From b1c495ffb3ee354a2e91bedc8b9e832cb7777043 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 14 Jul 2006 16:23:33 +0100 Subject: [PATCH 001/143] [ARM] Fix cats build Fix more fallout from 894673ee6122a3ce1958e1fe096901ba5356a96b. arch/arm/mach-footbridge/cats-hw.c: In function 'fixup_cats': arch/arm/mach-footbridge/cats-hw.c:81: error: 'ORIG_VIDEO_LINES' undeclared (first use in this function) ... Signed-off-by: Russell King --- arch/arm/mach-footbridge/cats-hw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-footbridge/cats-hw.c b/arch/arm/mach-footbridge/cats-hw.c index 5b64d5c5b967..ef6ccc8993e9 100644 --- a/arch/arm/mach-footbridge/cats-hw.c +++ b/arch/arm/mach-footbridge/cats-hw.c @@ -8,7 +8,7 @@ #include #include #include -#include +#include #include #include From 284c5e6e938f79906f2ccc8c4039863e6c739aaa Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Sat, 15 Jul 2006 21:08:07 +0100 Subject: [PATCH 002/143] [ARM] 3730/1: ep93xx: enable usb ohci driver in the defconfig Patch from Lennert Buytenhek Since the ep93xx ohci bits are merged upstream now, enable ohci in the ep93xx defconfig. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- arch/arm/configs/ep93xx_defconfig | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/arch/arm/configs/ep93xx_defconfig b/arch/arm/configs/ep93xx_defconfig index c0de6fcd488a..2948b4589a8b 100644 --- a/arch/arm/configs/ep93xx_defconfig +++ b/arch/arm/configs/ep93xx_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Sun Jul 9 15:21:30 2006 +# Linux kernel version: 2.6.18-rc1-git9 +# Sat Jul 15 15:08:10 2006 # CONFIG_ARM=y CONFIG_MMU=y @@ -30,6 +30,7 @@ CONFIG_SWAP=y CONFIG_SYSVIPC=y # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set CONFIG_SYSCTL=y # CONFIG_AUDIT is not set CONFIG_IKCONFIG=y @@ -749,7 +750,7 @@ CONFIG_VIDEO_V4L2=y # USB support # CONFIG_USB_ARCH_HAS_HCD=y -# CONFIG_USB_ARCH_HAS_OHCI is not set +CONFIG_USB_ARCH_HAS_OHCI=y # CONFIG_USB_ARCH_HAS_EHCI is not set CONFIG_USB=y CONFIG_USB_DEBUG=y @@ -766,6 +767,9 @@ CONFIG_USB_DYNAMIC_MINORS=y # USB Host Controller Drivers # # CONFIG_USB_ISP116X_HCD is not set +CONFIG_USB_OHCI_HCD=y +# CONFIG_USB_OHCI_BIG_ENDIAN is not set +CONFIG_USB_OHCI_LITTLE_ENDIAN=y # CONFIG_USB_SL811_HCD is not set # @@ -855,6 +859,7 @@ CONFIG_USB_SERIAL_CONSOLE=y CONFIG_USB_SERIAL_PL2303=y # CONFIG_USB_SERIAL_HP4X is not set # CONFIG_USB_SERIAL_SAFE is not set +# CONFIG_USB_SERIAL_SIERRAWIRELESS is not set # CONFIG_USB_SERIAL_TI is not set # CONFIG_USB_SERIAL_CYBERJACK is not set # CONFIG_USB_SERIAL_XIRCOM is not set @@ -871,7 +876,7 @@ CONFIG_USB_SERIAL_PL2303=y # CONFIG_USB_LEGOTOWER is not set # CONFIG_USB_LCD is not set # CONFIG_USB_LED is not set -# CONFIG_USB_CY7C63 is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set # CONFIG_USB_PHIDGETKIT is not set # CONFIG_USB_PHIDGETSERVO is not set @@ -916,6 +921,7 @@ CONFIG_RTC_INTF_DEV=y # CONFIG_RTC_DRV_X1205 is not set # CONFIG_RTC_DRV_DS1307 is not set # CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_ISL1208 is not set # CONFIG_RTC_DRV_DS1672 is not set # CONFIG_RTC_DRV_DS1742 is not set # CONFIG_RTC_DRV_PCF8563 is not set @@ -1023,7 +1029,6 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set From 4528e4da79675b4995e085046b8ffbe0415c3261 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 8 Jul 2006 20:17:26 +0900 Subject: [PATCH 003/143] [PATCH] libata: fix autopsy ehc->i.action and ehc->i.dev handling Commit 0662c58b3265f52f708a6d59476bc7862b01f9c0 updated ata_eh_autopsy() to OR determined action to ehc->i.action to preserve action mask set directly into ehc->i.action by nested functions. This broke action mask clearing on SENSE_VALID case causing revalidation and EH complete message on successful ATAPI CC. This patch removes two local variables - action and failed_dev - which cache ehc->i.action and ehc->i.dev respectively, and make the function directly modify ehc->i.* fields to remove aliasing issues. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/scsi/libata-eh.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) diff --git a/drivers/scsi/libata-eh.c b/drivers/scsi/libata-eh.c index 4b6aa30f4d68..36a801c83325 100644 --- a/drivers/scsi/libata-eh.c +++ b/drivers/scsi/libata-eh.c @@ -1276,8 +1276,6 @@ static int ata_eh_speed_down(struct ata_device *dev, int is_io, static void ata_eh_autopsy(struct ata_port *ap) { struct ata_eh_context *ehc = &ap->eh_context; - unsigned int action = ehc->i.action; - struct ata_device *failed_dev = NULL; unsigned int all_err_mask = 0; int tag, is_io = 0; u32 serror; @@ -1294,7 +1292,7 @@ static void ata_eh_autopsy(struct ata_port *ap) ehc->i.serror |= serror; ata_eh_analyze_serror(ap); } else if (rc != -EOPNOTSUPP) - action |= ATA_EH_HARDRESET; + ehc->i.action |= ATA_EH_HARDRESET; /* analyze NCQ failure */ ata_eh_analyze_ncq_error(ap); @@ -1315,7 +1313,7 @@ static void ata_eh_autopsy(struct ata_port *ap) qc->err_mask |= ehc->i.err_mask; /* analyze TF */ - action |= ata_eh_analyze_tf(qc, &qc->result_tf); + ehc->i.action |= ata_eh_analyze_tf(qc, &qc->result_tf); /* DEV errors are probably spurious in case of ATA_BUS error */ if (qc->err_mask & AC_ERR_ATA_BUS) @@ -1329,11 +1327,11 @@ static void ata_eh_autopsy(struct ata_port *ap) /* SENSE_VALID trumps dev/unknown error and revalidation */ if (qc->flags & ATA_QCFLAG_SENSE_VALID) { qc->err_mask &= ~(AC_ERR_DEV | AC_ERR_OTHER); - action &= ~ATA_EH_REVALIDATE; + ehc->i.action &= ~ATA_EH_REVALIDATE; } /* accumulate error info */ - failed_dev = qc->dev; + ehc->i.dev = qc->dev; all_err_mask |= qc->err_mask; if (qc->flags & ATA_QCFLAG_IO) is_io = 1; @@ -1342,25 +1340,22 @@ static void ata_eh_autopsy(struct ata_port *ap) /* enforce default EH actions */ if (ap->pflags & ATA_PFLAG_FROZEN || all_err_mask & (AC_ERR_HSM | AC_ERR_TIMEOUT)) - action |= ATA_EH_SOFTRESET; + ehc->i.action |= ATA_EH_SOFTRESET; else if (all_err_mask) - action |= ATA_EH_REVALIDATE; + ehc->i.action |= ATA_EH_REVALIDATE; /* if we have offending qcs and the associated failed device */ - if (failed_dev) { + if (ehc->i.dev) { /* speed down */ - action |= ata_eh_speed_down(failed_dev, is_io, all_err_mask); + ehc->i.action |= ata_eh_speed_down(ehc->i.dev, is_io, + all_err_mask); /* perform per-dev EH action only on the offending device */ - ehc->i.dev_action[failed_dev->devno] |= - action & ATA_EH_PERDEV_MASK; - action &= ~ATA_EH_PERDEV_MASK; + ehc->i.dev_action[ehc->i.dev->devno] |= + ehc->i.action & ATA_EH_PERDEV_MASK; + ehc->i.action &= ~ATA_EH_PERDEV_MASK; } - /* record autopsy result */ - ehc->i.dev = failed_dev; - ehc->i.action |= action; - DPRINTK("EXIT\n"); } From 7c8c2cff81b2b7f6dd3f9fd7b77033c1be5d7920 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 10 Jul 2006 23:18:23 +0900 Subject: [PATCH 004/143] [PATCH] libata: fix eh_skip_recovery condition * (ata_dev_absent() || ata_dev_ready()) test doesn't indicate SUSPENDED state properly. Fix it. * Link resuming resets shouldn't be skipped. Don't skip recovery on EHI_RESUME_LINK. This doesn't matter for host ports as EHI_RESUME always coincides with EHI_HOTPLUGGED which makes attached disabled devices vacant. However, PMP reset causes non-hotplug link-resuming resets which shouldn't be skipped. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/scsi/libata-eh.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/libata-eh.c b/drivers/scsi/libata-eh.c index 36a801c83325..1dcb2c13ffa9 100644 --- a/drivers/scsi/libata-eh.c +++ b/drivers/scsi/libata-eh.c @@ -1843,15 +1843,16 @@ static int ata_eh_skip_recovery(struct ata_port *ap) for (i = 0; i < ata_port_max_devices(ap); i++) { struct ata_device *dev = &ap->device[i]; - if (ata_dev_absent(dev) || ata_dev_ready(dev)) + if (!(dev->flags & ATA_DFLAG_SUSPENDED)) break; } if (i == ata_port_max_devices(ap)) return 1; - /* always thaw frozen port and recover failed devices */ - if (ap->pflags & ATA_PFLAG_FROZEN || ata_port_nr_enabled(ap)) + /* thaw frozen port, resume link and recover failed devices */ + if ((ap->pflags & ATA_PFLAG_FROZEN) || + (ehc->i.flags & ATA_EHI_RESUME_LINK) || ata_port_nr_enabled(ap)) return 0; /* skip if class codes for all vacant slots are ATA_DEV_NONE */ From 13abf50df209008b5d44075bafeeab42ace56aa6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 10 Jul 2006 23:18:46 +0900 Subject: [PATCH 005/143] [PATCH] libata: improve EH action and EHI flag handling Update ata_eh_about_to_do() and ata_eh_done() to improve EH action and EHI flag handling. * There are two types of EHI flags - one which expires on successful EH and the other which expires on a successful reset. Make this distinction clear. * Unlike other EH actions, reset actions are represented by two EH action masks and a EHI modifier. Implement correct about_to_do/done semantics for resets. That is, prior to reset, related EH info is sucked in from ehi and cleared, and after reset is complete, related EH info in ehc is cleared. These changes improve consistency and remove unnecessary EH actions caused by stale EH action masks and EHI flags. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/scsi/libata-eh.c | 33 ++++++++++++++++++++++++++++----- include/linux/libata.h | 4 +++- 2 files changed, 31 insertions(+), 6 deletions(-) diff --git a/drivers/scsi/libata-eh.c b/drivers/scsi/libata-eh.c index 1dcb2c13ffa9..29f59345305d 100644 --- a/drivers/scsi/libata-eh.c +++ b/drivers/scsi/libata-eh.c @@ -764,12 +764,27 @@ static void ata_eh_about_to_do(struct ata_port *ap, struct ata_device *dev, unsigned int action) { unsigned long flags; + struct ata_eh_info *ehi = &ap->eh_info; + struct ata_eh_context *ehc = &ap->eh_context; spin_lock_irqsave(ap->lock, flags); - ata_eh_clear_action(dev, &ap->eh_info, action); + /* Reset is represented by combination of actions and EHI + * flags. Suck in all related bits before clearing eh_info to + * avoid losing requested action. + */ + if (action & ATA_EH_RESET_MASK) { + ehc->i.action |= ehi->action & ATA_EH_RESET_MASK; + ehc->i.flags |= ehi->flags & ATA_EHI_RESET_MODIFIER_MASK; - if (!(ap->eh_context.i.flags & ATA_EHI_QUIET)) + /* make sure all reset actions are cleared & clear EHI flags */ + action |= ATA_EH_RESET_MASK; + ehi->flags &= ~ATA_EHI_RESET_MODIFIER_MASK; + } + + ata_eh_clear_action(dev, ehi, action); + + if (!(ehc->i.flags & ATA_EHI_QUIET)) ap->pflags |= ATA_PFLAG_RECOVERED; spin_unlock_irqrestore(ap->lock, flags); @@ -790,6 +805,12 @@ static void ata_eh_about_to_do(struct ata_port *ap, struct ata_device *dev, static void ata_eh_done(struct ata_port *ap, struct ata_device *dev, unsigned int action) { + /* if reset is complete, clear all reset actions & reset modifier */ + if (action & ATA_EH_RESET_MASK) { + action |= ATA_EH_RESET_MASK; + ap->eh_context.i.flags &= ~ATA_EHI_RESET_MODIFIER_MASK; + } + ata_eh_clear_action(dev, &ap->eh_context.i, action); } @@ -1478,6 +1499,9 @@ static int ata_eh_reset(struct ata_port *ap, int classify, ata_reset_fn_t reset; int i, did_followup_srst, rc; + /* about to reset */ + ata_eh_about_to_do(ap, NULL, ehc->i.action & ATA_EH_RESET_MASK); + /* Determine which reset to use and record in ehc->i.action. * prereset() may examine and modify it. */ @@ -1526,8 +1550,7 @@ static int ata_eh_reset(struct ata_port *ap, int classify, ata_port_printk(ap, KERN_INFO, "%s resetting port\n", reset == softreset ? "soft" : "hard"); - /* reset */ - ata_eh_about_to_do(ap, NULL, ATA_EH_RESET_MASK); + /* mark that this EH session started with reset */ ehc->i.flags |= ATA_EHI_DID_RESET; rc = ata_do_reset(ap, reset, classes); @@ -1590,7 +1613,7 @@ static int ata_eh_reset(struct ata_port *ap, int classify, postreset(ap, classes); /* reset successful, schedule revalidation */ - ata_eh_done(ap, NULL, ATA_EH_RESET_MASK); + ata_eh_done(ap, NULL, ehc->i.action & ATA_EH_RESET_MASK); ehc->i.action |= ATA_EH_REVALIDATE; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 6cc497a2b6da..66c3100c2b94 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -265,12 +265,14 @@ enum { /* ata_eh_info->flags */ ATA_EHI_HOTPLUGGED = (1 << 0), /* could have been hotplugged */ - ATA_EHI_RESUME_LINK = (1 << 1), /* need to resume link */ + ATA_EHI_RESUME_LINK = (1 << 1), /* resume link (reset modifier) */ ATA_EHI_NO_AUTOPSY = (1 << 2), /* no autopsy */ ATA_EHI_QUIET = (1 << 3), /* be quiet */ ATA_EHI_DID_RESET = (1 << 16), /* already reset this port */ + ATA_EHI_RESET_MODIFIER_MASK = ATA_EHI_RESUME_LINK, + /* max repeat if error condition is still set after ->error_handler */ ATA_EH_MAX_REPEAT = 5, From 59b693fbbec37765426ba0916920ff93d2e26eb3 Mon Sep 17 00:00:00 2001 From: Jay Cliburn Date: Thu, 20 Jul 2006 23:23:57 +0200 Subject: [PATCH 006/143] via-velocity: fix speed and link status reported by ethtool The via-velocity driver reports incorrect speed and link detected status as viewed by ethtool (and probably other tools). This patch fixes those incorrect reports and prettifies a long line. Signed-off-by: Jay Cliburn Signed-off-by: Francois Romieu --- drivers/net/via-velocity.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index f5b0078eb4ad..aa9cd92f46b2 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -2742,7 +2742,7 @@ static u32 check_connection_type(struct mac_regs __iomem * regs) if (PHYSR0 & PHYSR0_SPDG) status |= VELOCITY_SPEED_1000; - if (PHYSR0 & PHYSR0_SPD10) + else if (PHYSR0 & PHYSR0_SPD10) status |= VELOCITY_SPEED_10; else status |= VELOCITY_SPEED_100; @@ -2851,8 +2851,17 @@ static int velocity_get_settings(struct net_device *dev, struct ethtool_cmd *cmd u32 status; status = check_connection_type(vptr->mac_regs); - cmd->supported = SUPPORTED_TP | SUPPORTED_Autoneg | SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full | SUPPORTED_100baseT_Half | SUPPORTED_100baseT_Full | SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full; - if (status & VELOCITY_SPEED_100) + cmd->supported = SUPPORTED_TP | + SUPPORTED_Autoneg | + SUPPORTED_10baseT_Half | + SUPPORTED_10baseT_Full | + SUPPORTED_100baseT_Half | + SUPPORTED_100baseT_Full | + SUPPORTED_1000baseT_Half | + SUPPORTED_1000baseT_Full; + if (status & VELOCITY_SPEED_1000) + cmd->speed = SPEED_1000; + else if (status & VELOCITY_SPEED_100) cmd->speed = SPEED_100; else cmd->speed = SPEED_10; @@ -2896,7 +2905,7 @@ static u32 velocity_get_link(struct net_device *dev) { struct velocity_info *vptr = netdev_priv(dev); struct mac_regs __iomem * regs = vptr->mac_regs; - return BYTE_REG_BITS_IS_ON(PHYSR0_LINKGD, ®s->PHYSR0) ? 0 : 1; + return BYTE_REG_BITS_IS_ON(PHYSR0_LINKGD, ®s->PHYSR0) ? 1 : 0; } static void velocity_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) From 8419dc8a34d765f2ff1aa4051084d54cfeff09cf Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Mon, 24 Jul 2006 03:37:52 -0400 Subject: [PATCH 007/143] [libata] sata_promise: comment out duplicate PCI ID This is just the for-RC fix. A 'TODO' command is added, describing what's needed for the more-complete fix. --- drivers/scsi/sata_promise.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/scsi/sata_promise.c b/drivers/scsi/sata_promise.c index 64631bd38952..4776f4e55839 100644 --- a/drivers/scsi/sata_promise.c +++ b/drivers/scsi/sata_promise.c @@ -269,8 +269,15 @@ static const struct pci_device_id pdc_ata_pci_tbl[] = { { PCI_VENDOR_ID_PROMISE, 0x6629, PCI_ANY_ID, PCI_ANY_ID, 0, 0, board_20619 }, +/* TODO: remove all associated board_20771 code, as it completely + * duplicates board_2037x code, unless reason for separation can be + * divined. + */ +#if 0 { PCI_VENDOR_ID_PROMISE, 0x3570, PCI_ANY_ID, PCI_ANY_ID, 0, 0, board_20771 }, +#endif + { } /* terminate list */ }; From d8ca3d11c6611685ea7783b4acb623bf08496f94 Mon Sep 17 00:00:00 2001 From: Martin Michlmayr Date: Mon, 24 Jul 2006 21:30:01 +0100 Subject: [PATCH 008/143] [ARM] 3731/1: Allow IRQ definitions of IQ80331 and IQ80332 to co-exist Patch from Martin Michlmayr ARCH_IQ80331 and MACH_IQ80332 can be enabled at the same time but a header file makes certain IRQ definitions conditional, leading to the following compilation error when both platforms are enabled: arch/arm/mach-iop3xx/iq80332-pci.c: In function 'iq80332_map_irq': arch/arm/mach-iop3xx/iq80332-pci.c:54: error: 'IRQ_IQ80332_INTA' undeclared (first use in this function) arch/arm/mach-iop3xx/iq80332-pci.c:54: error: (Each undeclared identifier is reported only once arch/arm/mach-iop3xx/iq80332-pci.c:54: error: for each function it appears in.) arch/arm/mach-iop3xx/iq80332-pci.c:54: error: 'IRQ_IQ80332_INTB' undeclared (first use in this function) arch/arm/mach-iop3xx/iq80332-pci.c:54: error: 'IRQ_IQ80332_INTC' undeclared (first use in this function) arch/arm/mach-iop3xx/iq80332-pci.c:54: error: 'IRQ_IQ80332_INTD' undeclared (first use in this function) Signed-off-by: Martin Michlmayr Signed-off-by: Russell King --- include/asm-arm/arch-iop3xx/iop331-irqs.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/include/asm-arm/arch-iop3xx/iop331-irqs.h b/include/asm-arm/arch-iop3xx/iop331-irqs.h index 8ff73d487222..7135ad7e335e 100644 --- a/include/asm-arm/arch-iop3xx/iop331-irqs.h +++ b/include/asm-arm/arch-iop3xx/iop331-irqs.h @@ -91,7 +91,6 @@ #define NR_IRQS NR_IOP331_IRQS -#if defined(CONFIG_ARCH_IQ80331) /* * Interrupts available on the IQ80331 board */ @@ -111,7 +110,6 @@ #define IRQ_IQ80331_INTC IRQ_IOP331_XINT2 #define IRQ_IQ80331_INTD IRQ_IOP331_XINT3 -#elif defined(CONFIG_MACH_IQ80332) /* * Interrupts available on the IQ80332 board */ @@ -131,6 +129,4 @@ #define IRQ_IQ80332_INTC IRQ_IOP331_XINT2 #define IRQ_IQ80332_INTD IRQ_IOP331_XINT3 -#endif - #endif // _IOP331_IRQ_H_ From d1a8df9136ff55e554e11ce65854c282965be8f5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 3 Jul 2006 14:28:14 +0200 Subject: [PATCH 009/143] [POWERPC] fix up front-LED Kconfig This patch fixes the front-LED Kconfig issues I introduced while creating it. Apparently having a dependency isn't enough to have the select not evaluated or something like that. The patch also changes the default configuration for pmac32 select the default for the LED to be the IDE trigger. While I was at it, I completely updated the defconfig and also added snd-aoa to it. Signed-off-by: Johannes Berg Signed-off-by: Paul Mackerras --- arch/powerpc/configs/pmac32_defconfig | 112 ++++++++++++++++++++++---- drivers/ide/Kconfig | 14 ---- drivers/macintosh/Kconfig | 9 +++ drivers/macintosh/via-pmu-led.c | 2 +- 4 files changed, 106 insertions(+), 31 deletions(-) diff --git a/arch/powerpc/configs/pmac32_defconfig b/arch/powerpc/configs/pmac32_defconfig index addc79381c3b..3545af9896af 100644 --- a/arch/powerpc/configs/pmac32_defconfig +++ b/arch/powerpc/configs/pmac32_defconfig @@ -1,16 +1,18 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-rc5 -# Mon May 29 14:47:49 2006 +# Linux kernel version: 2.6.17 +# Mon Jul 3 14:20:49 2006 # # CONFIG_PPC64 is not set CONFIG_PPC32=y CONFIG_PPC_MERGE=y CONFIG_MMU=y CONFIG_GENERIC_HARDIRQS=y +CONFIG_IRQ_PER_CPU=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_PPC=y CONFIG_EARLY_PRINTK=y CONFIG_GENERIC_NVRAM=y @@ -29,6 +31,7 @@ CONFIG_CLASSIC32=y # CONFIG_PPC_82xx is not set # CONFIG_PPC_83xx is not set # CONFIG_PPC_85xx is not set +# CONFIG_PPC_86xx is not set # CONFIG_40x is not set # CONFIG_44x is not set # CONFIG_8xx is not set @@ -39,6 +42,7 @@ CONFIG_ALTIVEC=y CONFIG_PPC_STD_MMU=y CONFIG_PPC_STD_MMU_32=y # CONFIG_SMP is not set +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # # Code maturity level options @@ -72,10 +76,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y +CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y +CONFIG_VM_EVENT_COUNTERS=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -119,6 +125,9 @@ CONFIG_PPC_MULTIPLATFORM=y # CONFIG_APUS is not set # CONFIG_PPC_CHRP is not set CONFIG_PPC_PMAC=y +# CONFIG_PPC_CELL is not set +# CONFIG_PPC_CELL_NATIVE is not set +# CONFIG_UDBG_RTAS_CONSOLE is not set CONFIG_MPIC=y # CONFIG_PPC_RTAS is not set # CONFIG_MMIO_NVRAM is not set @@ -154,6 +163,7 @@ CONFIG_PREEMPT_NONE=y # CONFIG_PREEMPT is not set CONFIG_BINFMT_ELF=y CONFIG_BINFMT_MISC=m +CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y # CONFIG_KEXEC is not set CONFIG_ARCH_FLATMEM_ENABLE=y CONFIG_SELECT_MEMORY_MODEL=y @@ -164,6 +174,7 @@ CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y # CONFIG_SPARSEMEM_STATIC is not set CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set CONFIG_PROC_DEVICETREE=y # CONFIG_CMDLINE_BOOL is not set CONFIG_PM=y @@ -182,6 +193,7 @@ CONFIG_GENERIC_ISA_DMA=y CONFIG_PPC_INDIRECT_PCI=y CONFIG_PCI=y CONFIG_PCI_DOMAINS=y +# CONFIG_PCIEPORTBUS is not set # CONFIG_PCI_DEBUG is not set # @@ -256,6 +268,8 @@ CONFIG_INET_ESP=y # CONFIG_INET_IPCOMP is not set # CONFIG_INET_XFRM_TUNNEL is not set # CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set @@ -268,6 +282,7 @@ CONFIG_TCP_CONG_BIC=y # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set +# CONFIG_NETWORK_SECMARK is not set CONFIG_NETFILTER=y # CONFIG_NETFILTER_DEBUG is not set @@ -292,9 +307,11 @@ CONFIG_NETFILTER_XT_MATCH_MARK=m CONFIG_NETFILTER_XT_MATCH_POLICY=m CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m +# CONFIG_NETFILTER_XT_MATCH_QUOTA is not set CONFIG_NETFILTER_XT_MATCH_REALM=m CONFIG_NETFILTER_XT_MATCH_SCTP=m CONFIG_NETFILTER_XT_MATCH_STATE=m +# CONFIG_NETFILTER_XT_MATCH_STATISTIC is not set CONFIG_NETFILTER_XT_MATCH_STRING=m CONFIG_NETFILTER_XT_MATCH_TCPMSS=m @@ -313,6 +330,7 @@ CONFIG_IP_NF_TFTP=m CONFIG_IP_NF_AMANDA=m CONFIG_IP_NF_PPTP=m CONFIG_IP_NF_H323=m +# CONFIG_IP_NF_SIP is not set # CONFIG_IP_NF_QUEUE is not set CONFIG_IP_NF_IPTABLES=m CONFIG_IP_NF_MATCH_IPRANGE=m @@ -457,6 +475,7 @@ CONFIG_IRTTY_SIR=m # CONFIG_ALI_FIR is not set # CONFIG_VLSI_FIR is not set # CONFIG_VIA_FIR is not set +# CONFIG_MCS_FIR is not set CONFIG_BT=m CONFIG_BT_L2CAP=m CONFIG_BT_SCO=m @@ -500,6 +519,7 @@ CONFIG_WIRELESS_EXT=y CONFIG_PREVENT_FIRMWARE_BUILD=y CONFIG_FW_LOADER=y # CONFIG_DEBUG_DRIVER is not set +# CONFIG_SYS_HYPERVISOR is not set # # Connector - unified userspace <-> kernelspace linker @@ -600,7 +620,6 @@ CONFIG_BLK_DEV_PDC202XX_NEW=y CONFIG_BLK_DEV_IDE_PMAC=y CONFIG_BLK_DEV_IDE_PMAC_ATA100FIRST=y CONFIG_BLK_DEV_IDEDMA_PMAC=y -CONFIG_BLK_DEV_IDE_PMAC_BLINK=y # CONFIG_IDE_ARM is not set CONFIG_BLK_DEV_IDEDMA=y # CONFIG_IDEDMA_IVB is not set @@ -661,6 +680,7 @@ CONFIG_SCSI_AIC7XXX_OLD=m # CONFIG_MEGARAID_LEGACY is not set # CONFIG_MEGARAID_SAS is not set # CONFIG_SCSI_SATA is not set +# CONFIG_SCSI_HPTIOP is not set # CONFIG_SCSI_BUSLOGIC is not set # CONFIG_SCSI_DMX3191D is not set # CONFIG_SCSI_EATA is not set @@ -705,9 +725,7 @@ CONFIG_MD_LINEAR=m CONFIG_MD_RAID0=m CONFIG_MD_RAID1=m CONFIG_MD_RAID10=m -CONFIG_MD_RAID5=m -CONFIG_MD_RAID5_RESHAPE=y -CONFIG_MD_RAID6=m +# CONFIG_MD_RAID456 is not set CONFIG_MD_MULTIPATH=m CONFIG_MD_FAULTY=m CONFIG_BLK_DEV_DM=m @@ -750,7 +768,6 @@ CONFIG_IEEE1394_OHCI1394=m # CONFIG_IEEE1394_VIDEO1394=m CONFIG_IEEE1394_SBP2=m -# CONFIG_IEEE1394_SBP2_PHYS_DMA is not set # CONFIG_IEEE1394_ETH1394 is not set CONFIG_IEEE1394_DV1394=m CONFIG_IEEE1394_RAWIO=m @@ -766,9 +783,12 @@ CONFIG_IEEE1394_RAWIO=m CONFIG_ADB=y CONFIG_ADB_CUDA=y CONFIG_ADB_PMU=y +CONFIG_ADB_PMU_LED=y +CONFIG_ADB_PMU_LED_IDE=y CONFIG_PMAC_APM_EMU=m CONFIG_PMAC_MEDIABAY=y CONFIG_PMAC_BACKLIGHT=y +CONFIG_PMAC_BACKLIGHT_LEGACY=y CONFIG_INPUT_ADBHID=y CONFIG_MAC_EMUMOUSEBTN=y CONFIG_THERM_WINDTUNNEL=m @@ -858,6 +878,7 @@ CONFIG_PCNET32=y # CONFIG_CHELSIO_T1 is not set # CONFIG_IXGB is not set # CONFIG_S2IO is not set +# CONFIG_MYRI10GE is not set # # Token Ring devices @@ -908,6 +929,7 @@ CONFIG_APPLE_AIRPORT=m # Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support # CONFIG_PRISM54=m +# CONFIG_USB_ZD1201 is not set # CONFIG_HOSTAP is not set CONFIG_NET_WIRELESS=y @@ -998,6 +1020,7 @@ CONFIG_SERIO=y CONFIG_VT=y CONFIG_VT_CONSOLE=y CONFIG_HW_CONSOLE=y +# CONFIG_VT_HW_CONSOLE_BINDING is not set # CONFIG_SERIAL_NONSTANDARD is not set # @@ -1029,6 +1052,7 @@ CONFIG_LEGACY_PTY_COUNT=256 # Watchdog Cards # # CONFIG_WATCHDOG is not set +# CONFIG_HW_RANDOM is not set CONFIG_NVRAM=y CONFIG_GEN_RTC=y # CONFIG_GEN_RTC_X is not set @@ -1040,6 +1064,7 @@ CONFIG_GEN_RTC=y # Ftape, the floppy tape device driver # CONFIG_AGP=m +# CONFIG_AGP_SIS is not set # CONFIG_AGP_VIA is not set CONFIG_AGP_UNINORTH=m CONFIG_DRM=m @@ -1092,6 +1117,7 @@ CONFIG_I2C_ALGOBIT=y CONFIG_I2C_POWERMAC=y # CONFIG_I2C_MPC is not set # CONFIG_I2C_NFORCE2 is not set +# CONFIG_I2C_OCORES is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set # CONFIG_I2C_SAVAGE4 is not set @@ -1156,12 +1182,13 @@ CONFIG_VIDEO_V4L2=y # # Graphics support # +# CONFIG_FIRMWARE_EDID is not set CONFIG_FB=y CONFIG_FB_CFB_FILLRECT=y CONFIG_FB_CFB_COPYAREA=y CONFIG_FB_CFB_IMAGEBLIT=y CONFIG_FB_MACMODES=y -CONFIG_FB_FIRMWARE_EDID=y +CONFIG_FB_BACKLIGHT=y CONFIG_FB_MODE_HELPERS=y CONFIG_FB_TILEBLITTING=y # CONFIG_FB_CIRRUS is not set @@ -1178,6 +1205,7 @@ CONFIG_FB_IMSTT=y # CONFIG_FB_S1D13XXX is not set CONFIG_FB_NVIDIA=y CONFIG_FB_NVIDIA_I2C=y +CONFIG_FB_NVIDIA_BACKLIGHT=y # CONFIG_FB_RIVA is not set CONFIG_FB_MATROX=y CONFIG_FB_MATROX_MILLENIUM=y @@ -1187,12 +1215,15 @@ CONFIG_FB_MATROX_MYSTIQUE=y # CONFIG_FB_MATROX_MULTIHEAD is not set CONFIG_FB_RADEON=y CONFIG_FB_RADEON_I2C=y +CONFIG_FB_RADEON_BACKLIGHT=y # CONFIG_FB_RADEON_DEBUG is not set CONFIG_FB_ATY128=y +CONFIG_FB_ATY128_BACKLIGHT=y CONFIG_FB_ATY=y CONFIG_FB_ATY_CT=y # CONFIG_FB_ATY_GENERIC_LCD is not set CONFIG_FB_ATY_GX=y +CONFIG_FB_ATY_BACKLIGHT=y # CONFIG_FB_SAVAGE is not set # CONFIG_FB_SIS is not set # CONFIG_FB_NEOMAGIC is not set @@ -1221,7 +1252,11 @@ CONFIG_LOGO=y CONFIG_LOGO_LINUX_MONO=y CONFIG_LOGO_LINUX_VGA16=y CONFIG_LOGO_LINUX_CLUT224=y -# CONFIG_BACKLIGHT_LCD_SUPPORT is not set +CONFIG_BACKLIGHT_LCD_SUPPORT=y +CONFIG_BACKLIGHT_CLASS_DEVICE=y +CONFIG_BACKLIGHT_DEVICE=y +CONFIG_LCD_CLASS_DEVICE=m +CONFIG_LCD_DEVICE=y # # Sound @@ -1278,6 +1313,18 @@ CONFIG_SND_DUMMY=m # CONFIG_SND_CMIPCI is not set # CONFIG_SND_CS4281 is not set # CONFIG_SND_CS46XX is not set +# CONFIG_SND_DARLA20 is not set +# CONFIG_SND_GINA20 is not set +# CONFIG_SND_LAYLA20 is not set +# CONFIG_SND_DARLA24 is not set +# CONFIG_SND_GINA24 is not set +# CONFIG_SND_LAYLA24 is not set +# CONFIG_SND_MONA is not set +# CONFIG_SND_MIA is not set +# CONFIG_SND_ECHO3G is not set +# CONFIG_SND_INDIGO is not set +# CONFIG_SND_INDIGOIO is not set +# CONFIG_SND_INDIGODJ is not set # CONFIG_SND_EMU10K1 is not set # CONFIG_SND_EMU10K1X is not set # CONFIG_SND_ENS1370 is not set @@ -1314,6 +1361,17 @@ CONFIG_SND_DUMMY=m CONFIG_SND_POWERMAC=m CONFIG_SND_POWERMAC_AUTO_DRC=y +# +# Apple Onboard Audio driver +# +CONFIG_SND_AOA=m +CONFIG_SND_AOA_FABRIC_LAYOUT=m +CONFIG_SND_AOA_ONYX=m +CONFIG_SND_AOA_TAS=m +CONFIG_SND_AOA_TOONIE=m +CONFIG_SND_AOA_SOUNDBUS=m +CONFIG_SND_AOA_SOUNDBUS_I2S=m + # # USB devices # @@ -1355,6 +1413,7 @@ CONFIG_USB_DYNAMIC_MINORS=y CONFIG_USB_EHCI_HCD=m CONFIG_USB_EHCI_SPLIT_ISO=y CONFIG_USB_EHCI_ROOT_HUB_TT=y +# CONFIG_USB_EHCI_TT_NEWSCHED is not set # CONFIG_USB_ISP116X_HCD is not set CONFIG_USB_OHCI_HCD=y # CONFIG_USB_OHCI_BIG_ENDIAN is not set @@ -1431,7 +1490,6 @@ CONFIG_USB_NET_NET1080=m # CONFIG_USB_NET_RNDIS_HOST is not set # CONFIG_USB_NET_CDC_SUBSET is not set CONFIG_USB_NET_ZAURUS=m -# CONFIG_USB_ZD1201 is not set CONFIG_USB_MON=y # @@ -1499,10 +1557,12 @@ CONFIG_USB_EZUSB=y # CONFIG_USB_LEGOTOWER is not set # CONFIG_USB_LCD is not set # CONFIG_USB_LED is not set +# CONFIG_USB_CY7C63 is not set # CONFIG_USB_CYTHERM is not set # CONFIG_USB_PHIDGETKIT is not set # CONFIG_USB_PHIDGETSERVO is not set # CONFIG_USB_IDMOUSE is not set +CONFIG_USB_APPLEDISPLAY=m # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set # CONFIG_USB_TEST is not set @@ -1524,7 +1584,8 @@ CONFIG_USB_EZUSB=y # # LED devices # -# CONFIG_NEW_LEDS is not set +CONFIG_NEW_LEDS=y +CONFIG_LEDS_CLASS=y # # LED drivers @@ -1533,6 +1594,10 @@ CONFIG_USB_EZUSB=y # # LED Triggers # +CONFIG_LEDS_TRIGGERS=y +# CONFIG_LEDS_TRIGGER_TIMER is not set +CONFIG_LEDS_TRIGGER_IDE_DISK=y +# CONFIG_LEDS_TRIGGER_HEARTBEAT is not set # # InfiniBand support @@ -1548,6 +1613,19 @@ CONFIG_USB_EZUSB=y # # CONFIG_RTC_CLASS is not set +# +# DMA Engine support +# +# CONFIG_DMA_ENGINE is not set + +# +# DMA Clients +# + +# +# DMA Devices +# + # # File systems # @@ -1569,6 +1647,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set CONFIG_INOTIFY=y +CONFIG_INOTIFY_USER=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -1649,6 +1728,7 @@ CONFIG_RPCSEC_GSS_KRB5=y CONFIG_SMB_FS=m # CONFIG_SMB_NLS_DEFAULT is not set # CONFIG_CIFS is not set +# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -1732,6 +1812,7 @@ CONFIG_TEXTSEARCH=y CONFIG_TEXTSEARCH_KMP=m CONFIG_TEXTSEARCH_BM=m CONFIG_TEXTSEARCH_FSM=m +CONFIG_PLIST=y # # Instrumentation Support @@ -1744,12 +1825,15 @@ CONFIG_OPROFILE=y # # CONFIG_PRINTK_TIME is not set # CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=14 CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set # CONFIG_DEBUG_MUTEXES is not set +# CONFIG_DEBUG_RT_MUTEXES is not set +# CONFIG_RT_MUTEX_TESTER is not set # CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set @@ -1763,11 +1847,7 @@ CONFIG_XMON=y CONFIG_XMON_DEFAULT=y # CONFIG_BDI_SWITCH is not set CONFIG_BOOTX_TEXT=y -# CONFIG_PPC_EARLY_DEBUG_LPAR is not set -# CONFIG_PPC_EARLY_DEBUG_G5 is not set -# CONFIG_PPC_EARLY_DEBUG_RTAS is not set -# CONFIG_PPC_EARLY_DEBUG_MAPLE is not set -# CONFIG_PPC_EARLY_DEBUG_ISERIES is not set +# CONFIG_PPC_EARLY_DEBUG is not set # # Security options diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig index d1266fe2d1ab..53bba41f29bc 100644 --- a/drivers/ide/Kconfig +++ b/drivers/ide/Kconfig @@ -773,20 +773,6 @@ config BLK_DEV_IDEDMA_PMAC to transfer data to and from memory. Saying Y is safe and improves performance. -config BLK_DEV_IDE_PMAC_BLINK - bool "Blink laptop LED on drive activity (DEPRECATED)" - depends on BLK_DEV_IDE_PMAC && ADB_PMU - select ADB_PMU_LED - select LEDS_TRIGGERS - select LEDS_TRIGGER_IDE_DISK - help - This option enables the use of the sleep LED as a hard drive - activity LED. - This option is deprecated, it only selects ADB_PMU_LED and - LEDS_TRIGGER_IDE_DISK and changes the code in the new led class - device to default to the ide-disk trigger (which should be set - from userspace via sysfs). - config BLK_DEV_IDE_SWARM tristate "IDE for Sibyte evaluation boards" depends on SIBYTE_SB1xxx_SOC diff --git a/drivers/macintosh/Kconfig b/drivers/macintosh/Kconfig index f5fe7fb4b3ad..f18c956fd595 100644 --- a/drivers/macintosh/Kconfig +++ b/drivers/macintosh/Kconfig @@ -90,6 +90,15 @@ config ADB_PMU_LED and the ide-disk LED trigger and configure appropriately through sysfs. +config ADB_PMU_LED_IDE + bool "Use front LED as IDE LED by default" + depends on ADB_PMU_LED + select LEDS_TRIGGERS + select LEDS_TRIGGER_IDE_DISK + help + This option makes the front LED default to the IDE trigger + so that it blinks on IDE activity. + config PMAC_SMU bool "Support for SMU based PowerMacs" depends on PPC_PMAC64 diff --git a/drivers/macintosh/via-pmu-led.c b/drivers/macintosh/via-pmu-led.c index af8375ed0f5e..5189d5454b1f 100644 --- a/drivers/macintosh/via-pmu-led.c +++ b/drivers/macintosh/via-pmu-led.c @@ -74,7 +74,7 @@ static void pmu_led_set(struct led_classdev *led_cdev, static struct led_classdev pmu_led = { .name = "pmu-front-led", -#ifdef CONFIG_BLK_DEV_IDE_PMAC_BLINK +#ifdef CONFIG_ADB_PMU_LED_IDE .default_trigger = "ide-disk", #endif .brightness_set = pmu_led_set, From f0ca330bc4ce00377f940b786d7ec7d848dc4102 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 14:06:29 +1000 Subject: [PATCH 010/143] [POWERPC] Workaround Pegasos incorrect ISA "ranges" The Pegasos firmware doesn't create a valid "ranges" property for the ISA bridge, thus causing translation of ISA addresses and IO ports to fail. This fixes it, thus re-enabling proper early serial console to work on Pegasos. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/prom_init.c | 34 +++++++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index ebd501a59abd..b6c3ac20c14c 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -2030,6 +2030,39 @@ static void __init fixup_device_tree_maple(void) #define fixup_device_tree_maple() #endif +#ifdef CONFIG_PPC_CHRP +/* Pegasos lacks the "ranges" property in the isa node */ +static void __init fixup_device_tree_chrp(void) +{ + phandle isa; + u32 isa_ranges[6]; + char *name; + int rc; + + name = "/pci@80000000/isa@c"; + isa = call_prom("finddevice", 1, 1, ADDR(name)); + if (!PHANDLE_VALID(isa)) + return; + + rc = prom_getproplen(isa, "ranges"); + if (rc != 0 && rc != PROM_ERROR) + return; + + prom_printf("Fixing up missing ISA range on Pegasos...\n"); + + isa_ranges[0] = 0x1; + isa_ranges[1] = 0x0; + isa_ranges[2] = 0x01006000; + isa_ranges[3] = 0x0; + isa_ranges[4] = 0x0; + isa_ranges[5] = 0x00010000; + prom_setprop(isa, name, "ranges", + isa_ranges, sizeof(isa_ranges)); +} +#else +#define fixup_device_tree_chrp() +#endif + #if defined(CONFIG_PPC64) && defined(CONFIG_PPC_PMAC) static void __init fixup_device_tree_pmac(void) { @@ -2077,6 +2110,7 @@ static void __init fixup_device_tree_pmac(void) static void __init fixup_device_tree(void) { fixup_device_tree_maple(); + fixup_device_tree_chrp(); fixup_device_tree_pmac(); } From 21bd2e6696bb5c8b32b00a0ea597f49bfda0ddc6 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 14:07:42 +1000 Subject: [PATCH 011/143] [POWERPC] Fix 32 bits warning in prom_init.c A warning is hurting my eyes when building 32 bits kernels Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/prom_init.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index b6c3ac20c14c..462bced40c12 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -557,7 +557,9 @@ unsigned long prom_memparse(const char *ptr, const char **retptr) static void __init early_cmdline_parse(void) { struct prom_t *_prom = &RELOC(prom); +#ifdef CONFIG_PPC64 const char *opt; +#endif char *p; int l = 0; From 8cffc6ac66a2b251df2490702923611aa4ac1fc5 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 14:09:36 +1000 Subject: [PATCH 012/143] [POWERPC] Fix non-MPIC CHRPs with CONFIG_SMP set Pseudo-CHRP machines like Pegasos without an MPIC would crash at boot if CONFIG_SMP was set because the "smp_ops" pointer was set to MPIC related ops unconditionally. This patch makes it NULL on machines that don't support SMP and provides proper default behaviour in the callers when smp_ops is NULL. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/smp.c | 24 +++++++++++++++++------- arch/powerpc/platforms/chrp/setup.c | 12 ++++++++---- 2 files changed, 25 insertions(+), 11 deletions(-) diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 46c56cfd1b2f..6a9bc9ce54e0 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -144,13 +144,15 @@ void smp_message_recv(int msg, struct pt_regs *regs) void smp_send_reschedule(int cpu) { - smp_ops->message_pass(cpu, PPC_MSG_RESCHEDULE); + if (likely(smp_ops)) + smp_ops->message_pass(cpu, PPC_MSG_RESCHEDULE); } #ifdef CONFIG_DEBUGGER void smp_send_debugger_break(int cpu) { - smp_ops->message_pass(cpu, PPC_MSG_DEBUGGER_BREAK); + if (likely(smp_ops)) + smp_ops->message_pass(cpu, PPC_MSG_DEBUGGER_BREAK); } #endif @@ -158,7 +160,7 @@ void smp_send_debugger_break(int cpu) void crash_send_ipi(void (*crash_ipi_callback)(struct pt_regs *)) { crash_ipi_function_ptr = crash_ipi_callback; - if (crash_ipi_callback) { + if (crash_ipi_callback && smp_ops) { mb(); smp_ops->message_pass(MSG_ALL_BUT_SELF, PPC_MSG_DEBUGGER_BREAK); } @@ -220,6 +222,9 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, /* Can deadlock when called with interrupts disabled */ WARN_ON(irqs_disabled()); + if (unlikely(smp_ops == NULL)) + return -1; + data.func = func; data.info = info; atomic_set(&data.started, 0); @@ -357,7 +362,10 @@ void __init smp_prepare_cpus(unsigned int max_cpus) smp_store_cpu_info(boot_cpuid); cpu_callin_map[boot_cpuid] = 1; - max_cpus = smp_ops->probe(); + if (smp_ops) + max_cpus = smp_ops->probe(); + else + max_cpus = 1; smp_space_timers(max_cpus); @@ -453,7 +461,7 @@ void generic_mach_cpu_die(void) static int __devinit cpu_enable(unsigned int cpu) { - if (smp_ops->cpu_enable) + if (smp_ops && smp_ops->cpu_enable) return smp_ops->cpu_enable(cpu); return -ENOSYS; @@ -467,7 +475,8 @@ int __devinit __cpu_up(unsigned int cpu) if (!cpu_enable(cpu)) return 0; - if (smp_ops->cpu_bootable && !smp_ops->cpu_bootable(cpu)) + if (smp_ops == NULL || + (smp_ops->cpu_bootable && !smp_ops->cpu_bootable(cpu))) return -EINVAL; /* Make sure callin-map entry is 0 (can be leftover a CPU @@ -568,7 +577,8 @@ void __init smp_cpus_done(unsigned int max_cpus) old_mask = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(boot_cpuid)); - smp_ops->setup_cpu(boot_cpuid); + if (smp_ops) + smp_ops->setup_cpu(boot_cpuid); set_cpus_allowed(current, old_mask); diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c index 538e337d63e2..9c08ff322290 100644 --- a/arch/powerpc/platforms/chrp/setup.c +++ b/arch/powerpc/platforms/chrp/setup.c @@ -291,10 +291,6 @@ void __init chrp_setup_arch(void) pci_create_OF_bus_map(); -#ifdef CONFIG_SMP - smp_ops = &chrp_smp_ops; -#endif /* CONFIG_SMP */ - /* * Print the banner, then scroll down so boot progress * can be printed. -- Cort @@ -479,6 +475,14 @@ void __init chrp_init_IRQ(void) chrp_find_openpic(); chrp_find_8259(); +#ifdef CONFIG_SMP + /* Pegasos has no MPIC, those ops would make it crash. It might be an + * option to move setting them to after we probe the PIC though + */ + if (chrp_mpic != NULL) + smp_ops = &chrp_smp_ops; +#endif /* CONFIG_SMP */ + if (_chrp_type == _CHRP_Pegasos) ppc_md.get_irq = i8259_irq; From 171505dafef56e4c84c48399b36604f4be55cf36 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 14:11:23 +1000 Subject: [PATCH 013/143] [POWERPC] Fix default clock for udbg_16550 This patch makes it possible to provide 0 as the clock value for udbg_16550, making it default to the standard 1.8432Mhz clock Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/udbg_16550.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/udbg_16550.c b/arch/powerpc/kernel/udbg_16550.c index 0835b4841dea..2d17f2b8eda7 100644 --- a/arch/powerpc/kernel/udbg_16550.c +++ b/arch/powerpc/kernel/udbg_16550.c @@ -81,10 +81,14 @@ static int udbg_550_getc(void) void udbg_init_uart(void __iomem *comport, unsigned int speed, unsigned int clock) { - unsigned int dll, base_bauds = clock / 16; + unsigned int dll, base_bauds; + if (clock == 0) + clock = 1843200; if (speed == 0) speed = 9600; + + base_bauds = clock / 16; dll = base_bauds / speed; if (comport) { From 45507ff3a2c58bc4781688eac1a80979ea972dce Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 14:14:07 +1000 Subject: [PATCH 014/143] [POWERPC] Fix legacy_serial.c error handling on 32 bits The code in legacy_serial.c wouldn't properly compare OF translation results against OF_BAD_ADDR as it's using a phys_addr_t which is 32 bits on some 32-bit powerpc platforms. This fixes it by always using a u64 which is what is returned by the OF parsing routines. It also makes translation failure harmless for ISA serial ports. If they can't translate, we can't use the UART early, but we can still let the 8250 driver use it later on by using IO port accessors. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/legacy_serial.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/kernel/legacy_serial.c b/arch/powerpc/kernel/legacy_serial.c index 7e98e778b52f..359ab89748e0 100644 --- a/arch/powerpc/kernel/legacy_serial.c +++ b/arch/powerpc/kernel/legacy_serial.c @@ -112,7 +112,7 @@ static int __init add_legacy_port(struct device_node *np, int want_index, static int __init add_legacy_soc_port(struct device_node *np, struct device_node *soc_dev) { - phys_addr_t addr; + u64 addr; u32 *addrp; upf_t flags = UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_SHARE_IRQ; @@ -143,7 +143,7 @@ static int __init add_legacy_isa_port(struct device_node *np, u32 *reg; char *typep; int index = -1; - phys_addr_t taddr; + u64 taddr; DBG(" -> add_legacy_isa_port(%s)\n", np->full_name); @@ -165,10 +165,13 @@ static int __init add_legacy_isa_port(struct device_node *np, if (typep && *typep == 'S') index = simple_strtol(typep+1, NULL, 0) - 1; - /* Translate ISA address */ + /* Translate ISA address. If it fails, we still register the port + * with no translated address so that it can be picked up as an IO + * port later by the serial driver + */ taddr = of_translate_address(np, reg); if (taddr == OF_BAD_ADDR) - return -1; + taddr = 0; /* Add port, irq will be dealt with later */ return add_legacy_port(np, index, UPIO_PORT, reg[1], taddr, @@ -180,7 +183,7 @@ static int __init add_legacy_isa_port(struct device_node *np, static int __init add_legacy_pci_port(struct device_node *np, struct device_node *pci_dev) { - phys_addr_t addr, base; + u64 addr, base; u32 *addrp; unsigned int flags; int iotype, index = -1, lindex = 0; From 98c82472e98469ef23d7c680a0d6be1429540166 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 4 Jul 2006 17:07:18 +1000 Subject: [PATCH 015/143] [POWERPC] More offb/bootx fixes There were still some issues with offb when BootX doesn't provide a proper display node, this fixes them. This also re-instates the palette hacks that were disabled a couple of kernel versions ago when I converted to the new OF parsing, and shuffles some functions around to avoid prototypes. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/powermac/bootx_init.c | 35 ++- drivers/video/offb.c | 307 ++++++++++--------- 2 files changed, 185 insertions(+), 157 deletions(-) diff --git a/arch/powerpc/platforms/powermac/bootx_init.c b/arch/powerpc/platforms/powermac/bootx_init.c index 871b002c9f90..6a026c733f6a 100644 --- a/arch/powerpc/platforms/powermac/bootx_init.c +++ b/arch/powerpc/platforms/powermac/bootx_init.c @@ -181,13 +181,18 @@ static void __init bootx_add_chosen_props(unsigned long base, } static void __init bootx_add_display_props(unsigned long base, - unsigned long *mem_end) + unsigned long *mem_end, + int has_real_node) { boot_infos_t *bi = bootx_info; u32 tmp; - bootx_dt_add_prop("linux,boot-display", NULL, 0, mem_end); - bootx_dt_add_prop("linux,opened", NULL, 0, mem_end); + if (has_real_node) { + bootx_dt_add_prop("linux,boot-display", NULL, 0, mem_end); + bootx_dt_add_prop("linux,opened", NULL, 0, mem_end); + } else + bootx_dt_add_prop("linux,bootx-noscreen", NULL, 0, mem_end); + tmp = bi->dispDeviceDepth; bootx_dt_add_prop("linux,bootx-depth", &tmp, 4, mem_end); tmp = bi->dispDeviceRect[2] - bi->dispDeviceRect[0]; @@ -241,11 +246,6 @@ static void __init bootx_scan_dt_build_strings(unsigned long base, DBG(" detected display ! adding properties names !\n"); bootx_dt_add_string("linux,boot-display", mem_end); bootx_dt_add_string("linux,opened", mem_end); - bootx_dt_add_string("linux,bootx-depth", mem_end); - bootx_dt_add_string("linux,bootx-width", mem_end); - bootx_dt_add_string("linux,bootx-height", mem_end); - bootx_dt_add_string("linux,bootx-linebytes", mem_end); - bootx_dt_add_string("linux,bootx-addr", mem_end); strncpy(bootx_disp_path, namep, 255); } @@ -329,10 +329,13 @@ static void __init bootx_scan_dt_build_struct(unsigned long base, ppp = &pp->next; } - if (node == bootx_node_chosen) + if (node == bootx_node_chosen) { bootx_add_chosen_props(base, mem_end); - if (node == bootx_info->dispDeviceRegEntryOffset) - bootx_add_display_props(base, mem_end); + if (bootx_info->dispDeviceRegEntryOffset == 0) + bootx_add_display_props(base, mem_end, 0); + } + else if (node == bootx_info->dispDeviceRegEntryOffset) + bootx_add_display_props(base, mem_end, 1); /* do all our children */ cpp = &np->child; @@ -374,6 +377,14 @@ static unsigned long __init bootx_flatten_dt(unsigned long start) mem_end += 4; bootx_dt_strend = mem_end; bootx_scan_dt_build_strings(base, 4, &mem_end); + /* Add some strings */ + bootx_dt_add_string("linux,bootx-noscreen", &mem_end); + bootx_dt_add_string("linux,bootx-depth", &mem_end); + bootx_dt_add_string("linux,bootx-width", &mem_end); + bootx_dt_add_string("linux,bootx-height", &mem_end); + bootx_dt_add_string("linux,bootx-linebytes", &mem_end); + bootx_dt_add_string("linux,bootx-addr", &mem_end); + /* Wrap up strings */ hdr->off_dt_strings = bootx_dt_strbase - mem_start; hdr->dt_strings_size = bootx_dt_strend - bootx_dt_strbase; @@ -471,6 +482,7 @@ void __init bootx_init(unsigned long r3, unsigned long r4) if (bi->dispDeviceDepth == 16) bi->dispDeviceDepth = 15; + #ifdef CONFIG_BOOTX_TEXT ptr = (unsigned long)bi->logicalDisplayBase; ptr += bi->dispDeviceRect[1] * bi->dispDeviceRowBytes; @@ -508,6 +520,7 @@ void __init bootx_init(unsigned long r3, unsigned long r4) #ifdef CONFIG_BOOTX_TEXT btext_welcome(bi); #endif + /* New BootX enters kernel with MMU off, i/os are not allowed * here. This hack will have been done by the boostrap anyway. */ diff --git a/drivers/video/offb.c b/drivers/video/offb.c index ce5f3031b99b..0013311e0564 100644 --- a/drivers/video/offb.c +++ b/drivers/video/offb.c @@ -62,8 +62,6 @@ struct offb_par default_par; * Interface used by the world */ -int offb_init(void); - static int offb_setcolreg(u_int regno, u_int red, u_int green, u_int blue, u_int transp, struct fb_info *info); static int offb_blank(int blank, struct fb_info *info); @@ -72,11 +70,6 @@ static int offb_blank(int blank, struct fb_info *info); extern boot_infos_t *boot_infos; #endif -static void offb_init_nodriver(struct device_node *); -static void offb_init_fb(const char *name, const char *full_name, - int width, int height, int depth, int pitch, - unsigned long address, struct device_node *dp); - static struct fb_ops offb_ops = { .owner = THIS_MODULE, .fb_setcolreg = offb_setcolreg, @@ -229,123 +222,17 @@ static int offb_blank(int blank, struct fb_info *info) return 0; } - /* - * Initialisation - */ -int __init offb_init(void) +static void __iomem *offb_map_reg(struct device_node *np, int index, + unsigned long offset, unsigned long size) { - struct device_node *dp = NULL, *boot_disp = NULL; + struct resource r; - if (fb_get_options("offb", NULL)) - return -ENODEV; - - for (dp = NULL; (dp = of_find_node_by_type(dp, "display"));) { - if (get_property(dp, "linux,opened", NULL) && - get_property(dp, "linux,boot-display", NULL)) { - boot_disp = dp; - offb_init_nodriver(dp); - } - } - for (dp = NULL; (dp = of_find_node_by_type(dp, "display"));) { - if (get_property(dp, "linux,opened", NULL) && - dp != boot_disp) - offb_init_nodriver(dp); - } - - return 0; -} - - -static void __init offb_init_nodriver(struct device_node *dp) -{ - unsigned int len; - int i, width = 640, height = 480, depth = 8, pitch = 640; - unsigned int flags, rsize, addr_prop = 0; - unsigned long max_size = 0; - u64 rstart, address = OF_BAD_ADDR; - u32 *pp, *addrp, *up; - u64 asize; - - pp = (u32 *)get_property(dp, "linux,bootx-depth", &len); - if (pp == NULL) - pp = (u32 *)get_property(dp, "depth", &len); - if (pp && len == sizeof(u32)) - depth = *pp; - - pp = (u32 *)get_property(dp, "linux,bootx-width", &len); - if (pp == NULL) - pp = (u32 *)get_property(dp, "width", &len); - if (pp && len == sizeof(u32)) - width = *pp; - - pp = (u32 *)get_property(dp, "linux,bootx-height", &len); - if (pp == NULL) - pp = (u32 *)get_property(dp, "height", &len); - if (pp && len == sizeof(u32)) - height = *pp; - - pp = (u32 *)get_property(dp, "linux,bootx-linebytes", &len); - if (pp == NULL) - pp = (u32 *)get_property(dp, "linebytes", &len); - if (pp && len == sizeof(u32)) - pitch = *pp; - else - pitch = width * ((depth + 7) / 8); - - rsize = (unsigned long)pitch * (unsigned long)height; - - /* Ok, now we try to figure out the address of the framebuffer. - * - * Unfortunately, Open Firmware doesn't provide a standard way to do - * so. All we can do is a dodgy heuristic that happens to work in - * practice. On most machines, the "address" property contains what - * we need, though not on Matrox cards found in IBM machines. What I've - * found that appears to give good results is to go through the PCI - * ranges and pick one that is both big enough and if possible encloses - * the "address" property. If none match, we pick the biggest - */ - up = (u32 *)get_property(dp, "linux,bootx-addr", &len); - if (up == NULL) - up = (u32 *)get_property(dp, "address", &len); - if (up && len == sizeof(u32)) - addr_prop = *up; - - for (i = 0; (addrp = of_get_address(dp, i, &asize, &flags)) - != NULL; i++) { - int match_addrp = 0; - - if (!(flags & IORESOURCE_MEM)) - continue; - if (asize < rsize) - continue; - rstart = of_translate_address(dp, addrp); - if (rstart == OF_BAD_ADDR) - continue; - if (addr_prop && (rstart <= addr_prop) && - ((rstart + asize) >= (addr_prop + rsize))) - match_addrp = 1; - if (match_addrp) { - address = addr_prop; - break; - } - if (rsize > max_size) { - max_size = rsize; - address = OF_BAD_ADDR; - } - - if (address == OF_BAD_ADDR) - address = rstart; - } - if (address == OF_BAD_ADDR && addr_prop) - address = (u64)addr_prop; - if (address != OF_BAD_ADDR) { - /* kludge for valkyrie */ - if (strcmp(dp->name, "valkyrie") == 0) - address += 0x1000; - offb_init_fb(dp->name, dp->full_name, width, height, depth, - pitch, address, dp); - } + if (of_address_to_resource(np, index, &r)) + return 0; + if ((r.start + offset + size) > r.end) + return 0; + return ioremap(r.start + offset, size); } static void __init offb_init_fb(const char *name, const char *full_name, @@ -402,45 +289,39 @@ static void __init offb_init_fb(const char *name, const char *full_name, par->cmap_type = cmap_unknown; if (depth == 8) { - /* Palette hacks disabled for now */ -#if 0 if (dp && !strncmp(name, "ATY,Rage128", 11)) { - unsigned long regbase = dp->addrs[2].address; - par->cmap_adr = ioremap(regbase, 0x1FFF); - par->cmap_type = cmap_r128; + par->cmap_adr = offb_map_reg(dp, 2, 0, 0x1fff); + if (par->cmap_adr) + par->cmap_type = cmap_r128; } else if (dp && (!strncmp(name, "ATY,RageM3pA", 12) || !strncmp(name, "ATY,RageM3p12A", 14))) { - unsigned long regbase = - dp->parent->addrs[2].address; - par->cmap_adr = ioremap(regbase, 0x1FFF); - par->cmap_type = cmap_M3A; + par->cmap_adr = offb_map_reg(dp, 2, 0, 0x1fff); + if (par->cmap_adr) + par->cmap_type = cmap_M3A; } else if (dp && !strncmp(name, "ATY,RageM3pB", 12)) { - unsigned long regbase = - dp->parent->addrs[2].address; - par->cmap_adr = ioremap(regbase, 0x1FFF); - par->cmap_type = cmap_M3B; + par->cmap_adr = offb_map_reg(dp, 2, 0, 0x1fff); + if (par->cmap_adr) + par->cmap_type = cmap_M3B; } else if (dp && !strncmp(name, "ATY,Rage6", 9)) { - unsigned long regbase = dp->addrs[1].address; - par->cmap_adr = ioremap(regbase, 0x1FFF); - par->cmap_type = cmap_radeon; + par->cmap_adr = offb_map_reg(dp, 1, 0, 0x1fff); + if (par->cmap_adr) + par->cmap_type = cmap_radeon; } else if (!strncmp(name, "ATY,", 4)) { unsigned long base = address & 0xff000000UL; par->cmap_adr = ioremap(base + 0x7ff000, 0x1000) + 0xcc0; par->cmap_data = par->cmap_adr + 1; par->cmap_type = cmap_m64; - } else if (device_is_compatible(dp, "pci1014,b7")) { - unsigned long regbase = dp->addrs[0].address; - par->cmap_adr = ioremap(regbase + 0x6000, 0x1000); - par->cmap_type = cmap_gxt2000; + } else if (dp && device_is_compatible(dp, "pci1014,b7")) { + par->cmap_adr = offb_map_reg(dp, 0, 0x6000, 0x1000); + if (par->cmap_adr) + par->cmap_type = cmap_gxt2000; } -#endif - fix->visual = par->cmap_adr ? FB_VISUAL_PSEUDOCOLOR - : FB_VISUAL_STATIC_PSEUDOCOLOR; + fix->visual = (par->cmap_type != cmap_unknown) ? + FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_STATIC_PSEUDOCOLOR; } else - fix->visual = /* par->cmap_adr ? FB_VISUAL_DIRECTCOLOR - : */ FB_VISUAL_TRUECOLOR; + fix->visual = FB_VISUAL_TRUECOLOR; var->xoffset = var->yoffset = 0; switch (depth) { @@ -520,5 +401,139 @@ static void __init offb_init_fb(const char *name, const char *full_name, info->node, full_name); } + +static void __init offb_init_nodriver(struct device_node *dp, int no_real_node) +{ + unsigned int len; + int i, width = 640, height = 480, depth = 8, pitch = 640; + unsigned int flags, rsize, addr_prop = 0; + unsigned long max_size = 0; + u64 rstart, address = OF_BAD_ADDR; + u32 *pp, *addrp, *up; + u64 asize; + + pp = (u32 *)get_property(dp, "linux,bootx-depth", &len); + if (pp == NULL) + pp = (u32 *)get_property(dp, "depth", &len); + if (pp && len == sizeof(u32)) + depth = *pp; + + pp = (u32 *)get_property(dp, "linux,bootx-width", &len); + if (pp == NULL) + pp = (u32 *)get_property(dp, "width", &len); + if (pp && len == sizeof(u32)) + width = *pp; + + pp = (u32 *)get_property(dp, "linux,bootx-height", &len); + if (pp == NULL) + pp = (u32 *)get_property(dp, "height", &len); + if (pp && len == sizeof(u32)) + height = *pp; + + pp = (u32 *)get_property(dp, "linux,bootx-linebytes", &len); + if (pp == NULL) + pp = (u32 *)get_property(dp, "linebytes", &len); + if (pp && len == sizeof(u32)) + pitch = *pp; + else + pitch = width * ((depth + 7) / 8); + + rsize = (unsigned long)pitch * (unsigned long)height; + + /* Ok, now we try to figure out the address of the framebuffer. + * + * Unfortunately, Open Firmware doesn't provide a standard way to do + * so. All we can do is a dodgy heuristic that happens to work in + * practice. On most machines, the "address" property contains what + * we need, though not on Matrox cards found in IBM machines. What I've + * found that appears to give good results is to go through the PCI + * ranges and pick one that is both big enough and if possible encloses + * the "address" property. If none match, we pick the biggest + */ + up = (u32 *)get_property(dp, "linux,bootx-addr", &len); + if (up == NULL) + up = (u32 *)get_property(dp, "address", &len); + if (up && len == sizeof(u32)) + addr_prop = *up; + + /* Hack for when BootX is passing us */ + if (no_real_node) + goto skip_addr; + + for (i = 0; (addrp = of_get_address(dp, i, &asize, &flags)) + != NULL; i++) { + int match_addrp = 0; + + if (!(flags & IORESOURCE_MEM)) + continue; + if (asize < rsize) + continue; + rstart = of_translate_address(dp, addrp); + if (rstart == OF_BAD_ADDR) + continue; + if (addr_prop && (rstart <= addr_prop) && + ((rstart + asize) >= (addr_prop + rsize))) + match_addrp = 1; + if (match_addrp) { + address = addr_prop; + break; + } + if (rsize > max_size) { + max_size = rsize; + address = OF_BAD_ADDR; + } + + if (address == OF_BAD_ADDR) + address = rstart; + } + skip_addr: + if (address == OF_BAD_ADDR && addr_prop) + address = (u64)addr_prop; + if (address != OF_BAD_ADDR) { + /* kludge for valkyrie */ + if (strcmp(dp->name, "valkyrie") == 0) + address += 0x1000; + offb_init_fb(no_real_node ? "bootx" : dp->name, + no_real_node ? "display" : dp->full_name, + width, height, depth, pitch, address, + no_real_node ? dp : NULL); + } +} + +static int __init offb_init(void) +{ + struct device_node *dp = NULL, *boot_disp = NULL; + + if (fb_get_options("offb", NULL)) + return -ENODEV; + + /* Check if we have a MacOS display without a node spec */ + if (get_property(of_chosen, "linux,bootx-noscreen", NULL) != NULL) { + /* The old code tried to work out which node was the MacOS + * display based on the address. I'm dropping that since the + * lack of a node spec only happens with old BootX versions + * (users can update) and with this code, they'll still get + * a display (just not the palette hacks). + */ + offb_init_nodriver(of_chosen, 1); + } + + for (dp = NULL; (dp = of_find_node_by_type(dp, "display"));) { + if (get_property(dp, "linux,opened", NULL) && + get_property(dp, "linux,boot-display", NULL)) { + boot_disp = dp; + offb_init_nodriver(dp, 0); + } + } + for (dp = NULL; (dp = of_find_node_by_type(dp, "display"));) { + if (get_property(dp, "linux,opened", NULL) && + dp != boot_disp) + offb_init_nodriver(dp, 0); + } + + return 0; +} + + module_init(offb_init); MODULE_LICENSE("GPL"); From 30f30e13055d1e018c70d2b1b3ff86f8c25c511b Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 4 Jul 2006 17:13:23 +1000 Subject: [PATCH 016/143] [POWERPC] Fix mem= handling when the memory limit is > RMO size There's a bug in my cleaned up mem= handling, if the memory limit is larger than the RMO size we'll erroneously enlarge the RMO size. Fix is to only change the RMO size if the memory limit is less than the current RMO value. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/mm/lmb.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/mm/lmb.c b/arch/powerpc/mm/lmb.c index 4b17a7359924..716a2906a24d 100644 --- a/arch/powerpc/mm/lmb.c +++ b/arch/powerpc/mm/lmb.c @@ -320,7 +320,8 @@ void __init lmb_enforce_memory_limit(unsigned long memory_limit) break; } - lmb.rmo_size = lmb.memory.region[0].size; + if (lmb.memory.region[0].size < lmb.rmo_size) + lmb.rmo_size = lmb.memory.region[0].size; /* And truncate any reserves above the limit also. */ for (i = 0; i < lmb.reserved.cnt; i++) { From 34d926266500919a43520d3d52a4f7020b9f0687 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 26 Jul 2006 18:57:40 +0100 Subject: [PATCH 017/143] [ARM] Fix SMP booting Processor support files now use r6 in their CPU setup code, so we can't rely on r6 being preserved. Use r7 instead. Signed-off-by: Russell King --- arch/arm/kernel/head.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 2242f5f7cb7d..4fe386eea4b4 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -114,9 +114,9 @@ ENTRY(secondary_startup) * Use the page tables supplied from __cpu_up. */ adr r4, __secondary_data - ldmia r4, {r5, r6, r13} @ address to jump to after + ldmia r4, {r5, r7, r13} @ address to jump to after sub r4, r4, r5 @ mmu has been enabled - ldr r4, [r6, r4] @ get secondary_data.pgdir + ldr r4, [r7, r4] @ get secondary_data.pgdir adr lr, __enable_mmu @ return address add pc, r10, #12 @ initialise processor @ (return control reg) @@ -125,7 +125,7 @@ ENTRY(secondary_startup) * r6 = &secondary_data */ ENTRY(__secondary_switched) - ldr sp, [r6, #4] @ get secondary_data.stack + ldr sp, [r7, #4] @ get secondary_data.stack mov fp, #0 b secondary_start_kernel From 705630dbddf743e600d654f5dc721d899db91594 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 26 Jul 2006 20:16:39 +0100 Subject: [PATCH 018/143] [ARM] 3732/1: S3C24XX: tidy syntax in osiris and anubis machines Patch from Ben Dooks Tidy the syntax, such as missing ,'s on the end of struct entries, in the Osiris and Anubis machines. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/mach-anubis.c | 25 +++++++++++++------------ arch/arm/mach-s3c2410/mach-osiris.c | 20 ++++++++++---------- 2 files changed, 23 insertions(+), 22 deletions(-) diff --git a/arch/arm/mach-s3c2410/mach-anubis.c b/arch/arm/mach-s3c2410/mach-anubis.c index 4a92d6f92d6b..0a5a7e4a24ad 100644 --- a/arch/arm/mach-s3c2410/mach-anubis.c +++ b/arch/arm/mach-s3c2410/mach-anubis.c @@ -60,11 +60,12 @@ static struct map_desc anubis_iodesc[] __initdata = { .virtual = (u32)S3C24XX_VA_ISA_BYTE, .pfn = __phys_to_pfn(0x0), .length = SZ_4M, - .type = MT_DEVICE + .type = MT_DEVICE, }, { .virtual = (u32)S3C24XX_VA_ISA_WORD, .pfn = __phys_to_pfn(0x0), - .length = SZ_4M, MT_DEVICE + .length = SZ_4M, + .type = MT_DEVICE, }, /* we could possibly compress the next set down into a set of smaller tables @@ -78,12 +79,12 @@ static struct map_desc anubis_iodesc[] __initdata = { .virtual = (u32)ANUBIS_VA_CTRL1, .pfn = __phys_to_pfn(ANUBIS_PA_CTRL1), .length = SZ_4K, - .type = MT_DEVICE + .type = MT_DEVICE, }, { .virtual = (u32)ANUBIS_VA_CTRL2, .pfn = __phys_to_pfn(ANUBIS_PA_CTRL2), .length = SZ_4K, - .type =MT_DEVICE + .type = MT_DEVICE, }, /* IDE drives */ @@ -126,7 +127,7 @@ static struct s3c24xx_uart_clksrc anubis_serial_clocks[] = { .name = "pclk", .divisor = 1, .min_baud = 0, - .max_baud = 0. + .max_baud = 0, } }; @@ -139,7 +140,7 @@ static struct s3c2410_uartcfg anubis_uartcfgs[] __initdata = { .ulcon = ULCON, .ufcon = UFCON, .clocks = anubis_serial_clocks, - .clocks_size = ARRAY_SIZE(anubis_serial_clocks) + .clocks_size = ARRAY_SIZE(anubis_serial_clocks), }, [1] = { .hwport = 2, @@ -148,7 +149,7 @@ static struct s3c2410_uartcfg anubis_uartcfgs[] __initdata = { .ulcon = ULCON, .ufcon = UFCON, .clocks = anubis_serial_clocks, - .clocks_size = ARRAY_SIZE(anubis_serial_clocks) + .clocks_size = ARRAY_SIZE(anubis_serial_clocks), }, }; @@ -162,7 +163,7 @@ static struct mtd_partition anubis_default_nand_part[] = { [0] = { .name = "Boot Agent", .size = SZ_16K, - .offset = 0 + .offset = 0, }, [1] = { .name = "/boot", @@ -194,21 +195,21 @@ static struct s3c2410_nand_set anubis_nand_sets[] = { .nr_chips = 1, .nr_map = external_map, .nr_partitions = ARRAY_SIZE(anubis_default_nand_part), - .partitions = anubis_default_nand_part + .partitions = anubis_default_nand_part, }, [0] = { .name = "chip0", .nr_chips = 1, .nr_map = chip0_map, .nr_partitions = ARRAY_SIZE(anubis_default_nand_part), - .partitions = anubis_default_nand_part + .partitions = anubis_default_nand_part, }, [2] = { .name = "chip1", .nr_chips = 1, .nr_map = chip1_map, .nr_partitions = ARRAY_SIZE(anubis_default_nand_part), - .partitions = anubis_default_nand_part + .partitions = anubis_default_nand_part, }, }; @@ -313,7 +314,7 @@ static struct s3c24xx_board anubis_board __initdata = { .devices = anubis_devices, .devices_count = ARRAY_SIZE(anubis_devices), .clocks = anubis_clocks, - .clocks_count = ARRAY_SIZE(anubis_clocks) + .clocks_count = ARRAY_SIZE(anubis_clocks), }; static void __init anubis_map_io(void) diff --git a/arch/arm/mach-s3c2410/mach-osiris.c b/arch/arm/mach-s3c2410/mach-osiris.c index 858fd03c6bc5..e193ba69e652 100644 --- a/arch/arm/mach-s3c2410/mach-osiris.c +++ b/arch/arm/mach-s3c2410/mach-osiris.c @@ -67,12 +67,12 @@ static struct map_desc osiris_iodesc[] __initdata = { .virtual = (u32)OSIRIS_VA_CTRL1, .pfn = __phys_to_pfn(OSIRIS_PA_CTRL1), .length = SZ_16K, - .type = MT_DEVICE + .type = MT_DEVICE, }, { .virtual = (u32)OSIRIS_VA_CTRL2, .pfn = __phys_to_pfn(OSIRIS_PA_CTRL2), .length = SZ_16K, - .type = MT_DEVICE + .type = MT_DEVICE, }, }; @@ -91,7 +91,7 @@ static struct s3c24xx_uart_clksrc osiris_serial_clocks[] = { .name = "pclk", .divisor = 1, .min_baud = 0, - .max_baud = 0. + .max_baud = 0, } }; @@ -103,7 +103,7 @@ static struct s3c2410_uartcfg osiris_uartcfgs[] __initdata = { .ulcon = ULCON, .ufcon = UFCON, .clocks = osiris_serial_clocks, - .clocks_size = ARRAY_SIZE(osiris_serial_clocks) + .clocks_size = ARRAY_SIZE(osiris_serial_clocks), }, [1] = { .hwport = 1, @@ -112,7 +112,7 @@ static struct s3c2410_uartcfg osiris_uartcfgs[] __initdata = { .ulcon = ULCON, .ufcon = UFCON, .clocks = osiris_serial_clocks, - .clocks_size = ARRAY_SIZE(osiris_serial_clocks) + .clocks_size = ARRAY_SIZE(osiris_serial_clocks), }, }; @@ -126,7 +126,7 @@ static struct mtd_partition osiris_default_nand_part[] = { [0] = { .name = "Boot Agent", .size = SZ_16K, - .offset = 0 + .offset = 0, }, [1] = { .name = "/boot", @@ -158,21 +158,21 @@ static struct s3c2410_nand_set osiris_nand_sets[] = { .nr_chips = 1, .nr_map = external_map, .nr_partitions = ARRAY_SIZE(osiris_default_nand_part), - .partitions = osiris_default_nand_part + .partitions = osiris_default_nand_part, }, [0] = { .name = "chip0", .nr_chips = 1, .nr_map = chip0_map, .nr_partitions = ARRAY_SIZE(osiris_default_nand_part), - .partitions = osiris_default_nand_part + .partitions = osiris_default_nand_part, }, [2] = { .name = "chip1", .nr_chips = 1, .nr_map = chip1_map, .nr_partitions = ARRAY_SIZE(osiris_default_nand_part), - .partitions = osiris_default_nand_part + .partitions = osiris_default_nand_part, }, }; @@ -245,7 +245,7 @@ static struct s3c24xx_board osiris_board __initdata = { .devices = osiris_devices, .devices_count = ARRAY_SIZE(osiris_devices), .clocks = osiris_clocks, - .clocks_count = ARRAY_SIZE(osiris_clocks) + .clocks_count = ARRAY_SIZE(osiris_clocks), }; static void __init osiris_map_io(void) From dffb0605e6f34c138ae5662d957fe6634dd4d71e Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 26 Jul 2006 21:07:04 +0100 Subject: [PATCH 019/143] [ARM] 3733/2: S3C24XX: Remove old IDE registers in Anubis Patch from Ben Dooks Remove unused IDE static mapping, now being ioremap()d by the simtec IDE driver. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/mach-anubis.c | 24 ------------------------ 1 file changed, 24 deletions(-) diff --git a/arch/arm/mach-s3c2410/mach-anubis.c b/arch/arm/mach-s3c2410/mach-anubis.c index 0a5a7e4a24ad..60641d452db3 100644 --- a/arch/arm/mach-s3c2410/mach-anubis.c +++ b/arch/arm/mach-s3c2410/mach-anubis.c @@ -86,30 +86,6 @@ static struct map_desc anubis_iodesc[] __initdata = { .length = SZ_4K, .type = MT_DEVICE, }, - - /* IDE drives */ - - { - .virtual = (u32)ANUBIS_IDEPRI, - .pfn = __phys_to_pfn(S3C2410_CS3), - .length = SZ_1M, - .type = MT_DEVICE - }, { - .virtual = (u32)ANUBIS_IDEPRIAUX, - .pfn = __phys_to_pfn(S3C2410_CS3+(1<<26)), - .length = SZ_1M, - .type = MT_DEVICE - }, { - .virtual = (u32)ANUBIS_IDESEC, - .pfn = __phys_to_pfn(S3C2410_CS4), - .length = SZ_1M, - .type = MT_DEVICE - }, { - .virtual = (u32)ANUBIS_IDESECAUX, - .pfn = __phys_to_pfn(S3C2410_CS4+(1<<26)), - .length = SZ_1M, - .type = MT_DEVICE - }, }; #define UCON S3C2410_UCON_DEFAULT | S3C2410_UCON_UCLK From ba4ba8a69dcb446450b5ddeca48a7bd15783f4c2 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Thu, 27 Jul 2006 14:00:23 +0200 Subject: [PATCH 020/143] [S390] permanent subchannel busy conditions may cause I/O stall In special conditions where a subchannel rejects the HALT I/O- instruction with a busy indication (cc 2), I/O may stall. I/O request termination logic retries HALT I/O indefinitely because it expects HALT I/O to alter the subchannel status which is not true when cc 2 is returned. In case of a busy indication, try CLEAR I/O instruction immediately. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/device_fsm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index ac6e0c7e43d9..7a39e0b0386c 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -152,7 +152,8 @@ ccw_device_cancel_halt_clear(struct ccw_device *cdev) if (cdev->private->iretry) { cdev->private->iretry--; ret = cio_halt(sch); - return (ret == 0) ? -EBUSY : ret; + if (ret != -EBUSY) + return (ret == 0) ? -EBUSY : ret; } /* halt io unsuccessful. */ cdev->private->iretry = 255; /* 255 clear retries. */ From 17088229846c078aa936ca64912ab221d083aca1 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 27 Jul 2006 14:00:33 +0200 Subject: [PATCH 021/143] [S390] duplicate ccw devices in ccwgroup. Fail to create a ccwgroup device if a ccw device is passed in twice. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/ccwgroup.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/drivers/s390/cio/ccwgroup.c b/drivers/s390/cio/ccwgroup.c index f26a2ee3aad8..3cba6c9fab11 100644 --- a/drivers/s390/cio/ccwgroup.c +++ b/drivers/s390/cio/ccwgroup.c @@ -152,7 +152,6 @@ ccwgroup_create(struct device *root, struct ccwgroup_device *gdev; int i; int rc; - int del_drvdata; if (argc > 256) /* disallow dumb users */ return -EINVAL; @@ -163,7 +162,6 @@ ccwgroup_create(struct device *root, atomic_set(&gdev->onoff, 0); - del_drvdata = 0; for (i = 0; i < argc; i++) { gdev->cdev[i] = get_ccwdev_by_busid(cdrv, argv[i]); @@ -180,10 +178,8 @@ ccwgroup_create(struct device *root, rc = -EINVAL; goto free_dev; } - } - for (i = 0; i < argc; i++) gdev->cdev[i]->dev.driver_data = gdev; - del_drvdata = 1; + } gdev->creator_id = creator_id; gdev->count = argc; @@ -226,9 +222,9 @@ ccwgroup_create(struct device *root, free_dev: for (i = 0; i < argc; i++) if (gdev->cdev[i]) { - put_device(&gdev->cdev[i]->dev); - if (del_drvdata) + if (gdev->cdev[i]->dev.driver_data == gdev) gdev->cdev[i]->dev.driver_data = NULL; + put_device(&gdev->cdev[i]->dev); } kfree(gdev); return rc; From 468310a8a7af4f3933ade2700f01d493fa1a9754 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 27 Jul 2006 14:04:57 +0200 Subject: [PATCH 022/143] [S390] update default configuration Signed-off-by: Martin Schwidefsky --- arch/s390/defconfig | 44 +++++++++++++++++++++++++++++++++++--------- 1 file changed, 35 insertions(+), 9 deletions(-) diff --git a/arch/s390/defconfig b/arch/s390/defconfig index f4dfc10026d2..f1d4591eddbb 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,13 +1,16 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-rc1 -# Mon Apr 3 14:34:15 2006 +# Linux kernel version: 2.6.18-rc2 +# Thu Jul 27 13:51:07 2006 # CONFIG_MMU=y +CONFIG_LOCKDEP_SUPPORT=y +CONFIG_STACKTRACE_SUPPORT=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_S390=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # # Code maturity level options @@ -25,6 +28,7 @@ CONFIG_SWAP=y CONFIG_SYSVIPC=y CONFIG_POSIX_MQUEUE=y # CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set CONFIG_SYSCTL=y CONFIG_AUDIT=y # CONFIG_AUDITSYSCALL is not set @@ -43,10 +47,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y +CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y +CONFIG_VM_EVENT_COUNTERS=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -94,7 +100,6 @@ CONFIG_HOTPLUG_CPU=y CONFIG_DEFAULT_MIGRATION_COST=1000000 CONFIG_COMPAT=y CONFIG_SYSVIPC_COMPAT=y -CONFIG_BINFMT_ELF32=y # # Code generation options @@ -115,6 +120,7 @@ CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y # CONFIG_SPARSEMEM_STATIC is not set CONFIG_SPLIT_PTLOCK_CPUS=4 +CONFIG_RESOURCES_64BIT=y # # I/O subsystem configuration @@ -142,6 +148,7 @@ CONFIG_VIRT_CPU_ACCOUNTING=y # CONFIG_APPLDATA_BASE is not set CONFIG_NO_IDLE_HZ=y CONFIG_NO_IDLE_HZ_INIT=y +CONFIG_S390_HYPFS_FS=y CONFIG_KEXEC=y # @@ -174,6 +181,8 @@ CONFIG_IP_FIB_HASH=y # CONFIG_INET_IPCOMP is not set # CONFIG_INET_XFRM_TUNNEL is not set # CONFIG_INET_TUNNEL is not set +CONFIG_INET_XFRM_MODE_TRANSPORT=y +CONFIG_INET_XFRM_MODE_TUNNEL=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set @@ -186,7 +195,10 @@ CONFIG_IPV6=y # CONFIG_INET6_IPCOMP is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set +CONFIG_INET6_XFRM_MODE_TRANSPORT=y +CONFIG_INET6_XFRM_MODE_TUNNEL=y # CONFIG_IPV6_TUNNEL is not set +# CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set # @@ -263,6 +275,7 @@ CONFIG_NET_ESTIMATOR=y # Network testing # # CONFIG_NET_PKTGEN is not set +# CONFIG_NET_TCPPROBE is not set # CONFIG_HAMRADIO is not set # CONFIG_IRDA is not set # CONFIG_BT is not set @@ -276,6 +289,7 @@ CONFIG_STANDALONE=y CONFIG_PREVENT_FIRMWARE_BUILD=y # CONFIG_FW_LOADER is not set # CONFIG_DEBUG_DRIVER is not set +CONFIG_SYS_HYPERVISOR=y # # Connector - unified userspace <-> kernelspace linker @@ -334,6 +348,7 @@ CONFIG_BLK_DEV_NBD=m CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set @@ -359,9 +374,7 @@ CONFIG_MD_LINEAR=m CONFIG_MD_RAID0=m CONFIG_MD_RAID1=m # CONFIG_MD_RAID10 is not set -CONFIG_MD_RAID5=m -# CONFIG_MD_RAID5_RESHAPE is not set -# CONFIG_MD_RAID6 is not set +# CONFIG_MD_RAID456 is not set CONFIG_MD_MULTIPATH=m # CONFIG_MD_FAULTY is not set CONFIG_BLK_DEV_DM=y @@ -419,7 +432,8 @@ CONFIG_S390_TAPE_34XX=m # # Cryptographic devices # -CONFIG_Z90CRYPT=m +CONFIG_ZCRYPT=m +# CONFIG_ZCRYPT_MONOLITHIC is not set # # Network device support @@ -509,6 +523,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set CONFIG_INOTIFY=y +CONFIG_INOTIFY_USER=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -614,26 +629,36 @@ CONFIG_MSDOS_PARTITION=y # Instrumentation Support # # CONFIG_PROFILING is not set -# CONFIG_STATISTICS is not set +CONFIG_STATISTICS=y +CONFIG_KPROBES=y # # Kernel hacking # +CONFIG_TRACE_IRQFLAGS_SUPPORT=y # CONFIG_PRINTK_TIME is not set CONFIG_MAGIC_SYSRQ=y +# CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=17 # CONFIG_DETECT_SOFTLOCKUP is not set # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set CONFIG_DEBUG_PREEMPT=y -CONFIG_DEBUG_MUTEXES=y +# CONFIG_DEBUG_RT_MUTEXES is not set +# CONFIG_RT_MUTEX_TESTER is not set CONFIG_DEBUG_SPINLOCK=y +CONFIG_DEBUG_MUTEXES=y +# CONFIG_DEBUG_RWSEMS is not set +# CONFIG_DEBUG_LOCK_ALLOC is not set +# CONFIG_PROVE_LOCKING is not set CONFIG_DEBUG_SPINLOCK_SLEEP=y +# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set +# CONFIG_FRAME_POINTER is not set # CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y # CONFIG_RCU_TORTURE_TEST is not set @@ -688,3 +713,4 @@ CONFIG_CRYPTO=y # CONFIG_CRC16 is not set CONFIG_CRC32=m # CONFIG_LIBCRC32C is not set +CONFIG_PLIST=y From 92f282988b4ce3967ee8399f7d1184ebfa04e48b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 27 Jul 2006 16:49:21 -0700 Subject: [PATCH 023/143] [SPARC64]: Fix quad-float multiply emulation. Something is wrong with the 3-multiply (vs. 4-multiply) optimized version of _FP_MUL_MEAT_2_*(), so just use the slower version which actually computes correct values. Noticed by Rene Rebe Signed-off-by: David S. Miller --- include/asm-sparc64/sfp-machine.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-sparc64/sfp-machine.h b/include/asm-sparc64/sfp-machine.h index 5015bb8d6c32..89d42431efb5 100644 --- a/include/asm-sparc64/sfp-machine.h +++ b/include/asm-sparc64/sfp-machine.h @@ -34,7 +34,7 @@ #define _FP_MUL_MEAT_D(R,X,Y) \ _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm) #define _FP_MUL_MEAT_Q(R,X,Y) \ - _FP_MUL_MEAT_2_wide_3mul(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm) + _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm) #define _FP_DIV_MEAT_S(R,X,Y) _FP_DIV_MEAT_1_imm(S,R,X,Y,_FP_DIV_HELP_imm) #define _FP_DIV_MEAT_D(R,X,Y) _FP_DIV_MEAT_1_udiv_norm(D,R,X,Y) From 8279d2e6062673f31c59ad9e2104024123154267 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 6 Jul 2006 18:03:06 +1000 Subject: [PATCH 024/143] [POWERPC] Xserve G5 thermal control fixes The thermal control for the Xserve G5s had a few issues. For one, the way to program the RPM fans speeds into the FCU is different between it and the desktop models, which I didn't figure out until recently, and it was missing a control loop for the slots fan, running it too fast. Both of those problems were causing the machine to be much more noisy than necessary. This patch also changes the fixed value of the slots fan for desktop G5s to 40% instead of 50%. It seems to still have a pretty good airflow that way and is much less noisy. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- drivers/macintosh/therm_pm72.c | 218 ++++++++++++++++++++++++++++++--- drivers/macintosh/therm_pm72.h | 33 ++++- 2 files changed, 234 insertions(+), 17 deletions(-) diff --git a/drivers/macintosh/therm_pm72.c b/drivers/macintosh/therm_pm72.c index c1fe0b368f76..20bf67244e2c 100644 --- a/drivers/macintosh/therm_pm72.c +++ b/drivers/macintosh/therm_pm72.c @@ -95,6 +95,17 @@ * - Use min/max macros here or there * - Latest darwin updated U3H min fan speed to 20% PWM * + * July. 06, 2006 : 1.3 + * - Fix setting of RPM fans on Xserve G5 (they were going too fast) + * - Add missing slots fan control loop for Xserve G5 + * - Lower fixed slots fan speed from 50% to 40% on desktop G5s. We + * still can't properly implement the control loop for these, so let's + * reduce the noise a little bit, it appears that 40% still gives us + * a pretty good air flow + * - Add code to "tickle" the FCU regulary so it doesn't think that + * we are gone while in fact, the machine just didn't need any fan + * speed change lately + * */ #include @@ -121,7 +132,7 @@ #include "therm_pm72.h" -#define VERSION "1.2b2" +#define VERSION "1.3" #undef DEBUG @@ -146,6 +157,7 @@ static struct basckside_pid_params backside_params; static struct backside_pid_state backside_state; static struct drives_pid_state drives_state; static struct dimm_pid_state dimms_state; +static struct slots_pid_state slots_state; static int state; static int cpu_count; static int cpu_pid_type; @@ -154,7 +166,8 @@ static struct completion ctrl_complete; static int critical_state; static int rackmac; static s32 dimm_output_clamp; - +static int fcu_rpm_shift; +static int fcu_tickle_ticks; static DECLARE_MUTEX(driver_lock); /* @@ -495,13 +508,20 @@ static int start_fcu(void) rc = fan_write_reg(0x2e, &buf, 1); if (rc < 0) return -EIO; + rc = fan_read_reg(0, &buf, 1); + if (rc < 0) + return -EIO; + fcu_rpm_shift = (buf == 1) ? 2 : 3; + printk(KERN_DEBUG "FCU Initialized, RPM fan shift is %d\n", + fcu_rpm_shift); + return 0; } static int set_rpm_fan(int fan_index, int rpm) { unsigned char buf[2]; - int rc, id; + int rc, id, min, max; if (fcu_fans[fan_index].type != FCU_FAN_RPM) return -EINVAL; @@ -509,12 +529,15 @@ static int set_rpm_fan(int fan_index, int rpm) if (id == FCU_FAN_ABSENT_ID) return -EINVAL; - if (rpm < 300) - rpm = 300; - else if (rpm > 8191) - rpm = 8191; - buf[0] = rpm >> 5; - buf[1] = rpm << 3; + min = 2400 >> fcu_rpm_shift; + max = 56000 >> fcu_rpm_shift; + + if (rpm < min) + rpm = min; + else if (rpm > max) + rpm = max; + buf[0] = rpm >> (8 - fcu_rpm_shift); + buf[1] = rpm << fcu_rpm_shift; rc = fan_write_reg(0x10 + (id * 2), buf, 2); if (rc < 0) return -EIO; @@ -551,7 +574,7 @@ static int get_rpm_fan(int fan_index, int programmed) if (rc != 2) return -EIO; - return (buf[0] << 5) | buf[1] >> 3; + return (buf[0] << (8 - fcu_rpm_shift)) | buf[1] >> fcu_rpm_shift; } static int set_pwm_fan(int fan_index, int pwm) @@ -609,6 +632,26 @@ static int get_pwm_fan(int fan_index) return (buf[0] * 1000) / 2559; } +static void tickle_fcu(void) +{ + int pwm; + + pwm = get_pwm_fan(SLOTS_FAN_PWM_INDEX); + + DBG("FCU Tickle, slots fan is: %d\n", pwm); + if (pwm < 0) + pwm = 100; + + if (!rackmac) { + pwm = SLOTS_FAN_DEFAULT_PWM; + } else if (pwm < SLOTS_PID_OUTPUT_MIN) + pwm = SLOTS_PID_OUTPUT_MIN; + + /* That is hopefully enough to make the FCU happy */ + set_pwm_fan(SLOTS_FAN_PWM_INDEX, pwm); +} + + /* * Utility routine to read the CPU calibration EEPROM data * from the device-tree @@ -715,6 +758,9 @@ BUILD_SHOW_FUNC_INT(backside_fan_pwm, backside_state.pwm) BUILD_SHOW_FUNC_FIX(drives_temperature, drives_state.last_temp) BUILD_SHOW_FUNC_INT(drives_fan_rpm, drives_state.rpm) +BUILD_SHOW_FUNC_FIX(slots_temperature, slots_state.last_temp) +BUILD_SHOW_FUNC_INT(slots_fan_pwm, slots_state.pwm) + BUILD_SHOW_FUNC_FIX(dimms_temperature, dimms_state.last_temp) static DEVICE_ATTR(cpu0_temperature,S_IRUGO,show_cpu0_temperature,NULL); @@ -735,6 +781,9 @@ static DEVICE_ATTR(backside_fan_pwm,S_IRUGO,show_backside_fan_pwm,NULL); static DEVICE_ATTR(drives_temperature,S_IRUGO,show_drives_temperature,NULL); static DEVICE_ATTR(drives_fan_rpm,S_IRUGO,show_drives_fan_rpm,NULL); +static DEVICE_ATTR(slots_temperature,S_IRUGO,show_slots_temperature,NULL); +static DEVICE_ATTR(slots_fan_pwm,S_IRUGO,show_slots_fan_pwm,NULL); + static DEVICE_ATTR(dimms_temperature,S_IRUGO,show_dimms_temperature,NULL); /* @@ -1076,6 +1125,9 @@ static void do_monitor_cpu_rack(struct cpu_pid_state *state) fan_min = dimm_output_clamp; fan_min = max(fan_min, (int)state->mpu.rminn_intake_fan); + DBG(" CPU min mpu = %d, min dimm = %d\n", + state->mpu.rminn_intake_fan, dimm_output_clamp); + state->rpm = max(state->rpm, (int)fan_min); state->rpm = min(state->rpm, (int)state->mpu.rmaxn_intake_fan); state->intake_rpm = state->rpm; @@ -1374,7 +1426,8 @@ static void do_monitor_drives(struct drives_pid_state *state) DBG(" current rpm: %d\n", state->rpm); /* Get some sensor readings */ - temp = le16_to_cpu(i2c_smbus_read_word_data(state->monitor, DS1775_TEMP)) << 8; + temp = le16_to_cpu(i2c_smbus_read_word_data(state->monitor, + DS1775_TEMP)) << 8; state->last_temp = temp; DBG(" temp: %d.%03d, target: %d.%03d\n", FIX32TOPRINT(temp), FIX32TOPRINT(DRIVES_PID_INPUT_TARGET)); @@ -1575,7 +1628,7 @@ static int init_dimms_state(struct dimm_pid_state *state) } /* - * Dispose of the state data for the drives control loop + * Dispose of the state data for the DIMM control loop */ static void dispose_dimms_state(struct dimm_pid_state *state) { @@ -1588,6 +1641,127 @@ static void dispose_dimms_state(struct dimm_pid_state *state) state->monitor = NULL; } +/* + * Slots fan control loop + */ +static void do_monitor_slots(struct slots_pid_state *state) +{ + s32 temp, integral, derivative; + s64 integ_p, deriv_p, prop_p, sum; + int i, rc; + + if (--state->ticks != 0) + return; + state->ticks = SLOTS_PID_INTERVAL; + + DBG("slots:\n"); + + /* Check fan status */ + rc = get_pwm_fan(SLOTS_FAN_PWM_INDEX); + if (rc < 0) { + printk(KERN_WARNING "Error %d reading slots fan !\n", rc); + /* XXX What do we do now ? */ + } else + state->pwm = rc; + DBG(" current pwm: %d\n", state->pwm); + + /* Get some sensor readings */ + temp = le16_to_cpu(i2c_smbus_read_word_data(state->monitor, + DS1775_TEMP)) << 8; + state->last_temp = temp; + DBG(" temp: %d.%03d, target: %d.%03d\n", FIX32TOPRINT(temp), + FIX32TOPRINT(SLOTS_PID_INPUT_TARGET)); + + /* Store temperature and error in history array */ + state->cur_sample = (state->cur_sample + 1) % SLOTS_PID_HISTORY_SIZE; + state->sample_history[state->cur_sample] = temp; + state->error_history[state->cur_sample] = temp - SLOTS_PID_INPUT_TARGET; + + /* If first loop, fill the history table */ + if (state->first) { + for (i = 0; i < (SLOTS_PID_HISTORY_SIZE - 1); i++) { + state->cur_sample = (state->cur_sample + 1) % + SLOTS_PID_HISTORY_SIZE; + state->sample_history[state->cur_sample] = temp; + state->error_history[state->cur_sample] = + temp - SLOTS_PID_INPUT_TARGET; + } + state->first = 0; + } + + /* Calculate the integral term */ + sum = 0; + integral = 0; + for (i = 0; i < SLOTS_PID_HISTORY_SIZE; i++) + integral += state->error_history[i]; + integral *= SLOTS_PID_INTERVAL; + DBG(" integral: %08x\n", integral); + integ_p = ((s64)SLOTS_PID_G_r) * (s64)integral; + DBG(" integ_p: %d\n", (int)(integ_p >> 36)); + sum += integ_p; + + /* Calculate the derivative term */ + derivative = state->error_history[state->cur_sample] - + state->error_history[(state->cur_sample + SLOTS_PID_HISTORY_SIZE - 1) + % SLOTS_PID_HISTORY_SIZE]; + derivative /= SLOTS_PID_INTERVAL; + deriv_p = ((s64)SLOTS_PID_G_d) * (s64)derivative; + DBG(" deriv_p: %d\n", (int)(deriv_p >> 36)); + sum += deriv_p; + + /* Calculate the proportional term */ + prop_p = ((s64)SLOTS_PID_G_p) * (s64)(state->error_history[state->cur_sample]); + DBG(" prop_p: %d\n", (int)(prop_p >> 36)); + sum += prop_p; + + /* Scale sum */ + sum >>= 36; + + DBG(" sum: %d\n", (int)sum); + state->pwm = (s32)sum; + + state->pwm = max(state->pwm, SLOTS_PID_OUTPUT_MIN); + state->pwm = min(state->pwm, SLOTS_PID_OUTPUT_MAX); + + DBG("** DRIVES PWM: %d\n", (int)state->pwm); + set_pwm_fan(SLOTS_FAN_PWM_INDEX, state->pwm); +} + +/* + * Initialize the state structure for the slots bay fan control loop + */ +static int init_slots_state(struct slots_pid_state *state) +{ + state->ticks = 1; + state->first = 1; + state->pwm = 50; + + state->monitor = attach_i2c_chip(XSERVE_SLOTS_LM75, "slots_temp"); + if (state->monitor == NULL) + return -ENODEV; + + device_create_file(&of_dev->dev, &dev_attr_slots_temperature); + device_create_file(&of_dev->dev, &dev_attr_slots_fan_pwm); + + return 0; +} + +/* + * Dispose of the state data for the slots control loop + */ +static void dispose_slots_state(struct slots_pid_state *state) +{ + if (state->monitor == NULL) + return; + + device_remove_file(&of_dev->dev, &dev_attr_slots_temperature); + device_remove_file(&of_dev->dev, &dev_attr_slots_fan_pwm); + + detach_i2c_chip(state->monitor); + state->monitor = NULL; +} + + static int call_critical_overtemp(void) { char *argv[] = { critical_overtemp_path, NULL }; @@ -1617,14 +1791,17 @@ static int main_control_loop(void *x) goto out; } - /* Set the PCI fan once for now */ - set_pwm_fan(SLOTS_FAN_PWM_INDEX, SLOTS_FAN_DEFAULT_PWM); + /* Set the PCI fan once for now on non-RackMac */ + if (!rackmac) + set_pwm_fan(SLOTS_FAN_PWM_INDEX, SLOTS_FAN_DEFAULT_PWM); /* Initialize ADCs */ initialize_adc(&cpu_state[0]); if (cpu_state[1].monitor != NULL) initialize_adc(&cpu_state[1]); + fcu_tickle_ticks = FCU_TICKLE_TICKS; + up(&driver_lock); while (state == state_attached) { @@ -1634,6 +1811,12 @@ static int main_control_loop(void *x) down(&driver_lock); + /* Tickle the FCU just in case */ + if (--fcu_tickle_ticks < 0) { + fcu_tickle_ticks = FCU_TICKLE_TICKS; + tickle_fcu(); + } + /* First, we always calculate the new DIMMs state on an Xserve */ if (rackmac) do_monitor_dimms(&dimms_state); @@ -1654,7 +1837,9 @@ static int main_control_loop(void *x) } /* Then, the rest */ do_monitor_backside(&backside_state); - if (!rackmac) + if (rackmac) + do_monitor_slots(&slots_state); + else do_monitor_drives(&drives_state); up(&driver_lock); @@ -1696,6 +1881,7 @@ static void dispose_control_loops(void) dispose_cpu_state(&cpu_state[1]); dispose_backside_state(&backside_state); dispose_drives_state(&drives_state); + dispose_slots_state(&slots_state); dispose_dimms_state(&dimms_state); } @@ -1745,6 +1931,8 @@ static int create_control_loops(void) goto fail; if (rackmac && init_dimms_state(&dimms_state)) goto fail; + if (rackmac && init_slots_state(&slots_state)) + goto fail; if (!rackmac && init_drives_state(&drives_state)) goto fail; diff --git a/drivers/macintosh/therm_pm72.h b/drivers/macintosh/therm_pm72.h index fc7e9b7ecaf2..393cc9df94e1 100644 --- a/drivers/macintosh/therm_pm72.h +++ b/drivers/macintosh/therm_pm72.h @@ -105,6 +105,7 @@ static char * critical_overtemp_path = "/sbin/critical_overtemp"; #define DRIVES_DALLAS_ID 0x94 #define BACKSIDE_MAX_ID 0x98 #define XSERVE_DIMMS_LM87 0x25a +#define XSERVE_SLOTS_LM75 0x290 /* * Some MAX6690, DS1775, LM87 register definitions @@ -198,7 +199,7 @@ struct drives_pid_state #define SLOTS_FAN_PWM_DEFAULT_ID 2 #define SLOTS_FAN_PWM_INDEX 2 -#define SLOTS_FAN_DEFAULT_PWM 50 /* Do better here ! */ +#define SLOTS_FAN_DEFAULT_PWM 40 /* Do better here ! */ /* @@ -206,7 +207,7 @@ struct drives_pid_state */ #define DIMM_PID_G_d 0 #define DIMM_PID_G_p 0 -#define DIMM_PID_G_r 0x6553600 +#define DIMM_PID_G_r 0x06553600 #define DIMM_PID_INPUT_TARGET 3276800 #define DIMM_PID_INTERVAL 1 #define DIMM_PID_OUTPUT_MAX 14000 @@ -226,6 +227,31 @@ struct dimm_pid_state }; +/* + * PID factors for the Xserve Slots control loop + */ +#define SLOTS_PID_G_d 0 +#define SLOTS_PID_G_p 0 +#define SLOTS_PID_G_r 0x00100000 +#define SLOTS_PID_INPUT_TARGET 3200000 +#define SLOTS_PID_INTERVAL 1 +#define SLOTS_PID_OUTPUT_MAX 100 +#define SLOTS_PID_OUTPUT_MIN 20 +#define SLOTS_PID_HISTORY_SIZE 20 + +struct slots_pid_state +{ + int ticks; + struct i2c_client * monitor; + s32 sample_history[SLOTS_PID_HISTORY_SIZE]; + s32 error_history[SLOTS_PID_HISTORY_SIZE]; + int cur_sample; + s32 last_temp; + int first; + int pwm; +}; + + /* Desktops */ @@ -283,6 +309,9 @@ struct cpu_pid_state s32 pump_max; }; +/* Tickle FCU every 10 seconds */ +#define FCU_TICKLE_TICKS 10 + /* * Driver state */ From e272a2853459b1e7282a7d0b54114ffaa7b3980f Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 10 Jul 2006 16:44:54 +1000 Subject: [PATCH 025/143] [POWERPC] Add cpufreq support for Xserve G5 The Xserve G5 are capable of frequency switching like other desktop G5s. This enables it. It also fix a Kconfig issue which prevented from building the G5 cpufreq support if CONFIG_PMAC_SMU was not set (the first version of that driver only worked with SMU based macs, but this isn't the case anymore). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/Kconfig | 2 +- arch/powerpc/platforms/powermac/cpufreq_64.c | 78 ++++++++++++-------- 2 files changed, 50 insertions(+), 30 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 2643dbc3f289..13e583f16ede 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -504,7 +504,7 @@ config CPU_FREQ_PMAC config CPU_FREQ_PMAC64 bool "Support for some Apple G5s" - depends on CPU_FREQ && PMAC_SMU && PPC64 + depends on CPU_FREQ && PPC64 select CPU_FREQ_TABLE help This adds support for frequency switching on Apple iMac G5, diff --git a/arch/powerpc/platforms/powermac/cpufreq_64.c b/arch/powerpc/platforms/powermac/cpufreq_64.c index f08a14516139..7b1156ea5341 100644 --- a/arch/powerpc/platforms/powermac/cpufreq_64.c +++ b/arch/powerpc/platforms/powermac/cpufreq_64.c @@ -10,6 +10,8 @@ * that is iMac G5 and latest single CPU desktop. */ +#undef DEBUG + #include #include #include @@ -30,13 +32,7 @@ #include #include -#undef DEBUG - -#ifdef DEBUG -#define DBG(fmt...) printk(fmt) -#else -#define DBG(fmt...) -#endif +#define DBG(fmt...) pr_debug(fmt) /* see 970FX user manual */ @@ -82,8 +78,6 @@ static struct freq_attr* g5_cpu_freqs_attr[] = { /* Power mode data is an array of the 32 bits PCR values to use for * the various frequencies, retrieved from the device-tree */ -static u32 *g5_pmode_data; -static int g5_pmode_max; static int g5_pmode_cur; static void (*g5_switch_volt)(int speed_mode); @@ -93,6 +87,11 @@ static int (*g5_query_freq)(void); static DEFINE_MUTEX(g5_switch_mutex); +#ifdef CONFIG_PMAC_SMU + +static u32 *g5_pmode_data; +static int g5_pmode_max; + static struct smu_sdbp_fvt *g5_fvt_table; /* table of op. points */ static int g5_fvt_count; /* number of op. points */ static int g5_fvt_cur; /* current op. point */ @@ -209,6 +208,16 @@ static int g5_scom_query_freq(void) return i; } +/* + * Fake voltage switching for platforms with missing support + */ + +static void g5_dummy_switch_volt(int speed_mode) +{ +} + +#endif /* CONFIG_PMAC_SMU */ + /* * Platform function based voltage switching for PowerMac7,2 & 7,3 */ @@ -248,6 +257,9 @@ static int g5_pfunc_switch_freq(int speed_mode) struct pmf_args args; u32 done = 0; unsigned long timeout; + int rc; + + DBG("g5_pfunc_switch_freq(%d)\n", speed_mode); /* If frequency is going up, first ramp up the voltage */ if (speed_mode < g5_pmode_cur) @@ -255,9 +267,12 @@ static int g5_pfunc_switch_freq(int speed_mode) /* Do it */ if (speed_mode == CPUFREQ_HIGH) - pmf_call_one(pfunc_cpu_setfreq_high, NULL); + rc = pmf_call_one(pfunc_cpu_setfreq_high, NULL); else - pmf_call_one(pfunc_cpu_setfreq_low, NULL); + rc = pmf_call_one(pfunc_cpu_setfreq_low, NULL); + + if (rc) + printk(KERN_WARNING "cpufreq: pfunc switch error %d\n", rc); /* It's an irq GPIO so we should be able to just block here, * I'll do that later after I've properly tested the IRQ code for @@ -296,13 +311,6 @@ static int g5_pfunc_query_freq(void) return val ? CPUFREQ_HIGH : CPUFREQ_LOW; } -/* - * Fake voltage switching for platforms with missing support - */ - -static void g5_dummy_switch_volt(int speed_mode) -{ -} /* * Common interface to the cpufreq core @@ -375,6 +383,8 @@ static struct cpufreq_driver g5_cpufreq_driver = { }; +#ifdef CONFIG_PMAC_SMU + static int __init g5_neo2_cpufreq_init(struct device_node *cpus) { struct device_node *cpunode; @@ -525,6 +535,9 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus) return rc; } +#endif /* CONFIG_PMAC_SMU */ + + static int __init g5_pm72_cpufreq_init(struct device_node *cpus) { struct device_node *cpuid = NULL, *hwclock = NULL, *cpunode = NULL; @@ -533,6 +546,9 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus) u64 max_freq, min_freq, ih, il; int has_volt = 1, rc = 0; + DBG("cpufreq: Initializing for PowerMac7,2, PowerMac7,3 and" + " RackMac3,1...\n"); + /* Get first CPU node */ for (cpunode = NULL; (cpunode = of_get_next_child(cpus, cpunode)) != NULL;) { @@ -636,6 +652,15 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus) */ ih = *((u32 *)(eeprom + 0x10)); il = *((u32 *)(eeprom + 0x20)); + + /* Check for machines with no useful settings */ + if (il == ih) { + printk(KERN_WARNING "cpufreq: No low frequency mode available" + " on this model !\n"); + rc = -ENODEV; + goto bail; + } + min_freq = 0; if (ih != 0 && il != 0) min_freq = (max_freq * il) / ih; @@ -643,7 +668,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus) /* Sanity check */ if (min_freq >= max_freq || min_freq < 1000) { printk(KERN_ERR "cpufreq: Can't calculate low frequency !\n"); - rc = -ENODEV; + rc = -ENXIO; goto bail; } g5_cpu_freqs[0].frequency = max_freq; @@ -690,16 +715,10 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus) return rc; } -static int __init g5_rm31_cpufreq_init(struct device_node *cpus) -{ - /* NYI */ - return 0; -} - static int __init g5_cpufreq_init(void) { struct device_node *cpus; - int rc; + int rc = 0; cpus = of_find_node_by_path("/cpus"); if (cpus == NULL) { @@ -708,12 +727,13 @@ static int __init g5_cpufreq_init(void) } if (machine_is_compatible("PowerMac7,2") || - machine_is_compatible("PowerMac7,3")) + machine_is_compatible("PowerMac7,3") || + machine_is_compatible("RackMac3,1")) rc = g5_pm72_cpufreq_init(cpus); - else if (machine_is_compatible("RackMac3,1")) - rc = g5_rm31_cpufreq_init(cpus); +#ifdef CONFIG_PMAC_SMU else rc = g5_neo2_cpufreq_init(cpus); +#endif /* CONFIG_PMAC_SMU */ of_node_put(cpus); return rc; From d319a03bf87209b3914fdf2ede88a2161123b3eb Mon Sep 17 00:00:00 2001 From: Segher Boessenkool Date: Sat, 8 Jul 2006 02:37:23 +0200 Subject: [PATCH 026/143] [POWERPC] Fix new interrupt code (MPIC endianness) All U3/U4 based systems are big-endian, not all express it in their device trees. Signed-off-by: Segher Boessenkool Acked-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/maple/setup.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/powerpc/platforms/maple/setup.c b/arch/powerpc/platforms/maple/setup.c index cb528c9de4c3..2ecea3b21c7c 100644 --- a/arch/powerpc/platforms/maple/setup.c +++ b/arch/powerpc/platforms/maple/setup.c @@ -252,6 +252,8 @@ static void __init maple_init_IRQ(void) /* XXX Maple specific bits */ flags |= MPIC_BROKEN_U3 | MPIC_WANTS_RESET; + /* All U3/U4 are big-endian, older SLOF firmware doesn't encode this */ + flags |= MPIC_BIG_ENDIAN; /* Setup the openpic driver. More device-tree junks, we hard code no * ISUs for now. I'll have to revisit some stuffs with the folks doing From 96278d21000568a9261f016e8b2569a95a2d4c9e Mon Sep 17 00:00:00 2001 From: Segher Boessenkool Date: Sat, 8 Jul 2006 02:37:20 +0200 Subject: [PATCH 027/143] [POWERPC] Fix new interrupt code (MPIC detection) As the code comment already says, the Maple device-tree is incorrect here; make the Linux code detect the correct thing, too. Signed-off-by: Segher Boessenkool Acked-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/maple/setup.