From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Luck, Tony" Date: Wed, 27 Oct 2004 03:58:51 +0000 Subject: bk pull on ia64 linux tree Message-Id: <200410270358.i9R3wpS27056@unix-os.sc.intel.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, please do a bk pull http://lia64.bkbits.net/linux-ia64-release-2.6.10 This will update the files shown below. Thanks! -Tony arch/ia64/ia32/elfcore32.h | 1 arch/ia64/kernel/efi.c | 35 ++++++++---- arch/ia64/kernel/perfmon.c | 32 ++++------- arch/ia64/mm/discontig.c | 4 - arch/ia64/sn/kernel/idle.c | 12 +--- arch/ia64/sn/kernel/io_init.c | 4 + arch/ia64/sn/kernel/irq.c | 8 +- arch/ia64/sn/kernel/mca.c | 28 +++++----- arch/ia64/sn/kernel/setup.c | 51 ++++++++++++++++-- arch/ia64/sn/kernel/sn2/sn2_smp.c | 19 ++++--- arch/ia64/sn/kernel/sn2/sn_hwperf.c | 51 ++++++++---------- arch/ia64/sn/kernel/sn2/sn_proc_fs.c | 3 - drivers/ide/pci/sgiioc4.c | 12 ++-- drivers/serial/sn_console.c | 79 +---------------------------- include/asm-ia64/pgtable.h | 8 +- include/asm-ia64/sn/intr.h | 2 include/asm-ia64/sn/nodepda.h | 10 +++ include/asm-ia64/sn/sn2/sn_hwperf.h | 10 +++ include/asm-ia64/sn/sn_cpuid.h | 95 ++++++++++------------------------- include/asm-ia64/sn/sn_sal.h | 32 +++++++++++ 20 files changed, 239 insertions(+), 257 deletions(-) through these ChangeSets: (04/10/26 1.2185.1.12) [IA64] Need for GRANULEROUNDDOWN Alex's change for mem=/max_addr= now needs to include meminit.h Signed-off-by: Tony Luck (04/10/26 1.2185.1.11) [IA64-SGI] only allocate irq if the device can interrupt Signed-off-by: Patrick Gefre Acked-by: Mike Habeck Signed-off-by: Tony Luck (04/10/26 1.2185.1.10) [IA64-SGI] Delete simulator support from SN idle loop Delete hack for supporting simulator for SN platforms. This capability has been moved into the simulator environment & is no longer required in the kernel. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck (04/10/26 1.2185.1.9) [IA64-SGI] Update asm-ia64/sn/sn_cpuid.h macros SGI SN code currently makes assumptions about the bits in the LID register. These assumptions do not conform to the bit specifications from Intel. For example, SN currently assumes that bits [28:16] of the LID contain the physical node ID of a node. This patch eliminates these assumptions. A SAL call is now used to translate LID values to the NASID/subnode/slice values that are needed for SN platforms. The results of the SAL call are saved in the SN nodepda. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck (04/10/26 1.2185.1.8) [IA64] Delete obsolete code from SGI console driver Delete obsolete code that supports the SGI simulator. This support code has been moved into the fakeprom and is no longer required in the kernel. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck (04/10/26 1.2185.1.7) [IA64] fix pgtable.h comments Remove some statements from comments in whose correctness is page size dependent ... and based on using an 8k page size, which almost nobody actually uses. Signed-off-by: Jason Uhlenkott Signed-off-by: Tony Luck (04/10/26 1.2185.1.6) [IA64] Correct references from text/data to init.text/data. These errors were found by 'make buildcheck'. Signed-off-by: Keith Owens Signed-off-by: Tony Luck (04/10/26 1.2185.1.5) [IA64] sn_hwperf correctly handle bricks with multiple slabs The procfs handler for /proc/sgi_sn/sn_topology did not correctly handle multiple slabs in the same brick, e.g. a brick containing a compute node and an ionode. Signed-off-by: Mark Goodwin Signed-off-by: Tony Luck (04/10/26 1.2185.1.4) [IA64] Correct bit test for salinfo oem decode The valid.oem_data bit is in different positions in each section. Make the test for valid oem data a section specific test. Signed-off-by: Keith Owens Signed-off-by: Tony Luck (04/10/26 1.2185.1.3) [IA64] misc small patches for perfmon change-log: - update comment for pfm_do_fasync() - fix return value for pfm_smpl_buffer_alloc(). Now return ENOMEM instead of EAGAIN when buffer too big for RLIMIT_MEMLOCK. - added missing vm_pgoff initialization in pfm_smpl_buffer_alloc() - added flags tro debug print in pfm_write_pmcs - added seed and mask to debug print in pfm_write_pmds - shorten some of the debug prints - remove bogus sanity check from pfm_save_regs() in SMP. this could cause invalid PMU state restore signed-off-by: eranian@hpl.hp.com Signed-off-by: Tony Luck (04/10/26 1.2185.1.2) [IA64] efi.c: fix mem= & max_addr With this change, there's some extra fuzz introduced that a max_addr specification will get rounded down to a granule boundary and memory quantity, when using mem=, will be within a granule size of the requested amount. Let me know if anyone finds more problems with it. Signed-off-by: Alex Williamson Signed-off-by: Tony Luck (04/10/26 1.2000.10.2) [IA64] Need for KERNEL_DS & set_fs() definitions. Signed-off-by: Andrew Morton Signed-off-by: Tony Luck