- 13 10月, 2007 40 次提交
-
-
* Convert {ide_hwif_t,ide_pci_device_t}->host_flag to be u16. * Add IDE_HFLAG_POST_SET_MODE host flag to indicate the need to program the host for the transfer mode after programming the device. Set it in au1xxx-ide, amd74xx, cs5530, cs5535, pdc202xx_new, sc1200, pmac and via82cxxx host drivers. * Add IDE_HFLAG_NO_SET_MODE host flag to indicate the need to completely skip programming of host/device for the transfer mode ("smart" hosts). Set it in it821x host driver and check it in ide_tune_dma(). * Add ide_set_pio_mode()/ide_set_dma_mode() helpers and convert all direct ->set_pio_mode/->speedproc users to use these helpers. * Move ide_config_drive_speed() calls from ->set_pio_mode/->speedproc methods to callers. * Rename ->speedproc method to ->set_dma_mode, make it void and update all implementations accordingly. * Update ide_set_xfer_rate() comments. * Unexport ide_config_drive_speed(). v2: * Fix issues noticed by Sergei: - export ide_set_dma_mode() instead of moving ->set_pio_mode abuse wrt to setting DMA modes from sc1200_set_pio_mode() to do_special() - check IDE_HFLAG_NO_SET_MODE in ide_tune_dma() - check for (hwif->set_pio_mode) == NULL in ide_set_pio_mode() - check for (hwif->set_dma_mode) == NULL in ide_set_dma_mode() - return -1 from ide_set_{pio,dma}_mode() if ->set_{pio,dma}_mode == NULL - don't set ->set_{pio,dma}_mode on it821x in "smart" mode - fix build problem in pmac.c - minor fixes in au1xxx-ide.c/cs5530.c/siimage.c - improve patch description Changes in behavior caused by this patch: - HDIO_SET_PIO_MODE ioctl would now return -ENOSYS for attempts to change PIO mode if it821x controller is in "smart" mode - removal of two debugging printk-s (from cs5530.c and sc1200.c) - transfer modes 0x00-0x07 passed from user space may be programmed twice on the device (not really an issue since 0x00 is not supported correctly by any host driver ATM, 0x01 is not supported at all and 0x02-0x07 are invalid) Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Check ide_config_drive_speed() return value. * While at also call ide_config_drive_speed() if the transfer mode is XFER_PIO_SLOW (this case happens iff the transfer mode has already been set on the device by ide-proc.c::set_xfer_rate()) and remove redundant setting of ->{init,current}_speed. * Bump driver version. Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Set transfer mode on the device before programming the host controller for the new timings (matches what auide_tune_chipset() is doing wrt DMA modes). Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Add "good DMA drives" hack for icside to ide-dma.c::ide_find_dma_mode() (in the long-term it should be either removed or generalized for all hosts). * Use ide_tune_dma() in icside.c::icside_dma_check(). This results in the following changes in behavior: - pre-EIDE SWDMA modes are now also respected - drive->autodma is checked instead of hwif->autodma (doesn't really matter as icside sets both to "1") * Make ide-dma.c::__ide_dma_good_drive() static and drop "__" prefix. Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
由 Benjamin Herrenschmidt 提交于
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Add missing initialization of hwif->autodma and drive->autodma to pmac_ide_setup_dma(). * Use ide_tune_dma() in pmac_ide_dma_check(). While at it: * Fix pmac_ide_dma_check() return value if DMA mode is not programmed (should be "-1" otherwise ide_set_dma() will try to enable DMA). * Remove unnecessary drive->using_dma fiddling (->dma_off_quietly is always called before ide_set_dma() call and ide_set_dma() calls ->ide_dma_on if ->ide_dma_check returns "0"). v2: * No reason to blacklist all ide_floppy devices and the old code was always enabling DMA anyway (without even programming controller/device if the device was ide_floppy). Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Use ide_config_drive_speed() instead of pmac_ide_do_setfeature() and remove the latter, also ide-iops.c::__ide_wait_stat() could be static again. Since for IDE PMAC host driver IDE_CONTROL_REG is always true, device's ->quirk_list is always zero and ->ide_dma_host_{on,off} are nops than the only changes in behavior are: * if PIO mode is set then ->dma_off_queitly is called to disable DMA * if setting transfer mode fails ide_dump_status() is called to dump status v2: * IDE PMAC controllers allow separate PIO and DMA timings and PPC userland depends on this fact, and calls "hdparm -p" without calling "hdparm -d". Therefore to compensate for DMA being disabled by ide_config_drive_speed() for PIO modes: - add IDE_HFLAG_SET_PIO_MODE_KEEP_DMA flag and set it in PMAC host driver - add handling of the new flag to ide-io.c::do_special() Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Upper layers are responsible for controlling nIEN so don't clear nIEN after command execution in pmac_ide_do_setfeature(). Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Use __ide_wait_stat() instead of wait_for_ready() in pmac_ide_do_setfeature(). While at it do following changes to match __ide_wait_stat() call in ide_config_drive_speed(): * Wait WAIT_CMD time (20 sec) instead of 2 sec for device to clear BUSY_STAT. * Check DRQ_STAT bit (shouldn't be set for good device status). Also remove no longer needed wait_for_ready() from ide-iops.c. Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
Don't check for good device status before executing the command in pmac_ide_do_setfeature() (ide_config_drive_speed() doesn't do this). It is a job of upper layers to guarantee that the device is ready to accept new command before we get here. Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Split off checking of the status register from ide_wait_stat() to __ide_wait_stat() helper. * Use the new helper in ide_config_drive_speed(). The only change in the functionality is that the function now fails if after 20 sec (WAIT_CMD) device is still busy (BUSY_STAT bit is set) while previously instead of failing the function continued with checking for the correct device status (which would give the device additional 10 usec to clear BUSY_STAT bit). * Remove stale comment for ide_config_drive_speed(). * Remove duplicate comment for ide_wait_stat() from <linux/ide.h>. Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Fix pmac_ide_dma_check() to use pmac_ide_tune_chipset() and remove no longer necessary pmac_ide_{m,u}dma_enable(). * While at it remove some dead code from pmac_ide_dma_check() (leftovers from conversion to use ide_max_dma_mode()). There should be no functionality changes caused by this patch. v2: * Fix compile by replacing "id" with "drive->id" in pmac_ide_dma_check() (Noticed by Ben). Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
pmac_ide_do_setfeature() contains matching nIEN setting/clearing so this Device Control register messing in pmac_ide_dma_check() is totally unnecessary. Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Move adjusting of cycle time for devices providing explicit DMA cycle time from pmac_ide_mdma_enable() to set_timings_mdma(). * Remove no longer needed drive_cycle_time argument. * BUG() if unsupported speed argument value is passed (shouldn't happen). * Matching access/recovery timings always exist so remove redundant check. * Make set_timings_mdma() void. * Update pmac_ide_tune_chipset()'s comment. Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
* Don't check check for pmif == NULL (it should never be NULL if we got here). * Make a local copy of the timings and set the pmif->timings[] only after setting the transfer mode on the device (otherwise SELECT_DRIVE() call in pmac_ide_do_setfeature() will program new timings before the transfer mode is set on the device - this was pointed out by Sergei). This change makes pmac_ide_tune_chipset() behavior match this of pmac_ide_{m,u}dma_enable(). Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
kauai_lookup_timing() should always return non-zero return value: * BUG() in kauai_lookup_timing() if the timing info cannot be found. * Remove code checking for zero return value from all callers. Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: NSergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
由 Adrian Bunk 提交于
This patch removes the unused EXPORT_SYMBOL_GPL(ide_acpi_set_state) Signed-off-by: NAdrian Bunk <bunk@kernel.org> Cc: Shaohua Li <shaohua.li@intel.com> Cc: Len Brown <lenb@kernel.org> Signed-off-by: NAndrew Morton <akpm@linux-foundation.org> Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
We need to set hwif->chipset or IDE PCI host drivers may try to claim our ide_hwifs[] slot. Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
-
由 Jeff Garzik 提交于
Fix DMI const-ification fallout that appeared when merging subsystem trees. Signed-off-by: NJeff Garzik <jgarzik@redhat.com> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Nick Piggin 提交于
