923a65e501d44859552d70080b4 Author: Rafa³ Bilski Date: Mon Jul 3 07:19:05 2006 +0200 [CPUFREQ] Longhaul - Hook into ACPI C states. Minimal change necessary for hardware support. Changes in longhaul.c: - most important - now C3 state is causing transition, - code responsible for clearing "bus master" bit removed, - protect bcr2 transition in the same way as longhaul. Signed-off-by: Rafa³ Bilski Signed-off-by: Dave Jones commit e7e0d4a19371446553ddd14a9c637b779e14d670 Author: Mattia Dongili Date: Wed Jul 5 23:12:20 2006 +0200 [CPUFREQ] return error when failing to set minfreq I just stumbled on this bug/feature, this is how to reproduce it: # echo 450000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq # echo 450000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq # echo powersave > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor # cpufreq-info -p 450000 450000 powersave # echo 1800000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq ; echo $? 0 # cpufreq-info -p 450000 450000 powersave Here it is. The kernel refuses to set a min_freq higher than the max_freq but it allows a max_freq lower than min_freq (lowering min_freq also). This behaviour is pretty straightforward (but undocumented) and it doesn't return an error altough failing to accomplish the requested action (set min_freq). The problem (IMO) is basically that userspace is not allowed to set a full policy atomically while the kernel always does that thus it must enforce an ordering on operations. The attached patch returns -EINVAL if trying to increase frequencies starting from scaling_min_freq and documents the correct ordering of writes. Signed-off-by: Mattia Dongili Signed-off-by: Dominik Brodowski Signed-off-by: Dave Jones -- --- diff --git a/Documentation/cpu-freq/user-guide.txt b/Documentation/cpu-freq/user-guide.txt index 7fedc00..555c8cf 100644 --- a/Documentation/cpu-freq/user-guide.txt +++ b/Documentation/cpu-freq/user-guide.txt @@ -153,10 +153,13 @@ scaling_governor, and by "echoing" the that some governors won't load - they only work on some specific architectures or processors. -scaling_min_freq and +scaling_min_freq and scaling_max_freq show the current "policy limits" (in kHz). By echoing new values into these files, you can change these limits. + NOTE: when setting a policy you need to + first set scaling_max_freq, then + scaling_min_freq. If you have selected the "userspace" governor which allows you to diff --git a/arch/i386/kernel/cpu/cpufreq/Kconfig b/arch/i386/kernel/cpu/cpufreq/Kconfig index e44a4c6..bb7a890 100644 --- a/arch/i386/kernel/cpu/cpufreq/Kconfig +++ b/arch/i386/kernel/cpu/cpufreq/Kconfig @@ -96,6 +96,7 @@ config X86_POWERNOW_K8_ACPI config X86_GX_SUSPMOD tristate "Cyrix MediaGX/NatSemi Geode Suspend Modulation" + depends on PCI help This add the CPUFreq driver for NatSemi Geode processors which support suspend modulation. @@ -202,7 +203,7 @@ config X86_LONGRUN config X86_LONGHAUL tristate "VIA Cyrix III Longhaul" select CPU_FREQ_TABLE - depends on BROKEN + depends on ACPI_PROCESSOR && EXPERIMENTAL help This adds the CPUFreq driver for VIA Samuel/CyrixIII, VIA Cyrix Samuel/C3, VIA Cyrix Ezra and VIA Cyrix Ezra-T diff --git a/arch/i386/kernel/cpu/cpufreq/longhaul.c b/arch/i386/kernel/cpu/cpufreq/longhaul.c index 146f607..60d470e 100644 --- a/arch/i386/kernel/cpu/cpufreq/longhaul.c +++ b/arch/i386/kernel/cpu/cpufreq/longhaul.c @@ -29,11 +29,13 @@ #include #include #include #include -#include #include #include #include +#include +#include +#include #include "longhaul.h" @@ -56,6 +58,8 @@ static int minvid, maxvid; static unsigned int minmult, maxmult; static int can_scale_voltage; static int vrmrev; +static struct acpi_processor *pr = NULL; +static struct acpi_processor_cx *cx = NULL; /* Module parameters */ static int dont_scale_voltage; @@ -118,84 +122,65 @@ static int longhaul_get_cpu_mult(void) return eblcr_table[invalue]; } +/* For processor with BCR2 MSR */ -static void do_powersaver(union msr_longhaul *longhaul, - unsigned int clock_ratio_index) +static void do_longhaul1(int cx_address, unsigned int clock_ratio_index) { - struct pci_dev *dev; - unsigned long flags; - unsigned int tmp_mask; - int version; - int i; - u16 pci_cmd; - u16 cmd_state[64]; + union msr_bcr2 bcr2; + u32 t; - switch (cpu_model) { - case CPU_EZRA_T: - version = 3; - break; - case CPU_NEHEMIAH: - version = 0xf; - break; - default: - return; - } + rdmsrl(MSR_VIA_BCR2, bcr2.val); + /* Enable software clock multiplier */ + bcr2.bits.ESOFTBF = 1; + bcr2.bits.CLOCKMUL = clock_ratio_index; - rdmsrl(MSR_VIA_LONGHAUL, longhaul->val); - longhaul->bits.SoftBusRatio = clock_ratio_index & 0xf; - longhaul->bits.SoftBusRatio4 = (clock_ratio_index & 0x10) >> 4; - longhaul->bits.EnableSoftBusRatio = 1; - longhaul->bits.RevisionKey = 0; + /* Sync to timer tick */ + safe_halt(); + ACPI_FLUSH_CPU_CACHE(); + /* Change frequency on next halt or sleep */ + wrmsrl(MSR_VIA_BCR2, bcr2.val); + /* Invoke C3 */ + inb(cx_address); + /* Dummy op - must do something useless after P_LVL3 read */ + t = inl(acpi_fadt.xpm_tmr_blk.address); + + /* Disable software clock multiplier */ + local_irq_disable(); + rdmsrl(MSR_VIA_BCR2, bcr2.val); + bcr2.bits.ESOFTBF = 0; + wrmsrl(MSR_VIA_BCR2, bcr2.val); +} - preempt_disable(); - local_irq_save(flags); +/* For processor with Longhaul MSR */ - /* - * get current pci bus master state for all devices - * and clear bus master bit - */ - dev = NULL; - i = 0; - do { - dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev); - if (dev != NULL) { - pci_read_config_word(dev, PCI_COMMAND, &pci_cmd); - cmd_state[i++] = pci_cmd; - pci_cmd &= ~PCI_COMMAND_MASTER; - pci_write_config_word(dev, PCI_COMMAND, pci_cmd); - } - } while (dev != NULL); +static void do_powersaver(int cx_address, unsigned int clock_ratio_index) +{ + union msr_longhaul longhaul; + u32 t; - tmp_mask=inb(0x21); /* works on C3. save mask. */ - outb(0xFE,0x21); /* TMR0 only */ - outb(0xFF,0x80); /* delay */ + rdmsrl(MSR_VIA_LONGHAUL, longhaul.val); + longhaul.bits.RevisionKey = longhaul.bits.RevisionID; + longhaul.bits.SoftBusRatio = clock_ratio_index & 0xf; + longhaul.bits.SoftBusRatio4 = (clock_ratio_index & 0x10) >> 4; + longhaul.bits.EnableSoftBusRatio = 1; + /* Sync to timer tick */ safe_halt(); - wrmsrl(MSR_VIA_LONGHAUL, longhaul->val); - halt(); - + ACPI_FLUSH_CPU_CACHE(); + /* Change frequency on next halt or sleep */ + wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); + /* Invoke C3 */ + inb(cx_address); + /* Dummy op - must do something useless after P_LVL3 read */ + t = inl(acpi_fadt.xpm_tmr_blk.address); + + /* Disable bus ratio bit */ local_irq_disable(); - - outb(tmp_mask,0x21); /* restore mask */ - - /* restore pci bus master state for all devices */ - dev = NULL; - i = 0; - do { - dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev); - if (dev != NULL) { - pci_cmd = cmd_state[i++]; - pci_write_config_byte(dev, PCI_COMMAND, pci_cmd); - } - } while (dev != NULL); - local_irq_restore(flags); - preempt_enable(); - - /* disable bus ratio bit */ - rdmsrl(MSR_VIA_LONGHAUL, longhaul->val); - longhaul->bits.EnableSoftBusRatio = 0; - longhaul->bits.RevisionKey = version; - wrmsrl(MSR_VIA_LONGHAUL, longhaul->val); + longhaul.bits.RevisionKey = longhaul.bits.RevisionID; + longhaul.bits.EnableSoftBusRatio = 0; + longhaul.bits.EnableSoftBSEL = 0; + longhaul.bits.EnableSoftVID = 0; + wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); } /** @@ -209,9 +194,9 @@ static void longhaul_setstate(unsigned i { int speed, mult; struct cpufreq_freqs freqs; - union msr_longhaul longhaul; - union msr_bcr2 bcr2; static unsigned int old_ratio=-1; + unsigned long flags; + unsigned int pic1_mask, pic2_mask; if (old_ratio == clock_ratio_index) return; @@ -234,6 +219,20 @@ static void longhaul_setstate(unsigned i dprintk ("Setting to FSB:%dMHz Mult:%d.