kernel.git/arch/powerpc/platforms/pseries, branch linux-2.6.19.y

[POWERPC] Use 4kB iommu pages even on 64kB-page systems

2006-11-01T03:52:48Z

The 10Gigabit ethernet device drivers appear to be able to chew up all 256MB of TCE mappings on pSeries systems, as evidenced by numerous error messages: iommu_alloc failed, tbl c0000000010d5c48 vaddr c0000000d875eff0 npages 1 Some experimentation indicates that this is essentially because one 1500 byte ethernet MTU gets mapped as a 64K DMA region when the large 64K pages are enabled. Thus, it doesn't take much to exhaust all of the available DMA mappings for a high-speed card. This patch changes the iommu allocator to work with its own unique, distinct page size. Although the patch is long, its actually quite simple: it just #defines a distinct IOMMU_PAGE_SIZE and then uses this in all the places that matter. As a side effect, it also dramatically improves network performance on platforms with H-calls on iommu translation inserts/removes (since we no longer call it 16 times for a 1500 bytes packet when the iommu HW is still 4k). In the future, we might want to make the IOMMU_PAGE_SIZE a variable in the iommu_table instance, thus allowing support for different HW page sizes in the iommu itself. Signed-off-by: Linas Vepstas Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Acked-by: Stephen Rothwell Signed-off-by: Paul Mackerras

[POWERPC] Fix up after irq changes

2006-10-07T12:08:26Z

Remove struct pt_regs * from all handlers. Also remove the regs argument from get_irq() functions. Compile tested with arch/powerpc/config/* and arch/ppc/configs/prep_defconfig Signed-off-by: Olaf Hering Signed-off-by: Paul Mackerras

[POWERPC] linux,tce-size property is 32 bits

2006-10-06T11:10:41Z

The "linux,tce-size" property is only 32 bits (see prom_initialize_tce_table() in arch/powerpc/kernel/prom_init.c). Treating it as an unsigned long in iommu_table_setparms() leads to access beyond the end of the property's buffer, so we pass garbage to the memset() in that function. [boot]0020 XICS Init i8259 legacy interrupt controller initialized [boot]0021 XICS Done PID hash table entries: 4096 (order: 12, 32768 bytes) cpu 0x0: Vector: 300 (Data Access) at [c0000000fe783850] pc: c000000000035e90: .memset+0x60/0xfc lr: c000000000044fa4: .iommu_table_setparms+0xb0/0x158 sp: c0000000fe783ad0 msr: 9000000000009032 dar: c000000100000000 dsisr: 42010000 current = 0xc00000000450e810 paca = 0xc000000000411580 pid = 1, comm = swapper enter ? for help [link register ] c000000000044fa4 .iommu_table_setparms+0xb0/0x158 [c0000000fe783ad0] c000000000044f4c .iommu_table_setparms+0x58/0x158 (unreliable) [c0000000fe783b70] c00000000004529c .iommu_bus_setup_pSeries+0x1c4/0x254 [c0000000fe783c00] c00000000002b8ac .do_bus_setup+0x3c/0xe4 [c0000000fe783c80] c00000000002c924 .pcibios_fixup_bus+0x64/0xd8 [c0000000fe783d00] c0000000001a2d5c .pci_scan_child_bus+0x6c/0x10c [c0000000fe783da0] c00000000002be28 .scan_phb+0x17c/0x1b4 [c0000000fe783e40] c0000000003cfa00 .pcibios_init+0x58/0x19c [c0000000fe783ec0] c0000000000094b4 .init+0x1e8/0x3d8 [c0000000fe783f90] c000000000026e54 .kernel_thread+0x4c/0x68 Signed-off-by: Nathan Lynch Acked-by: Olof Johansson Signed-off-by: Paul Mackerras

IRQ: Maintain regs pointer globally rather than passing to IRQ handlers

2006-10-05T14:10:12Z

Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)

[POWERPC] Fix a printk in pseries_mpic_init_IRQ

2006-10-04T04:52:35Z

This should probably say "mpic" to save confusion. Signed-off-by: Michael Ellerman Acked-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras

Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc

2006-10-03T15:52:26Z

* git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: (29 commits) [POWERPC] Fix rheap alignment problem [POWERPC] Use check_legacy_ioport() for ISAPnP [POWERPC] Avoid NULL pointer in gpio1_interrupt [POWERPC] Enable generic rtc hook for the MPC8349 mITX [POWERPC] Add powerpc get/set_rtc_time interface to new generic rtc class [POWERPC] Create a "wrapper" script and use it in arch/powerpc/boot [POWERPC] fix spin lock nesting in hvc_iseries [POWERPC] EEH failure to mark pci slot as frozen. [POWERPC] update powerpc defconfig files after libata kconfig breakage [POWERPC] enable sysrq in pmac32_defconfig [POWERPC] UPIO_TSI cleanup [POWERPC] rewrite mkprep and mkbugboot in sane C [POWERPC] maple/pci iomem annotations [POWERPC] powerpc oprofile __user annotations [POWERPC] cell spufs iomem annotations [POWERPC] NULL noise removal: spufs [POWERPC] ppc math-emu needs -fno-builtin-fabs for math.c and fabs.c [POWERPC] update mpc8349_itx_defconfig and remove some debug settings [POWERPC] Always call cede in pseries dedicated idle loop [POWERPC] Fix loop logic in irq_alloc_virt() ...

[PATCH] restore parport_pc probing on powermac

2006-10-03T15:03:41Z

The last change for partport_pc did fix the common case for all PowerMacs, but it broke the case for PCI multiport IO cards. In fact, the config option CONFIG_PARPORT_PC_SUPERIO=y lead to a hard crash when cups probed the parport driver. It enables the winbond and smsc probing. Remove the PARPORT_BASE check again, parport_pc_find_nonpci_ports() will take care of it. All powerpc configs should have CONFIG_PARPORT_PC_SUPERIO=n, the code did not find anything on the chrp boards we tested it on. Tested on a G4/466 with a PCI card: 0001:10:13.0 Serial controller: Timedia Technology Co Ltd PCI2S550 (Dual 16550 UART) (rev 01) (prog-if 02 [16550]) Subsystem: Timedia Technology Co Ltd Unknown device 5079 Control: I/O+ Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping+ SERR- FastB2B- Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Adam Belay Cc: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

[PATCH] namespaces: utsname: use init_utsname when appropriate

2006-10-02T14:57:21Z

In some places, particularly drivers and __init code, the init utsns is the appropriate one to use. This patch replaces those with a the init_utsname helper. Changes: Removed several uses of init_utsname(). Hope I picked all the right ones in net/ipv4/ipconfig.c. These are now changed to utsname() (the per-process namespace utsname) in the previous patch (2/7) [akpm@osdl.org: CIFS fix] Signed-off-by: Serge E. Hallyn Cc: Kirill Korotaev Cc: "Eric W. Biederman" Cc: Herbert Poetzl Cc: Andrey Savochkin Cc: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

[PATCH] update legacy io handling for pmac

2006-10-01T07:39:23Z

ppc can boot one single binary on prep, chrp and pmac boards. ppc64 can boot one single binary on pseries and G5 boards. pmac has no legacy io, probing for PC style legacy hardware (or accessing the legacy io area regulary) may lead to a hard crash: * add check for parport_pc, exit on pmac. 32bit chrp has no ->check_legacy_ioport, the probe is always called. 64bit chrp has check_legacy_ioport, check for a "parallel" node * add check for isapnp, only PReP boards may have real ISA slots. 32bit PReP will have no ->check_legacy_ioport, the probe is always called. * update code in i8042_platform_init. Run ->check_legacy_ioport first, always call request_region. No functional change. Remove whitespace before i8042_reset init. Signed-off-by: Olaf Hering Acked-by: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Adam Belay Cc: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

[PATCH] pidspace: is_init()

2006-09-29T16:18:12Z

This is an updated version of Eric Biederman's is_init() patch. (http://lkml.org/lkml/2006/2/6/280). It applies cleanly to 2.6.18-rc3 and replaces a few more instances of ->pid == 1 with is_init(). Further, is_init() checks pid and thus removes dependency on Eric's other patches for now. Eric's original description: There are a lot of places in the kernel where we test for init because we give it special properties. Most significantly init must not die. This results in code all over the kernel test ->pid == 1. Introduce is_init to capture this case. With multiple pid spaces for all of the cases affected we are looking for only the first process on the system, not some other process that has pid == 1. Signed-off-by: Eric W. Biederman Signed-off-by: Sukadev Bhattiprolu Cc: Dave Hansen Cc: Serge Hallyn Cc: Cedric Le Goater Cc: Acked-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds