OpenCloudOS-Kernel/arch/powerpc/kernel/setup_64.c

987 lines
25 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
*
* Common boot and setup code.
*
* Copyright (C) 2001 PPC64 Team, IBM Corp
*/
#include <linux/export.h>
#include <linux/string.h>
#include <linux/sched.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/reboot.h>
#include <linux/delay.h>
#include <linux/initrd.h>
#include <linux/seq_file.h>
#include <linux/ioport.h>
#include <linux/console.h>
#include <linux/utsname.h>
#include <linux/tty.h>
#include <linux/root_dev.h>
#include <linux/notifier.h>
#include <linux/cpu.h>
#include <linux/unistd.h>
#include <linux/serial.h>
#include <linux/serial_8250.h>
mm: remove include/linux/bootmem.h Move remaining definitions and declarations from include/linux/bootmem.h into include/linux/memblock.h and remove the redundant header. The includes were replaced with the semantic patch below and then semi-automated removal of duplicated '#include <linux/memblock.h> @@ @@ - #include <linux/bootmem.h> + #include <linux/memblock.h> [sfr@canb.auug.org.au: dma-direct: fix up for the removal of linux/bootmem.h] Link: http://lkml.kernel.org/r/20181002185342.133d1680@canb.auug.org.au [sfr@canb.auug.org.au: powerpc: fix up for removal of linux/bootmem.h] Link: http://lkml.kernel.org/r/20181005161406.73ef8727@canb.auug.org.au [sfr@canb.auug.org.au: x86/kaslr, ACPI/NUMA: fix for linux/bootmem.h removal] Link: http://lkml.kernel.org/r/20181008190341.5e396491@canb.auug.org.au Link: http://lkml.kernel.org/r/1536927045-23536-30-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-31 06:09:49 +08:00
#include <linux/memblock.h>
#include <linux/pci.h>
#include <linux/lockdep.h>
#include <linux/memory.h>
#include <linux/nmi.h>
#include <asm/debugfs.h>
#include <asm/io.h>
#include <asm/kdump.h>
#include <asm/prom.h>
#include <asm/processor.h>
#include <asm/pgtable.h>
#include <asm/smp.h>
#include <asm/elf.h>
#include <asm/machdep.h>
#include <asm/paca.h>
#include <asm/time.h>
#include <asm/cputable.h>
#include <asm/dt_cpu_ftrs.h>
#include <asm/sections.h>
#include <asm/btext.h>
#include <asm/nvram.h>
#include <asm/setup.h>
#include <asm/rtas.h>
#include <asm/iommu.h>
#include <asm/serial.h>
#include <asm/cache.h>
#include <asm/page.h>
#include <asm/mmu.h>
#include <asm/firmware.h>
#include <asm/xmon.h>
#include <asm/udbg.h>
#include <asm/kexec.h>
#include <asm/code-patching.h>
#include <asm/livepatch.h>
#include <asm/opal.h>
#include <asm/cputhreads.h>
#include <asm/hw_irq.h>
#include <asm/feature-fixups.h>
#include <asm/kup.h>
#include <asm/early_ioremap.h>
#include "setup.h"
int spinning_secondaries;
u64 ppc64_pft_size;
struct ppc64_caches ppc64_caches = {
.l1d = {
.block_size = 0x40,
.log_block_size = 6,
},
.l1i = {
.block_size = 0x40,
.log_block_size = 6
},
};
EXPORT_SYMBOL_GPL(ppc64_caches);
#if defined(CONFIG_PPC_BOOK3E) && defined(CONFIG_SMP)
void __init setup_tlb_core_data(void)
{
int cpu;
BUILD_BUG_ON(offsetof(struct tlb_core_data, lock) != 0);
for_each_possible_cpu(cpu) {
int first = cpu_first_thread_sibling(cpu);
/*
* If we boot via kdump on a non-primary thread,
* make sure we point at the thread that actually
* set up this TLB.
*/
if (cpu_first_thread_sibling(boot_cpuid) == first)
first = boot_cpuid;
paca_ptrs[cpu]->tcd_ptr = &paca_ptrs[first]->tcd;
/*
* If we have threads, we need either tlbsrx.
* or e6500 tablewalk mode, or else TLB handlers
* will be racy and could produce duplicate entries.
