summaryrefslogtreecommitdiffstats
path: root/drivers/platform/x86/intel_ips.c
Commit message (Collapse)AuthorAgeFilesLines
* x86: remove 32-bit versions of readq()/writeq()Roland Dreier2011-05-251-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The presense of a writeq() implementation on 32-bit x86 that splits the 64-bit write into two 32-bit writes turns out to break the mpt2sas driver (and in general is risky for drivers as was discussed in <http://lkml.kernel.org/r/adaab6c1h7c.fsf@cisco.com>). To fix this, revert 2c5643b1c5c7 ("x86: provide readq()/writeq() on 32-bit too") and follow-on cleanups. This unfortunately leads to pushing non-atomic definitions of readq() and write() to various x86-only drivers that in the meantime started using the definitions in the x86 version of <asm/io.h>. However as discussed exhaustively, this is actually the right thing to do, because the right way to split a 64-bit transaction is hardware dependent and therefore belongs in the hardware driver (eg mpt2sas needs a spinlock to make sure no other accesses occur in between the two halves of the access). Build tested on 32- and 64-bit x86 allmodconfig. Link: http://lkml.kernel.org/r/x86-32-writeq-is-broken@mdm.bga.com Acked-by: Hitoshi Mitake <h.mitake@gmail.com> Cc: Kashyap Desai <Kashyap.Desai@lsi.com> Cc: Len Brown <lenb@kernel.org> Cc: Ravi Anand <ravi.anand@qlogic.com> Cc: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Cc: Matthew Garrett <mjg@redhat.com> Cc: Jason Uhlenkott <juhlenko@akamai.com> Acked-by: James Bottomley <James.Bottomley@parallels.com> Acked-by: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* ips: use interruptible waits in ips-monitorJesse Barnes2011-03-281-1/+1
| | | | | | | | | | This is what I intended to do since: 1) the driver handles variable waits just fine, and 2) interruptible waits aren't reported as load in the load avg. Reported-and-tested-by: Andreas Hartmann <andihartmann@freenet.de> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* intel_ips: fix sparse non-ANSI function warningRandy Dunlap2011-01-101-1/+1
| | | | | | | | | | Fix sparse warning for non-ANSI function declaration: drivers/platform/x86/intel_ips.c:1477:25: warning: non-ANSI function declaration of function 'ips_link_to_i915_driver' Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Matthew Garrett <mjg@redhat.com> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* drm/i915, intel_ips: When i915 loads after IPS, make IPS relink to i915.Eric Anholt2010-12-231-3/+33
| | | | | | | | | | | | | | The IPS driver is designed to be able to run detached from i915 and just not enable GPU turbo in that case, in order to avoid module dependencies between the two drivers. This means that we don't know what the load order between the two is going to be, and we had previously only supported IPS after (optionally) i915, but not i915 after IPS. If the wrong order was chosen, you'd get no GPU turbo, and something like half the possible graphics performance. Signed-off-by: Eric Anholt <eric@anholt.net> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: stable@kernel.org
* IPS driver: Fix limit clamping when reducing CPU powerMatthew Garrett2010-10-051-1/+1
| | | | | | | Values here are in internal units rather than Watts, so we shouldn't perform any conversion. Signed-off-by: Matthew Garrett <mjg@redhat.com>
* [PATCH 2/2] IPS driver: disable CPU turboJesse Barnes2010-10-051-3/+15
| | | | | | | | | | The undocumented interface we're using for reading CPU power seems to be overreporting power. Until we figure out how to correct it, disable CPU turbo and power reporting to be safe. This will keep the CPU within default limits and still allow us to increase GPU frequency as needed. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* IPS driver: apply BIOS provided CPU limit if different from defaultJesse Barnes2010-10-051-3/+4
| | | | | | | | | The BIOS may hand us a lower CPU power limit than the default for a given SKU. We should use it in case the platform isn't designed to dissapate the full TDP of a given part. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* intel_ips -- ensure we do not enable gpu turbo mode without driver linkageAndy Whitcroft2010-10-051-2/+4
| | | | | | | | | | | | | | | | Both when polling the current turbo status (in poll_turbo_status mode) and when handling thermal events (in ips_irq_handler) the current status of GPU turbo is updated to match the hardware status. However if during driver initialisation we were unable aquire linkage to the i915 driver enabling GPU turbo will lead to an oops on the first attempt to determine GPU busy status. Ensure that we do not enable GPU turbo unless we have driver linkage. BugLink: http://bugs.launchpad.net/bugs/632430 Cc: stable@kernel.org Signed-off-by: Andy Whitcroft <apw@canonical.com> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* intel_ips: Print MCP limit exceeded values.Tim Gardner2010-10-051-6/+18