According to latest memory ordering specification documents from Intel and AMD, both manufacturers are committed to in-order loads from cacheable memory for the x86 architecture. Hence, smp_rmb() may be a simple barrier. Also according to those documents, and according to existing practice in Linux (eg. spin_unlock doesn't enforce ordering), stores to cacheable memory are visible in program order too. Special string stores are safe -- their constituent stores may be out of order, but they must complete in order WRT surrounding stores. Nontemporal stores to WB memory can go out of order, and so they should be fenced explicitly to make them appear in-order WRT other stores. Hence, smp_wmb() may be a simple barrier. http://developer.intel.com/products/processor/manuals/318147.pdf http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/24593.pdf In userspace microbenchmarks on a core2 system, fence instructions range anywhere from around 15 cycles to 50, which may not be totally insignificant in performance critical paths (code size will go down too). However the primary motivation for this is to have the canonical barrier implementation for x86 architecture. smp_rmb on buggy pentium pros remains a locked op, which is apparently required. Signed-off-by: NNick Piggin <npiggin@suse.de> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Nick Piggin 提交于
wmb() on x86 must always include a barrier, because stores can go out of order in many cases when dealing with devices (eg. WC memory). Signed-off-by: NNick Piggin <npiggin@suse.de> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Nick Piggin 提交于
movnt* instructions are not strongly ordered with respect to other stores, so if we are to assume stores are strongly ordered in the rest of the 64 bit code, we must fence these off (see similar examples in 32 bit code). [ The AMD memory ordering document seems to say that nontemporal stores can also pass earlier regular stores, so maybe we need sfences _before_ movnt* everywhere too? ] Signed-off-by: NNick Piggin <npiggin@suse.de> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Linus Torvalds 提交于
IOW, it needs to depend on both CONFIG_BLOCK and CONFIG_COMPAT. Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev由 Linus Torvalds 提交于
* 'upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev: (119 commits) [libata] struct pci_dev related cleanups libata: use ata_exec_internal() for PMP register access libata: implement ATA_PFLAG_RESETTING libata: add @timeout to ata_exec_internal[_sg]() ahci: fix notification handling ahci: clean up PORT_IRQ_BAD_PMP enabling ahci: kill leftover from enabling NCQ over PMP libata: wrap schedule_timeout_uninterruptible() in loop libata: skip suppress reporting if ATA_EHI_QUIET libata: clear ehi description after initial host report pata_jmicron: match vendor and class code only libata: add ST9160821AS / 3.ALD to NCQ blacklist pata_acpi: ACPI driver support libata-core: Expose gtm methods for driver use libata: add HDT722516DLA380 to NCQ blacklist libata: blacklist NCQ on Seagate Barracuda ST380817AS [libata] Turn on ACPI by default libata_scsi: Fix ATAPI transfer lengths libata: correct handling of SRST reset sequences libata: Integrate ACPI-based PATA/SATA hotplug - version 5 ...
-
由 Andi Kleen 提交于
Since there is no x86-64 architecture anymore it cannot be maintained. Signed-off-by: NAndi Kleen <ak@suse.de> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Linus Torvalds 提交于
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6: (37 commits) PCI: merge almost all of pci_32.h and pci_64.h together PCI: X86: Introduce and enable PCI domain support PCI: Add 'nodomains' boot option, and pci_domains_supported global PCI: modify PCI bridge control ISA flag for clarity PCI: use _CRS for PCI resource allocation PCI: avoid P2P prefetch window for expansion ROMs PCI: skip ISA ioresource alignment on some systems PCI: remove transparent bridge sizing pci: write file size to inode on proc bus file write pci: use size stored in proc_dir_entry for proc bus files pci: implement "pci=noaer" PCI: fix IDE legacy mode resources MSI: Use correct data offset for 32-bit MSI in read_msi_msg() PCI: Fix incorrect argument order to list_add_tail() in PCI dynamic ID code PCI: i386: Compaq EVO N800c needs PCI bus renumbering PCI: Remove no longer correct documentation regarding MSI vector assignment PCI: re-enable onboard sound on "MSI K8T Neo2-FIR" PCI: quirk_vt82c586_acpi: Omit reading PCI revision ID PCI: quirk amd_8131_mmrbc: Omit reading pci revision ID cpqphp: Use PCI_CLASS_REVISION instead of PCI_REVISION_ID for read ...