%dx (%s)\n", fsb, mult/10, mult%10, print_speed(speed/1000)); + preempt_disable(); + local_irq_save(flags); + + pic2_mask = inb(0xA1); + pic1_mask = inb(0x21); /* works on C3. save mask. */ + outb(0xFF,0xA1); /* Overkill */ + outb(0xFE,0x21); /* TMR0 only */ + + /* Disable bus master arbitration */ + if (pr->flags.bm_check) { + acpi_set_register(ACPI_BITREG_ARB_DISABLE, 1, + ACPI_MTX_DO_NOT_LOCK); + } + switch (longhaul_version) { /* @@ -245,20 +244,7 @@ static void longhaul_setstate(unsigned i */ case TYPE_LONGHAUL_V1: case TYPE_LONGHAUL_V2: - rdmsrl (MSR_VIA_BCR2, bcr2.val); - /* Enable software clock multiplier */ - bcr2.bits.ESOFTBF = 1; - bcr2.bits.CLOCKMUL = clock_ratio_index; - local_irq_disable(); - wrmsrl (MSR_VIA_BCR2, bcr2.val); - safe_halt(); - - /* Disable software clock multiplier */ - rdmsrl (MSR_VIA_BCR2, bcr2.val); - bcr2.bits.ESOFTBF = 0; - local_irq_disable(); - wrmsrl (MSR_VIA_BCR2, bcr2.val); - local_irq_enable(); + do_longhaul1(cx->address, clock_ratio_index); break; /* @@ -273,10 +259,22 @@ static void longhaul_setstate(unsigned i * to work in practice. */ case TYPE_POWERSAVER: - do_powersaver(&longhaul, clock_ratio_index); + do_powersaver(cx->address, clock_ratio_index); break; } + /* Enable bus master arbitration */ + if (pr->flags.bm_check) { + acpi_set_register(ACPI_BITREG_ARB_DISABLE, 0, + ACPI_MTX_DO_NOT_LOCK); + } + + outb(pic2_mask,0xA1); /* restore mask */ + outb(pic1_mask,0x21); + + local_irq_restore(flags); + preempt_enable(); + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); } @@ -324,9 +322,11 @@ static int guess_fsb(void) static int __init longhaul_get_ranges(void) { unsigned long invalue; - unsigned int multipliers[32]= { - 50,30,40,100,55,35,45,95,90,70,80,60,120,75,85,65, - -1,110,120,-1,135,115,125,105,130,150,160,140,-1,155,-1,145 }; + unsigned int ezra_t_multipliers[32]= { + 90, 30, 40, 100, 55, 35, 45, 95, + 50, 70, 80, 60, 120, 75, 85, 65, + -1, 110, 120, -1, 135, 115, 125, 105, + 130, 150, 160, 140, -1, 155, -1, 145 }; unsigned int j, k = 0; union msr_longhaul longhaul; unsigned long lo, hi; @@ -355,13 +355,13 @@ static int __init longhaul_get_ranges(vo invalue = longhaul.bits.MaxMHzBR; if (longhaul.bits.MaxMHzBR4) invalue += 16; - maxmult=multipliers[invalue]; + maxmult=ezra_t_multipliers[invalue]; invalue = longhaul.bits.MinMHzBR; if (longhaul.bits.MinMHzBR4 == 1) minmult = 30; else - minmult = multipliers[invalue]; + minmult = ezra_t_multipliers[invalue]; fsb = eblcr_fsb_table_v2[longhaul.bits.MaxMHzFSB]; break; } @@ -527,6 +527,18 @@ static unsigned int longhaul_get(unsigne return calc_speed(longhaul_get_cpu_mult()); } +static acpi_status longhaul_walk_callback(acpi_handle obj_handle, + u32 nesting_level, + void *context, void **return_value) +{ + struct acpi_device *d; + + if ( acpi_bus_get_device(obj_handle, &d) ) { + return 0; + } + *return_value = (void *)acpi_driver_data(d); + return 1; +} static int __init longhaul_cpu_init(struct cpufreq_policy *policy) { @@ -534,6 +546,15 @@ static int __init longhaul_cpu_init(stru char *cpuname=NULL; int ret; + /* Check ACPI support for C3 state */ + acpi_walk_namespace(ACPI_TYPE_PROCESSOR, ACPI_ROOT_OBJECT, ACPI_UINT32_MAX, + &longhaul_walk_callback, NULL, (void *)&pr); + if (pr == NULL) goto err_acpi; + + cx = &pr->power.states[ACPI_STATE_C3]; + if (cx->address == 0 || cx->latency > 1000) goto err_acpi; + + /* Now check what we have on this motherboard */ switch (c->x86_model) { case 6: cpu_model = CPU_SAMUEL; @@ -634,6 +655,10 @@ static int __init longhaul_cpu_init(stru cpufreq_frequency_table_get_attr(longhaul_table, policy->cpu); return 0; + +err_acpi: + printk(KERN_ERR PFX "No ACPI support for CPU frequency changes.