| | | | | | | | | Print some interesting values when MCP limits are exceeded. Signed-off-by: Tim Gardner <tim.gardner@canonical.com> Cc: Matthew Garrett <mjg@redhat.com> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* IPS driver: verify BIOS provided limitsJesse Barnes2010-10-051-0/+23
| | | | | | | | They're optional. If not present or sane, we should use the CPU defaults. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* IPS driver: don't toggle CPU turbo on unsupported CPUsJesse Barnes2010-10-051-5/+9
| | | | | | | | | If the CPU doesn't support turbo, don't try to enable/disable it. http://bugzilla.kernel.org/show_bug.cgi?id=18742 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* NULL pointer might be used in ips_monitor()minskey guo2010-10-051-9/+13
| | | | | | | | | | | | The patch is to create ips_adjust thread before ips_monitor begins to run because the latter will kthread_stop() or wake up the former via ips->adjust pointer. Without this change, it is possible that ips->adjust is NULL when kthread_stop() or wake_up_process() is called in ips_monitor(). Signed-off-by: minskey guo <chaohong.guo@intel.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* Release symbol on error-handling path of ips_get_i915_syms()minskey guo2010-10-051-1/+1
| | | | | | | | | In ips_get_i915_syms(), the symbol i915_gpu_busy() is not released when error occurs. Signed-off-by: minskey guo <chaohong.guo@intel.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* old_cpu_power is wrongly divided by 65535 in ips_monitor()minskey guo2010-10-051-1/+1
| | | | | | | | | The variable old_cpu_power is used to save the value of THM_CEC register. In get_cpu_power(), it will be divided by 65535. Signed-off-by: minskey guo <chaohong.guo@intel.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* seqno mask of THM_ITV register is 16bitminskey guo2010-10-051-1/+1
| | | | | | | | | The mask of sequence number in THM_ITV register is 16bit width instead of 8bit. Signed-off-by: minskey guo <chaohong.guo@intel.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* intel_ips: potential null dereferenceDan Carpenter2010-08-161-1/+3
| | | | | | | | | There is a potential NULL dereference of "limits." We can just return NULL earlier to avoid it. The caller already handles NULL returns. Signed-off-by: Dan Carpenter <error27@gmail.com> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* x86: intel_ips: do not use PCI resources before pci_enable_device()Kulikov Vasiliy2010-08-161-5/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | IRQ and resource[] may not have correct values until after PCI hotplug setup occurs at pci_enable_device() time. The semantic match that finds this problem is as follows: // <smpl> @@ identifier x; identifier request ~= "pci_request.*|pci_resource.*"; @@ ( * x->irq | * x->resource | * request(x, ...) ) ... *pci_enable_device(x) // </smpl> Signed-off-by: Kulikov Vasiliy <segooon@gmail.com> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* ips driver: make it less chattyJesse Barnes2010-08-031-2/+2
| | | | | | | | | We don't need a dev_warn when we exceed a thermal or power limit as we'll handle it appropriately by clamping down on the CPU, GPU or both as needed. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* X86: intel_ips, check for kzalloc properlyJiri Slaby2010-08-031-4/+5
| | | | | | | | | | | | | | | | | Stanse found that there are two NULL checks missing in ips_monitor. So check their value too and bail out appropriately if the allocation failed. While at it, add one more kfree to the fail path. It is not necessary now, but may be needed in the future when a new allocation is added. And for completeness. Also remove unneeded initialization of the variables. They are all set right after their declaration. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: Matthew Garrett <mjg@redhat.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
* IPS driver: add GPU busy and turbo checkingJesse Barnes2010-08-031-13/+17
| | | | | | | | | Be sure to enable GPU turbo by default at load time and check GPU busy and MCP exceeded status correctly. Also fix up CPU power comparison and work around buggy MCH temp reporting. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
* x86 platform driver: intelligent power sharing driverJesse Barnes2010-08-031-0/+1655
Intel Core i3/5 platforms with integrated graphics support both CPU and GPU turbo mode. CPU turbo mode is opportunistic: the CPU will use any available power to increase core frequencies if thermal headroom is available. The GPU side is more manual however; the graphics driver must monitor GPU power and temperature and coordinate with a core thermal driver to take advantage of available thermal and power headroom in the package. The intelligent power sharing (IPS) driver is intended to coordinate this activity by monitoring MCP (multi-chip package) temperature and power, allowing the CPU and/or GPU to increase their power consumption, and thus performance, when possible. The goal is to maximize performance within a given platform's TDP (thermal design point). Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Matthew Garrett <mjg@redhat.com>
OpenPOWER on IntegriCloud