-
由 Linus Torvalds 提交于
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6: (75 commits) PM: merge device power-management source files sysfs: add copyrights kobject: update the copyrights kset: add some kerneldoc to help describe what these strange things are Driver core: rename ktype_edd and ktype_efivar Driver core: rename ktype_driver Driver core: rename ktype_device Driver core: rename ktype_class driver core: remove subsystem_init() sysfs: move sysfs file poll implementation to sysfs_open_dirent sysfs: implement sysfs_open_dirent sysfs: move sysfs_dirent->s_children into sysfs_dirent->s_dir sysfs: make sysfs_root a regular directory dirent sysfs: open code sysfs_attach_dentry() sysfs: make s_elem an anonymous union sysfs: make bin attr open get active reference of parent too sysfs: kill unnecessary NULL pointer check in sysfs_release() sysfs: kill unnecessary sysfs_get() in open paths sysfs: reposition sysfs_dirent->s_mode. sysfs: kill sysfs_update_file() ...
-
由 Linus Torvalds 提交于
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6: (142 commits) USB: fix race in autosuspend reschedule atmel_usba_udc: Keep track of the device status USB: Nikon D40X unusual_devs entry USB: serial core should respect driver requirements USB: documentation for USB power management USB: skip autosuspended devices during system resume USB: mutual exclusion for EHCI init and port resets USB: allow usbstorage to have LUNS greater than 2Tb USB: Adding support for SHARP WS011SH to ipaq.c USB: add atmel_usba_udc driver USB: ohci SSB bus glue USB: ehci build fixes on au1xxx, ppc-soc USB: add runtime frame_no quirk for big-endian OHCI USB: funsoft: Fix termios USB: visor: termios bits USB: unusual_devs entry for Nikon DSC D2Xs USB: re-remove <linux/usb_sl811.h> USB: move <linux/usb_gadget.h> to <linux/usb/gadget.h> USB: Export URB statistics for powertop USB: serial gadget: Disable endpoints on unload ...
-
由 Linus Torvalds 提交于
* master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreq: [CPUFREQ] Don't take semaphore in cpufreq_quick_get() [CPUFREQ] Support different families in fid/did to frequency conversion [CPUFREQ] cpufreq_stats: misc cpuinit section annotations [CPUFREQ] implement !CONFIG_CPU_FREQ stub for cpufreq_unregister_notifier() [CPUFREQ] mark hotplug notifier callback as __cpuinit [CPUFREQ] Only check for transition latency on problematic governors (kconfig fix) [CPUFREQ] allow ondemand and conservative cpufreq governors to be used as default [CPUFREQ] move policy's governor initialisation out of low-level drivers into cpufreq core [CPUFREQ] Longhaul - Add support for PM133 northbridge [CPUFREQ] x86: use num_online_nodes to get physical cpus numbers for
-
git://git.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-x86由 Linus Torvalds 提交于
* git://git.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-x86: (40 commits) x86: HPET add another ICH7 PCI id x86: HPET force enable ICH5 suspend/resume fix x86: HPET force enable for ICH5 x86: HPET try to activate force detected hpet x86: HPET force enable o ICH7 and later x86: HPET restructure hpet code for hpet force enable clock events: allow replacement of broadcast timer i386/x8664: cleanup the shared hpet code i386: Remove the useless #ifdef in i8253.h ACPI: remove the now unused ifdef code jiffies: remove unused macros x86_64: cleanup apic.c after clock events switch x86_64: remove now unused code x86: unify timex.h variants x86: kill 8253pit.h x86: disable apic timer for AMD C1E enabled CPUs x86: Fix irq0 / local apic timer accounting x86_64: convert to clock events x86_64: Add (not yet used) clock event functions x86_64: prepare idle loop for dynamic ticks ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2由 Linus Torvalds 提交于
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2: (23 commits) ocfs2: Optionally return filldir errors ocfs2: Write support for directories with inline data ocfs2: Read support for directories with inline data ocfs2: Write support for inline data ocfs2: Read support for inline data ocfs2: Structure updates for inline data ocfs2: Cleanup dirent size check ocfs2: Rename cleanups ocfs2: Provide convenience function for ino lookup ocfs2: Implement ocfs2_empty_dir() as a caller of ocfs2_dir_foreach() ocfs2: Remove open coded readdir() ocfs2: Pass raw u64 to filldir ocfs2: Abstract out core dir listing functionality ocfs2: Move directory manipulation code into dir.c ocfs2: Small refactor of truncate zeroing code ocfs2: move nonsparse hole-filling into ocfs2_write_begin() ocfs2: Sync ocfs2_fs.h with ocfs2-tools [PATCH] fs/ocfs2/: removed unneeded initial value and function's return value ocfs2: Implement show_options() ocfs2: Clear slot map when umounting a local volume ...