* Should we panic instead?
*/
WARN_ONCE(smt_enabled_at_boot >= 2 &&
!mmu_has_feature(MMU_FTR_USE_TLBRSRV) &&
book3e_htw_mode != PPC_HTW_E6500,
"%s: unsupported MMU configuration\n", __func__);
}
}
#endif
#ifdef CONFIG_SMP
static char *smt_enabled_cmdline;
/* Look for ibm,smt-enabled OF option */
void __init check_smt_enabled(void)
{
struct device_node *dn;
const char *smt_option;
/* Default to enabling all threads */
smt_enabled_at_boot = threads_per_core;
/* Allow the command line to overrule the OF option */
if (smt_enabled_cmdline) {
if (!strcmp(smt_enabled_cmdline, "on"))
smt_enabled_at_boot = threads_per_core;
else if (!strcmp(smt_enabled_cmdline, "off"))
smt_enabled_at_boot = 0;
else {
int smt;
int rc;
rc = kstrtoint(smt_enabled_cmdline, 10, &smt);
if (!rc)
smt_enabled_at_boot =
min(threads_per_core, smt);
}
} else {
dn = of_find_node_by_path("/options");
if (dn) {
smt_option = of_get_property(dn, "ibm,smt-enabled",
NULL);
if (smt_option) {
if (!strcmp(smt_option, "on"))
smt_enabled_at_boot = threads_per_core;
else if (!strcmp(smt_option, "off"))
smt_enabled_at_boot = 0;
}
of_node_put(dn);
}
}
}
/* Look for smt-enabled= cmdline option */
static int __init early_smt_enabled(char *p)
{
smt_enabled_cmdline = p;
return 0;
}
early_param("smt-enabled", early_smt_enabled);
#endif /* CONFIG_SMP */
/** Fix up paca fields required for the boot cpu */
static void __init fixup_boot_paca(void)
{
/* The boot cpu is started */
get_paca()->cpu_start = 1;
/* Allow percpu accesses to work until we setup percpu data */
get_paca()->data_offset = 0;
/* Mark interrupts disabled in PACA */
irq_soft_mask_set(IRQS_DISABLED);
}
static void __init configure_exceptions(void)
{
/*
* Setup the trampolines from the lowmem exception vectors
* to the kdump kernel when not using a relocatable kernel.
*/
setup_kdump_trampoline();
/* Under a PAPR hypervisor, we need hypercalls */
if (firmware_has_feature(FW_FEATURE_SET_MODE)) {
/* Enable AIL if possible */
pseries_enable_reloc_on_exc();
/*
* Tell the hypervisor that we want our exceptions to
* be taken in little endian mode.
*
* We don't call this for big endian as our calling convention
* makes us always enter in BE, and the call may fail under
* some circumstances with kdump.
*/
#ifdef __LITTLE_ENDIAN__
pseries_little_endian_exceptions();
#endif
} else {
/* Set endian mode using OPAL */
if (firmware_has_feature(FW_FEATURE_OPAL))
opal_configure_cores();
/* AIL on native is done in cpu_ready_for_interrupts() */
}
}
static void cpu_ready_for_interrupts(void)
{
/*
* Enable AIL if supported, and we are in hypervisor mode. This
* is called once for every processor.
*
* If we are not in hypervisor mode the job is done once for
* the whole partition in configure_exceptions().
*/
if (cpu_has_feature(CPU_FTR_HVMODE) &&
cpu_has_feature(CPU_FTR_ARCH_207S)) {
unsigned long lpcr = mfspr(SPRN_LPCR);
mtspr(SPRN_LPCR, lpcr | LPCR_AIL_3);
}
/*
* Set HFSCR:TM based on CPU features:
* In the special case of TM no suspend (P9N DD2.1), Linux is
* told TM is off via the dt-ftrs but told to (partially) use
* it via OPAL_REINIT_CPUS_TM_SUSPEND_DISABLED. So HFSCR[TM]
* will be off from dt-ftrs but we need to turn it on for the
* no suspend case.
*/
if (cpu_has_feature(CPU_FTR_HVMODE)) {
if (cpu_has_feature(CPU_FTR_TM_COMP))
mtspr(SPRN_HFSCR, mfspr(SPRN_HFSCR) | HFSCR_TM);
else
mtspr(SPRN_HFSCR, mfspr(SPRN_HFSCR) & ~HFSCR_TM);
}
/* Set IR and DR in PACA MSR */
get_paca()->kernel_msr = MSR_KERNEL;
}
unsigned long spr_default_dscr = 0;
void __init record_spr_defaults(void)
{
if (early_cpu_has_feature(CPU_FTR_DSCR))
spr_default_dscr = mfspr(SPRN_DSCR);
}
/*
* Early initialization entry point. This is called by head.S
* with MMU translation disabled. We rely on the "feature" of
* the CPU that ignores the top 2 bits of the address in real
* mode so we can access kernel globals normally provided we
* only toy with things in the RMO region. From here, we do
* some early parsing of the device-tree to setup out MEMBLOCK
* data structures, and allocate & initialize the hash table
* and segment tables so we can start running with translation
* enabled.
*
* It is this function which will call the probe() callback of
* the various platform types and copy the matching one to the
* global ppc_md structure. Your platform can eventually do
* some very early initializations from the probe() routine, but
* this is not recommended, be very careful as, for example, the
* device-tree is not accessible via normal means at this point.
*/
powerpc/64: Prevent stack protection in early boot The previous commit reduced the amount of code that is run before we setup a paca. However there are still a few remaining functions that run with no paca, or worse, with an arbitrary value in r13 that will be used as a paca pointer. In particular the stack protector canary is stored in the paca, so if stack protector is activated for any of these functions we will read the stack canary from wherever r13 points. If r13 happens to point outside of memory we will get a machine check / checkstop. For example if we modify initialise_paca() to trigger stack protection, and then boot in the mambo simulator with r13 poisoned in skiboot before calling the kernel: DEBUG: 19952232: (19952232): INSTRUCTION: PC=0xC0000000191FC1E8: [0x3C4C006D]: addis r2,r12,0x6D [fetch] DEBUG: 19952236: (19952236): INSTRUCTION: PC=0xC00000001807EAD8: [0x7D8802A6]: mflr r12 [fetch] FATAL ERROR: 19952276: (19952276): Check Stop for 0:0: Machine Check with ME bit of MSR off DEBUG: 19952276: (19952276): INSTRUCTION: PC=0xC0000000191FCA7C: [0xE90D0CF8]: ld r8,0xCF8(r13) [Instruction Failed] INFO: 19952276: (19952277): ** Execution stopped: Mambo Error, Machine Check Stop, ** systemsim % bt pc: 0xC0000000191FCA7C initialise_paca+0x54 lr: 0xC0000000191FC22C early_setup+0x44 stack:0x00000000198CBED0 0x0 +0x0 stack:0x00000000198CBF00 0xC0000000191FC22C early_setup+0x44 stack:0x00000000198CBF90 0x1801C968 +0x1801C968 So annotate the relevant functions to ensure stack protection is never enabled for them. Fixes: 06ec27aea9fc ("powerpc/64: add stack protector support") Cc: stable@vger.kernel.org # v4.20+ Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200320032116.1024773-2-mpe@ellerman.id.au
2020-03-20 11:21:16 +08:00
void __init __nostackprotector early_setup(unsigned long dt_ptr)
{
static __initdata struct paca_struct boot_paca;
/* -------- printk is _NOT_ safe to use here ! ------- */
powerpc/64: Setup a paca before parsing device tree etc. Currently we set up the paca after parsing the device tree for CPU features. Prior to that, r13 contains random data, which means there is random data in r13 while we're running the generic dt parsing code. This random data varies depending on whether we boot through a vmlinux or a zImage: for the vmlinux case it's usually around zero, but for zImages we see random values like 912a72603d420015. This is poor practice, and can also lead to difficult-to-debug crashes. For example, when kcov is enabled, the kcov instrumentation attempts to read preempt_count out of the current task, which goes via the paca. This then crashes in the zImage case. Similarly stack protector can cause crashes if r13 is bogus, by reading from the stack canary in the paca. To resolve this: - move the paca setup to before the CPU feature parsing. - because we no longer have access to CPU feature flags in paca setup, change the HV feature test in the paca setup path to consider the actual value of the MSR rather than the CPU feature. Translations get switched on once we leave early_setup, so I think we'd already catch any other cases where the paca or task aren't set up. Boot tested on a P9 guest and host. Fixes: fb0b0a73b223 ("powerpc: Enable kcov") Fixes: 06ec27aea9fc ("powerpc/64: add stack protector support") Cc: stable@vger.kernel.org # v4.20+ Reviewed-by: Andrew Donnellan <ajd@linux.ibm.com> Suggested-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Daniel Axtens <dja@axtens.net> [mpe: Reword comments & change log a bit to mention stack protector] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200320032116.1024773-1-mpe@ellerman.id.au
2020-03-20 11:21:15 +08:00
/*
* Assume we're on cpu 0 for now.
*
* We need to load a PACA very early for a few reasons.
*
* The stack protector canary is stored in the paca, so as soon as we
* call any stack protected code we need r13 pointing somewhere valid.
*
* If we are using kcov it will call in_task() in its instrumentation,
* which relies on the current task from the PACA.
*
* dt_cpu_ftrs_init() calls into generic OF/fdt code, as well as
* printk(), which can trigger both stack protector and kcov.
*
* percpu variables and spin locks also use the paca.
*
* So set up a temporary paca. It will be replaced below once we know
* what CPU we are on.
*/
initialise_paca(&boot_paca, 0);
setup_paca(&boot_paca);
fixup_boot_paca();
/* -------- printk is now safe to use ------- */
powerpc/64: Setup a paca before parsing device tree etc. Currently we set up the paca after parsing the device tree for CPU features. Prior to that, r13 contains random data, which means there is random data in r13 while we're running the generic dt parsing code. This random data varies depending on whether we boot through a vmlinux or a zImage: for the vmlinux case it's usually around zero, but for zImages we see random values like 912a72603d420015. This is poor practice, and can also lead to difficult-to-debug crashes. For example, when kcov is enabled, the kcov instrumentation attempts to read preempt_count out of the current task, which goes via the paca. This then crashes in the zImage case. Similarly stack protector can cause crashes if r13 is bogus, by reading from the stack canary in the paca. To resolve this: - move the paca setup to before the CPU feature parsing. - because we no longer have access to CPU feature flags in paca setup, change the HV feature test in the paca setup path to consider the actual value of the MSR rather than the CPU feature. Translations get switched on once we leave early_setup, so I think we'd already catch any other cases where the paca or task aren't set up. Boot tested on a P9 guest and host. Fixes: fb0b0a73b223 ("powerpc: Enable kcov") Fixes: 06ec27aea9fc ("powerpc/64: add stack protector support") Cc: stable@vger.kernel.org # v4.20+ Reviewed-by: Andrew Donnellan <ajd@linux.ibm.com> Suggested-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Daniel Axtens <dja@axtens.net> [mpe: Reword comments & change log a bit to mention stack protector] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200320032116.1024773-1-mpe@ellerman.id.au
2020-03-20 11:21:15 +08:00
/* Try new device tree based feature discovery ... */
if (!dt_cpu_ftrs_init(__va(dt_ptr)))
/* Otherwise use the old style CPU table */
identify_cpu(0, mfspr(SPRN_PVR));
/* Enable early debugging if any specified (see udbg.h) */
udbg_early_init();
udbg_printf(" -> %s(), dt_ptr: 0x%lx\n", __func__, dt_ptr);
/*
* Do early initialization using the flattened device
* tree, such as retrieving the physical memory map or
* calculating/retrieving the hash table size.
*/
early_init_devtree(__va(dt_ptr));
/* Now we know the logical id of our boot cpu, setup the paca. */
if (boot_cpuid != 0) {
/* Poison paca_ptrs[0] again if it's not the boot cpu */
memset(&paca_ptrs[0], 0x88, sizeof(paca_ptrs[0]));
}
setup_paca(paca_ptrs[boot_cpuid]);
fixup_boot_paca();
/*
* Configure exception handlers. This include setting up trampolines
* if needed, setting exception endian mode, etc...
*/
configure_exceptions();
/*
* Configure Kernel Userspace Protection. This needs to happen before
* feature fixups for platforms that implement this using features.
*/
setup_kup();
/* Apply all the dynamic patching */
apply_feature_fixups();
setup_feature_keys();
early_ioremap_setup();
/* Initialize the hash table or TLB handling */
early_init_mmu();
/*
* After firmware and early platform setup code has set things up,
* we note the SPR values for configurable control/performance
* registers, and use those as initial defaults.
*/
record_spr_defaults();
/*
* At this point, we can let interrupts switch to virtual mode
* (the MMU has been setup), so adjust the MSR in the PACA to
* have IR and DR set and enable AIL if it exists
*/
cpu_ready_for_interrupts();
/*
* We enable ftrace here, but since we only support DYNAMIC_FTRACE, it
* will only actually get enabled on the boot cpu much later once
* ftrace itself has been initialized.
*/
this_cpu_enable_ftrace();
udbg_printf(" <- %s()\n", __func__);
#ifdef CONFIG_PPC_EARLY_DEBUG_BOOTX
/*
* This needs to be done *last* (after the above udbg_printf() even)
*
* Right after we return from this function, we turn on the MMU
* which means the real-mode access trick that btext does will
* no longer work, it needs to switch to using a real MMU
* mapping. This call will ensure that it does
*/
btext_map();
#endif /* CONFIG_PPC_EARLY_DEBUG_BOOTX */
}
#ifdef CONFIG_SMP
void early_setup_secondary(void)
{
/* Mark interrupts disabled in PACA */
irq_soft_mask_set(IRQS_DISABLED);
/* Initialize the hash table or TLB handling */
early_init_mmu_secondary();
/* Perform any KUP setup that is per-cpu */
setup_kup();
/*
* At this point, we can let interrupts switch to virtual mode
* (the MMU has been setup), so adjust the MSR in the PACA to
* have IR and DR set.
*/
cpu_ready_for_interrupts();
}
#endif /* CONFIG_SMP */
void panic_smp_self_stop(void)
{
hard_irq_disable();
spin_begin();
while (1)
spin_cpu_relax();
}
#if defined(CONFIG_SMP) || defined(CONFIG_KEXEC_CORE)
static bool use_spinloop(void)
{
if (IS_ENABLED(CONFIG_PPC_BOOK3S)) {
/*
* See comments in head_64.S -- not all platforms insert
* secondaries at __secondary_hold and wait at the spin
* loop.
*/
if (firmware_has_feature(FW_FEATURE_OPAL))
return false;
return true;
}
/*
* When book3e boots from kexec, the ePAPR spin table does
* not get used.
*/
return of_property_read_bool(of_chosen, "linux,booted-from-kexec");
}
void smp_release_cpus(void)
{
unsigned long *ptr;
int i;
if (!use_spinloop())
return;
/* All secondary cpus are spinning on a common spinloop, release them
* all now so they can start to spin on their individual paca
* spinloops. For non SMP kernels, the secondary cpus never get out
* of the common spinloop.
powerpc: Make it possible to move the interrupt handlers away from the kernel This changes the way that the exception prologs transfer control to the handlers in 64-bit kernels with the aim of making it possible to have the prologs separate from the main body of the kernel. Now, instead of computing the address of the handler by taking the top 32 bits of the paca address (to get the 0xc0000000........ part) and ORing in something in the bottom 16 bits, we get the base address of the kernel by doing a load from the paca and add an offset. This also replaces an mfmsr and an ori to compute the MSR value for the handler with a load from the paca. That makes it unnecessary to have a separate version of EXCEPTION_PROLOG_PSERIES that forces 64-bit mode. We can no longer use a direct branches in the exception prolog code, which means that the SLB miss handlers can't branch directly to .slb_miss_realmode any more. Instead we have to compute the address and do an indirect branch. This is conditional on CONFIG_RELOCATABLE; for non-relocatable kernels we use a direct branch as before. (A later change will allow CONFIG_RELOCATABLE to be set on 64-bit powerpc.) Since the secondary CPUs on pSeries start execution in the first 0x100 bytes of real memory and then have to get to wherever the kernel is, we can't use a direct branch to get there. Instead this changes __secondary_hold_spinloop from a flag to a function pointer. When it is set to a non-NULL value, the secondary CPUs jump to the function pointed to by that value. Finally this eliminates one code difference between 32-bit and 64-bit by making __secondary_hold be the text address of the secondary CPU spinloop rather than a function descriptor for it. Signed-off-by: Paul Mackerras <paulus@samba.org>
2008-08-30 09:40:24 +08:00
*/
ptr = (unsigned long *)((unsigned long)&__secondary_hold_spinloop
- PHYSICAL_START);
*ptr = ppc_function_entry(generic_secondary_smp_init);
/* And wait a bit for them to catch up */
for (i = 0; i < 100000; i++) {
mb();
HMT_low();
if (spinning_secondaries == 0)
break;
udelay(1);
}
pr_debug("spinning_secondaries = %d\n", spinning_secondaries);
}
#endif /* CONFIG_SMP || CONFIG_KEXEC_CORE */
/*
* Initialize some remaining members of the ppc64_caches and systemcfg
* structures
* (at least until we get rid of them completely). This is mostly some
* cache informations about the CPU that will be used by cache flush
* routines and/or provided to userland
*/
static void init_cache_info(struct ppc_cache_info *info, u32 size, u32 lsize,
u32 bsize, u32 sets)
{
info->size = size;
info->sets = sets;
info->line_size = lsize;
info->block_size = bsize;
info->log_block_size = __ilog2(bsize);
if (bsize)
info->blocks_per_page = PAGE_SIZE / bsize;
else
info->blocks_per_page = 0;
if (sets == 0)
info->assoc = 0xffff;
else
info->assoc = size / (sets * lsize);
}
static bool __init parse_cache_info(struct device_node *np,
bool icache,
struct ppc_cache_info *info)
{
static const char *ipropnames[] __initdata = {
"i-cache-size",
"i-cache-sets",
"i-cache-block-size",
"i-cache-line-size",
};
static const char *dpropnames[] __initdata = {
"d-cache-size",
"d-cache-sets",
"d-cache-block-size",
"d-cache-line-size",
};
const char **propnames = icache ? ipropnames : dpropnames;
const __be32 *sizep, *lsizep, *bsizep, *setsp;
u32 size, lsize, bsize, sets;
bool success = true;
size = 0;
sets = -1u;
lsize = bsize = cur_cpu_spec->dcache_bsize;
sizep = of_get_property(np, propnames[0], NULL);
if (sizep != NULL)
size = be32_to_cpu(*sizep);
setsp = of_get_property(np, propnames[1], NULL);
if (setsp != NULL)
sets = be32_to_cpu(*setsp);
bsizep = of_get_property(np, propnames[2], NULL);
lsizep = of_get_property(np, propnames[3], NULL);
if (bsizep == NULL)
bsizep = lsizep;
if (lsizep != NULL)
lsize = be32_to_cpu(*lsizep);
if (bsizep != NULL)
bsize = be32_to_cpu(*bsizep);
if (sizep == NULL || bsizep == NULL || lsizep == NULL)
success = false;
/*
* OF is weird .. it represents fully associative caches
* as "1 way" which doesn't make much sense and doesn't
* leave room for direct mapped. We'll assume that 0
* in OF means direct mapped for that reason.
*/
if (sets == 1)
sets = 0;
else if (sets == 0)
sets = 1;
init_cache_info(info, size, lsize, bsize, sets);
return success;
}
void __init initialize_cache_info(void)
{
struct device_node *cpu = NULL, *l2, *l3 = NULL;
u32 pvr;
/*
* All shipping POWER8 machines have a firmware bug that
* puts incorrect information in the device-tree. This will
* be (hopefully) fixed for future chips but for now hard
* code the values if we are running on one of these
*/
pvr = PVR_VER(mfspr(SPRN_PVR));
if (pvr == PVR_POWER8 || pvr == PVR_POWER8E ||
pvr == PVR_POWER8NVL) {
/* size lsize blk sets */
init_cache_info(&ppc64_caches.l1i, 0x8000, 128, 128, 32);
init_cache_info(&ppc64_caches.l1d, 0x10000, 128, 128, 64);
init_cache_info(&ppc64_caches.l2, 0x80000, 128, 0, 512);
init_cache_info(&ppc64_caches.l3, 0x800000, 128, 0, 8192);
} else
cpu = of_find_node_by_type(NULL, "cpu");
/*
* We're assuming *all* of the CPUs have the same
* d-cache and i-cache sizes... -Peter
*/
if (cpu) {
if (!parse_cache_info(cpu, false, &ppc64_caches.l1d))
pr_warn("Argh, can't find dcache properties !\n");
if (!parse_cache_info(cpu, true, &ppc64_caches.l1i))
pr_warn("Argh, can't find icache properties !\n");
/*
* Try to find the L2 and L3 if any. Assume they are
* unified and use the D-side properties.