\n"); + return -ENODEV; } static int __devexit longhaul_cpu_exit(struct cpufreq_policy *policy) @@ -666,6 +691,18 @@ static int __init longhaul_init(void) if (c->x86_vendor != X86_VENDOR_CENTAUR || c->x86 != 6) return -ENODEV; +#ifdef CONFIG_SMP + if (num_online_cpus() > 1) { + printk(KERN_ERR PFX "More than 1 CPU detected, longhaul disabled.\n"); + return -ENODEV; + } +#endif +#ifdef CONFIG_X86_IO_APIC + if (cpu_has_apic) { + printk(KERN_ERR PFX "APIC detected. Longhaul is currently broken in this configuration.\n"); + return -ENODEV; + } +#endif switch (c->x86_model) { case 6 ... 9: return cpufreq_register_driver(&longhaul_driver); @@ -699,6 +736,6 @@ MODULE_AUTHOR ("Dave Jones name,CPUFREQ_NAME_LEN)) + return t; + + return NULL; +} + /** * cpufreq_parse_governor - parse a governor string */ static int cpufreq_parse_governor (char *str_governor, unsigned int *policy, struct cpufreq_governor **governor) { + int err = -EINVAL; + if (!cpufreq_driver) - return -EINVAL; + goto out; + if (cpufreq_driver->setpolicy) { if (!strnicmp(str_governor, "performance", CPUFREQ_NAME_LEN)) { *policy = CPUFREQ_POLICY_PERFORMANCE; - return 0; + err = 0; } else if (!strnicmp(str_governor, "powersave", CPUFREQ_NAME_LEN)) { *policy = CPUFREQ_POLICY_POWERSAVE; - return 0; + err = 0; } - return -EINVAL; - } else { + } else if (cpufreq_driver->target) { struct cpufreq_governor *t; + mutex_lock(&cpufreq_governor_mutex); - if (!cpufreq_driver || !cpufreq_driver->target) - goto out; - list_for_each_entry(t, &cpufreq_governor_list, governor_list) { - if (!strnicmp(str_governor,t->name,CPUFREQ_NAME_LEN)) { - *governor = t; + + t = __find_governor(str_governor); + + if (t == NULL) { + char *name = kasprintf(GFP_KERNEL, "cpufreq_%s", str_governor); + + if (name) { + int ret; + mutex_unlock(&cpufreq_governor_mutex); - return 0; + ret = request_module(name); + mutex_lock(&cpufreq_governor_mutex); + + if (ret == 0) + t = __find_governor(str_governor); } + + kfree(name); } -out: + + if (t != NULL) { + *governor = t; + err = 0; + } + mutex_unlock(&cpufreq_governor_mutex); } - return -EINVAL; + out: + return err; } @@ -1277,23 +1307,21 @@ EXPORT_SYMBOL_GPL(cpufreq_governor); int cpufreq_register_governor(struct cpufreq_governor *governor) { - struct cpufreq_governor *t; + int err; if (!governor) return -EINVAL; mutex_lock(&cpufreq_governor_mutex); - list_for_each_entry(t, &cpufreq_governor_list, governor_list) { - if (!strnicmp(governor->name,t->name,CPUFREQ_NAME_LEN)) { - mutex_unlock(&cpufreq_governor_mutex); - return -EBUSY; - } + err = -EBUSY; + if (__find_governor(governor->name) == NULL) { + err = 0; + list_add(&governor->governor_list, &cpufreq_governor_list); } - list_add(&governor->governor_list, &cpufreq_governor_list); mutex_unlock(&cpufreq_governor_mutex); - return 0; + return err; } EXPORT_SYMBOL_GPL(cpufreq_register_governor); @@ -1352,6 +1380,11 @@ static int __cpufreq_set_policy(struct c memcpy(&policy->cpuinfo, &data->cpuinfo, sizeof(struct cpufreq_cpuinfo)); + if (policy->min > data->min && policy->min > policy->max) { + ret = -EINVAL; + goto error_out; + } + /* verify the cpu speed can be set within this limit */ ret = cpufreq_driver->verify(policy); if (ret)