From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Mosberger Date: Fri, 09 Apr 2004 16:05:27 +0000 Subject: bk pull on ia64 linux tree Message-Id: <200404091605.i39G5RPi009402@napali.hpl.hp.com> List-Id: References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org Hi Linus, I'm running behind with 2.6.5 but since there is a largish perfmon-update pending, I thought I should get this out before 2.6.5-rc1. Please do a bk pull http://lia64.bkbits.net/to-linus-2.5 This will update the files shown below. Thanks! --david arch/ia64/Kconfig | 4 arch/ia64/hp/common/sba_iommu.c | 13 arch/ia64/hp/sim/Kconfig | 2 arch/ia64/kernel/acpi.c | 24 - arch/ia64/kernel/perfmon.c | 615 ++++++++++++++++--------------- arch/ia64/kernel/perfmon_mckinley.h | 13 arch/ia64/kernel/sys_ia64.c | 8 arch/ia64/mm/hugetlbpage.c | 1 arch/ia64/pci/pci.c | 1 arch/ia64/sn/io/hwgfs/interface.c | 6 arch/ia64/sn/io/machvec/pci_bus_cvlink.c | 26 - arch/ia64/sn/io/machvec/pci_dma.c | 12 arch/ia64/sn/io/sn2/pcibr/pcibr_slot.c | 3 drivers/char/sn_serial.c | 4 include/asm-ia64/machvec_hpzx1.h | 3 include/asm-ia64/perfmon.h | 7 include/asm-ia64/processor.h | 8 17 files changed, 403 insertions(+), 347 deletions(-) through these ChangeSets: (04/04/08 1.1789) ia64: Make acpi.c compile again: there was an implicit declaration mismatch because the external declaration isn't in the arch- independent ACPI bits yet. (04/04/08 1.1788) [PATCH] ia64: perfmon update Here is a new perfmon patch. It is important because it fixes the problem of the close() when the file descriptor is shared between two related processes. The good thing is that it simplifies a lot the cleanup of the sampling buffer. Here is the ChangeLog: - fix bug in pfm_close() when the descriptor is shared between related processed. Introduce a pfm_flush() called for each invocation of close(). pfm_close() only called for the last user. - fix pfm_restore_monitoring() to also reload the debug registers. They could be modified while monitoring is masked. - fix pfm_close() to clear ctx_fl_is_sampling. - fix a bug in pfm_handle_work() which could cause the wrong PMD to be reset. - converted PROTECT_CTX/UNPROTECT_CTX into local_irq_save/restore to keep context protection but allow IPI to proceed. - updated pfm_syswide_force_stop() to use local_irq_save/restore now that the context is protected from the caller side. - updated pfm_mck_pmc_check() to check if context is loaded before checking for special IBR/DBR combinations. Clearing the debug registers is not needed when the context is not yet loaded. - updated perfmon.h to have to correct prototype definitions for the pfm_mod_*() functions. - got rid of the PFM_CTX_TERMINATED state. - cleanup the DPRINT() statements to remove explicit output of current->pid. This is done systematically by the macros. - added a systctl entry (expert_mode) to bypass read/write checks on PMC/PMD. As its name indicates this is for experts ONLY. Must be root to toggle /proc/sys entry. - corrected pfm_mod_*() to check against the current task. - removed pfm_mod_fast_read_pmds(). It is never needed. - added pfm_mod_write_ibrs() and pfm_mod_write_dbrs(). (04/04/08 1.1787) [PATCH] ia64: allow simscsi to be a module Requiring CONFIG_HP_SIMSCSI to be either "y" or "n" breaks allmodconfig, because simscsi ends up built-in, while scsi itself is a module. So allow simscsi to be a module also. (04/04/08 1.1786) [PATCH] ia64: ACPI IRQ cleanup (arch part) Here's the ia64 part of the ACPI IRQ cleanup I proposed here: http://www.gelato.unsw.edu.au/linux-ia64/0403/8979.html After the arch bits are in, I'll post the corresponding ACPI changes. I removed the "Found IRQ" printk now because when the ACPI change goes in, dev->irq won't be initialized until *after* acpi_pci_irq_enable(). (04/04/08 1.1785) [PATCH] ia64: put URLs in documentation files inside angle-brackets Patch by Petri T. Koistinen. (04/04/05 1.1678.1.13) [PATCH] ia64: setup max dma addr earlier on hp boxes sba_iommu was setting up MAX_DMA_ADDRESS way too late to do any good. This patch makes it get setup via platform_setup, so it's ready for paging_init(). All pages should show up in zone DMA now. Against latest 2.6. (04/04/05 1.1678.1.12) [PATCH] ia64: Missing include in hugetlbpage.c This fixes a missing include file in arch/ia64/mm/hugetlbpage.c in 2.6.5. module.h is needed for EXPORT_SYMBOL. (04/04/05 1.1678.1.11) [PATCH] ia64: Missing overflow check in mmap Calling mmap with len = -1 was silently accepted. The test in the generic code was fixed in July 2003, but the fix didn't make it into the ia64- specific code. (04/04/01 1.1678.1.10) [PATCH] ia64: bring ia64 nodes online properly Andi's libnuma exposed the fact that we don't bring nodes online properly as we discover them in ia64. Here's a quick patch to fix that. Tested with Andi's numademo program; seems to work fine. (04/04/01 1.1678.1.9) [PATCH] ia64: small SN2 interface.c patch Change the calls to link_path_walk() to path_walk(). link_path_walk() doesn't initialze current->total_link_count and path_walk() does. Without the initialization, path_walks start to fail when 40 sym links have been followed. (04/04/01 1.1678.1.8) [PATCH] ia64: fix sn_serial warning In my fixes to quiet the sn_serial driver, I made sn_debug_printf only get used when the DEBUG define was turned on in the driver but forgot to #ifdef out the function definition, causing an 'defined but not used' warning. This patch fixes that. (04/04/01 1.1678.1.7) [PATCH] ia64: clean up pcibr_slot.c warnings A few warnings crept into pcibr_slot.c that this patch takes care of. - forward declare pcibr_slot_pwr so it can be used w/o warning - fix argument passed to pcibr_slot_attach in pcibr_slot_enable (04/03/26 1.1678.1.6) [PATCH] ia64: only scan SN2 PCI busses that exist sn_pci_init was unconditionally scanning 255 PCI busses, even though only a few are actually present on most systems. This patch adds a check to make sure a given bus exists before doing the scan, which saves a little memory and also makes /sys/devices reflect reality a bit more closely. (04/03/26 1.1678.1.5) [PATCH] ia64: implement SN2 version of dma_mapping_error() On sn2, dma_mapping_error is trivial. pci_map_single was already returning 0 in case of failure (0 is an invalid DMA address on sn2), so make dma_mapping_error check for that. (04/03/25 1.1678.1.4) ia64: Patch by Stephane Eranian: remove irq_stat from cpuinfo_ia64. irq_stat isn't needed anymore since its members got rolled into thread_info.preempt_count.