*/
l2 = of_find_next_cache_node(cpu);
of_node_put(cpu);
if (l2) {
parse_cache_info(l2, false, &ppc64_caches.l2);
l3 = of_find_next_cache_node(l2);
of_node_put(l2);
}
if (l3) {
parse_cache_info(l3, false, &ppc64_caches.l3);
of_node_put(l3);
}
}
/* For use by binfmt_elf */
dcache_bsize = ppc64_caches.l1d.block_size;
icache_bsize = ppc64_caches.l1i.block_size;
cur_cpu_spec->dcache_bsize = dcache_bsize;
cur_cpu_spec->icache_bsize = icache_bsize;
}
/*
* This returns the limit below which memory accesses to the linear
* mapping are guarnateed not to cause an architectural exception (e.g.,
* TLB or SLB miss fault).
*
* This is used to allocate PACAs and various interrupt stacks that
* that are accessed early in interrupt handlers that must not cause
* re-entrant interrupts.
*/
__init u64 ppc64_bolted_size(void)
{
#ifdef CONFIG_PPC_BOOK3E
/* Freescale BookE bolts the entire linear mapping */
/* XXX: BookE ppc64_rma_limit setup seems to disagree? */
if (early_mmu_has_feature(MMU_FTR_TYPE_FSL_E))
return linear_map_top;
/* Other BookE, we assume the first GB is bolted */
return 1ul << 30;
#else
/* BookS radix, does not take faults on linear mapping */
if (early_radix_enabled())
return ULONG_MAX;
/* BookS hash, the first segment is bolted */
if (early_mmu_has_feature(MMU_FTR_1T_SEGMENT))
return 1UL << SID_SHIFT_1T;
return 1UL << SID_SHIFT;
#endif
}
static void *__init alloc_stack(unsigned long limit, int cpu)
{
void *ptr;
BUILD_BUG_ON(STACK_INT_FRAME_SIZE % 16);
ptr = memblock_alloc_try_nid(THREAD_SIZE, THREAD_ALIGN,
MEMBLOCK_LOW_LIMIT, limit,
early_cpu_to_node(cpu));
if (!ptr)
panic("cannot allocate stacks");
return ptr;
}
void __init irqstack_early_init(void)
{
u64 limit = ppc64_bolted_size();
unsigned int i;
/*
* Interrupt stacks must be in the first segment since we
* cannot afford to take SLB misses on them. They are not
* accessed in realmode.
*/
for_each_possible_cpu(i) {
softirq_ctx[i] = alloc_stack(limit, i);
hardirq_ctx[i] = alloc_stack(limit, i);
}
}
#ifdef CONFIG_PPC_BOOK3E
void __init exc_lvl_early_init(void)
{
unsigned int i;
for_each_possible_cpu(i) {
void *sp;
sp = alloc_stack(ULONG_MAX, i);
critirq_ctx[i] = sp;
paca_ptrs[i]->crit_kstack = sp + THREAD_SIZE;
sp = alloc_stack(ULONG_MAX, i);
dbgirq_ctx[i] = sp;
paca_ptrs[i]->dbg_kstack = sp + THREAD_SIZE;
sp = alloc_stack(ULONG_MAX, i);
mcheckirq_ctx[i] = sp;
paca_ptrs[i]->mc_kstack = sp + THREAD_SIZE;
}
if (cpu_has_feature(CPU_FTR_DEBUG_LVL_EXC))
patch_exception(0x040, exc_debug_debug_book3e);
}
#endif
/*
* Stack space used when we detect a bad kernel stack pointer, and
* early in SMP boots before relocation is enabled. Exclusive emergency
* stack for machine checks.