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/platforms/maple/setup.c b/arch/powerpc/platforms/maple/setup.c index 2ecea3b21c7c..57567dfb9819 100644 --- a/arch/powerpc/platforms/maple/setup.c +++ b/arch/powerpc/platforms/maple/setup.c @@ -221,10 +221,17 @@ static void __init maple_init_IRQ(void) * in Maple device-tree where the type of the controller is * open-pic and not interrupt-controller */ - for_each_node_by_type(np, "open-pic") { - mpic_node = np; - break; - } + + for_each_node_by_type(np, "interrupt-controller") + if (device_is_compatible(np, "open-pic")) { + mpic_node = np; + break; + } + if (mpic_node == NULL) + for_each_node_by_type(np, "open-pic") { + mpic_node = np; + break; + } if (mpic_node == NULL) { printk(KERN_ERR "Failed to locate the MPIC interrupt controller\n"); From 4c962b5a42e5fc24e2d5d73896732cf3578da260 Mon Sep 17 00:00:00 2001 From: Danny van Dyk Date: Tue, 11 Jul 2006 22:25:28 +0100 Subject: [PATCH 028/143] [POWERPC] PMAC_APM_EMU should depend on ADB_PMU This patch fixes undefined refereneces to pmu_ symbols on 2.6.17. Signed-Off-By: Danny van Dyk Signed-off-by: Daniel Drake Signed-off-by: Paul Mackerras --- drivers/macintosh/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/macintosh/Kconfig b/drivers/macintosh/Kconfig index f18c956fd595..44708b5e582d 100644 --- a/drivers/macintosh/Kconfig +++ b/drivers/macintosh/Kconfig @@ -109,7 +109,7 @@ config PMAC_SMU config PMAC_APM_EMU tristate "APM emulation" - depends on PPC_PMAC && PPC32 && PM + depends on PPC_PMAC && PPC32 && PM && ADB_PMU config PMAC_MEDIABAY bool "Support PowerBook hotswap media bay" From b8cfac4c2f3d12d0f4cbe6f992d945f2fdfc098d Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 27 Jul 2006 17:57:32 -0700 Subject: [PATCH 029/143] [SPARC64]: Fix typo in pgprot_noncached(). The sun4v code sequence was or'ing in the sun4u pte bits by mistake. Signed-off-by: David S. Miller --- include/asm-sparc64/pgtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h index 03f5bc9b6bec..1ba19eb34ce3 100644 --- a/include/asm-sparc64/pgtable.h +++ b/include/asm-sparc64/pgtable.h @@ -339,7 +339,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t prot) " .section .sun4v_2insn_patch, \"ax\"\n" " .word 661b\n" " andn %0, %4, %0\n" - " or %0, %3, %0\n" + " or %0, %5, %0\n" " .previous\n" : "=r" (val) : "0" (val), "i" (_PAGE_CP_4U | _PAGE_CV_4U), "i" (_PAGE_E_4U), From 96ba989d22de779ca19ca214e2b2e53a4ca86b7b Mon Sep 17 00:00:00 2001 From: Bob Breuer Date: Thu, 27 Jul 2006 22:08:01 -0700 Subject: [PATCH 030/143] [SPARC]: Defer clock_probe to fs_initcall() From: Bob Breuer That way all the of_driver bits will be ready. Signed-off-by: David S. Miller --- arch/sparc/kernel/time.c | 74 +++++++++++++++++++++++----------------- 1 file changed, 42 insertions(+), 32 deletions(-) diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c index 04eb1eab6e3e..845081b01267 100644 --- a/arch/sparc/kernel/time.c +++ b/arch/sparc/kernel/time.c @@ -225,6 +225,32 @@ static __inline__ int has_low_battery(void) return (data1 == data2); /* Was the write blocked? */ } +static void __init mostek_set_system_time(void) +{ + unsigned int year, mon, day, hour, min, sec; + struct mostek48t02 *mregs; + + mregs = (struct mostek48t02 *)mstk48t02_regs; + if(!mregs) { + prom_printf("Something wrong, clock regs not mapped yet.\n"); + prom_halt(); + } + spin_lock_irq(&mostek_lock); + mregs->creg |= MSTK_CREG_READ; + sec = MSTK_REG_SEC(mregs); + min = MSTK_REG_MIN(mregs); + hour = MSTK_REG_HOUR(mregs); + day = MSTK_REG_DOM(mregs); + mon = MSTK_REG_MONTH(mregs); + year = MSTK_CVT_YEAR( MSTK_REG_YEAR(mregs) ); + xtime.tv_sec = mktime(year, mon, day, hour, min, sec); + xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ); + set_normalized_timespec(&wall_to_monotonic, + -xtime.tv_sec, -xtime.tv_nsec); + mregs->creg &= ~MSTK_CREG_READ; + spin_unlock_irq(&mostek_lock); +} + /* Probe for the real time clock chip on Sun4 */ static __inline__ void sun4_clock_probe(void) { @@ -273,6 +299,7 @@ static __inline__ void sun4_clock_probe(void) #endif } +#ifndef CONFIG_SUN4 static int __devinit clock_probe(struct of_device *op, const struct of_device_id *match) { struct device_node *dp = op->node; @@ -307,6 +334,8 @@ static int __devinit clock_probe(struct of_device *op, const struct of_device_id if (mostek_read(mstk48t02_regs + MOSTEK_SEC) & MSTK_STOP) kick_start_clock(); + mostek_set_system_time(); + return 0; } @@ -325,56 +354,37 @@ static struct of_platform_driver clock_driver = { /* Probe for the mostek real time clock chip. */ -static void clock_init(void) +static int __init clock_init(void) { - of_register_driver(&clock_driver, &of_bus_type); + return of_register_driver(&clock_driver, &of_bus_type); } +/* Must be after subsys_initcall() so that busses are probed. Must + * be before device_initcall() because things like the RTC driver + * need to see the clock registers. + */ +fs_initcall(clock_init); +#endif /* !CONFIG_SUN4 */ + void __init sbus_time_init(void) { - unsigned int year, mon, day, hour, min, sec; - struct mostek48t02 *mregs; - -#ifdef CONFIG_SUN4 - int temp; - struct intersil *iregs; -#endif BTFIXUPSET_CALL(bus_do_settimeofday, sbus_do_settimeofday, BTFIXUPCALL_NORM); btfixup(); if (ARCH_SUN4) sun4_clock_probe(); - else - clock_init(); sparc_init_timers(timer_interrupt); #ifdef CONFIG_SUN4 if(idprom->id_machtype == (SM_SUN4 | SM_4_330)) { -#endif - mregs = (struct mostek48t02 *)mstk48t02_regs; - if(!mregs) { - prom_printf("Something wrong, clock regs not mapped yet.\n"); - prom_halt(); - } - spin_lock_irq(&mostek_lock); - mregs->creg |= MSTK_CREG_READ; - sec = MSTK_REG_SEC(mregs); - min = MSTK_REG_MIN(mregs); - hour = MSTK_REG_HOUR(mregs); - day = MSTK_REG_DOM(mregs); - mon = MSTK_REG_MONTH(mregs); - year = MSTK_CVT_YEAR( MSTK_REG_YEAR(mregs) ); - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); - xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ); - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - mregs->creg &= ~MSTK_CREG_READ; - spin_unlock_irq(&mostek_lock); -#ifdef CONFIG_SUN4 + mostek_set_system_time(); } else if(idprom->id_machtype == (SM_SUN4 | SM_4_260) ) { /* initialise the intersil on sun4 */ + unsigned int year, mon, day, hour, min, sec; + int temp; + struct intersil *iregs; iregs=intersil_clock; if(!iregs) { From 361934849e9c0418950bedf667732f36337d88b9 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 08:54:59 +0200 Subject: [PATCH 031/143] [PATCH] ide: option to disable cache flushes for buggy drives Some drives claim they support cache flushing, but get seriously confused if you try. Add this option to be able to boot with barriers enabled by default. Signed-off-by: Jens Axboe --- drivers/ide/ide-disk.c | 2 +- drivers/ide/ide.c | 5 ++++- include/linux/ide.h | 1 + 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index f712e4cfd9dc..7cf3eb023521 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -776,7 +776,7 @@ static void update_ordered(ide_drive_t *drive) * not available so we don't need to recheck that. */ capacity = idedisk_capacity(drive); - barrier = ide_id_has_flush_cache(id) && + barrier = ide_id_has_flush_cache(id) && !drive->noflush && (drive->addressing == 0 || capacity <= (1ULL << 28) || ide_id_has_flush_cache_ext(id)); diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 05fbd9298db7..defd4b4bd374 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -1539,7 +1539,7 @@ static int __init ide_setup(char *s) const char *hd_words[] = { "none", "noprobe", "nowerr", "cdrom", "serialize", "autotune", "noautotune", "minus8", "swapdata", "bswap", - "minus11", "remap", "remap63", "scsi", NULL }; + "noflush", "remap", "remap63", "scsi", NULL }; unit = s[2] - 'a'; hw = unit / MAX_DRIVES; unit = unit % MAX_DRIVES; @@ -1578,6 +1578,9 @@ static int __init ide_setup(char *s) case -10: /* "bswap" */ drive->bswap = 1; goto done; + case -11: /* noflush */ + drive->noflush = 1; + goto done; case -12: /* "remap" */ drive->remap_0_to_1 = 1; goto done; diff --git a/include/linux/ide.h b/include/linux/ide.h index dc7abef10965..99620451d958 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -571,6 +571,7 @@ typedef struct ide_drive_s { u8 waiting_for_dma; /* dma currently in progress */ u8 unmask; /* okay to unmask other irqs */ u8 bswap; /* byte swap data */ + u8 noflush; /* don't attempt flushes */ u8 dsc_overlap; /* DSC overlap */ u8 nice1; /* give potential excess bandwidth */ From 0a8348d08677ad77ee353f96eb8745c693a05a13 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 08:58:26 +0200 Subject: [PATCH 032/143] [PATCH] ide: if the id fields looks screwy, disable DMA It's the safer choice. Originally due to a bug in itx821x, but a generally sound thing to do. Signed-off-by: Jens Axboe --- drivers/ide/ide-dma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c index 98918fb6b2ce..7c3a13e1cf64 100644 --- a/drivers/ide/ide-dma.c +++ b/drivers/ide/ide-dma.c @@ -750,7 +750,7 @@ void ide_dma_verbose(ide_drive_t *drive) goto bug_dma_off; printk(", DMA"); } else if (id->field_valid & 1) { - printk(", BUG"); + goto bug_dma_off; } return; bug_dma_off: From 71ef51cc1756d1c56b57c70e7cc27a3559c81ee6 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:02:17 +0200 Subject: [PATCH 033/143] [PATCH] it821x: fix ide dma setup bug Only enable dma for a valid speed setting. Signed-off-by: Jens Axboe --- drivers/ide/pci/it821x.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c index 3cb04424d351..e9bad185968a 100644 --- a/drivers/ide/pci/it821x.c +++ b/drivers/ide/pci/it821x.c @@ -498,9 +498,14 @@ static int config_chipset_for_dma (ide_drive_t *drive) { u8 speed = ide_dma_speed(drive, it821x_ratemask(drive)); - config_it821x_chipset_for_pio(drive, !speed); - it821x_tune_chipset(drive, speed); - return ide_dma_enable(drive); + if (speed) { + config_it821x_chipset_for_pio(drive, 0); + it821x_tune_chipset(drive, speed); + + return ide_dma_enable(drive); + } + + return 0; } /** From a75ad3c27a6ad78c4306cac939938050dcde54f3 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:04:09 +0200 Subject: [PATCH 034/143] [PATCH] scsi: kill overeager "not-ready" messages HAL and friends have a tendency to trigger this one all the time. It's not really interesting, so kill it. The vendor kernels all do anyways. Signed-off-by: Jens Axboe --- drivers/scsi/scsi_ioctl.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c index a89c4115cfba..32293f451669 100644 --- a/drivers/scsi/scsi_ioctl.c +++ b/drivers/scsi/scsi_ioctl.c @@ -110,11 +110,8 @@ static int ioctl_internal_command(struct scsi_device *sdev, char *cmd, sshdr.asc, sshdr.ascq); break; case NOT_READY: /* This happens if there is no disc in drive */ - if (sdev->removable && (cmd[0] != TEST_UNIT_READY)) { - printk(KERN_INFO "Device not ready. Make sure" - " there is a disc in the drive.\n"); + if (sdev->removable) break; - } case UNIT_ATTENTION: if (sdev->removable) { sdev->changed = 1; From 2a293b7d5aa2f0d1e3d87b642f7ac263c2d664e3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 28 Jul 2006 17:04:26 +1000 Subject: [PATCH 035/143] [XFS] All xfs_disk_dquot_t values are (as the name says) disk endian. Before putting them into struct statfs they should be endian-swapped. SGI-PV: 954580 SGI-Modid: xfs-linux-melb:xfs-kern:26550a Signed-off-by: Christoph Hellwig Signed-off-by: Nathan Scott --- fs/xfs/quota/xfs_qm_bhv.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index e95e99f7168f..f137856c3261 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -217,17 +217,24 @@ xfs_qm_statvfs( return 0; dp = &dqp->q_core; - limit = dp->d_blk_softlimit ? dp->d_blk_softlimit : dp->d_blk_hardlimit; + limit = dp->d_blk_softlimit ? + be64_to_cpu(dp->d_blk_softlimit) : + be64_to_cpu(dp->d_blk_hardlimit); if (limit && statp->f_blocks > limit) { statp->f_blocks = limit; - statp->f_bfree = (statp->f_blocks > dp->d_bcount) ? - (statp->f_blocks - dp->d_bcount) : 0; + statp->f_bfree = + (statp->f_blocks > be64_to_cpu(dp->d_bcount)) ? + (statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0; } - limit = dp->d_ino_softlimit ? dp->d_ino_softlimit : dp->d_ino_hardlimit; + + limit = dp->d_ino_softlimit ? + be64_to_cpu(dp->d_ino_softlimit) : + be64_to_cpu(dp->d_ino_hardlimit); if (limit && statp->f_files > limit) { statp->f_files = limit; - statp->f_ffree = (statp->f_files > dp->d_icount) ? - (statp->f_ffree - dp->d_icount) : 0; + statp->f_ffree = + (statp->f_files > be64_to_cpu(dp->d_icount)) ? + (statp->f_ffree - be64_to_cpu(dp->d_icount)) : 0; } xfs_qm_dqput(dqp); From f5faad799475c4058416264f672bb33bf8b5ef41 Mon Sep 17 00:00:00 2001 From: Nathan Scott Date: Fri, 28 Jul 2006 17:04:44 +1000 Subject: [PATCH 036/143] [XFS] Fix remount vs no/barrier options by ensuring we clear unwanted flags from iclog buffers before submitting them for writing. SGI-PV: 954772 SGI-Modid: xfs-linux-melb:xfs-kern:26605a Signed-off-by: Nathan Scott --- fs/xfs/linux-2.6/xfs_buf.h | 4 ++-- fs/xfs/xfs_log.c | 12 ++++++------ 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index ceda3a2859d2..7858703ed84c 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -246,8 +246,8 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *); #define BUF_BUSY XBF_DONT_BLOCK #define XFS_BUF_BFLAGS(bp) ((bp)->b_flags) -#define XFS_BUF_ZEROFLAGS(bp) \ - ((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI)) +#define XFS_BUF_ZEROFLAGS(bp) ((bp)->b_flags &= \ + ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI|XBF_ORDERED)) #define XFS_BUF_STALE(bp) ((bp)->b_flags |= XFS_B_STALE) #define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XFS_B_STALE) diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index e730328636c3..21ac1a67e3e0 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1413,7 +1413,7 @@ xlog_sync(xlog_t *log, ops = iclog->ic_header.h_num_logops; INT_SET(iclog->ic_header.h_num_logops, ARCH_CONVERT, ops); - bp = iclog->ic_bp; + bp = iclog->ic_bp; ASSERT(XFS_BUF_FSPRIVATE2(bp, unsigned long) == (unsigned long)1); XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)2); XFS_BUF_SET_ADDR(bp, BLOCK_LSN(INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT))); @@ -1430,15 +1430,14 @@ xlog_sync(xlog_t *log, } XFS_BUF_SET_PTR(bp, (xfs_caddr_t) &(iclog->ic_header), count); XFS_BUF_SET_FSPRIVATE(bp, iclog); /* save for later */ + XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); /* * Do an ordered write for the log block. - * - * It may not be needed to flush the first split block in the log wrap - * case, but do it anyways to be safe -AK + * Its unnecessary to flush the first split block in the log wrap case. */ - if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) + if (!split && (log->l_mp->m_flags & XFS_MOUNT_BARRIER)) XFS_BUF_ORDERED(bp); ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1); @@ -1460,7 +1459,7 @@ xlog_sync(xlog_t *log, return error; } if (split) { - bp = iclog->ic_log->l_xbuf; + bp = iclog->ic_log->l_xbuf; ASSERT(XFS_BUF_FSPRIVATE2(bp, unsigned long) == (unsigned long)1); XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)2); @@ -1468,6 +1467,7 @@ xlog_sync(xlog_t *log, XFS_BUF_SET_PTR(bp, (xfs_caddr_t)((__psint_t)&(iclog->ic_header)+ (__psint_t)count), split); XFS_BUF_SET_FSPRIVATE(bp, iclog); + XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) From b2ea401bac39e75ebb64038609ed22efbc799905 Mon Sep 17 00:00:00 2001 From: Nathan Scott Date: Fri, 28 Jul 2006 17:05:13 +1000 Subject: [PATCH 037/143] [XFS] Fix a barrier related forced shutdown on mounts with quota enabled. SGI-PV: 912426 SGI-Modid: xfs-linux-melb:xfs-kern:26622a Signed-off-by: Nathan Scott --- fs/xfs/linux-2.6/xfs_super.c | 7 +++++++ fs/xfs/xfs_vfsops.c | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 9bdef9d51900..4754f342a5d3 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -314,6 +314,13 @@ xfs_mountfs_check_barriers(xfs_mount_t *mp) return; } + if (xfs_readonly_buftarg(mp->m_ddev_targp)) { + xfs_fs_cmn_err(CE_NOTE, mp, + "Disabling barriers, underlying device is readonly"); + mp->m_flags &= ~XFS_MOUNT_BARRIER; + return; + } + error = xfs_barrier_test(mp); if (error) { xfs_fs_cmn_err(CE_NOTE, mp, diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 6c96391f3f1a..b427d220a169 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -515,7 +515,7 @@ xfs_mount( if (error) goto error2; - if ((mp->m_flags & XFS_MOUNT_BARRIER) && !(vfsp->vfs_flag & VFS_RDONLY)) + if (mp->m_flags & XFS_MOUNT_BARRIER) xfs_mountfs_check_barriers(mp); error = XFS_IOINIT(vfsp, args, flags); From 41ff715abc49324fb2cb20e66bc4e0290cfdbe51 Mon Sep 17 00:00:00 2001 From: Nathan Scott Date: Fri, 28 Jul 2006 17:05:51 +1000 Subject: [PATCH 038/143] [XFS] Ensure bulkstat from an invalid inode number gets caught always with EINVAL. SGI-PV: 953819 SGI-Modid: xfs-linux-melb:xfs-kern:26629a Signed-off-by: Nathan Scott --- fs/xfs/xfs_inode.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 86c1bf0bba9e..1f8ecff8553a 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -334,10 +334,9 @@ xfs_itobp( #if !defined(__KERNEL__) ni = 0; #elif defined(DEBUG) - ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : - (BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog); + ni = BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog; #else /* usual case */ - ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : 1; + ni = 1; #endif for (i = 0; i < ni; i++) { @@ -348,11 +347,15 @@ xfs_itobp( (i << mp->m_sb.sb_inodelog)); di_ok = INT_GET(dip->di_core.di_magic, ARCH_CONVERT) == XFS_DINODE_MAGIC && XFS_DINODE_GOOD_VERSION(INT_GET(dip->di_core.di_version, ARCH_CONVERT)); - if (unlikely(XFS_TEST_ERROR(!di_ok, mp, XFS_ERRTAG_ITOBP_INOTOBP, - XFS_RANDOM_ITOBP_INOTOBP))) { + if (unlikely(XFS_TEST_ERROR(!di_ok, mp, + XFS_ERRTAG_ITOBP_INOTOBP, + XFS_RANDOM_ITOBP_INOTOBP))) { + if (imap_flags & XFS_IMAP_BULKSTAT) { + xfs_trans_brelse(tp, bp); + return XFS_ERROR(EINVAL); + } #ifdef DEBUG - if (!(imap_flags & XFS_IMAP_BULKSTAT)) - cmn_err(CE_ALERT, + cmn_err(CE_ALERT, "Device %s - bad inode magic/vsn " "daddr %lld #%d (magic=%x)", XFS_BUFTARG_NAME(mp->m_ddev_targp), From 93853fd0d492524e9172297d8e8b8364dc2c4c59 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 28 Jul 2006 01:09:40 -0700 Subject: [PATCH 039/143] [SUNLANCE]: fix compilation on sparc-UP Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- drivers/net/sunlance.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c index 1ef9fd39a79a..0e3fdf7c6dd3 100644 --- a/drivers/net/sunlance.c +++ b/drivers/net/sunlance.c @@ -1537,7 +1537,7 @@ static int __init sparc_lance_init(void) { if ((idprom->id_machtype == (SM_SUN4|SM_4_330)) || (idprom->id_machtype == (SM_SUN4|SM_4_470))) { - memset(&sun4_sdev, 0, sizeof(sdev)); + memset(&sun4_sdev, 0, sizeof(struct sbus_dev)); sun4_sdev.reg_addrs[0].phys_addr = sun4_eth_physaddr; sun4_sdev.irqs[0] = 6; return sparc_lance_probe_one(&sun4_sdev, NULL, NULL); @@ -1547,16 +1547,16 @@ static int __init sparc_lance_init(void) static int __exit sunlance_sun4_remove(void) { - struct lance_private *lp = dev_get_drvdata(&sun4_sdev->dev); + struct lance_private *lp = dev_get_drvdata(&sun4_sdev.ofdev.dev); struct net_device *net_dev = lp->dev; unregister_netdevice(net_dev); - lance_free_hwresources(root_lance_dev); + lance_free_hwresources(lp); free_netdev(net_dev); - dev_set_drvdata(&sun4_sdev->dev, NULL); + dev_set_drvdata(&sun4_sdev.ofdev.dev, NULL); return 0; } From 9fc0a92c7ebcad96467d62077497ce195d9b741c Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Wed, 19 Jul 2006 10:34:05 +0200 Subject: [PATCH 040/143] [POWERPC] force 64bit mode in fwnmi handlers to workaround firmware bugs The firmware of POWER4 and JS20 systems does not switch the cpu to 64bit mode when the registered system_reset and machine_check handlers get called. If a 32bit process runs on that cpu at the time of the event, the cpu remains in 32bit mode. xmon and kdump can not deal with it, the result is an error like 'Bad kernel stack pointer fff2aad0 at 3200'. xmon just loses some register info, but booting the kdump kernel usually fails. Both handlers are not hot paths. Duplicate the EXCEPTION_PROLOG_PSERIES macro and add two instructions to switch to 64bit: li r11,5; rldimi r10,r11,61,0; Signed-off-by: Olaf Hering Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/head_64.S | 35 +++++++++++++++++++++++++++++++++-- 1 file changed, 33 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index e16eb2a33173..6ff3cf506088 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -191,6 +191,37 @@ exception_marker: ori reg,reg,(label)@l; /* virt addr of handler ... */ #endif +/* + * Equal to EXCEPTION_PROLOG_PSERIES, except that it forces 64bit mode. + * The firmware calls the registered system_reset_fwnmi and + * machine_check_fwnmi handlers in 32bit mode if the cpu happens to run + * a 32bit application at the time of the event. + * This firmware bug is present on POWER4 and JS20. + */ +#define EXCEPTION_PROLOG_PSERIES_FORCE_64BIT(area, label) \ + mfspr r13,SPRN_SPRG3; /* get paca address into r13 */ \ + std r9,area+EX_R9(r13); /* save r9 - r12 */ \ + std r10,area+EX_R10(r13); \ + std r11,area+EX_R11(r13); \ + std r12,area+EX_R12(r13); \ + mfspr r9,SPRN_SPRG1; \ + std r9,area+EX_R13(r13); \ + mfcr r9; \ + clrrdi r12,r13,32; /* get high part of &label */ \ + mfmsr r10; \ + /* force 64bit mode */ \ + li r11,5; /* MSR_SF_LG|MSR_ISF_LG */ \ + rldimi r10,r11,61,0; /* insert into top 3 bits */ \ + /* done 64bit mode */ \ + mfspr r11,SPRN_SRR0; /* save SRR0 */ \ + LOAD_HANDLER(r12,label) \ + ori r10,r10,MSR_IR|MSR_DR|MSR_RI; \ + mtspr SPRN_SRR0,r12; \ + mfspr r12,SPRN_SRR1; /* and SRR1 */ \ + mtspr SPRN_SRR1,r10; \ + rfid; \ + b . /* prevent speculative execution */ + #define EXCEPTION_PROLOG_PSERIES(area, label) \ mfspr r13,SPRN_SPRG3; /* get paca address into r13 */ \ std r9,area+EX_R9(r13); /* save r9 - r12 */ \ @@ -604,14 +635,14 @@ slb_miss_user_pseries: system_reset_fwnmi: HMT_MEDIUM mtspr SPRN_SPRG1,r13 /* save r13 */ - EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common) + EXCEPTION_PROLOG_PSERIES_FORCE_64BIT(PACA_EXGEN, system_reset_common) .globl machine_check_fwnmi .align 7 machine_check_fwnmi: HMT_MEDIUM mtspr SPRN_SPRG1,r13 /* save r13 */ - EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) + EXCEPTION_PROLOG_PSERIES_FORCE_64BIT(PACA_EXMC, machine_check_common) #ifdef CONFIG_PPC_ISERIES /*** ISeries-LPAR interrupt handlers ***/ From facf014792093d95e308b5d6ce9bc86d3c90e5b1 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Tue, 25 Jul 2006 16:15:16 -0400 Subject: [PATCH 041/143] [PATCH] i386: switch_to(): misplaced parentheses Recent changes in i386 __switch_to() have a misplaced closing parenthesis causing an unlikely() to terminate early. Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Acked-by: Steven Rostedt Signed-off-by: Linus Torvalds --- arch/i386/kernel/process.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 923bb292f47f..8657c739656a 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -690,8 +690,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas /* * Now maybe handle debug registers and/or IO bitmaps */ - if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW)) - || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) + if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW) + || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))) __switch_to_xtra(next_p, tss); disable_tsc(prev_p, next_p); From d5a2601734bcc740ee78dc4cb0c56b5687da7bd9 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 28 Jul 2006 14:44:42 +0200 Subject: [PATCH 042/143] [PATCH] i386/x86-64: Add user_mode checks to profile_pc for oprofile Fixes a obscure user space triggerable crash during oprofiling. Oprofile calls profile_pc from NMIs even when user_mode(regs) is not true and the program counter is inside the kernel lock section. This opens a race - when a user program jumps to a kernel lock address and a NMI happens before the illegal page fault exception is raised and the program has a unmapped esp or ebp then the kernel could oops. NMIs have a higher priority than exceptions so that could happen. Add user_mode checks to i386/x86-64 profile_pc to prevent that. Cc: John Levon Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/time.c | 2 +- arch/x86_64/kernel/time.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index 8705c0f05788..edd00f6cee37 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -135,7 +135,7 @@ unsigned long profile_pc(struct pt_regs *regs) { unsigned long pc = instruction_pointer(regs); - if (in_lock_functions(pc)) + if (!user_mode_vm(regs) && in_lock_functions(pc)) return *(unsigned long *)(regs->ebp + 4); return pc; diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index b9ff75992c16..e0341c6808e5 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -193,7 +193,7 @@ unsigned long profile_pc(struct pt_regs *regs) is just accounted to the spinlock function. Better would be to write these functions in assembler again and check exactly. */ - if (in_lock_functions(pc)) { + if (!user_mode(regs) && in_lock_functions(pc)) { char *v = *(char **)regs->rsp; if ((v >= _stext && v <= _etext) || (v >= _sinittext && v <= _einittext) || From 0e92da4acb763272c6060f0b14adc2377b627d07 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 28 Jul 2006 14:44:45 +0200 Subject: [PATCH 043/143] [PATCH] x86_64: Don't clobber r8-r11 in int 0x80 handler When int 0x80 is called from long mode r8-r11 would leak out of the kernel (or rather they would be filled with some values from the kernel stack). I don't think it's a security issue because the values come from the fixed stack frame which should be near always user registers from a previous interrupt. Still better fix it. Longer term the register save macros need to be cleaned up to avoid such mistakes in the future. Original analysis from Richard Brunner, fix by me. Cc: Richard.Brunner@amd.com Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32entry.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index 9b5bb413a6e9..5d4a7d125ed0 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -103,7 +103,7 @@ ENTRY(ia32_sysenter_target) pushq %rax CFI_ADJUST_CFA_OFFSET 8 cld - SAVE_ARGS 0,0,1 + SAVE_ARGS 0,0,0 /* no need to do an access_ok check here because rbp has been 32bit zero extended */ 1: movl (%rbp),%r9d From a4045dff782a8692637c24a0222120082c887caa Mon Sep 17 00:00:00 2001 From: bibo mao Date: Fri, 28 Jul 2006 14:44:48 +0200 Subject: [PATCH 044/143] [PATCH] x86_64: Enlarge debug stack for nested kprobes In x86_64 platform, INT1 and INT3 trap stack is IST stack called DEBUG_STACK, when INT1/INT3 trap happens, system will switch to DEBUG_STACK by hardware. Current DEBUG_STACK size is 4K, when int1/int3 trap happens, kernel will minus current DEBUG_STACK IST value by 4k. But if int3/int1 trap is nested, it will destroy other vector's IST stack. This patch modifies this, it sets DEBUG_STACK size as 8K and allows two level of nested int1/int3 trap. Kprobe DEBUG_STACK may be nested, because kprobe handler may be probed by other kprobes. Thanks jbeulich for pointing out error in the first patch. [AK: nested kprobes are pretty dubious. Hopefully one nest will be enough. This will cost 8K per CPU (4K more than before)] Signed-off-by: bibo, mao Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-x86_64/page.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h index f7bf875aae40..10f346165cab 100644 --- a/include/asm-x86_64/page.h +++ b/include/asm-x86_64/page.h @@ -19,7 +19,7 @@ #define EXCEPTION_STACK_ORDER 0 #define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER) -#define DEBUG_STACK_ORDER EXCEPTION_STACK_ORDER +#define DEBUG_STACK_ORDER (EXCEPTION_STACK_ORDER + 1) #define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER) #define IRQSTACK_ORDER 2 From b13761ecd1d9977d2083da243e051e9f29097aef Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 28 Jul 2006 14:44:51 +0200 Subject: [PATCH 045/143] [PATCH] x86_64: Dump leftover backtrace entries when dwarf2 unwinder got stuck The dwarf2 unwinder currently often gets stuck because a lot of assembly code doesn't have proper dwarf2 annotiation yet. This currently often happens with __down. Should fix this by adding proper dwarf2 annotation to all inline assembly. However until that's done we need a quick fix for 2.6.18 to avoid incomplete backtraces. So when this happens dump the rest of the stack with the old unwinder instead of silently not dumping it. There was already a optional "both" mode that dumped both, but that was too ugly. I also clarified the headers for the different backtraces a bit. Also add a clear error message for missing dwarf2 annotation that people can work on. And I removed a dead variable left over from Ingo's changes. Cc: mingo@elte.hu Cc: jbeulich@novell.com Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/traps.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index eb39a2775236..f7a9d1421078 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -254,7 +254,6 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s { const unsigned cpu = safe_smp_processor_id(); unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; - int i = 11; unsigned used = 0; printk("\nCall Trace:\n"); @@ -275,11 +274,20 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (unwind_init_blocked(&info, tsk) == 0) unw_ret = show_trace_unwind(&info, NULL); } - if (unw_ret > 0) { - if (call_trace > 0) + if (unw_ret > 0 && !arch_unw_user_mode(&info)) { +#ifdef CONFIG_STACK_UNWIND + unsigned long rip = info.regs.rip; + print_symbol("DWARF2 unwinder stuck at %s\n", rip); + if (call_trace == 1) { + printk("Leftover inexact backtrace:\n"); + stack = (unsigned long *)info.regs.rsp; + } else if (call_trace > 1) return; - printk("Legacy call trace:"); - i = 18; + else + printk("Full inexact backtrace again:\n"); +#else + printk("Inexact backtrace:\n"); +#endif } } @@ -1118,8 +1126,10 @@ static int __init call_trace_setup(char *s) call_trace = -1; else if (strcmp(s, "both") == 0) call_trace = 0; - else if (strcmp(s, "new") == 0) + else if (strcmp(s, "newfallback") == 0) call_trace = 1; + else if (strcmp(s, "new") == 0) + call_trace = 2; return 1; } __setup("call_trace=", call_trace_setup); From b783fd925cdd56d24d164e5bdcb072f2a67aedf4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 28 Jul 2006 14:44:54 +0200 Subject: [PATCH 046/143] [PATCH] x86_64: Document backtracer selection options Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- Documentation/x86_64/boot-options.txt | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt index 6887d44d2661..6da24e7a56cb 100644 --- a/Documentation/x86_64/boot-options.txt +++ b/Documentation/x86_64/boot-options.txt @@ -238,6 +238,13 @@ Debugging pagefaulttrace Dump all page faults. Only useful for extreme debugging and will create a lot of output. + call_trace=[old|both|newfallback|new] + old: use old inexact backtracer + new: use new exact dwarf2 unwinder + both: print entries from both + newfallback: use new unwinder but fall back to old if it gets + stuck (default) + Misc noreplacement Don't replace instructions with more appropriate ones From c97d20a6c51067a38f53680d9609b4cf2867d077 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 28 Jul 2006 14:44:57 +0200 Subject: [PATCH 047/143] [PATCH] i386: Do backtrace fallback too Similar patch to earlier x86-64 patch. When the dwarf2 unwinder fails dump the left over stack with the old unwinder. Also some clarifications in the headers. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 313ac1f7dc5a..3facc8fcb91e 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -187,10 +187,21 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, if (unwind_init_blocked(&info, task) == 0) unw_ret = show_trace_unwind(&info, log_lvl); } - if (unw_ret > 0) { - if (call_trace > 0) + if (unw_ret > 0 && !arch_unw_user_mode(&info)) { +#ifdef CONFIG_STACK_UNWIND + print_symbol("DWARF2 unwinder stuck at %s\n", + UNW_PC(info.regs)); + if (call_trace == 1) { + printk("Leftover inexact backtrace:\n"); + if (UNW_SP(info.regs)) + stack = (void *)UNW_SP(info.regs); + } else if (call_trace > 1) return; - printk("%sLegacy call trace:\n", log_lvl); + else + printk("Full inexact backtrace again:\n"); +#else + printk("Inexact backtrace:\n"); +#endif } } From 627371d73cdd04ed23fe098755b4f855138ad9e0 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 29 Jul 2006 05:16:20 +0200 Subject: [PATCH 048/143] [PATCH] pi-futex: robust-futex exit crash fix Fix pi_state->list handling bugs: list handling mishap, locking error. Plus add more debug checks and fix a few style issues i noticed while debugging this. (reported by Ulrich Drepper and Jakub Jelinek.) Signed-off-by: Ingo Molnar Signed-off-by: Linus Torvalds --- kernel/futex.c | 32 ++++++++++++++++++++++++-------- 1 file changed, 24 insertions(+), 8 deletions(-) diff --git a/kernel/futex.c b/kernel/futex.c index cf0c8e21d1ab..f59003b1d8f9 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -415,15 +415,15 @@ static struct task_struct * futex_find_get_task(pid_t pid) */ void exit_pi_state_list(struct task_struct *curr) { - struct futex_hash_bucket *hb; struct list_head *next, *head = &curr->pi_state_list; struct futex_pi_state *pi_state; + struct futex_hash_bucket *hb; union futex_key key; /* * We are a ZOMBIE and nobody can enqueue itself on * pi_state_list anymore, but we have to be careful - * versus waiters unqueueing themselfs + * versus waiters unqueueing themselves: */ spin_lock_irq(&curr->pi_lock); while (!list_empty(head)) { @@ -431,21 +431,24 @@ void exit_pi_state_list(struct task_struct *curr) next = head->next; pi_state = list_entry(next, struct futex_pi_state, list); key = pi_state->key; + hb = hash_futex(&key); spin_unlock_irq(&curr->pi_lock); - hb = hash_futex(&key); spin_lock(&hb->lock); spin_lock_irq(&curr->pi_lock); + /* + * We dropped the pi-lock, so re-check whether this + * task still owns the PI-state: + */ if (head->next != next) { spin_unlock(&hb->lock); continue; } - list_del_init(&pi_state->list); - WARN_ON(pi_state->owner != curr); - + WARN_ON(list_empty(&pi_state->list)); + list_del_init(&pi_state->list); pi_state->owner = NULL; spin_unlock_irq(&curr->pi_lock); @@ -470,7 +473,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, struct futex_q *me) head = &hb->chain; list_for_each_entry_safe(this, next, head, list) { - if (match_futex (&this->key, &me->key)) { + if (match_futex(&this->key, &me->key)) { /* * Another waiter already exists - bump up * the refcount and return its pi_state: @@ -482,6 +485,8 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, struct futex_q *me) if (unlikely(!pi_state)) return -EINVAL; + WARN_ON(!atomic_read(&pi_state->refcount)); + atomic_inc(&pi_state->refcount); me->pi_state = pi_state; @@ -510,6 +515,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, struct futex_q *me) pi_state->key = me->key; spin_lock_irq(&p->pi_lock); + WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &p->pi_state_list); pi_state->owner = p; spin_unlock_irq(&p->pi_lock); @@ -584,9 +590,17 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) if (curval != uval) return -EINVAL; - list_del_init(&pi_state->owner->pi_state_list); + spin_lock_irq(&pi_state->owner->pi_lock); + WARN_ON(list_empty(&pi_state->list)); + list_del_init(&pi_state->list); + spin_unlock_irq(&pi_state->owner->pi_lock); + + spin_lock_irq(&new_owner->pi_lock); + WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &new_owner->pi_state_list); pi_state->owner = new_owner; + spin_unlock_irq(&new_owner->pi_lock); + rt_mutex_unlock(&pi_state->pi_mutex); return 0; @@ -1236,6 +1250,7 @@ static int do_futex_lock_pi(u32 __user *uaddr, int detect, int trylock, /* Owner died? */ if (q.pi_state->owner != NULL) { spin_lock_irq(&q.pi_state->owner->pi_lock); + WARN_ON(list_empty(&q.pi_state->list)); list_del_init(&q.pi_state->list); spin_unlock_irq(&q.pi_state->owner->pi_lock); } else @@ -1244,6 +1259,7 @@ static int do_futex_lock_pi(u32 __user *uaddr, int detect, int trylock, q.pi_state->owner = current; spin_lock_irq(¤t->pi_lock); + WARN_ON(!list_empty(&q.pi_state->list)); list_add(&q.pi_state->list, ¤t->pi_state_list); spin_unlock_irq(¤t->pi_lock); From e3f2ddeac718c768fdac4b7fe69d465172f788a8 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 29 Jul 2006 05:17:57 +0200 Subject: [PATCH 049/143] [PATCH] pi-futex: robust-futex exit Fix robust PI-futexes to be properly unlocked on unexpected exit. For this to work the kernel has to know whether a futex is a PI or a non-PI one, because the semantics are different. Since the space in relevant glibc data structures is extremely scarce, the best solution is to encode the 'PI' information in bit 0 of the robust list pointer. Existing (non-PI) glibc robust futexes have this bit always zero, so the ABI is kept. New glibc with PI-robust-futexes will set this bit. Further fixes from Thomas Gleixner Signed-off-by: Ingo Molnar Signed-off-by: Ulrich Drepper Signed-off-by: Thomas Gleixner Signed-off-by: Linus Torvalds --- include/linux/futex.h | 3 +- kernel/futex.c | 91 +++++++++++++++++++++++++++++-------------- kernel/futex_compat.c | 34 +++++++++++----- 3 files changed, 89 insertions(+), 39 deletions(-) diff --git a/include/linux/futex.h b/include/linux/futex.h index 34c3a215f2cd..d097b5b72bc6 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -96,7 +96,8 @@ struct robust_list_head { long do_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout, u32 __user *uaddr2, u32 val2, u32 val3); -extern int handle_futex_death(u32 __user *uaddr, struct task_struct *curr); +extern int +handle_futex_death(u32 __user *uaddr, struct task_struct *curr, int pi); #ifdef CONFIG_FUTEX extern void exit_robust_list(struct task_struct *curr); diff --git a/kernel/futex.c b/kernel/futex.c index f59003b1d8f9..dda2049692a2 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -495,10 +495,13 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, struct futex_q *me) } /* - * We are the first waiter - try to look up the real owner and - * attach the new pi_state to it: + * We are the first waiter - try to look up the real owner and attach + * the new pi_state to it, but bail out when the owner died bit is set + * and TID = 0: */ pid = uval & FUTEX_TID_MASK; + if (!pid && (uval & FUTEX_OWNER_DIED)) + return -ESRCH; p = futex_find_get_task(pid); if (!p) return -ESRCH; @@ -579,16 +582,17 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) * kept enabled while there is PI state around. We must also * preserve the owner died bit.) */ - newval = (uval & FUTEX_OWNER_DIED) | FUTEX_WAITERS | new_owner->pid; + if (!(uval & FUTEX_OWNER_DIED)) { + newval = FUTEX_WAITERS | new_owner->pid; - inc_preempt_count(); - curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); - - if (curval == -EFAULT) - return -EFAULT; - if (curval != uval) - return -EINVAL; + inc_preempt_count(); + curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); + dec_preempt_count(); + if (curval == -EFAULT) + return -EFAULT; + if (curval != uval) + return -EINVAL; + } spin_lock_irq(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); @@ -1443,9 +1447,11 @@ static int futex_unlock_pi(u32 __user *uaddr) * again. If it succeeds then we can return without waking * anyone else up: */ - inc_preempt_count(); - uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); - dec_preempt_count(); + if (!(uval & FUTEX_OWNER_DIED)) { + inc_preempt_count(); + uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); + dec_preempt_count(); + } if (unlikely(uval == -EFAULT)) goto pi_faulted; @@ -1478,9 +1484,11 @@ static int futex_unlock_pi(u32 __user *uaddr) /* * No waiters - kernel unlocks the futex: */ - ret = unlock_futex_pi(uaddr, uval); - if (ret == -EFAULT) - goto pi_faulted; + if (!(uval & FUTEX_OWNER_DIED)) { + ret = unlock_futex_pi(uaddr, uval); + if (ret == -EFAULT) + goto pi_faulted; + } out_unlock: spin_unlock(&hb->lock); @@ -1699,9 +1707,9 @@ sys_get_robust_list(int pid, struct robust_list_head __user **head_ptr, * Process a futex-list entry, check whether it's owned by the * dying task, and do notification if so: */ -int handle_futex_death(u32 __user *uaddr, struct task_struct *curr) +int handle_futex_death(u32 __user *uaddr, struct task_struct *curr, int pi) { - u32 uval, nval; + u32 uval, nval, mval; retry: if (get_user(uval, uaddr)) @@ -1718,20 +1726,44 @@ int handle_futex_death(u32 __user *uaddr, struct task_struct *curr) * thread-death.) The rest of the cleanup is done in * userspace. */ - nval = futex_atomic_cmpxchg_inatomic(uaddr, uval, - uval | FUTEX_OWNER_DIED); + mval = (uval & FUTEX_WAITERS) | FUTEX_OWNER_DIED; + nval = futex_atomic_cmpxchg_inatomic(uaddr, uval, mval); + if (nval == -EFAULT) return -1; if (nval != uval) goto retry; - if (uval & FUTEX_WAITERS) - futex_wake(uaddr, 1); + /* + * Wake robust non-PI futexes here. The wakeup of + * PI futexes happens in exit_pi_state(): + */ + if (!pi) { + if (uval & FUTEX_WAITERS) + futex_wake(uaddr, 1); + } } return 0; } +/* + * Fetch a robust-list pointer. Bit 0 signals PI futexes: + */ +static inline int fetch_robust_entry(struct robust_list __user **entry, + struct robust_list __user **head, int *pi) +{ + unsigned long uentry; + + if (get_user(uentry, (unsigned long *)head)) + return -EFAULT; + + *entry = (void *)(uentry & ~1UL); + *pi = uentry & 1; + + return 0; +} + /* * Walk curr->robust_list (very carefully, it's a userspace list!) * and mark any locks found there dead, and notify any waiters. @@ -1742,14 +1774,14 @@ void exit_robust_list(struct task_struct *curr) { struct robust_list_head __user *head = curr->robust_list; struct robust_list __user *entry, *pending; - unsigned int limit = ROBUST_LIST_LIMIT; + unsigned int limit = ROBUST_LIST_LIMIT, pi, pip; unsigned long futex_offset; /* * Fetch the list head (which was registered earlier, via * sys_set_robust_list()): */ - if (get_user(entry, &head->list.next)) + if (fetch_robust_entry(&entry, &head->list.next, &pi)) return; /* * Fetch the relative futex offset: @@ -1760,10 +1792,11 @@ void exit_robust_list(struct task_struct *curr) * Fetch any possibly pending lock-add first, and handle it * if it exists: */ - if (get_user(pending, &head->list_op_pending)) + if (fetch_robust_entry(&pending, &head->list_op_pending, &pip)) return; + if (pending) - handle_futex_death((void *)pending + futex_offset, curr); + handle_futex_death((void *)pending + futex_offset, curr, pip); while (entry != &head->list) { /* @@ -1772,12 +1805,12 @@ void exit_robust_list(struct task_struct *curr) */ if (entry != pending) if (handle_futex_death((void *)entry + futex_offset, - curr)) + curr, pi)) return; /* * Fetch the next entry in the list: */ - if (get_user(entry, &entry->next)) + if (fetch_robust_entry(&entry, &entry->next, &pi)) return; /* * Avoid excessively long or circular lists: diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index d1d92b441fb7..d1aab1a452cc 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -12,6 +12,23 @@ #include + +/* + * Fetch a robust-list pointer. Bit 0 signals PI futexes: + */ +static inline int +fetch_robust_entry(compat_uptr_t *uentry, struct robust_list __user **entry, + compat_uptr_t *head, int *pi) +{ + if (get_user(*uentry, head)) + return -EFAULT; + + *entry = compat_ptr((*uentry) & ~1); + *pi = (unsigned int)(*uentry) & 1; + + return 0; +} + /* * Walk curr->robust_list (very carefully, it's a userspace list!) * and mark any locks found there dead, and notify any waiters. @@ -22,17 +39,16 @@ void compat_exit_robust_list(struct task_struct *curr) { struct compat_robust_list_head __user *head = curr->compat_robust_list; struct robust_list __user *entry, *pending; + unsigned int limit = ROBUST_LIST_LIMIT, pi; compat_uptr_t uentry, upending; - unsigned int limit = ROBUST_LIST_LIMIT; compat_long_t futex_offset; /* * Fetch the list head (which was registered earlier, via * sys_set_robust_list()): */ - if (get_user(uentry, &head->list.next)) + if (fetch_robust_entry(&uentry, &entry, &head->list.next, &pi)) return; - entry = compat_ptr(uentry); /* * Fetch the relative futex offset: */ @@ -42,11 +58,11 @@ void compat_exit_robust_list(struct task_struct *curr) * Fetch any possibly pending lock-add first, and handle it * if it exists: */ - if (get_user(upending, &head->list_op_pending)) + if (fetch_robust_entry(&upending, &pending, + &head->list_op_pending, &pi)) return; - pending = compat_ptr(upending); if (upending) - handle_futex_death((void *)pending + futex_offset, curr); + handle_futex_death((void *)pending + futex_offset, curr, pi); while (compat_ptr(uentry) != &head->list) { /* @@ -55,15 +71,15 @@ void compat_exit_robust_list(struct task_struct *curr) */ if (entry != pending) if (handle_futex_death((void *)entry + futex_offset, - curr)) + curr, pi)) return; /* * Fetch the next entry in the list: */ - if (get_user(uentry, (compat_uptr_t *)&entry->next)) + if (fetch_robust_entry(&uentry, &entry, + (compat_uptr_t *)&entry->next, &pi)) return; - entry = compat_ptr(uentry); /* * Avoid excessively long or circular lists: */ From 9a71db721a2cbb9921b929b2699ab181f5a3c6c0 Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Fri, 21 Jul 2006 15:49:32 -0400 Subject: [PATCH 050/143] [PATCH] myri10ge - Always do a dummy RDMA after loading the firmware Always do a dummy RDMA after loading the firmware to work around buggy PCIe chipsets which do not implement resending properly. This is so cheap as to be almost free, and should never have been conditional on the tx boundary != 4096. Signed-off-by: Brice Goglin Signed-off-by: Jeff Garzik --- drivers/net/myri10ge/myri10ge.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 07ca9480a6fe..c3e52c806b13 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -620,7 +620,7 @@ static int myri10ge_load_firmware(struct myri10ge_priv *mgp) return -ENXIO; } dev_info(&mgp->pdev->dev, "handoff confirmed\n"); - myri10ge_dummy_rdma(mgp, mgp->tx.boundary != 4096); + myri10ge_dummy_rdma(mgp, 1); return 0; } From 187ff3b8575569668cde9021b9fe70386a4131fb Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 19 Jul 2006 14:08:42 -0700 Subject: [PATCH 051/143] [PATCH] skge: chip clock rate typo Okay, Fix both typo's in one patch .The impact is that the incorrect value was being computed for blinking LED and interrupt moderation values. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik --- drivers/net/skge.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 82200bfaa8ed..7de9a07b2ac2 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -516,10 +516,7 @@ static int skge_set_pauseparam(struct net_device *dev, /* Chip internal frequency for clock calculations */ static inline u32 hwkhz(const struct skge_hw *hw) { - if (hw->chip_id == CHIP_ID_GENESIS) - return 53215; /* or: 53.125 MHz */ - else - return 78215; /* or: 78.125 MHz */ + return (hw->chip_id == CHIP_ID_GENESIS) ? 53125 : 78125; } /* Chip HZ to microseconds */ From a6a38a66224c7c578cfed2f584b440c81af0c3ae Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Sat, 29 Jul 2006 08:29:26 +0100 Subject: [PATCH 052/143] [ARM] 3736/1: xscale: don't mis-report 80219 as an iop32x Patch from Lennert Buytenhek The IOP 80219 xscale CPU is a stripped down version of the IOP32x. But the fact that the 80219 and IOP32x are very similar doesn't mean that they need to share a cpu table entry. It's also somewhat confusing for the end user to see the 80219 reported as an IOP32x, so this patch splits the IOP32x cpu table entry to make a separate entry for the 80219. Signed-off-by: Lennert Buytenhek Signed-off-by: Dan Williams Signed-off-by: Russell King --- arch/arm/mm/proc-xscale.S | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S index 521538671f4c..561bff73a036 100644 --- a/arch/arm/mm/proc-xscale.S +++ b/arch/arm/mm/proc-xscale.S @@ -536,6 +536,11 @@ cpu_80200_name: .asciz "XScale-80200" .size cpu_80200_name, . - cpu_80200_name + .type cpu_80219_name, #object +cpu_80219_name: + .asciz "XScale-80219" + .size cpu_80219_name, . - cpu_80219_name + .type cpu_8032x_name, #object cpu_8032x_name: .asciz "XScale-IOP8032x Family" @@ -613,10 +618,33 @@ __80200_proc_info: .long xscale_cache_fns .size __80200_proc_info, . - __80200_proc_info + .type __80219_proc_info,#object +__80219_proc_info: + .long 0x69052e20 + .long 0xffffffe0 + .long PMD_TYPE_SECT | \ + PMD_SECT_BUFFERABLE | \ + PMD_SECT_CACHEABLE | \ + PMD_SECT_AP_WRITE | \ + PMD_SECT_AP_READ + .long PMD_TYPE_SECT | \ + PMD_SECT_AP_WRITE | \ + PMD_SECT_AP_READ + b __xscale_setup + .long cpu_arch_name + .long cpu_elf_name + .long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP + .long cpu_80219_name + .long xscale_processor_functions + .long v4wbi_tlb_fns + .long xscale_mc_user_fns + .long xscale_cache_fns + .size __80219_proc_info, . - __80219_proc_info + .type __8032x_proc_info,#object __8032x_proc_info: .long 0x69052420 - .long 0xfffff5e0 @ mask should accomodate IOP80219 also + .long 0xffffffe0 .long PMD_TYPE_SECT | \ PMD_SECT_BUFFERABLE | \ PMD_SECT_CACHEABLE | \ From 7b1df1985c7a4e6d5fcab0ed71a2cd1d4ad3be41 Mon Sep 17 00:00:00 2001 From: "George G. Davis" Date: Sat, 29 Jul 2006 08:29:27 +0100 Subject: [PATCH 053/143] [ARM] 3737/1: Export ARM copy/clear_user_page symbols Patch from George G. Davis As reported by various folks on the ARM Linux kernel mailing list, the video-buf.ko driver has undefined references on all ARM machines which use it as observed during `make modules`: Warning: "v4wb_clear_user_page" [drivers/media/video/video-buf.ko] undefined! Similar warnings exist for all ARM machines which use this driver. So this change adds the missing EXPORT_SYMBOLs to allow using this driver as a module. Signed-off-by: George G. Davis Signed-off-by: Russell King --- arch/arm/mm/proc-syms.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/arm/mm/proc-syms.c b/arch/arm/mm/proc-syms.c index 6c5f0fe578a5..ab143557e688 100644 --- a/arch/arm/mm/proc-syms.c +++ b/arch/arm/mm/proc-syms.c @@ -13,6 +13,7 @@ #include #include #include +#include #ifndef MULTI_CPU EXPORT_SYMBOL(cpu_dcache_clean_area); @@ -30,6 +31,13 @@ EXPORT_SYMBOL(__cpuc_coherent_kern_range); EXPORT_SYMBOL(cpu_cache); #endif +#ifndef MULTI_USER +EXPORT_SYMBOL(__cpu_clear_user_page); +EXPORT_SYMBOL(__cpu_copy_user_page); +#else +EXPORT_SYMBOL(cpu_user); +#endif + /* * No module should need to touch the TLB (and currently * no modules do. We export this for "loadkernel" support From ceaccbd2a6a82b97bd15938fc7ffe180754cbe6c Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Sat, 29 Jul 2006 08:29:30 +0100 Subject: [PATCH 054/143] [ARM] 3734/1: Fix the unused variable warning in __iounmap() Patch from Catalin Marinas This patch adds #ifdef around some variables in the arch/arm/mm/ioremap.c file. Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/mm/ioremap.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c index dba7dddfe57d..88a999df0ab3 100644 --- a/arch/arm/mm/ioremap.c +++ b/arch/arm/mm/ioremap.c @@ -363,7 +363,9 @@ EXPORT_SYMBOL(__ioremap); void __iounmap(void __iomem *addr) { +#ifndef CONFIG_SMP struct vm_struct **p, *tmp; +#endif unsigned int section_mapping = 0; addr = (void __iomem *)(PAGE_MASK & (unsigned long)addr); From 260f659b232b17889e3f0c9bf411675898b222c2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:34 +0200 Subject: [PATCH 055/143] [PATCH] x86_64: Update defconfig Update defconfig Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/defconfig | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 83d389b8ebd8..840d5d93d5cc 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-git22 -# Tue Jul 4 14:24:40 2006 +# Linux kernel version: 2.6.18-rc2 +# Tue Jul 18 17:13:20 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -37,6 +37,7 @@ CONFIG_SWAP=y CONFIG_SYSVIPC=y CONFIG_POSIX_MQUEUE=y # CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set CONFIG_SYSCTL=y # CONFIG_AUDIT is not set CONFIG_IKCONFIG=y @@ -413,6 +414,7 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set @@ -1195,7 +1197,7 @@ CONFIG_USB_MON=y # CONFIG_USB_LEGOTOWER is not set # CONFIG_USB_LCD is not set # CONFIG_USB_LED is not set -# CONFIG_USB_CY7C63 is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set # CONFIG_USB_PHIDGETKIT is not set # CONFIG_USB_PHIDGETSERVO is not set @@ -1373,7 +1375,6 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set From 0e5f61b00c577da698fb00cd9c91a96b79044dfd Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:37 +0200 Subject: [PATCH 056/143] [PATCH] x86_64: On Intel systems when CPU has C3 don't use TSC On Intel systems generally the TSC stops in C3 or deeper, so don't use it there. Follows similar logic on i386. This should fix problems on Meroms. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/time.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index e0341c6808e5..7a9b18224182 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -28,6 +28,7 @@ #include #ifdef CONFIG_ACPI #include /* for PM timer frequency */ +#include #endif #include #include @@ -953,11 +954,18 @@ __cpuinit int unsynchronized_tsc(void) #ifdef CONFIG_SMP if (apic_is_clustered_box()) return 1; - /* Intel systems are normally all synchronized. Exceptions - are handled in the check above. */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - return 0; #endif + /* Most intel systems have synchronized TSCs except for + multi node systems */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) { +#ifdef CONFIG_ACPI + /* But TSC doesn't tick in C3 so don't use it there */ + if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 100) + return 1; +#endif + return 0; + } + /* Assume multi socket systems are not synchronized */ return num_present_cpus() > 1; } From 089bbbcb36979166131868a89ca5f4e695d6637d Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Sat, 29 Jul 2006 21:42:40 +0200 Subject: [PATCH 057/143] [PATCH] x86_64: Calgary IOMMU - fix off by one error Fixed off-by-one error in detect_calgary and calgary_init which will cause arrays to overflow. Also, removed impossible to hit BUG_ON. Signed-off-by: Jon Mason Signed-off-by: Muli Ben-Yehuda Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/pci-calgary.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index e71ed53b08fb..92744abff133 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -812,7 +812,7 @@ static int __init calgary_init(void) int i, ret = -ENODEV; struct pci_dev *dev = NULL; - for (i = 0; i <= num_online_nodes() * MAX_NUM_OF_PHBS; i++) { + for (i = 0; i < num_online_nodes() * MAX_NUM_OF_PHBS; i++) { dev = pci_get_device(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CALGARY, dev); @@ -890,9 +890,8 @@ void __init detect_calgary(void) specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE); for (bus = 0, table_idx = 0; - bus <= num_online_nodes() * MAX_PHB_BUS_NUM; + bus < num_online_nodes() * MAX_PHB_BUS_NUM; bus++) { - BUG_ON(bus > MAX_NUMNODES * MAX_PHB_BUS_NUM); if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY) continue; if (test_bit(bus, translation_disabled)) { @@ -1002,7 +1001,7 @@ static int __init calgary_parse_options(char *p) if (p == endp) break; - if (bridge <= (num_online_nodes() * MAX_PHB_BUS_NUM)) { + if (bridge < (num_online_nodes() * MAX_PHB_BUS_NUM)) { printk(KERN_INFO "Calgary: disabling " "translation for PHB 0x%x\n", bridge); set_bit(bridge, translation_disabled); From d2105b10fe0f460c388fe4e09226313f519d8c00 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Sat, 29 Jul 2006 21:42:43 +0200 Subject: [PATCH 058/143] [PATCH] x86_64: Calgary IOMMU - Multi-Node NULL pointer dereference fix Calgary hits a NULL pointer dereference when booting in a multi-chassis NUMA system. See Redhat bugzilla number 198498, found by Konrad Rzeszutek (konradr@redhat.com). There are many issues that had to be resolved to fix this problem. Firstly when I originally wrote the code to handle NUMA systems, I had a large misunderstanding that was not corrected until now. That was that I thought the "number of nodes online" referred to number of physical systems connected. So that if NUMA was disabled, there would only be 1 node and it would only show that node's PCI bus. In reality if NUMA is disabled, the system displays all of the connected chassis as one node but is only ignorant of the delays in accessing main memory. Therefore, references to num_online_nodes() and MAX_NUMNODES are incorrect and need to be set to the maximum number of nodes that can be accessed (which are 8). I created a variable, MAX_NUM_CHASSIS, and set it to 8 to fix this. Secondly, when walking the PCI in detect_calgary, the code only checked the first "slot" when looking to see if a device is present. This will work for most cases, but unfortunately it isn't always the case. In the NUMA MXE drawers, there are USB devices present on the 3rd slot (with slot 1 being empty). So, to work around this, all slots (up to 8) are scanned to see if there are any devices present. Lastly, the bus is being enumerated on large systems in a different way the we originally thought. This throws the ugly logic we had out the window. To more elegantly handle this, I reorganized the kva array to be sparse (which removed the need to have any bus number to kva slot logic in tce.c) and created a secondary space array to contain the bus number to phb mapping. With these changes Calgary boots on an x460 with 4 nodes with and without NUMA enabled. Signed-off-by: Jon Mason Signed-off-by: Muli Ben-Yehuda Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/pci-calgary.c | 76 ++++++++++++++++++-------------- arch/x86_64/kernel/tce.c | 4 +- include/asm-x86_64/calgary.h | 5 --- 3 files changed, 45 insertions(+), 40 deletions(-) diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index 92744abff133..146924ba5df5 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -85,7 +85,8 @@ #define CSR_AGENT_MASK 0xffe0ffff #define MAX_NUM_OF_PHBS 8 /* how many PHBs in total? */ -#define MAX_PHB_BUS_NUM (MAX_NUM_OF_PHBS * 2) /* max dev->bus->number */ +#define MAX_NUM_CHASSIS 8 /* max number of chassis */ +#define MAX_PHB_BUS_NUM (MAX_NUM_OF_PHBS * MAX_NUM_CHASSIS * 2) /* max dev->bus->number */ #define PHBS_PER_CALGARY 4 /* register offsets in Calgary's internal register space */ @@ -110,7 +111,8 @@ static const unsigned long phb_offsets[] = { 0xB000 /* PHB3 */ }; -void* tce_table_kva[MAX_NUM_OF_PHBS * MAX_NUMNODES]; +static char bus_to_phb[MAX_PHB_BUS_NUM]; +void* tce_table_kva[MAX_PHB_BUS_NUM]; unsigned int specified_table_size = TCE_TABLE_SIZE_UNSPECIFIED; static int translate_empty_slots __read_mostly = 0; static int calgary_detected __read_mostly = 0; @@ -119,7 +121,7 @@ static int calgary_detected __read_mostly = 0; * the bitmap of PHBs the user requested that we disable * translation on. */ -static DECLARE_BITMAP(translation_disabled, MAX_NUMNODES * MAX_PHB_BUS_NUM); +static DECLARE_BITMAP(translation_disabled, MAX_PHB_BUS_NUM); static void tce_cache_blast(struct iommu_table *tbl); @@ -452,7 +454,7 @@ static struct dma_mapping_ops calgary_dma_ops = { static inline int busno_to_phbid(unsigned char num) { - return bus_to_phb(num) % PHBS_PER_CALGARY; + return bus_to_phb[num]; } static inline unsigned long split_queue_offset(unsigned char num) @@ -812,7 +814,7 @@ static int __init calgary_init(void) int i, ret = -ENODEV; struct pci_dev *dev = NULL; - for (i = 0; i < num_online_nodes() * MAX_NUM_OF_PHBS; i++) { + for (i = 0; i < MAX_PHB_BUS_NUM; i++) { dev = pci_get_device(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CALGARY, dev); @@ -822,7 +824,7 @@ static int __init calgary_init(void) calgary_init_one_nontraslated(dev); continue; } - if (!tce_table_kva[i] && !translate_empty_slots) { + if (!tce_table_kva[dev->bus->number] && !translate_empty_slots) { pci_dev_put(dev); continue; } @@ -842,7 +844,7 @@ static int __init calgary_init(void) pci_dev_put(dev); continue; } - if (!tce_table_kva[i] && !translate_empty_slots) + if (!tce_table_kva[dev->bus->number] && !translate_empty_slots) continue; calgary_disable_translation(dev); calgary_free_tar(dev); @@ -876,9 +878,10 @@ static inline int __init determine_tce_table_size(u64 ram) void __init detect_calgary(void) { u32 val; - int bus, table_idx; + int bus; void *tbl; - int detected = 0; + int calgary_found = 0; + int phb = -1; /* * if the user specified iommu=off or iommu=soft or we found @@ -889,37 +892,46 @@ void __init detect_calgary(void) specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE); - for (bus = 0, table_idx = 0; - bus < num_online_nodes() * MAX_PHB_BUS_NUM; - bus++) { + for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) { + int dev; + + tce_table_kva[bus] = NULL; + bus_to_phb[bus] = -1; + if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY) continue; + + /* + * There are 4 PHBs per Calgary chip. Set phb to which phb (0-3) + * it is connected to releative to the clagary chip. + */ + phb = (phb + 1) % PHBS_PER_CALGARY; + if (test_bit(bus, translation_disabled)) { printk(KERN_INFO "Calgary: translation is disabled for " "PHB 0x%x\n", bus); /* skip this phb, don't allocate a tbl for it */ - tce_table_kva[table_idx] = NULL; - table_idx++; continue; } /* - * scan the first slot of the PCI bus to see if there - * are any devices present + * Scan the slots of the PCI bus to see if there is a device present. + * The parent bus will be the zero-ith device, so start at 1. */ - val = read_pci_config(bus, 1, 0, 0); - if (val != 0xffffffff || translate_empty_slots) { - tbl = alloc_tce_table(); - if (!tbl) - goto cleanup; - detected = 1; - } else - tbl = NULL; - - tce_table_kva[table_idx] = tbl; - table_idx++; + for (dev = 1; dev < 8; dev++) { + val = read_pci_config(bus, dev, 0, 0); + if (val != 0xffffffff || translate_empty_slots) { + tbl = alloc_tce_table(); + if (!tbl) + goto cleanup; + tce_table_kva[bus] = tbl; + bus_to_phb[bus] = phb; + calgary_found = 1; + break; + } + } } - if (detected) { + if (calgary_found) { iommu_detected = 1; calgary_detected = 1; printk(KERN_INFO "PCI-DMA: Calgary IOMMU detected. " @@ -928,9 +940,9 @@ void __init detect_calgary(void) return; cleanup: - for (--table_idx; table_idx >= 0; --table_idx) - if (tce_table_kva[table_idx]) - free_tce_table(tce_table_kva[table_idx]); + for (--bus; bus >= 0; --bus) + if (tce_table_kva[bus]) + free_tce_table(tce_table_kva[bus]); } int __init calgary_iommu_init(void) @@ -1001,7 +1013,7 @@ static int __init calgary_parse_options(char *p) if (p == endp) break; - if (bridge < (num_online_nodes() * MAX_PHB_BUS_NUM)) { + if (bridge < MAX_PHB_BUS_NUM) { printk(KERN_INFO "Calgary: disabling " "translation for PHB 0x%x\n", bridge); set_bit(bridge, translation_disabled); diff --git a/arch/x86_64/kernel/tce.c b/arch/x86_64/kernel/tce.c index d3a9e79e954c..5530dda3f27a 100644 --- a/arch/x86_64/kernel/tce.c +++ b/arch/x86_64/kernel/tce.c @@ -96,7 +96,6 @@ static inline unsigned int table_size_to_number_of_entries(unsigned char size) static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl) { unsigned int bitmapsz; - unsigned int tce_table_index; unsigned long bmppages; int ret; @@ -105,8 +104,7 @@ static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl) /* set the tce table size - measured in entries */ tbl->it_size = table_size_to_number_of_entries(specified_table_size); - tce_table_index = bus_to_phb(tbl->it_busno); - tbl->it_base = (unsigned long)tce_table_kva[tce_table_index]; + tbl->it_base = (unsigned long)tce_table_kva[dev->bus->number]; if (!tbl->it_base) { printk(KERN_ERR "Calgary: iommu_table_setparms: " "no table allocated?!