-
由 Linus Torvalds 提交于
* 'isdn-cleanups' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/misc-2.6: [ISDN] HiSax diva: split setup into three smaller functions [ISDN] HiSax sedlbauer: move ISAPNP and PCI code into functions of their own [ISDN] HiSax elsa: split huge setup function into four smaller functions [ISDN] HiSax avm_pci: split setup into three smaller functions [ISDN] Remove CONFIG_PCI ifdefs from 100% PCI source code
-
由 Greg Kroah-Hartman 提交于
It was just duplicated code... Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Jeff Garzik 提交于
* fix bug in pci_read() and pci_write() which prevented PCI domain support from working (hardcoded domain 0). * unconditionally enable CONFIG_PCI_DOMAINS * implement pci_domain_nr() and pci_proc_domain(), as required of all arches when CONFIG_PCI_DOMAINS is enabled. * store domain in struct pci_sysdata, as assigned by ACPI * support "pci=nodomains" Signed-off-by: NJeff Garzik <jgarzik@redhat.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Jeff Garzik 提交于
* Introduce pci_domains_supported global, hardcoded to zero if !CONFIG_PCI_DOMAINS. * Introduce 'nodomains' boot option, which clears pci_domains_supported on platforms that enable it by default (x86, x86-64, and others when they are converted to use this). Signed-off-by: NJeff Garzik <jgarzik@redhat.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Gary Hade 提交于
Modify PCI Bridge Control ISA flag for clarity This patch changes PCI_BRIDGE_CTL_NO_ISA to PCI_BRIDGE_CTL_ISA and modifies it's clarifying comment and locations where used. The change reduces the chance of future confusion since it makes the set/unset meaning of the bit the same in both the bridge control register and bridge_ctl field of the pci_bus struct. Signed-off-by: NGary Hade <garyhade@us.ibm.com> Acked-by: NLinas Vepstas <linas@austin.ibm.com> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Gary Hade 提交于
Use _CRS for PCI resource allocation This patch resolves an issue where incorrect PCI memory and i/o ranges are being assigned to hotplugged PCI devices on some IBM systems. The resource mis-allocation not only makes the PCI device unuseable but often makes the entire system unuseable due to resulting machine checks. The hotplug capable PCI slots on the affected systems are not located under a standard P2P bridge but are instead located under PCI root bridges or subtractive decode P2P bridges. For example, the IBM x3850 contains 2 hotplug capable PCI-X slots and 4 hotplug capable PCIe slots with the PCI-X slots each located under a PCI root bridge and the PCIe slots each located under a subtractive decode P2P bridge. The current i386/x86_64 PCI resource allocation code does not use _CRS returned resource information. No other resource information source is available for slots that are not below a standard P2P bridge so incorrect ranges are being allocated from e820 hole causing the bad result. This patch causes the kernel to use _CRS returned resource info. It is roughly based on a change provided by Matthew Wilcox for the ia64 kernel in 2005. Due to possible buggy BIOS factor and possible yet to be discovered kernel issues the function is disabled by default and can be enabled with pci=use_crs. Signed-off-by: NGary Hade <gary.hade@us.ibm.com> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Gary Hade 提交于
Avoid creating P2P prefetch window for expansion ROMs Because of the future possibility that P2P prefetch windows will contain address ranges above 4GB some BIOSes are providing space in the P2P non-prefetch windows for expansion ROMs. This is due to expansion ROM BAR 32-bit limitation. When expansion ROM BARs without BIOS assigned address(es) are currently found behind a P2P bridge, the kernel attempts to create a P2P prefetch window for them even though space for them has already been provided in the non-prefetch window. _CRS on some systems with certain resource conservation conscious BIOSes may not provide the extra 1MB or more memory resource needed for the expansion ROM motivated prefetch window causing resource allocation errors. This change corrects the problem by removing IORESOURCE_PREFETCH from the expansion ROM flags initialization. It also removes IORESOURCE_CACHEABLE which seems inappropriate if only non-cacheable memory is available. Signed-off-by: NGary Hade <gary.hade@us.ibm.com> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-