\n"); diff --git a/include/asm-x86_64/calgary.h b/include/asm-x86_64/calgary.h index fbfb50136edb..4e3919524240 100644 --- a/include/asm-x86_64/calgary.h +++ b/include/asm-x86_64/calgary.h @@ -60,9 +60,4 @@ static inline int calgary_iommu_init(void) { return 1; } static inline void detect_calgary(void) { return; } #endif -static inline unsigned int bus_to_phb(unsigned char busno) -{ - return ((busno % 15 == 0) ? 0 : busno / 2 + 1); -} - #endif /* _ASM_X86_64_CALGARY_H */ From 355540f3338e1408dac98413f05d612a76d4f5e3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:46 +0200 Subject: [PATCH 059/143] [PATCH] x86_64: Revert k8-bus.c northbridge access change As Travis Betak points out it accesses the wrong northbridge subfunction now. Switch back to the old code. Cc: "Travis Betak" Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/pci/k8-bus.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/x86_64/pci/k8-bus.c b/arch/x86_64/pci/k8-bus.c index b50a7c7c47f8..3acf60ded2a0 100644 --- a/arch/x86_64/pci/k8-bus.c +++ b/arch/x86_64/pci/k8-bus.c @@ -2,7 +2,6 @@ #include #include #include -#include /* * This discovers the pcibus <-> node mapping on AMD K8. @@ -19,6 +18,7 @@ #define NR_LDT_BUS_NUMBER_REGISTERS 3 #define SECONDARY_LDT_BUS_NUMBER(dword) ((dword >> 8) & 0xFF) #define SUBORDINATE_LDT_BUS_NUMBER(dword) ((dword >> 16) & 0xFF) +#define PCI_DEVICE_ID_K8HTCONFIG 0x1100 /** * fill_mp_bus_to_cpumask() @@ -28,7 +28,8 @@ __init static int fill_mp_bus_to_cpumask(void) { - int i, j, k; + struct pci_dev *nb_dev = NULL; + int i, j; u32 ldtbus, nid; static int lbnr[3] = { LDT_BUS_NUMBER_REGISTER_0, @@ -36,9 +37,8 @@ fill_mp_bus_to_cpumask(void) LDT_BUS_NUMBER_REGISTER_2 }; - cache_k8_northbridges(); - for (k = 0; k < num_k8_northbridges; k++) { - struct pci_dev *nb_dev = k8_northbridges[k]; + while ((nb_dev = pci_get_device(PCI_VENDOR_ID_AMD, + PCI_DEVICE_ID_K8HTCONFIG, nb_dev))) { pci_read_config_dword(nb_dev, NODE_ID_REGISTER, &nid); for (i = 0; i < NR_LDT_BUS_NUMBER_REGISTERS; i++) { From 65f87d8a8a6e1b560c61951d0a68ed80f7c8ff19 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:49 +0200 Subject: [PATCH 060/143] [PATCH] x86_64: Fix swiotlb=force It was broken before. But having it is important as possible hardware bug workaround. And previously there was no way to force swiotlb if there is another IOMMU. Side effect is that iommu=force won't force swiotlb anymore even if there isn't another IOMMU. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/pci-swiotlb.c | 5 +++-- include/asm-x86_64/swiotlb.h | 2 ++ 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c index ebdb77fe2057..6a55f87ba97f 100644 --- a/arch/x86_64/kernel/pci-swiotlb.c +++ b/arch/x86_64/kernel/pci-swiotlb.c @@ -31,9 +31,10 @@ struct dma_mapping_ops swiotlb_dma_ops = { void pci_swiotlb_init(void) { /* don't initialize swiotlb if iommu=off (no_iommu=1) */ - if (!iommu_detected && !no_iommu && - (end_pfn > MAX_DMA32_PFN || force_iommu)) + if (!iommu_detected && !no_iommu && end_pfn > MAX_DMA32_PFN) swiotlb = 1; + if (swiotlb_force) + swiotlb = 1; if (swiotlb) { printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n"); swiotlb_init(); diff --git a/include/asm-x86_64/swiotlb.h b/include/asm-x86_64/swiotlb.h index 5f9a01805821..ba94ab3d2673 100644 --- a/include/asm-x86_64/swiotlb.h +++ b/include/asm-x86_64/swiotlb.h @@ -42,6 +42,8 @@ extern void swiotlb_free_coherent (struct device *hwdev, size_t size, extern int swiotlb_dma_supported(struct device *hwdev, u64 mask); extern void swiotlb_init(void); +extern int swiotlb_force; + #ifdef CONFIG_SWIOTLB extern int swiotlb; #else From 70583161e878c8e9baaace71ba000b46c86b89cc Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:52 +0200 Subject: [PATCH 061/143] [PATCH] i386: Fix up backtrace fallback patch I didn't test all compilation combinations. Shame on me. And fix a missing option in the boot option following x86-64 (Jan Beulich) Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 3facc8fcb91e..021f8fdc7512 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -190,11 +190,11 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, if (unw_ret > 0 && !arch_unw_user_mode(&info)) { #ifdef CONFIG_STACK_UNWIND print_symbol("DWARF2 unwinder stuck at %s\n", - UNW_PC(info.regs)); + UNW_PC(&info)); if (call_trace == 1) { printk("Leftover inexact backtrace:\n"); - if (UNW_SP(info.regs)) - stack = (void *)UNW_SP(info.regs); + if (UNW_SP(&info)) + stack = (void *)UNW_SP(&info); } else if (call_trace > 1) return; else @@ -1249,8 +1249,10 @@ static int __init call_trace_setup(char *s) call_trace = -1; else if (strcmp(s, "both") == 0) call_trace = 0; - else if (strcmp(s, "new") == 0) + else if (strcmp(s, "newfallback") == 0) call_trace = 1; + else if (strcmp(s, "new") == 2) + call_trace = 2; return 1; } __setup("call_trace=", call_trace_setup); From b83a8e64fd1aecf021111d22c062c97a3241d0c4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 29 Jul 2006 21:42:55 +0200 Subject: [PATCH 062/143] [PATCH] MM: Remove rogue readahead printk For some reason it triggers always with NFS root and spams the kernel logs of my nfs root boxes a lot. Signed-off-by: Andi Kleen Acked-by: Trond Myklebust Signed-off-by: Linus Torvalds --- mm/filemap.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index d087fc3d3281..b9a60c43b61a 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -849,8 +849,6 @@ static void shrink_readahead_size_eio(struct file *filp, return; ra->ra_pages /= 4; - printk(KERN_WARNING "Reducing readahead size to %luK\n", - ra->ra_pages << (PAGE_CACHE_SHIFT - 10)); } /** From b6ff50833ad43a8ebd9b16bf53c334f7aaf33c41 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 29 Jul 2006 23:15:36 -0700 Subject: [PATCH 063/143] Linux v2.6.18-rc3 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 1dd58d35d72c..c9b7dbb64c71 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 18 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME=Crazed Snow-Weasel # *DOCUMENTATION* From 9b9b72fe703bd5b89533815eb7eb682cdaf60a6e Mon Sep 17 00:00:00 2001 From: Jake Moilanen Date: Tue, 25 Jul 2006 15:11:47 -0500 Subject: [PATCH 064/143] [POWERPC] Use H_CEDE on non-SMT On the JS21 systems, they have the SPLPAR hypertas set, but are not SMT capable. So, they are not making the H_CEDE call. This is causing the hypervisor to have to queue up work for the hdecr, taking an excessive amount of time in maintenance code, and causing jitter on the box. Making the H_CEDE call helps alleviate that problem. Signed-off-by: Jake Moilanen Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/pseries/setup.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index 54a52437265c..71c634e0b87c 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -501,7 +501,8 @@ static void pseries_dedicated_idle_sleep(void) } /* - * Cede if the other thread is not idle, so that it can + * If not SMT, cede processor. If CPU is running SMT + * cede if the other thread is not idle, so that it can * go single-threaded. If the other thread is idle, * we ask the hypervisor if it has pending work it * wants to do and cede if it does. Otherwise we keep @@ -514,7 +515,8 @@ static void pseries_dedicated_idle_sleep(void) * very low priority. The cede enables interrupts, which * doesn't matter here. */ - if (!lppaca[cpu ^ 1].idle || poll_pending() == H_PENDING) + if (!cpu_has_feature(CPU_FTR_SMT) || !lppaca[cpu ^ 1].idle + || poll_pending() == H_PENDING) cede_processor(); out: From 8d950cb8896fc95a9444d190885779438bb9d01c Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Tue, 25 Jul 2006 14:05:01 -0700 Subject: [PATCH 065/143] [POWERPC] Minor comment fix for misc_64.S A minor comment fix for misc_64.S from Takao Shinohara. Signed-off-by: Geoff Levand Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/misc_64.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index bfb407fc1aa1..e3ed21cd3d94 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -687,7 +687,7 @@ _GLOBAL(kexec_sequence) /* clear out hardware hash page table and tlb */ ld r5,0(r27) /* deref function descriptor */ mtctr r5 - bctrl /* ppc_md.hash_clear_all(void); */ + bctrl /* ppc_md.hpte_clear_all(void); */ /* * kexec image calling is: From f712c0c7e1796f92e45e4de144e247816d974b8f Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Sun, 30 Jul 2006 03:02:59 -0700 Subject: [PATCH 066/143] [PATCH] sched: build_sched_domains() fix Use the correct groups while initializing sched groups power for allnodes_domain. This fixes the crash observed while creating exclusive cpusets. Signed-off-by: Suresh Siddha Reported-and-tested-by: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sched.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/sched.c b/kernel/sched.c index b44b9a43b0fc..41a571806ce0 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6494,7 +6494,12 @@ static int build_sched_domains(const cpumask_t *cpu_map) for (i = 0; i < MAX_NUMNODES; i++) init_numa_sched_groups_power(sched_group_nodes[i]); - init_numa_sched_groups_power(sched_group_allnodes); + if (sched_group_allnodes) { + int group = cpu_to_allnodes_group(first_cpu(*cpu_map)); + struct sched_group *sg = &sched_group_allnodes[group]; + + init_numa_sched_groups_power(sg); + } #endif /* Attach the domains */ From 2ccb48ebb4de139eef4fcefd5f2bb823cb0d81b9 Mon Sep 17 00:00:00 2001 From: Neil Brown Date: Sun, 30 Jul 2006 03:03:01 -0700 Subject: [PATCH 067/143] [PATCH] ext3: avoid triggering ext3_error on bad NFS file handle The inode number out of an NFS file handle gets passed eventually to ext3_get_inode_block() without any checking. If ext3_get_inode_block() allows it to trigger an error, then bad filehandles can have unpleasant effect - ext3_error() will usually cause a forced read-only remount, or a panic if `errors=panic' was used. So remove the call to ext3_error there and put a matching check in ext3/namei.c where inode numbers are read off storage. [akpm@osdl.org: fix off-by-one error] Signed-off-by: Neil Brown Signed-off-by: Jan Kara Cc: Marcel Holtmann Cc: Cc: "Stephen C. Tweedie" Cc: Eric Sandeen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/inode.c | 13 +++++++------ fs/ext3/namei.c | 15 +++++++++++++-- include/linux/ext3_fs.h | 9 +++++++++ 3 files changed, 29 insertions(+), 8 deletions(-) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index f804d5e9d60c..ab034d3053ea 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -2402,14 +2402,15 @@ static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, struct buffer_head *bh; struct ext3_group_desc * gdp; - - if ((ino != EXT3_ROOT_INO && ino != EXT3_JOURNAL_INO && - ino != EXT3_RESIZE_INO && ino < EXT3_FIRST_INO(sb)) || - ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) { - ext3_error(sb, "ext3_get_inode_block", - "bad inode number: %lu", ino); + if (!ext3_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ return 0; } + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); if (block_group >= EXT3_SB(sb)->s_groups_count) { ext3_error(sb,"ext3_get_inode_block","group >= groups count"); diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index d9176dba3698..2aa7101b27cd 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1000,7 +1000,12 @@ static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, str if (bh) { unsigned long ino = le32_to_cpu(de->inode); brelse (bh); - inode = iget(dir->i_sb, ino); + if (!ext3_valid_inum(dir->i_sb, ino)) { + ext3_error(dir->i_sb, "ext3_lookup", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(dir->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); @@ -1028,7 +1033,13 @@ struct dentry *ext3_get_parent(struct dentry *child) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); brelse(bh); - inode = iget(child->d_inode->i_sb, ino); + + if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { + ext3_error(child->d_inode->i_sb, "ext3_get_parent", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(child->d_inode->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 5607e6457a65..9f9cce7bd86d 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -492,6 +492,15 @@ static inline struct ext3_inode_info *EXT3_I(struct inode *inode) { return container_of(inode, struct ext3_inode_info, vfs_inode); } + +static inline int ext3_valid_inum(struct super_block *sb, unsigned long ino) +{ + return ino == EXT3_ROOT_INO || + ino == EXT3_JOURNAL_INO || + ino == EXT3_RESIZE_INO || + (ino >= EXT3_FIRST_INO(sb) && + ino <= le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)); +} #else /* Assume that user mode programs are passing in an ext3fs superblock, not * a kernel struct super_block. This will allow us to call the feature-test From 822cfbff2ef49a08d1b9618d50f81b475d4f936c Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Sun, 30 Jul 2006 03:03:04 -0700 Subject: [PATCH 068/143] [PATCH] Process Events: Fix biarch compatibility issue. use __u64 timestamp Events sent by Process Events Connector from a 64-bit kernel are not binary compatible with a 32-bit userspace program because the "timestamp" field (struct timespec) is not arch independent. This affects the fields that follow "timestamp" as they will be be off by 8 bytes. This is a problem for 32-bit userspace programs running with 64-bit kernels on ppc64, s390, x86-64.. any "biarch" system. Matt had submitted a different solution to lkml as an RFC earlier. We have since switched to a solution recommended by Evgeniy Polyakov. This patch fixes the problem by changing the timestamp to be a __u64, which stores the number of nanoseconds. Tested on a x86_64 system with both 32 bit application and 64 bit application and on a i386 system. Signed-off-by: Chandra Seetharaman Signed-off-by: Matt Helsley Cc: Evgeniy Polyakov Cc: Guillaume Thouvenin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/connector/cn_proc.c | 20 +++++++++++++++----- include/linux/cn_proc.h | 3 ++- 2 files changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index 498aa37bca22..3ece69231343 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -51,6 +51,7 @@ void proc_fork_connector(struct task_struct *task) struct cn_msg *msg; struct proc_event *ev; __u8 buffer[CN_PROC_MSG_SIZE]; + struct timespec ts; if (atomic_read(&proc_event_num_listeners) < 1) return; @@ -58,7 +59,8 @@ void proc_fork_connector(struct task_struct *task) msg = (struct cn_msg*)buffer; ev = (struct proc_event*)msg->data; get_seq(&msg->seq, &ev->cpu); - ktime_get_ts(&ev->timestamp); /* get high res monotonic timestamp */ + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + ev->timestamp_ns = timespec_to_ns(&ts); ev->what = PROC_EVENT_FORK; ev->event_data.fork.parent_pid = task->real_parent->pid; ev->event_data.fork.parent_tgid = task->real_parent->tgid; @@ -76,6 +78,7 @@ void proc_exec_connector(struct task_struct *task) { struct cn_msg *msg; struct proc_event *ev; + struct timespec ts; __u8 buffer[CN_PROC_MSG_SIZE]; if (atomic_read(&proc_event_num_listeners) < 1) @@ -84,7 +87,8 @@ void proc_exec_connector(struct task_struct *task) msg = (struct cn_msg*)buffer; ev = (struct proc_event*)msg->data; get_seq(&msg->seq, &ev->cpu); - ktime_get_ts(&ev->timestamp); + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + ev->timestamp_ns = timespec_to_ns(&ts); ev->what = PROC_EVENT_EXEC; ev->event_data.exec.process_pid = task->pid; ev->event_data.exec.process_tgid = task->tgid; @@ -100,6 +104,7 @@ void proc_id_connector(struct task_struct *task, int which_id) struct cn_msg *msg; struct proc_event *ev; __u8 buffer[CN_PROC_MSG_SIZE]; + struct timespec ts; if (atomic_read(&proc_event_num_listeners) < 1) return; @@ -118,7 +123,8 @@ void proc_id_connector(struct task_struct *task, int which_id) } else return; get_seq(&msg->seq, &ev->cpu); - ktime_get_ts(&ev->timestamp); + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + ev->timestamp_ns = timespec_to_ns(&ts); memcpy(&msg->id, &cn_proc_event_id, sizeof(msg->id)); msg->ack = 0; /* not used */ @@ -131,6 +137,7 @@ void proc_exit_connector(struct task_struct *task) struct cn_msg *msg; struct proc_event *ev; __u8 buffer[CN_PROC_MSG_SIZE]; + struct timespec ts; if (atomic_read(&proc_event_num_listeners) < 1) return; @@ -138,7 +145,8 @@ void proc_exit_connector(struct task_struct *task) msg = (struct cn_msg*)buffer; ev = (struct proc_event*)msg->data; get_seq(&msg->seq, &ev->cpu); - ktime_get_ts(&ev->timestamp); + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + ev->timestamp_ns = timespec_to_ns(&ts); ev->what = PROC_EVENT_EXIT; ev->event_data.exit.process_pid = task->pid; ev->event_data.exit.process_tgid = task->tgid; @@ -164,6 +172,7 @@ static void cn_proc_ack(int err, int rcvd_seq, int rcvd_ack) struct cn_msg *msg; struct proc_event *ev; __u8 buffer[CN_PROC_MSG_SIZE]; + struct timespec ts; if (atomic_read(&proc_event_num_listeners) < 1) return; @@ -171,7 +180,8 @@ static void cn_proc_ack(int err, int rcvd_seq, int rcvd_ack) msg = (struct cn_msg*)buffer; ev = (struct proc_event*)msg->data; msg->seq = rcvd_seq; - ktime_get_ts(&ev->timestamp); + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + ev->timestamp_ns = timespec_to_ns(&ts); ev->cpu = -1; ev->what = PROC_EVENT_NONE; ev->event_data.ack.err = err; diff --git a/include/linux/cn_proc.h b/include/linux/cn_proc.h index dbb7769009be..1c86d65bc4b9 100644 --- a/include/linux/cn_proc.h +++ b/include/linux/cn_proc.h @@ -57,7 +57,8 @@ struct proc_event { PROC_EVENT_EXIT = 0x80000000 } what; __u32 cpu; - struct timespec timestamp; + __u64 __attribute__((aligned(8))) timestamp_ns; + /* Number of nano seconds since system boot */ union { /* must be last field of proc_event struct */ struct { __u32 err; From 5e44ef238b7eb607532e89249e7b2523faf77a92 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 30 Jul 2006 03:03:07 -0700 Subject: [PATCH 069/143] [PATCH] drivers/char/pc8736x_gpio.c: unexport a static struct A static struct mustn't be exported. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/pc8736x_gpio.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/char/pc8736x_gpio.c b/drivers/char/pc8736x_gpio.c index 645eb81cb5a9..84e5a68635f1 100644 --- a/drivers/char/pc8736x_gpio.c +++ b/drivers/char/pc8736x_gpio.c @@ -221,7 +221,6 @@ static struct nsc_gpio_ops pc8736x_gpio_ops = { .gpio_change = pc8736x_gpio_change, .gpio_current = pc8736x_gpio_current }; -EXPORT_SYMBOL(pc8736x_gpio_ops); static int pc8736x_gpio_open(struct inode *inode, struct file *file) { From 15a647eba94c3da27ccc666bea72e7cca06b2d19 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Sun, 30 Jul 2006 03:03:08 -0700 Subject: [PATCH 070/143] [PATCH] genirq: {en,dis}able_irq_wake() need refcounting too IRQs need refcounting and a state flag to track whether the the IRQ should be enabled or disabled as a "normal IRQ" source after a series of calls to {en,dis}able_irq(). For shared IRQs, the IRQ must be enabled so long as at least one driver needs it active. Likewise, IRQs need the same support to track whether the IRQ should be enabled or disabled as a "wakeup event" source after a series of calls to {en,dis}able_irq_wake(). For shared IRQs, the IRQ must be enabled as a wakeup source during sleep so long as at least one driver needs it. But right now they _don't have_ that refcounting ... which means sharing a wakeup-capable IRQ can't work correctly in some configurations. This patch adds the refcount and flag mechanisms to set_irq_wake() -- which is what {en,dis}able_irq_wake() call -- and minimal documentation of what the irq wake mechanism does. Drivers relying on the older (broken) "toggle" semantics will trigger a warning; that'll be a handful of drivers on ARM systems. Signed-off-by: David Brownell Acked-by: Ingo Molnar Acked-by: Thomas Gleixner Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irq.h | 3 +++ kernel/irq/manage.c | 28 ++++++++++++++++++++++++++-- 2 files changed, 29 insertions(+), 2 deletions(-) diff --git a/include/linux/irq.h b/include/linux/irq.h index b48eae32dc61..6e59ac05ef2a 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -58,6 +58,7 @@ #define IRQ_NOREQUEST 0x04000000 /* IRQ cannot be requested */ #define IRQ_NOAUTOEN 0x08000000 /* IRQ will not be enabled on request irq */ #define IRQ_DELAYED_DISABLE 0x10000000 /* IRQ disable (masking) happens delayed. */ +#define IRQ_WAKEUP 0x20000000 /* IRQ triggers system wakeup */ struct proc_dir_entry; @@ -124,6 +125,7 @@ struct irq_chip { * @action: the irq action chain * @status: status information * @depth: disable-depth, for nested irq_disable() calls + * @wake_depth: enable depth, for multiple set_irq_wake() callers * @irq_count: stats field to detect stalled irqs * @irqs_unhandled: stats field for spurious unhandled interrupts * @lock: locking for SMP @@ -147,6 +149,7 @@ struct irq_desc { unsigned int status; /* IRQ status */ unsigned int depth; /* nested irq disables */ + unsigned int wake_depth; /* nested wake enables */ unsigned int irq_count; /* For detecting broken IRQs */ unsigned int irqs_unhandled; spinlock_t lock; diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 4e461438e48b..92be519eff26 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -137,16 +137,40 @@ EXPORT_SYMBOL(enable_irq); * @irq: interrupt to control * @on: enable/disable power management wakeup * - * Enable/disable power management wakeup mode + * Enable/disable power management wakeup mode, which is + * disabled by default. Enables and disables must match, + * just as they match for non-wakeup mode support. + * + * Wakeup mode lets this IRQ wake the system from sleep + * states like "suspend to RAM". */ int set_irq_wake(unsigned int irq, unsigned int on) { struct irq_desc *desc = irq_desc + irq; unsigned long flags; int ret = -ENXIO; + int (*set_wake)(unsigned, unsigned) = desc->chip->set_wake; + /* wakeup-capable irqs can be shared between drivers that + * don't need to have the same sleep mode behaviors. + */ spin_lock_irqsave(&desc->lock, flags); - if (desc->chip->set_wake) + if (on) { + if (desc->wake_depth++ == 0) + desc->status |= IRQ_WAKEUP; + else + set_wake = NULL; + } else { + if (desc->wake_depth == 0) { + printk(KERN_WARNING "Unbalanced IRQ %d " + "wake disable\n", irq); + WARN_ON(1); + } else if (--desc->wake_depth == 0) + desc->status &= ~IRQ_WAKEUP; + else + set_wake = NULL; + } + if (set_wake) ret = desc->chip->set_wake(irq, on); spin_unlock_irqrestore(&desc->lock, flags); return ret; From 7d94dddd438bcba97db44f120da39bb001b5249f Mon Sep 17 00:00:00 2001 From: Shailabh Nagar Date: Sun, 30 Jul 2006 03:03:10 -0700 Subject: [PATCH 071/143] [PATCH] make taskstats sending completely independent of delay accounting on/off status Complete the separation of delay accounting and taskstats by ignoring the return value of delay accounting functions that fill in parts of taskstats before it is sent out (either in response to a command or as part of a task exit). Also make delayacct_add_tsk return silently when delay accounting is turned off rather than treat it as an error. Signed-off-by: Shailabh Nagar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/delayacct.h | 4 +--- kernel/taskstats.c | 8 +++----- 2 files changed, 4 insertions(+), 8 deletions(-) diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h index 7e8b6011b8f3..8a284cc6fd5f 100644 --- a/include/linux/delayacct.h +++ b/include/linux/delayacct.h @@ -80,9 +80,7 @@ static inline void delayacct_blkio_end(void) static inline int delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) { - if (likely(!delayacct_on)) - return -EINVAL; - if (!tsk->delays) + if (likely(!delayacct_on) || !tsk->delays) return 0; return __delayacct_add_tsk(d, tsk); } diff --git a/kernel/taskstats.c b/kernel/taskstats.c index f45179ce028e..b4c737a11408 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -177,7 +177,7 @@ static int send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) static int fill_pid(pid_t pid, struct task_struct *pidtsk, struct taskstats *stats) { - int rc; + int rc = 0; struct task_struct *tsk = pidtsk; if (!pidtsk) { @@ -196,12 +196,10 @@ static int fill_pid(pid_t pid, struct task_struct *pidtsk, * Each accounting subsystem adds calls to its functions to * fill in relevant parts of struct taskstsats as follows * - * rc = per-task-foo(stats, tsk); - * if (rc) - * goto err; + * per-task-foo(stats, tsk); */ - rc = delayacct_add_tsk(stats, tsk); + delayacct_add_tsk(stats, tsk); stats->version = TASKSTATS_VERSION; /* Define err: label here if needed */ From d94a041519f3ab1ac023bf917619cd8c4a7d3c01 Mon Sep 17 00:00:00 2001 From: Shailabh Nagar Date: Sun, 30 Jul 2006 03:03:11 -0700 Subject: [PATCH 072/143] [PATCH] taskstats: free skb, avoid returns in send_cpu_listeners Add a missing freeing of skb in the case there are no listeners at all. Also remove the returning of error values by the function as it is unused by the sole caller. Signed-off-by: Shailabh Nagar Signed-off-by: Chandra Seetharaman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index b4c737a11408..e78187657330 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -121,46 +121,45 @@ static int send_reply(struct sk_buff *skb, pid_t pid) /* * Send taskstats data in @skb to listeners registered for @cpu's exit data */ -static int send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) +static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) { struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data); struct listener_list *listeners; struct listener *s, *tmp; struct sk_buff *skb_next, *skb_cur = skb; void *reply = genlmsg_data(genlhdr); - int rc, ret, delcount = 0; + int rc, delcount = 0; rc = genlmsg_end(skb, reply); if (rc < 0) { nlmsg_free(skb); - return rc; + return; } rc = 0; listeners = &per_cpu(listener_array, cpu); down_read(&listeners->sem); - list_for_each_entry_safe(s, tmp, &listeners->list, list) { + list_for_each_entry(s, &listeners->list, list) { skb_next = NULL; if (!list_is_last(&s->list, &listeners->list)) { skb_next = skb_clone(skb_cur, GFP_KERNEL); - if (!skb_next) { - nlmsg_free(skb_cur); - rc = -ENOMEM; + if (!skb_next) break; - } } - ret = genlmsg_unicast(skb_cur, s->pid); - if (ret == -ECONNREFUSED) { + rc = genlmsg_unicast(skb_cur, s->pid); + if (rc == -ECONNREFUSED) { s->valid = 0; delcount++; - rc = ret; } skb_cur = skb_next; } up_read(&listeners->sem); + if (skb_cur) + nlmsg_free(skb_cur); + if (!delcount) - return rc; + return; /* Delete invalidated entries */ down_write(&listeners->sem); @@ -171,7 +170,6 @@ static int send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) } } up_write(&listeners->sem); - return rc; } static int fill_pid(pid_t pid, struct task_struct *pidtsk, From 163ecdff060f2fa9e8f5238882fd0137493556a6 Mon Sep 17 00:00:00 2001 From: Shailabh Nagar Date: Sun, 30 Jul 2006 03:03:11 -0700 Subject: [PATCH 073/143] [PATCH] delay accounting: temporarily enable by default Enable delay accounting by default so that feature gets coverage testing without requiring special measures. Earlier, it was off by default and had to be enabled via a boot time param. This patch reverses the default behaviour to improve coverage testing. It can be removed late in the kernel development cycle if its believed users shouldn't have to incur any cost if they don't want delay accounting. Or it can be retained forever if the utility of the stats is deemed common enough to warrant keeping the feature on. Signed-off-by: Shailabh Nagar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/accounting/delay-accounting.txt | 10 ++++++---- Documentation/kernel-parameters.txt | 4 ++-- include/linux/delayacct.h | 4 ++-- kernel/delayacct.c | 8 ++++---- 4 files changed, 14 insertions(+), 12 deletions(-) diff --git a/Documentation/accounting/delay-accounting.txt b/Documentation/accounting/delay-accounting.txt index be215e58423b..1443cd71d263 100644 --- a/Documentation/accounting/delay-accounting.txt +++ b/Documentation/accounting/delay-accounting.txt @@ -64,11 +64,13 @@ Compile the kernel with CONFIG_TASK_DELAY_ACCT=y CONFIG_TASKSTATS=y -Enable the accounting at boot time by adding -the following to the kernel boot options - delayacct +Delay accounting is enabled by default at boot up. +To disable, add + nodelayacct +to the kernel boot options. The rest of the instructions +below assume this has not been done. -and after the system has booted up, use a utility +After the system has booted up, use a utility similar to getdelays.c to access the delays seen by a given task or a task group (tgid). The utility also allows a given command to be diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e11f7728ec6f..b50595a0550f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -448,8 +448,6 @@ running once the system is up. Format: [,] See also Documentation/networking/decnet.txt. - delayacct [KNL] Enable per-task delay accounting - dhash_entries= [KNL] Set number of hash buckets for dentry cache. @@ -1031,6 +1029,8 @@ running once the system is up. nocache [ARM] + nodelayacct [KNL] Disable per-task delay accounting + nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects. noexec [IA-64] diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h index 8a284cc6fd5f..11487b6e7127 100644 --- a/include/linux/delayacct.h +++ b/include/linux/delayacct.h @@ -55,7 +55,7 @@ static inline void delayacct_tsk_init(struct task_struct *tsk) { /* reinitialize in case parent's non-null pointer was dup'ed*/ tsk->delays = NULL; - if (unlikely(delayacct_on)) + if (delayacct_on) __delayacct_tsk_init(tsk); } @@ -80,7 +80,7 @@ static inline void delayacct_blkio_end(void) static inline int delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) { - if (likely(!delayacct_on) || !tsk->delays) + if (!delayacct_on || !tsk->delays) return 0; return __delayacct_add_tsk(d, tsk); } diff --git a/kernel/delayacct.c b/kernel/delayacct.c index f05392d64267..57ca3730205d 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -19,15 +19,15 @@ #include #include -int delayacct_on __read_mostly; /* Delay accounting turned on/off */ +int delayacct_on __read_mostly = 1; /* Delay accounting turned on/off */ kmem_cache_t *delayacct_cache; -static int __init delayacct_setup_enable(char *str) +static int __init delayacct_setup_disable(char *str) { - delayacct_on = 1; + delayacct_on = 0; return 1; } -__setup("delayacct", delayacct_setup_enable); +__setup("nodelayacct", delayacct_setup_disable); void delayacct_init(void) { From 31925c8857ba17c11129b766a980ff7c87780301 Mon Sep 17 00:00:00 2001 From: Peter Korsgaard Date: Sun, 30 Jul 2006 03:03:12 -0700 Subject: [PATCH 074/143] [PATCH] Fix ppc32 zImage inflate The recent zlib update (commit 4f3865fb57a04db7cca068fed1c15badc064a302) broke ppc32 zImage decompression as it tries to decompress to address zero and the updated zlib_inflate checks that strm->next_out isn't a null pointer. This little patch fixes it. [rpurdie@rpsys.net: add comment] Signed-off-by: Peter Korsgaard Acked-by: Tom Rini Signed-off-by: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/zlib_inflate/inflate.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/lib/zlib_inflate/inflate.c b/lib/zlib_inflate/inflate.c index 7f922dccf1a5..fceb97c3aff7 100644 --- a/lib/zlib_inflate/inflate.c +++ b/lib/zlib_inflate/inflate.c @@ -347,7 +347,10 @@ int zlib_inflate(z_streamp strm, int flush) static const unsigned short order[19] = /* permutation of code lengths */ {16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; - if (strm == NULL || strm->state == NULL || strm->next_out == NULL || + /* Do not check for strm->next_out == NULL here as ppc zImage + inflates to strm->next_out = 0 */ + + if (strm == NULL || strm->state == NULL || (strm->next_in == NULL && strm->avail_in != 0)) return Z_STREAM_ERROR; From a5ca63cb60892c1703998daf80d2d1b04cb70103 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 30 Jul 2006 03:03:16 -0700 Subject: [PATCH 075/143] [PATCH] mce section fix mce_disabled cannot be __initdata - we access it during APM resume. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/mcheck/mce.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/i386/kernel/cpu/mcheck/mce.h b/arch/i386/kernel/cpu/mcheck/mce.h index dc2416dfef15..84fd4cf7d0fb 100644 --- a/arch/i386/kernel/cpu/mcheck/mce.h +++ b/arch/i386/kernel/cpu/mcheck/mce.h @@ -9,6 +9,6 @@ void winchip_mcheck_init(struct cpuinfo_x86 *c); /* Call the installed machine check handler for this CPU setup. */ extern fastcall void (*machine_check_vector)(struct pt_regs *, long error_code); -extern int mce_disabled __initdata; +extern int mce_disabled; extern int nr_mce_banks; From d1bbf14f37261c2c0dba71404602e1ddcec069d2 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sun, 30 Jul 2006 03:03:16 -0700 Subject: [PATCH 076/143] [PATCH] knfsd: Fix stale file handle problem with subtree_checking. A recent commit (7fc90ec93a5eb71f4b08403baf5ba7176b3ec6b1) moved the call to nfsd_setuser out of the 'find a dentry for a filehandle' branch of fh_verify so that it would always be called. This had the unfortunately side-effect of moving *after* the call to decode_fh, so the prober fsuid was not set when nfsd_acceptable was called, the 'permission' check did the wrong thing. This patch moves the nfsd_setuser call back where it was, and add as call in the other branch of the if. Cc: "J. Bruce Fields" Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsfh.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index ecc439d2565f..501d83884530 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -187,6 +187,11 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) goto out; } + /* Set user creds for this exportpoint */ + error = nfserrno(nfsd_setuser(rqstp, exp)); + if (error) + goto out; + /* * Look up the dentry using the NFS file handle. */ @@ -241,16 +246,17 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) dprintk("nfsd: fh_verify - just checking\n"); dentry = fhp->fh_dentry; exp = fhp->fh_export; + /* Set user creds for this exportpoint; necessary even + * in the "just checking" case because this may be a + * filehandle that was created by fh_compose, and that + * is about to be used in another nfsv4 compound + * operation */ + error = nfserrno(nfsd_setuser(rqstp, exp)); + if (error) + goto out; } cache_get(&exp->h); - /* Set user creds for this exportpoint; necessary even in the "just - * checking" case because this may be a filehandle that was created by - * fh_compose, and that is about to be used in another nfsv4 compound - * operation */ - error = nfserrno(nfsd_setuser(rqstp, exp)); - if (error) - goto out; error = nfsd_mode_check(rqstp, dentry->d_inode->i_mode, type); if (error) From ddccef3b5ec906ff181171e8ffad4fcb996792fd Mon Sep 17 00:00:00 2001 From: Darrel Goeddel Date: Sun, 30 Jul 2006 03:03:17 -0700 Subject: [PATCH 077/143] [PATCH] selinux: fix memory leak This patch fixes a memory leak when a policydb structure is destroyed. Signed-off-by: Darrel Goeddel Signed-off-by: Stephen Smalley Acked-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- security/selinux/ss/policydb.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c index 0111990ba837..f03960e697ce 100644 --- a/security/selinux/ss/policydb.c +++ b/security/selinux/ss/policydb.c @@ -644,10 +644,18 @@ void policydb_destroy(struct policydb *p) kfree(lra); for (rt = p->range_tr; rt; rt = rt -> next) { - kfree(lrt); + if (lrt) { + ebitmap_destroy(&lrt->range.level[0].cat); + ebitmap_destroy(&lrt->range.level[1].cat); + kfree(lrt); + } lrt = rt; } - kfree(lrt); + if (lrt) { + ebitmap_destroy(&lrt->range.level[0].cat); + ebitmap_destroy(&lrt->range.level[1].cat); + kfree(lrt); + } if (p->type_attr_map) { for (i = 0; i < p->p_types.nprim; i++) From 851f8a6906b71f7a19043d4d722dd4ffab7aeafc Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Sun, 30 Jul 2006 03:03:18 -0700 Subject: [PATCH 078/143] [PATCH] selinux: fix bug in security_compute_sid Initializes newcontext sooner to allow for its destruction in all cases. Signed-off-by: Venkat Yekkirala Signed-off-by: Stephen Smalley Acked-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- security/selinux/ss/services.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index d2e80e62ff0c..85e429884393 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -833,6 +833,8 @@ static int security_compute_sid(u32 ssid, goto out; } + context_init(&newcontext); + POLICY_RDLOCK; scontext = sidtab_search(&sidtab, ssid); @@ -850,8 +852,6 @@ static int security_compute_sid(u32 ssid, goto out_unlock; } - context_init(&newcontext); - /* Set the user identity. */ switch (specified) { case AVTAB_TRANSITION: From c35a7261eaf0e57924e2c56c6d72dc44ee9f3634 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 30 Jul 2006 03:03:19 -0700 Subject: [PATCH 079/143] [PATCH] synchronize_tsc() fixes - Move the tsc synchronisation variables into a struct, mark it __initdata - local `realdelta' wants to be 64-bit - Print the skew for negative skews, as well as for positive ones - remove dead code Cc: john stultz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/smpboot.c | 62 ++++++++++++++++++++------------------ 1 file changed, 33 insertions(+), 29 deletions(-) diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 6f5fea05f1d7..f948419c888a 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -212,14 +212,20 @@ static void __devinit smp_store_cpu_info(int id) * then we print a warning if not, and always resync. */ -static atomic_t tsc_start_flag = ATOMIC_INIT(0); -static atomic_t tsc_count_start = ATOMIC_INIT(0); -static atomic_t tsc_count_stop = ATOMIC_INIT(0); -static unsigned long long tsc_values[NR_CPUS]; +static struct { + atomic_t start_flag; + atomic_t count_start; + atomic_t count_stop; + unsigned long long values[NR_CPUS]; +} tsc __initdata = { + .start_flag = ATOMIC_INIT(0), + .count_start = ATOMIC_INIT(0), + .count_stop = ATOMIC_INIT(0), +}; #define NR_LOOPS 5 -static void __init synchronize_tsc_bp (void) +static void __init synchronize_tsc_bp(void) { int i; unsigned long long t0; @@ -233,7 +239,7 @@ static void __init synchronize_tsc_bp (void) /* convert from kcyc/sec to cyc/usec */ one_usec = cpu_khz / 1000; - atomic_set(&tsc_start_flag, 1); + atomic_set(&tsc.start_flag, 1); wmb(); /* @@ -250,16 +256,16 @@ static void __init synchronize_tsc_bp (void) /* * all APs synchronize but they loop on '== num_cpus' */ - while (atomic_read(&tsc_count_start) != num_booting_cpus()-1) + while (atomic_read(&tsc.count_start) != num_booting_cpus()-1) cpu_relax(); - atomic_set(&tsc_count_stop, 0); + atomic_set(&tsc.count_stop, 0); wmb(); /* * this lets the APs save their current TSC: */ - atomic_inc(&tsc_count_start); + atomic_inc(&tsc.count_start); - rdtscll(tsc_values[smp_processor_id()]); + rdtscll(tsc.values[smp_processor_id()]); /* * We clear the TSC in the last loop: */ @@ -269,56 +275,54 @@ static void __init synchronize_tsc_bp (void) /* * Wait for all APs to leave the synchronization point: */ - while (atomic_read(&tsc_count_stop) != num_booting_cpus()-1) + while (atomic_read(&tsc.count_stop) != num_booting_cpus()-1) cpu_relax(); - atomic_set(&tsc_count_start, 0); + atomic_set(&tsc.count_start, 0); wmb(); - atomic_inc(&tsc_count_stop); + atomic_inc(&tsc.count_stop); } sum = 0; for (i = 0; i < NR_CPUS; i++) { if (cpu_isset(i, cpu_callout_map)) { - t0 = tsc_values[i]; + t0 = tsc.values[i]; sum += t0; } } avg = sum; do_div(avg, num_booting_cpus()); - sum = 0; for (i = 0; i < NR_CPUS; i++) { if (!cpu_isset(i, cpu_callout_map)) continue; - delta = tsc_values[i] - avg; + delta = tsc.values[i] - avg; if (delta < 0) delta = -delta; /* * We report bigger than 2 microseconds clock differences. */ if (delta > 2*one_usec) { - long realdelta; + long long realdelta; + if (!buggy) { buggy = 1; printk("\n"); } realdelta = delta; do_div(realdelta, one_usec); - if (tsc_values[i] < avg) + if (tsc.values[i] < avg) realdelta = -realdelta; - if (realdelta > 0) - printk(KERN_INFO "CPU#%d had %ld usecs TSC " + if (realdelta) + printk(KERN_INFO "CPU#%d had %Ld usecs TSC " "skew, fixed it up.\n", i, realdelta); } - - sum += delta; } if (!buggy) printk("passed.\n"); } -static void __init synchronize_tsc_ap (void) +static void __init synchronize_tsc_ap(void) { int i; @@ -327,20 +331,20 @@ static void __init synchronize_tsc_ap (void) * this gets called, so we first wait for the BP to * finish SMP initialization: */ - while (!atomic_read(&tsc_start_flag)) + while (!atomic_read(&tsc.start_flag)) cpu_relax(); for (i = 0; i < NR_LOOPS; i++) { - atomic_inc(&tsc_count_start); - while (atomic_read(&tsc_count_start) != num_booting_cpus()) + atomic_inc(&tsc.count_start); + while (atomic_read(&tsc.count_start) != num_booting_cpus()) cpu_relax(); - rdtscll(tsc_values[smp_processor_id()]); + rdtscll(tsc.values[smp_processor_id()]); if (i == NR_LOOPS-1) write_tsc(0, 0); - atomic_inc(&tsc_count_stop); - while (atomic_read(&tsc_count_stop) != num_booting_cpus()) + atomic_inc(&tsc.count_stop); + while (atomic_read(&tsc.count_stop) != num_booting_cpus()) cpu_relax(); } } From 2a8a3d5b65e86ec1dfef7d268c64a909eab94af7 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 30 Jul 2006 03:03:20 -0700 Subject: [PATCH 080/143] [PATCH] machine_kexec.c: Fix the description of segment handling One of my original comments in machine_kexec was unclear and this should fix it. Signed-off-by: Eric W. Biederman Cc: Andi Kleen Acked-by: Horms Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/machine_kexec.c | 13 +++++-------- arch/x86_64/kernel/machine_kexec.c | 13 +++++-------- 2 files changed, 10 insertions(+), 16 deletions(-) diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c index 511abe52a94e..6b1ae6ba76f0 100644 --- a/arch/i386/kernel/machine_kexec.c +++ b/arch/i386/kernel/machine_kexec.c @@ -189,14 +189,11 @@ NORET_TYPE void machine_kexec(struct kimage *image) memcpy((void *)reboot_code_buffer, relocate_new_kernel, relocate_new_kernel_size); - /* The segment registers are funny things, they are - * automatically loaded from a table, in memory wherever you - * set them to a specific selector, but this table is never - * accessed again you set the segment to a different selector. - * - * The more common model is are caches where the behide - * the scenes work is done, but is also dropped at arbitrary - * times. + /* The segment registers are funny things, they have both a + * visible and an invisible part. Whenever the visible part is + * set to a specific selector, the invisible part is loaded + * with from a table in memory. At no other time is the + * descriptor table in memory accessed. * * I take advantage of this here by force loading the * segments, before I zap the gdt with an invalid value. diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c index 83fb24a02821..106076b370fc 100644 --- a/arch/x86_64/kernel/machine_kexec.c +++ b/arch/x86_64/kernel/machine_kexec.c @@ -207,14 +207,11 @@ NORET_TYPE void machine_kexec(struct kimage *image) __flush_tlb(); - /* The segment registers are funny things, they are - * automatically loaded from a table, in memory wherever you - * set them to a specific selector, but this table is never - * accessed again unless you set the segment to a different selector. - * - * The more common model are caches where the behide - * the scenes work is done, but is also dropped at arbitrary - * times. + /* The segment registers are funny things, they have both a + * visible and an invisible part. Whenever the visible part is + * set to a specific selector, the invisible part is loaded + * with from a table in memory. At no other time is the + * descriptor table in memory accessed. * * I take advantage of this here by force loading the * segments, before I zap the gdt with an invalid value. From c39df470e04822965d945473d831786ab79dd4cd Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Sun, 30 Jul 2006 03:03:21 -0700 Subject: [PATCH 081/143] [PATCH] kprobe-booster: disable in preemptible kernel The kprobe-booster's safety check against preemption does not work well now, because the preemption count has been modified by read_rcu_lock() in atomic_notifier_call_chain() before we check it. So, I'd like to prevent boosting kprobe temporarily if the kernel is preemptable. Now we are searching for the good solution. Signed-off-by: Masami Hiramatsu Cc: Ananth N Mavinakayanahalli Cc: Prasanna S Panchamukhi Cc: Anil S Keshavamurthy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index de2e16e561c0..afe6505ca0b3 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -256,11 +256,6 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) int ret = 0; kprobe_opcode_t *addr; struct kprobe_ctlblk *kcb; -#ifdef CONFIG_PREEMPT - unsigned pre_preempt_count = preempt_count(); -#else - unsigned pre_preempt_count = 1; -#endif addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t)); @@ -338,13 +333,15 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) return 1; ss_probe: - if (pre_preempt_count && p->ainsn.boostable == 1 && !p->post_handler){ +#ifndef CONFIG_PREEMPT + if (p->ainsn.boostable == 1 && !p->post_handler){ /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); regs->eip = (unsigned long)p->ainsn.insn; preempt_enable_no_resched(); return 1; } +#endif prepare_singlestep(p, regs); kcb->kprobe_status = KPROBE_HIT_SS; return 1; From c8e5429e49dcf5243a2dc293cb11f53d65d5fca9 Mon Sep 17 00:00:00 2001 From: Arthur Othieno Date: Sun, 30 Jul 2006 03:03:22 -0700 Subject: [PATCH 082/143] [PATCH] i386: fix CONFIG_EFI help It is described as being experimental, but doesn't actually depend on EXPERIMENTAL. Change the text. Signed-off-by: Arthur Othieno Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index daa75ce4b777..f71fb4a029cb 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -672,7 +672,7 @@ config MTRR See for more information. config EFI - bool "Boot from EFI support (EXPERIMENTAL)" + bool "Boot from EFI support" depends on ACPI default n ---help--- From 25c8716cb08dea386c7d6220b82eba732ccbf976 Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Sun, 30 Jul 2006 03:03:23 -0700 Subject: [PATCH 083/143] [PATCH] arch/alpha: Use ARRAY_SIZE macro Use ARRAY_SIZE macro instead of sizeof(x)/sizeof(x[0]) and remove a duplicate of the macro. Also remove some trailing whitespaces and needless braces. Signed-off-by: Tobias Klauser Cc: Richard Henderson Cc: Ivan Kokshaysky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/err_ev7.c | 8 ++---- arch/alpha/kernel/osf_sys.c | 4 +-- arch/alpha/kernel/setup.c | 50 ++++++++++++++++----------------- arch/alpha/kernel/sys_ruffian.c | 8 +++--- arch/alpha/kernel/time.c | 2 +- 5 files changed, 34 insertions(+), 38 deletions(-) diff --git a/arch/alpha/kernel/err_ev7.c b/arch/alpha/kernel/err_ev7.c index bf52ba691957..fed6b3d1b803 100644 --- a/arch/alpha/kernel/err_ev7.c +++ b/arch/alpha/kernel/err_ev7.c @@ -274,16 +274,14 @@ ev7_process_pal_subpacket(struct el_subpacket *header) struct el_subpacket_handler ev7_pal_subpacket_handler = SUBPACKET_HANDLER_INIT(EL_CLASS__PAL, ev7_process_pal_subpacket); -void +void ev7_register_error_handlers(void) { int i; - for(i = 0; - i= sizeof(sysinfo_table)/sizeof(char *)) { + if (offset >= ARRAY_SIZE(sysinfo_table)) { /* Digital UNIX has a few unpublished interfaces here */ printk("sysinfo(%d)", command); goto out; } - + down_read(&uts_sem); res = sysinfo_table[offset]; len = strlen(res)+1; diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index 2cb9c4380113..fd4a8fa0c93d 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -114,8 +114,6 @@ struct alpha_machine_vector alpha_mv; int alpha_using_srm; #endif -#define N(a) (sizeof(a)/sizeof(a[0])) - static struct alpha_machine_vector *get_sysvec(unsigned long, unsigned long, unsigned long); static struct alpha_machine_vector *get_sysvec_byname(const char *); @@ -240,7 +238,7 @@ reserve_std_resources(void) standard_io_resources[0].start = RTC_PORT(0); standard_io_resources[0].end = RTC_PORT(0) + 0x10; - for (i = 0; i < N(standard_io_resources); ++i) + for (i = 0; i < ARRAY_SIZE(standard_io_resources); ++i) request_resource(io, standard_io_resources+i); } @@ -918,13 +916,13 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu) /* Search the system tables first... */ vec = NULL; - if (type < N(systype_vecs)) { + if (type < ARRAY_SIZE(systype_vecs)) { vec = systype_vecs[type]; } else if ((type > ST_API_BIAS) && - (type - ST_API_BIAS) < N(api_vecs)) { + (type - ST_API_BIAS) < ARRAY_SIZE(api_vecs)) { vec = api_vecs[type - ST_API_BIAS]; } else if ((type > ST_UNOFFICIAL_BIAS) && - (type - ST_UNOFFICIAL_BIAS) < N(unofficial_vecs)) { + (type - ST_UNOFFICIAL_BIAS) < ARRAY_SIZE(unofficial_vecs)) { vec = unofficial_vecs[type - ST_UNOFFICIAL_BIAS]; } @@ -938,11 +936,11 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu) switch (type) { case ST_DEC_ALCOR: - if (member < N(alcor_indices)) + if (member < ARRAY_SIZE(alcor_indices)) vec = alcor_vecs[alcor_indices[member]]; break; case ST_DEC_EB164: - if (member < N(eb164_indices)) + if (member < ARRAY_SIZE(eb164_indices)) vec = eb164_vecs[eb164_indices[member]]; /* PC164 may show as EB164 variation with EV56 CPU, but, since no true EB164 had anything but EV5... */ @@ -950,24 +948,24 @@ get_sysvec(unsigned long type, unsigned long variation, unsigned long cpu) vec = &pc164_mv; break; case ST_DEC_EB64P: - if (member < N(eb64p_indices)) + if (member < ARRAY_SIZE(eb64p_indices)) vec = eb64p_vecs[eb64p_indices[member]]; break; case ST_DEC_EB66: - if (member < N(eb66_indices)) + if (member < ARRAY_SIZE(eb66_indices)) vec = eb66_vecs[eb66_indices[member]]; break; case ST_DEC_MARVEL: - if (member < N(marvel_indices)) + if (member < ARRAY_SIZE(marvel_indices)) vec = marvel_vecs[marvel_indices[member]]; break; case ST_DEC_TITAN: vec = titan_vecs[0]; /* default */ - if (member < N(titan_indices)) + if (member < ARRAY_SIZE(titan_indices)) vec = titan_vecs[titan_indices[member]]; break; case ST_DEC_TSUNAMI: - if (member < N(tsunami_indices)) + if (member < ARRAY_SIZE(tsunami_indices)) vec = tsunami_vecs[tsunami_indices[member]]; break; case ST_DEC_1000: @@ -1039,7 +1037,7 @@ get_sysvec_byname(const char *name) size_t i; - for (i = 0; i < N(all_vecs); ++i) { + for (i = 0; i < ARRAY_SIZE(all_vecs); ++i) { struct alpha_machine_vector *mv = all_vecs[i]; if (strcasecmp(mv->vector_name, name) == 0) return mv; @@ -1055,13 +1053,13 @@ get_sysnames(unsigned long type, unsigned long variation, unsigned long cpu, /* If not in the tables, make it UNKNOWN, else set type name to family */ - if (type < N(systype_names)) { + if (type < ARRAY_SIZE(systype_names)) { *type_name = systype_names[type]; } else if ((type > ST_API_BIAS) && - (type - ST_API_BIAS) < N(api_names)) { + (type - ST_API_BIAS) < ARRAY_SIZE(api_names)) { *type_name = api_names[type - ST_API_BIAS]; } else if ((type > ST_UNOFFICIAL_BIAS) && - (type - ST_UNOFFICIAL_BIAS) < N(unofficial_names)) { + (type - ST_UNOFFICIAL_BIAS) < ARRAY_SIZE(unofficial_names)) { *type_name = unofficial_names[type - ST_UNOFFICIAL_BIAS]; } else { *type_name = sys_unknown; @@ -1083,7 +1081,7 @@ get_sysnames(unsigned long type, unsigned long variation, unsigned long cpu, default: /* default to variation "0" for now */ break; case ST_DEC_EB164: - if (member < N(eb164_indices)) + if (member < ARRAY_SIZE(eb164_indices)) *variation_name = eb164_names[eb164_indices[member]]; /* PC164 may show as EB164 variation, but with EV56 CPU, so, since no true EB164 had anything but EV5... */ @@ -1091,32 +1089,32 @@ get_sysnames(unsigned long type, unsigned long variation, unsigned long cpu, *variation_name = eb164_names[1]; /* make it PC164 */ break; case ST_DEC_ALCOR: - if (member < N(alcor_indices)) + if (member < ARRAY_SIZE(alcor_indices)) *variation_name = alcor_names[alcor_indices[member]]; break; case ST_DEC_EB64P: - if (member < N(eb64p_indices)) + if (member < ARRAY_SIZE(eb64p_indices)) *variation_name = eb64p_names[eb64p_indices[member]]; break; case ST_DEC_EB66: - if (member < N(eb66_indices)) + if (member < ARRAY_SIZE(eb66_indices)) *variation_name = eb66_names[eb66_indices[member]]; break; case ST_DEC_MARVEL: - if (member < N(marvel_indices)) + if (member < ARRAY_SIZE(marvel_indices)) *variation_name = marvel_names[marvel_indices[member]]; break; case ST_DEC_RAWHIDE: - if (member < N(rawhide_indices)) + if (member < ARRAY_SIZE(rawhide_indices)) *variation_name = rawhide_names[rawhide_indices[member]]; break; case ST_DEC_TITAN: *variation_name = titan_names[0]; /* default */ - if (member < N(titan_indices)) + if (member < ARRAY_SIZE(titan_indices)) *variation_name = titan_names[titan_indices[member]]; break; case ST_DEC_TSUNAMI: - if (member < N(tsunami_indices)) + if (member < ARRAY_SIZE(tsunami_indices)) *variation_name = tsunami_names[tsunami_indices[member]]; break; } @@ -1211,7 +1209,7 @@ show_cpuinfo(struct seq_file *f, void *slot) cpu_index = (unsigned) (cpu->type - 1); cpu_name = "Unknown"; - if (cpu_index < N(cpu_names)) + if (cpu_index < ARRAY_SIZE(cpu_names)) cpu_name = cpu_names[cpu_index]; get_sysnames(hwrpb->sys_type, hwrpb->sys_variation, diff --git a/arch/alpha/kernel/sys_ruffian.c b/arch/alpha/kernel/sys_ruffian.c index 78c30decf3ff..5b99cf3cd69c 100644 --- a/arch/alpha/kernel/sys_ruffian.c +++ b/arch/alpha/kernel/sys_ruffian.c @@ -182,16 +182,16 @@ static unsigned long __init ruffian_get_bank_size(unsigned long offset) { unsigned long bank_addr, bank, ret = 0; - + /* Valid offsets are: 0x800, 0x840 and 0x880 since Ruffian only uses three banks. */ bank_addr = (unsigned long)PYXIS_MCR + offset; bank = *(vulp)bank_addr; - + /* Check BANK_ENABLE */ if (bank & 0x01) { static unsigned long size[] __initdata = { - 0x40000000UL, /* 0x00, 1G */ + 0x40000000UL, /* 0x00, 1G */ 0x20000000UL, /* 0x02, 512M */ 0x10000000UL, /* 0x04, 256M */ 0x08000000UL, /* 0x06, 128M */ @@ -203,7 +203,7 @@ ruffian_get_bank_size(unsigned long offset) }; bank = (bank & 0x1e) >> 1; - if (bank < sizeof(size)/sizeof(*size)) + if (bank < ARRAY_SIZE(size)) ret = size[bank]; } diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c index 50eccde2dcd8..b191cc759737 100644 --- a/arch/alpha/kernel/time.c +++ b/arch/alpha/kernel/time.c @@ -233,7 +233,7 @@ validate_cc_value(unsigned long cc) index = cpu->type & 0xffffffff; /* If index out of bounds, no way to validate. */ - if (index >= sizeof(cpu_hz)/sizeof(cpu_hz[0])) + if (index >= ARRAY_SIZE(cpu_hz)) return cc; /* If index contains no data, no way to validate. */ From a9ad965ea9a6d719daf333847a2ceb0e363994bd Mon Sep 17 00:00:00 2001 From: "bibo, mao" Date: Sun, 30 Jul 2006 03:03:26 -0700 Subject: [PATCH 084/143] [PATCH] IA64: kprobe invalidate icache of jump buffer Kprobe inserts breakpoint instruction in probepoint and then jumps to instruction slot when breakpoint is hit, the instruction slot icache must be consistent with dcache. Here is the patch which invalidates instruction slot icache area. Without this patch, in some machines there will be fault when executing instruction slot where icache content is inconsistent with dcache. Signed-off-by: bibo,mao Acked-by: "Luck, Tony" Acked-by: Keshavamurthy Anil S Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/kprobes.c | 9 +++++++++ include/asm-i386/kprobes.h | 1 + include/asm-ia64/kprobes.h | 1 + include/asm-powerpc/kprobes.h | 1 + include/asm-sparc64/kprobes.h | 1 + include/asm-x86_64/kprobes.h | 1 + kernel/kprobes.c | 1 + 7 files changed, 15 insertions(+) diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 00d9c83b8020..781960f80b6f 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -448,11 +448,20 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) return 0; } +void __kprobes flush_insn_slot(struct kprobe *p) +{ + unsigned long arm_addr; + + arm_addr = ((unsigned long)&p->opcode.bundle) & ~0xFULL; + flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t)); +} + void __kprobes arch_arm_kprobe(struct kprobe *p) { unsigned long addr = (unsigned long)p->addr; unsigned long arm_addr = addr & ~0xFULL; + flush_insn_slot(p); memcpy((char *)arm_addr, &p->ainsn.insn.bundle, sizeof(bundle_t)); flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t)); } diff --git a/include/asm-i386/kprobes.h b/include/asm-i386/kprobes.h index 0730a20f6db8..8774d06689da 100644 --- a/include/asm-i386/kprobes.h +++ b/include/asm-i386/kprobes.h @@ -45,6 +45,7 @@ typedef u8 kprobe_opcode_t; #define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 0 +#define flush_insn_slot(p) do { } while (0) void arch_remove_kprobe(struct kprobe *p); void kretprobe_trampoline(void); diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h index 2418a787c405..938904910115 100644 --- a/include/asm-ia64/kprobes.h +++ b/include/asm-ia64/kprobes.h @@ -125,5 +125,6 @@ static inline void jprobe_return(void) } extern void invalidate_stacked_regs(void); extern void flush_register_stack(void); +extern void flush_insn_slot(struct kprobe *p); #endif /* _ASM_KPROBES_H */ diff --git a/include/asm-powerpc/kprobes.h b/include/asm-powerpc/kprobes.h index 2d0af52c823d..34e1f89a5fa0 100644 --- a/include/asm-powerpc/kprobes.h +++ b/include/asm-powerpc/kprobes.h @@ -51,6 +51,7 @@ typedef unsigned int kprobe_opcode_t; #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 1 +#define flush_insn_slot(p) do { } while (0) void kretprobe_trampoline(void); extern void arch_remove_kprobe(struct kprobe *p); diff --git a/include/asm-sparc64/kprobes.h b/include/asm-sparc64/kprobes.h index 15065af566c2..c9f5c34d318c 100644 --- a/include/asm-sparc64/kprobes.h +++ b/include/asm-sparc64/kprobes.h @@ -13,6 +13,7 @@ typedef u32 kprobe_opcode_t; #define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry #define arch_remove_kprobe(p) do {} while (0) #define ARCH_INACTIVE_KPROBE_COUNT 0 +#define flush_insn_slot(p) do { } while (0) /* Architecture specific copy of original instruction*/ struct arch_specific_insn { diff --git a/include/asm-x86_64/kprobes.h b/include/asm-x86_64/kprobes.h index d36febd9bb18..cf5317898fb0 100644 --- a/include/asm-x86_64/kprobes.h +++ b/include/asm-x86_64/kprobes.h @@ -47,6 +47,7 @@ typedef u8 kprobe_opcode_t; void kretprobe_trampoline(void); extern void arch_remove_kprobe(struct kprobe *p); +#define flush_insn_slot(p) do { } while (0) /* Architecture specific copy of original instruction*/ struct arch_specific_insn { diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 64aab081153b..3f57dfdc8f92 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -393,6 +393,7 @@ static int __kprobes add_new_kprobe(struct kprobe *old_p, struct kprobe *p) static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p) { copy_kprobe(p, ap); + flush_insn_slot(ap); ap->addr = p->addr; ap->pre_handler = aggr_pre_handler; ap->fault_handler = aggr_fault_handler; From 66f6b24be91fcc31259e8219286ae462ae64ea55 Mon Sep 17 00:00:00 2001 From: Miles Bader Date: Sun, 30 Jul 2006 03:03:27 -0700 Subject: [PATCH 085/143] [PATCH] v850: Remove symbol exports which duplicate global ones Signed-off-by: Miles Bader Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/v850/kernel/v850_ksyms.c | 16 ---------------- 1 file changed, 16 deletions(-) diff --git a/arch/v850/kernel/v850_ksyms.c b/arch/v850/kernel/v850_ksyms.c index c03ad6ed61cc..67bc48e57c60 100644 --- a/arch/v850/kernel/v850_ksyms.c +++ b/arch/v850/kernel/v850_ksyms.c @@ -21,9 +21,6 @@ EXPORT_SYMBOL (trap_table); /* platform dependent support */ EXPORT_SYMBOL (kernel_thread); -EXPORT_SYMBOL (enable_irq); -EXPORT_SYMBOL (disable_irq); -EXPORT_SYMBOL (disable_irq_nosync); EXPORT_SYMBOL (__bug); /* Networking helper routines. */ @@ -33,22 +30,9 @@ EXPORT_SYMBOL (ip_compute_csum); EXPORT_SYMBOL (ip_fast_csum); /* string / mem functions */ -EXPORT_SYMBOL (strcpy); -EXPORT_SYMBOL (strncpy); -EXPORT_SYMBOL (strcat); -EXPORT_SYMBOL (strncat); -EXPORT_SYMBOL (strcmp); -EXPORT_SYMBOL (strncmp); -EXPORT_SYMBOL (strchr); -EXPORT_SYMBOL (strlen); -EXPORT_SYMBOL (strnlen); -EXPORT_SYMBOL (strrchr); -EXPORT_SYMBOL (strstr); EXPORT_SYMBOL (memset); EXPORT_SYMBOL (memcpy); EXPORT_SYMBOL (memmove); -EXPORT_SYMBOL (memcmp); -EXPORT_SYMBOL (memscan); /* semaphores */ EXPORT_SYMBOL (__down); From a268cefebceeb2046dfdfa301f041c2468536852 Mon Sep 17 00:00:00 2001 From: Miles Bader Date: Sun, 30 Jul 2006 03:03:28 -0700 Subject: [PATCH 086/143] [PATCH] v850: call init_page_count() instead of set_page_count() Signed-off-by: Miles Bader Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/v850/kernel/setup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/v850/kernel/setup.c b/arch/v850/kernel/setup.c index 62bdb8d29fc0..1bf672a25692 100644 --- a/arch/v850/kernel/setup.c +++ b/arch/v850/kernel/setup.c @@ -1,8 +1,8 @@ /* * arch/v850/kernel/setup.c -- Arch-dependent initialization functions * - * Copyright (C) 2001,02,03,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,05 Miles Bader + * Copyright (C) 2001,02,03,05,06 NEC Electronics Corporation + * Copyright (C) 2001,02,03,05,06 Miles Bader * * This file is subject to the terms and conditions of the GNU General * Public License. See the file COPYING in the main directory of this @@ -190,7 +190,7 @@ void free_initmem (void) for (addr = start; addr < end; addr += PAGE_SIZE) { struct page *page = virt_to_page (addr); ClearPageReserved (page); - set_page_count (page, 1); + init_page_count (page); __free_page (page); total_ram_pages++; } From 0e1dfc66b6ec94984a4778132147a8aa36461d58 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 30 Jul 2006 03:03:28 -0700 Subject: [PATCH 087/143] [PATCH] invalidate_bdev() speedup We can immediately bail from invalidate_bdev() if the blockdev has no pagecache. This solves the huge IPI storms which hald is causing on the big ia64 machines when it polls CDROM drives. Acked-by: Jes Sorensen Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/buffer.c b/fs/buffer.c index 3660dcb97591..71649ef9b658 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -470,13 +470,18 @@ __find_get_block_slow(struct block_device *bdev, sector_t block) pass does the actual I/O. */ void invalidate_bdev(struct block_device *bdev, int destroy_dirty_buffers) { + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) + return; + invalidate_bh_lrus(); /* * FIXME: what about destroy_dirty_buffers? * We really want to use invalidate_inode_pages2() for * that, but not until that's cleaned up. */ - invalidate_inode_pages(bdev->bd_inode->i_mapping); + invalidate_inode_pages(mapping); } /* From 1e86240f3fd33f8052141778ad3dffe2c96184d7 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Sun, 30 Jul 2006 03:03:29 -0700 Subject: [PATCH 088/143] [PATCH] IDE: Touch NMI watchdog during resume from STR When resuming from suspend-to-RAM, the NMI watchdog detects a lockup in ide_wait_not_busy. Here's a screenshot of the trace taken by a digital camera: http://www.uamt.feec.vutbr.cz/rizeni/pom/DSC03510-2.JPG Let's touch the NMI watchdog in ide_wait_not_busy. The system then resumes correctly from STR. [akpm@osdl.org: modular build fix] Signed-off-by: Michal Schmidt Acked-by: Alan Cox Cc: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/nmi.c | 1 + drivers/ide/ide-iops.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 2dd928a84645..acb351478e42 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -575,6 +575,7 @@ void touch_nmi_watchdog (void) */ touch_softlockup_watchdog(); } +EXPORT_SYMBOL(touch_nmi_watchdog); extern void die_nmi(struct pt_regs *, const char *msg); diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c index 657165297dc7..77703acaec17 100644 --- a/drivers/ide/ide-iops.c +++ b/drivers/ide/ide-iops.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -1243,6 +1244,7 @@ int ide_wait_not_busy(ide_hwif_t *hwif, unsigned long timeout) if (stat == 0xff) return -ENODEV; touch_softlockup_watchdog(); + touch_nmi_watchdog(); } return -EBUSY; } From 8e185d83398b2708f50207bc12d0e0a70bd70767 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sun, 30 Jul 2006 03:03:30 -0700 Subject: [PATCH 089/143] [PATCH] lockdep: annotate pktcdvd natural device hierarchy The pkt_*_dev functions operate on not-this-blockdevice, and that is sufficiently checked at setup time. As a result there is a natural hierarchy, which needs nesting annotations Signed-off-by: Arjan van de Ven Cc: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/pktcdvd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index bde2c64b6346..451b996bba91 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -2577,19 +2577,19 @@ static int pkt_ctl_ioctl(struct inode *inode, struct file *file, unsigned int cm case PKT_CTRL_CMD_SETUP: if (!capable(CAP_SYS_ADMIN)) return -EPERM; - mutex_lock(&ctl_mutex); + mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING); ret = pkt_setup_dev(&ctrl_cmd); mutex_unlock(&ctl_mutex); break; case PKT_CTRL_CMD_TEARDOWN: if (!capable(CAP_SYS_ADMIN)) return -EPERM; - mutex_lock(&ctl_mutex); + mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING); ret = pkt_remove_dev(&ctrl_cmd); mutex_unlock(&ctl_mutex); break; case PKT_CTRL_CMD_STATUS: - mutex_lock(&ctl_mutex); + mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING); pkt_get_status(&ctrl_cmd); mutex_unlock(&ctl_mutex); break; From e4b57e0842621f597d744b193ea325d62724596b Mon Sep 17 00:00:00 2001 From: Michal Feix Date: Sun, 30 Jul 2006 03:03:31 -0700 Subject: [PATCH 090/143] [PATCH] nbd: Check magic before doing anything else We should check magic sequence in reply packet before trying to find request with it's request handle. This also solves the problem with "Unexpected reply" message beeing logged, when packet with invalid magic is received. Signed-off-by: Michal Feix Acked-by: Paul Clements Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/nbd.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 0a1b1ea36ddc..31a0b605f261 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -300,6 +300,15 @@ static struct request *nbd_read_stat(struct nbd_device *lo) lo->disk->disk_name, result); goto harderror; } + + if (ntohl(reply.magic) != NBD_REPLY_MAGIC) { + printk(KERN_ERR "%s: Wrong magic (0x%lx)\n", + lo->disk->disk_name, + (unsigned long)ntohl(reply.magic)); + result = -EPROTO; + goto harderror; + } + req = nbd_find_request(lo, reply.handle); if (unlikely(IS_ERR(req))) { result = PTR_ERR(req); @@ -312,13 +321,6 @@ static struct request *nbd_read_stat(struct nbd_device *lo) goto harderror; } - if (ntohl(reply.magic) != NBD_REPLY_MAGIC) { - printk(KERN_ERR "%s: Wrong magic (0x%lx)\n", - lo->disk->disk_name, - (unsigned long)ntohl(reply.magic)); - result = -EPROTO; - goto harderror; - } if (ntohl(reply.error)) { printk(KERN_ERR "%s: Other side returned error (%d)\n", lo->disk->disk_name, ntohl(reply.error)); From f0df33bcab4b687f0f18b9ebd69d374642e46a0e Mon Sep 17 00:00:00 2001 From: Michal Feix Date: Sun, 30 Jul 2006 03:03:32 -0700 Subject: [PATCH 091/143] [PATCH] nbd: Abort request on data reception failure When reading from nbd device, we need to receive all the data after receiving reply packet from the server - otherwise such request will never be ended. If socket is closed right after accepting reply control packet and in the middle of waiting for read data, nbd_read_stat() returns NULL and nbd_end_request() is not called. This patch fixes it. Signed-off-by: Michal Feix Acked-by: Paul Clements Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/nbd.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 31a0b605f261..bdbade9a5cf5 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -341,7 +341,8 @@ static struct request *nbd_read_stat(struct nbd_device *lo) printk(KERN_ERR "%s: Receive data failed (result %d)\n", lo->disk->disk_name, result); - goto harderror; + req->errors++; + return req; } dprintk(DBG_RX, "%s: request %p: got %d bytes data\n", lo->disk->disk_name, req, bvec->bv_len); From b8bdb460b7ecf08a4fed6e8b5b6b3fe874587aaa Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Sun, 30 Jul 2006 03:03:33 -0700 Subject: [PATCH 092/143] [PATCH] always define IRQ_PER_CPU Reduce the likelihood of someone accidentally introducing namespace collisions. Signed-off-by: Yoichi Yuasa Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/irq.h b/include/linux/irq.h index 6e59ac05ef2a..fbf6d901e9c2 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -47,8 +47,8 @@ #define IRQ_WAITING 0x00200000 /* IRQ not yet seen - for autodetection */ #define IRQ_LEVEL 0x00400000 /* IRQ level triggered */ #define IRQ_MASKED 0x00800000 /* IRQ masked - shouldn't be seen again */ +#define IRQ_PER_CPU 0x01000000 /* IRQ is per CPU */ #ifdef CONFIG_IRQ_PER_CPU -# define IRQ_PER_CPU 0x01000000 /* IRQ is per CPU */ # define CHECK_IRQ_PER_CPU(var) ((var) & IRQ_PER_CPU) #else # define CHECK_IRQ_PER_CPU(var) 0 From cea6a4ba8acfba6f59cc9ed71e0d05cb770b9d9c Mon Sep 17 00:00:00 2001 From: Horms Date: Sun, 30 Jul 2006 03:03:34 -0700 Subject: [PATCH 093/143] [PATCH] panic_on_oops: remove ssleep() This patch is part of an effort to unify the panic_on_oops behaviour across all architectures that implement it. It was pointed out to me by Andi Kleen that if an oops has occured in interrupt context, then calling sleep() in the oops path will only cause a panic, and that it would be really better for it not to be in the path at all. This patch removes the ssleep() call and reworks the console message accordinly. I have a slght concern that the resulting console message is too long, feedback welcome. For powerpc it also unifies the 32bit and 64bit behaviour. Fror x86_64, this patch only updates the console message, as ssleep() is already not present. Signed-off-by: Horms Acked-by: Paul Mackerras Cc: Russell King Cc: "Luck, Tony" Cc: Andi Kleen Cc: Chris Zankel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/traps.c | 7 ++----- arch/i386/kernel/traps.c | 8 +++----- arch/ia64/kernel/traps.c | 7 ++----- arch/powerpc/kernel/traps.c | 10 +++------- arch/x86_64/kernel/traps.c | 2 +- arch/xtensa/kernel/traps.c | 8 +++----- 6 files changed, 14 insertions(+), 28 deletions(-) diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 35a052fc177a..4e29dd03e582 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -232,11 +232,8 @@ NORET_TYPE void die(const char *str, struct pt_regs *regs, int err) bust_spinlocks(0); spin_unlock_irq(&die_lock); - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); - panic("Fatal exception"); - } + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); do_exit(SIGSEGV); } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 021f8fdc7512..0d4005dc06c5 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -453,11 +453,9 @@ void die(const char * str, struct pt_regs * regs, long err) if (in_interrupt()) panic("Fatal exception in interrupt"); - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); - panic("Fatal exception"); - } + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); + oops_exit(); do_exit(SIGSEGV); } diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index e7bbb0f40aa2..5a0420464c6c 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -117,11 +117,8 @@ die (const char *str, struct pt_regs *regs, long err) die.lock_owner = -1; spin_unlock_irq(&die.lock); - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); - panic("Fatal exception"); - } + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); do_exit(SIGSEGV); } diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 3c668078e524..2105767fcc57 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -150,13 +150,9 @@ int die(const char *str, struct pt_regs *regs, long err) if (in_interrupt()) panic("Fatal exception in interrupt"); - if (panic_on_oops) { -#ifdef CONFIG_PPC64 - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); -#endif - panic("Fatal exception"); - } + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); + do_exit(err); return 0; diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index f7a9d1421078..4e9938dee060 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -529,7 +529,7 @@ void __kprobes oops_end(unsigned long flags) /* Nest count reaches zero, release the lock. */ spin_unlock_irqrestore(&die_lock, flags); if (panic_on_oops) - panic("Oops"); + panic("Fatal exception: panic_on_oops"); } void __kprobes __die(const char * str, struct pt_regs * regs, long err) diff --git a/arch/xtensa/kernel/traps.c b/arch/xtensa/kernel/traps.c index 27e409089a7b..9734960a2451 100644 --- a/arch/xtensa/kernel/traps.c +++ b/arch/xtensa/kernel/traps.c @@ -487,11 +487,9 @@ void die(const char * str, struct pt_regs * regs, long err) if (in_interrupt()) panic("Fatal exception in interrupt"); - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); - panic("Fatal exception"); - } + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); + do_exit(err); } From 8c78f3075dab4be279e283f901f00e33ce44890a Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Sun, 30 Jul 2006 03:03:35 -0700 Subject: [PATCH 094/143] [PATCH] cpu hotplug: replace __devinit* with __cpuinit* for cpu notifications Few of the callback functions and notifier blocks that are associated with cpu notifications incorrectly have __devinit and __devinitdata. They should be __cpuinit and __cpuinitdata instead. It makes no functional difference but wastes text area when CONFIG_HOTPLUG is enabled and CONFIG_HOTPLUG_CPU is not. This patch fixes all those instances. Signed-off-by: Chandra Seetharaman Cc: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/sysfs.c | 4 ++-- kernel/hrtimer.c | 4 ++-- kernel/rcupdate.c | 4 ++-- kernel/softirq.c | 4 ++-- kernel/softlockup.c | 4 ++-- kernel/timer.c | 4 ++-- mm/slab.c | 2 +- 7 files changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index 010435095550..fec228cd0163 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -278,7 +278,7 @@ static void unregister_cpu_online(unsigned int cpu) } #endif /* CONFIG_HOTPLUG_CPU */ -static int __devinit sysfs_cpu_notify(struct notifier_block *self, +static int __cpuinit sysfs_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned int)(long)hcpu; @@ -296,7 +296,7 @@ static int __devinit sysfs_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __devinitdata sysfs_cpu_nb = { +static struct notifier_block __cpuinitdata sysfs_cpu_nb = { .notifier_call = sysfs_cpu_notify, }; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index d17766d40dab..be989efc7856 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -835,7 +835,7 @@ static void migrate_hrtimers(int cpu) } #endif /* CONFIG_HOTPLUG_CPU */ -static int __devinit hrtimer_cpu_notify(struct notifier_block *self, +static int __cpuinit hrtimer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -859,7 +859,7 @@ static int __devinit hrtimer_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __devinitdata hrtimers_nb = { +static struct notifier_block __cpuinitdata hrtimers_nb = { .notifier_call = hrtimer_cpu_notify, }; diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 759805c9859a..436ab35f6fa7 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -548,7 +548,7 @@ static void __devinit rcu_online_cpu(int cpu) tasklet_init(&per_cpu(rcu_tasklet, cpu), rcu_process_callbacks, 0UL); } -static int __devinit rcu_cpu_notify(struct notifier_block *self, +static int __cpuinit rcu_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -565,7 +565,7 @@ static int __devinit rcu_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __devinitdata rcu_nb = { +static struct notifier_block __cpuinitdata rcu_nb = { .notifier_call = rcu_cpu_notify, }; diff --git a/kernel/softirq.c b/kernel/softirq.c index 0f08a84ae307..aab880677ce0 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -547,7 +547,7 @@ static void takeover_tasklets(unsigned int cpu) } #endif /* CONFIG_HOTPLUG_CPU */ -static int __devinit cpu_callback(struct notifier_block *nfb, +static int __cpuinit cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -587,7 +587,7 @@ static int __devinit cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -static struct notifier_block __devinitdata cpu_nfb = { +static struct notifier_block __cpuinitdata cpu_nfb = { .notifier_call = cpu_callback }; diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 6b76caa22981..03e6a2b0b787 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -104,7 +104,7 @@ static int watchdog(void * __bind_cpu) /* * Create/destroy watchdog threads as CPUs come and go: */ -static int __devinit +static int __cpuinit cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; @@ -142,7 +142,7 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block __devinitdata cpu_nfb = { +static struct notifier_block __cpuinitdata cpu_nfb = { .notifier_call = cpu_callback }; diff --git a/kernel/timer.c b/kernel/timer.c index 05809c2e2fd6..ee319fc69f40 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1688,7 +1688,7 @@ static void __devinit migrate_timers(int cpu) } #endif /* CONFIG_HOTPLUG_CPU */ -static int __devinit timer_cpu_notify(struct notifier_block *self, +static int __cpuinit timer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -1708,7 +1708,7 @@ static int __devinit timer_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __devinitdata timers_nb = { +static struct notifier_block __cpuinitdata timers_nb = { .notifier_call = timer_cpu_notify, }; diff --git a/mm/slab.c b/mm/slab.c index 0f20843beffd..252972ff6495 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1106,7 +1106,7 @@ static inline int cache_free_alien(struct kmem_cache *cachep, void *objp) #endif -static int __devinit cpuup_callback(struct notifier_block *nfb, +static int __cpuinit cpuup_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { long cpu = (long)hcpu; From 7c7165c90801609b70492e50b2a9c69a677c573a Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Sun, 30 Jul 2006 03:03:36 -0700 Subject: [PATCH 095/143] [PATCH] cpu hotplug: fix hotplug cpu documentation for proper usage Update hotplug cpu documentation to clearly state when to use register_cpu_notifier() and register_hotcpu_notifier. Signed-off-by: Chandra Seetharaman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/cpu-hotplug.txt | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt index 1bcf69996c9d..bc107cb157a8 100644 --- a/Documentation/cpu-hotplug.txt +++ b/Documentation/cpu-hotplug.txt @@ -251,16 +251,24 @@ A: This is what you would need in your kernel code to receive notifications. return NOTIFY_OK; } - static struct notifier_block foobar_cpu_notifer = + static struct notifier_block __cpuinitdata foobar_cpu_notifer = { .notifier_call = foobar_cpu_callback, }; +You need to call register_cpu_notifier() from your init function. +Init functions could be of two types: +1. early init (init function called when only the boot processor is online). +2. late init (init function called _after_ all the CPUs are online). -In your init function, +For the first case, you should add the following to your init function register_cpu_notifier(&foobar_cpu_notifier); +For the second case, you should add the following to your init function + + register_hotcpu_notifier(&foobar_cpu_notifier); + You can fail PREPARE notifiers if something doesn't work to prepare resources. This will stop the activity and send a following CANCELED event back. From be6b5a3505fa0cd54c3b5959a39293f47c648980 Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Sun, 30 Jul 2006 03:03:37 -0700 Subject: [PATCH 096/143] [PATCH] cpu hotplug: use hotplug version of registration in late inits Use hotplug version of register_cpu_notifier in late init functions. Signed-off-by: Chandra Seetharaman Cc: "Luck, Tony" Cc: Martin Schwidefsky Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/intel_cacheinfo.c | 2 +- arch/ia64/kernel/palinfo.c | 13 ++++++------- arch/ia64/kernel/topology.c | 2 +- arch/s390/appldata/appldata_base.c | 10 +++++----- arch/x86_64/kernel/mce.c | 12 +++++------- arch/x86_64/kernel/mce_amd.c | 19 +++++++------------ drivers/base/topology.c | 2 +- 7 files changed, 26 insertions(+), 34 deletions(-) diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c index e9f0b928b0a9..5c43be47587f 100644 --- a/arch/i386/kernel/cpu/intel_cacheinfo.c +++ b/arch/i386/kernel/cpu/intel_cacheinfo.c @@ -759,7 +759,7 @@ static int __cpuinit cache_sysfs_init(void) if (num_cache_leaves == 0) return 0; - register_cpu_notifier(&cacheinfo_cpu_notifier); + register_hotcpu_notifier(&cacheinfo_cpu_notifier); for_each_online_cpu(i) { cacheinfo_cpu_callback(&cacheinfo_cpu_notifier, CPU_ONLINE, diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c index 3f5bac59209a..ab5b52413e91 100644 --- a/arch/ia64/kernel/palinfo.c +++ b/arch/ia64/kernel/palinfo.c @@ -958,9 +958,9 @@ remove_palinfo_proc_entries(unsigned int hcpu) } } -static int __cpuinit palinfo_cpu_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) +#ifdef CONFIG_HOTPLUG_CPU +static int palinfo_cpu_callback(struct notifier_block *nfb, + unsigned long action, void *hcpu) { unsigned int hotcpu = (unsigned long)hcpu; @@ -968,20 +968,19 @@ static int __cpuinit palinfo_cpu_callback(struct notifier_block *nfb, case CPU_ONLINE: create_palinfo_proc_entries(hotcpu); break; -#ifdef CONFIG_HOTPLUG_CPU case CPU_DEAD: remove_palinfo_proc_entries(hotcpu); break; -#endif } return NOTIFY_OK; } -static struct notifier_block __cpuinitdata palinfo_cpu_notifier = +static struct notifier_block palinfo_cpu_notifier = { .notifier_call = palinfo_cpu_callback, .priority = 0, }; +#endif static int __init palinfo_init(void) @@ -1020,7 +1019,7 @@ palinfo_exit(void) /* * Unregister from cpu notifier callbacks */ - unregister_cpu_notifier(&palinfo_cpu_notifier); + unregister_hotcpu_notifier(&palinfo_cpu_notifier); } module_init(palinfo_init); diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index b146f1cfad31..d24fa393b182 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -435,7 +435,7 @@ static int __cpuinit cache_sysfs_init(void) (void *)(long)i); } - register_cpu_notifier(&cache_cpu_notifier); + register_hotcpu_notifier(&cache_cpu_notifier); return 0; } diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index d0f82c995af6..6a4b5f9715c9 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -617,7 +617,8 @@ appldata_offline_cpu(int cpu) spin_unlock(&appldata_timer_lock); } -static int __cpuinit +#ifdef CONFIG_HOTPLUG_CPU +static int appldata_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -625,20 +626,19 @@ appldata_cpu_notify(struct notifier_block *self, case CPU_ONLINE: appldata_online_cpu((long) hcpu); break; -#ifdef CONFIG_HOTPLUG_CPU case CPU_DEAD: appldata_offline_cpu((long) hcpu); break; -#endif default: break; } return NOTIFY_OK; } -static struct notifier_block __devinitdata appldata_nb = { +static struct notifier_block appldata_nb = { .notifier_call = appldata_cpu_notify, }; +#endif /* * appldata_init() @@ -662,7 +662,7 @@ static int __init appldata_init(void) appldata_online_cpu(i); /* Register cpu hotplug notifier */ - register_cpu_notifier(&appldata_nb); + register_hotcpu_notifier(&appldata_nb); appldata_sysctl_header = register_sysctl_table(appldata_dir_table, 1); #ifdef MODULE diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 88845674c661..4e017fb30fb3 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -615,7 +615,7 @@ static __cpuinit int mce_create_device(unsigned int cpu) } #ifdef CONFIG_HOTPLUG_CPU -static __cpuinit void mce_remove_device(unsigned int cpu) +static void mce_remove_device(unsigned int cpu) { int i; @@ -626,10 +626,9 @@ static __cpuinit void mce_remove_device(unsigned int cpu) sysdev_remove_file(&per_cpu(device_mce,cpu), &attr_check_interval); sysdev_unregister(&per_cpu(device_mce,cpu)); } -#endif /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -static __cpuinit int +static int mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -638,18 +637,17 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) case CPU_ONLINE: mce_create_device(cpu); break; -#ifdef CONFIG_HOTPLUG_CPU case CPU_DEAD: mce_remove_device(cpu); break; -#endif } return NOTIFY_OK; } -static struct notifier_block __cpuinitdata mce_cpu_notifier = { +static struct notifier_block mce_cpu_notifier = { .notifier_call = mce_cpu_callback, }; +#endif static __init int mce_init_device(void) { @@ -664,7 +662,7 @@ static __init int mce_init_device(void) mce_create_device(i); } - register_cpu_notifier(&mce_cpu_notifier); + register_hotcpu_notifier(&mce_cpu_notifier); misc_register(&mce_log_device); return err; } diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c index db2acbf7ad28..883fe747f64c 100644 --- a/arch/x86_64/kernel/mce_amd.c +++ b/arch/x86_64/kernel/mce_amd.c @@ -558,7 +558,7 @@ static __cpuinit int threshold_create_device(unsigned int cpu) * of shared sysfs dir/files, and rest of the cores will be symlinked to it. */ -static __cpuinit void deallocate_threshold_block(unsigned int cpu, +static void deallocate_threshold_block(unsigned int cpu, unsigned int bank) { struct threshold_block *pos = NULL; @@ -578,7 +578,7 @@ static __cpuinit void deallocate_threshold_block(unsigned int cpu, per_cpu(threshold_banks, cpu)[bank]->blocks = NULL; } -static __cpuinit void threshold_remove_bank(unsigned int cpu, int bank) +static void threshold_remove_bank(unsigned int cpu, int bank) { int i = 0; struct threshold_bank *b; @@ -618,7 +618,7 @@ static __cpuinit void threshold_remove_bank(unsigned int cpu, int bank) per_cpu(threshold_banks, cpu)[bank] = NULL; } -static __cpuinit void threshold_remove_device(unsigned int cpu) +static void threshold_remove_device(unsigned int cpu) { unsigned int bank; @@ -629,14 +629,8 @@ static __cpuinit void threshold_remove_device(unsigned int cpu) } } -#else /* !CONFIG_HOTPLUG_CPU */ -static void threshold_remove_device(unsigned int cpu) -{ -} -#endif - /* get notified when a cpu comes on/off */ -static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb, +static int threshold_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { /* cpu was unsigned int to begin with */ @@ -659,9 +653,10 @@ static int __cpuinit threshold_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -static struct notifier_block threshold_cpu_notifier __cpuinitdata = { +static struct notifier_block threshold_cpu_notifier = { .notifier_call = threshold_cpu_callback, }; +#endif /* CONFIG_HOTPLUG_CPU */ static __init int threshold_init_device(void) { @@ -673,7 +668,7 @@ static __init int threshold_init_device(void) if (err) return err; } - register_cpu_notifier(&threshold_cpu_notifier); + register_hotcpu_notifier(&threshold_cpu_notifier); return 0; } diff --git a/drivers/base/topology.c b/drivers/base/topology.c index c2d621632383..3ef9d514b916 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -139,7 +139,7 @@ static int __cpuinit topology_sysfs_init(void) (void *)(long)i); } - register_cpu_notifier(&topology_cpu_notifier); + register_hotcpu_notifier(&topology_cpu_notifier); return 0; } From 6ea24f9ad18c65cc179593b5cc2a88cdadf8cc0c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Sun, 30 Jul 2006 03:03:38 -0700 Subject: [PATCH 097/143] [PATCH] fix bad macro param in timer.c We have #define INDEX(N) (base->timer_jiffies >> (TVR_BITS + N * TVN_BITS)) & TVN_MASK and it's used via list = varray[i + 1]->vec + (INDEX(i + 1)); So, due to underparenthesisation, this INDEX(i+1) is now a ... (TVR_BITS + i + 1 * TVN_BITS)) ... So this bugfix changes behaviour. It worked before by sheer luck: "If i was anything but 0, it was broken. But this was only used by s390 and arm. Since it was for the next interrupt, could that next interrupt be a problem (going into the second cascade)? But it was probably seldom wrong. That is, this would fail if the next interrupt was in the second cascade, and was wrapped. Which may never of happened. Also if it did happen, it would have just missed the interrupt. If an interrupt was missed, and no one was there to miss it, was it really missed :-)" Signed-off-by: Steven Rostedt Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/timer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/timer.c b/kernel/timer.c index ee319fc69f40..726caf9a0a69 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -408,7 +408,7 @@ static int cascade(tvec_base_t *base, tvec_t *tv, int index) * This function cascades all vectors and executes all expired timer * vectors. */ -#define INDEX(N) (base->timer_jiffies >> (TVR_BITS + N * TVN_BITS)) & TVN_MASK +#define INDEX(N) ((base->timer_jiffies >> (TVR_BITS + (N) * TVN_BITS)) & TVN_MASK) static inline void __run_timers(tvec_base_t *base) { From 2d7d253548cffdce80f4e03664686e9ccb1b0ed7 Mon Sep 17 00:00:00 2001 From: Jim Houston Date: Sun, 30 Jul 2006 03:03:39 -0700 Subject: [PATCH 098/143] [PATCH] fix cond_resched() fix In cond_resched_lock() it calls __resched_legal() before dropping the spin lock. __resched_legal() will always finds the preempt_count non-zero and will prevent the call to __cond_resched(). The attached patch adds a parameter to __resched_legal() with the expected preempt_count value. Cc: Ingo Molnar Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sched.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index 41a571806ce0..de440b220b4c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4456,9 +4456,9 @@ asmlinkage long sys_sched_yield(void) return 0; } -static inline int __resched_legal(void) +static inline int __resched_legal(int expected_preempt_count) { - if (unlikely(preempt_count())) + if (unlikely(preempt_count() != expected_preempt_count)) return 0; if (unlikely(system_state != SYSTEM_RUNNING)) return 0; @@ -4484,7 +4484,7 @@ static void __cond_resched(void) int __sched cond_resched(void) { - if (need_resched() && __resched_legal()) { + if (need_resched() && __resched_legal(0)) { __cond_resched(); return 1; } @@ -4510,7 +4510,7 @@ int cond_resched_lock(spinlock_t *lock) ret = 1; spin_lock(lock); } - if (need_resched() && __resched_legal()) { + if (need_resched() && __resched_legal(1)) { spin_release(&lock->dep_map, 1, _THIS_IP_); _raw_spin_unlock(lock); preempt_enable_no_resched(); @@ -4526,7 +4526,7 @@ int __sched cond_resched_softirq(void) { BUG_ON(!in_softirq()); - if (need_resched() && __resched_legal()) { + if (need_resched() && __resched_legal(0)) { raw_local_irq_disable(); _local_bh_enable(); raw_local_irq_enable(); From 2b54960bdf8fbb57d94dd61f4ac7513535ca7168 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 30 Jul 2006 03:03:40 -0700 Subject: [PATCH 099/143] [PATCH] fix kernel-api doc for kernel/resource.c insert_resource() was unexported, so kernel-doc needs to be told to search kernel/resource.c for internal functions instead of exported functions so that it won't report an error. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/DocBook/kernel-api.tmpl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index 1ae4dc0fd856..db67ba6f5b1d 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -300,7 +300,7 @@ X!Ekernel/module.c Resources Management -!Ekernel/resource.c +!Ikernel/resource.c MTRR Handling From 0129a057b652b33dba6406a940af061cf0f36f79 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 30 Jul 2006 03:03:41 -0700 Subject: [PATCH 100/143] [PATCH] kernel-doc: ignore __devinit Ignore __devinit in function definitions so that kernel-doc won't fail on them. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/kernel-doc | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/kernel-doc b/scripts/kernel-doc index f9460a6218de..c9ca0c23bd91 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -1518,6 +1518,7 @@ sub dump_function($$) { $prototype =~ s/^asmlinkage +//; $prototype =~ s/^inline +//; $prototype =~ s/^__inline__ +//; + $prototype =~ s/__devinit +//; $prototype =~ s/^#define +//; #ak added $prototype =~ s/__attribute__ \(\([a-z,]*\)\)//; From d75763d24063cafe28ace8863560da9c968ee099 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 30 Jul 2006 03:03:41 -0700 Subject: [PATCH 101/143] [PATCH] pci/search: cleanups, add to kernel-api.tmpl Clean up kernel-doc comments in drivers/pci/search.c (line sizes and typos). Enable that source file in DocBook/kernel-api.tmpl. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/DocBook/kernel-api.tmpl | 4 +-- drivers/pci/search.c | 46 +++++++++++++++------------ 2 files changed, 26 insertions(+), 24 deletions(-) diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index db67ba6f5b1d..356e34e6ce33 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -312,9 +312,7 @@ X!Ekernel/module.c !Edrivers/pci/pci-driver.c !Edrivers/pci/remove.c !Edrivers/pci/pci-acpi.c - +!Edrivers/pci/search.c !Edrivers/pci/msi.c !Edrivers/pci/bus.c