*/
void __init emergency_stack_init(void)
{
u64 limit;
unsigned int i;
/*
* Emergency stacks must be under 256MB, we cannot afford to take
* SLB misses on them. The ABI also requires them to be 128-byte
* aligned.
*
* Since we use these as temporary stacks during secondary CPU
* bringup, machine check, system reset, and HMI, we need to get
* at them in real mode. This means they must also be within the RMO
* region.
*
* The IRQ stacks allocated elsewhere in this file are zeroed and
* initialized in kernel/irq.c. These are initialized here in order
* to have emergency stacks available as early as possible.
*/
limit = min(ppc64_bolted_size(), ppc64_rma_size);
for_each_possible_cpu(i) {
paca_ptrs[i]->emergency_sp = alloc_stack(limit, i) + THREAD_SIZE;
#ifdef CONFIG_PPC_BOOK3S_64
/* emergency stack for NMI exception handling. */
paca_ptrs[i]->nmi_emergency_sp = alloc_stack(limit, i) + THREAD_SIZE;
/* emergency stack for machine check exception handling. */
paca_ptrs[i]->mc_emergency_sp = alloc_stack(limit, i) + THREAD_SIZE;
#endif
}
}
[PATCH] powerpc/64: per cpu data optimisations The current ppc64 per cpu data implementation is quite slow. eg: lhz 11,18(13) /* smp_processor_id() */ ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */ ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */ sldi 11,11,3 /* form index into __per_cpu_offset */ mr 10,9 ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */ ldx 0,10,9 /* load per cpu data */ 5 loads for something that is supposed to be fast, pretty awful. One reason for the large number of loads is that we have to synthesize 2 64bit constants (per_cpu__variable_name and __per_cpu_offset). By putting __per_cpu_offset into the paca we can avoid the 2 loads associated with it: ld 11,56(13) /* paca->data_offset */ ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */ ldx 0,9,11 /* load per cpu data Longer term we can should be able to do even better than 3 loads. If per_cpu__variable_name wasnt a 64bit constant and paca->data_offset was in a register we could cut it down to one load. A suggestion from Rusty is to use gcc's __thread extension here. In order to do this we would need to free up r13 (the __thread register and where the paca currently is). So far Ive had a few unsuccessful attempts at doing that :) The patch also allocates per cpu memory node local on NUMA machines. This patch from Rusty has been sitting in my queue _forever_ but stalled when I hit the compiler bug. Sorry about that. Finally I also only allocate per cpu data for possible cpus, which comes straight out of the x86-64 port. On a pseries kernel (with NR_CPUS == 128) and 4 possible cpus we see some nice gains: total used free shared buffers cached Mem: 4012228 212860 3799368 0 0 162424 total used free shared buffers cached Mem: 4016200 212984 3803216 0 0 162424 A saving of 3.75MB. Quite nice for smaller machines. Note: we now have to be careful of per cpu users that touch data for !possible cpus. At this stage it might be worth making the NUMA and possible cpu optimisations generic, but per cpu init is done so early we have to be careful that all architectures have their possible map setup correctly. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-11 10:16:44 +08:00
#ifdef CONFIG_SMP
#define PCPU_DYN_SIZE ()
static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align)
[PATCH] powerpc/64: per cpu data optimisations The current ppc64 per cpu data implementation is quite slow. eg: lhz 11,18(13) /* smp_processor_id() */ ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */ ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */ sldi 11,11,3 /* form index into __per_cpu_offset */ mr 10,9 ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */ ldx 0,10,9 /* load per cpu data */ 5 loads for something that is supposed to be fast, pretty awful. One reason for the large number of loads is that we have to synthesize 2 64bit constants (per_cpu__variable_name and __per_cpu_offset). By putting __per_cpu_offset into the paca we can avoid the 2 loads associated with it: ld 11,56(13) /* paca->data_offset */ ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */ ldx 0,9,11 /* load per cpu data Longer term we can should be able to do even better than 3 loads. If per_cpu__variable_name wasnt a 64bit constant and paca->data_offset was in a register we could cut it down to one load. A suggestion from Rusty is to use gcc's __thread extension here. In order to do this we would need to free up r13 (the __thread register and where the paca currently is). So far Ive had a few unsuccessful attempts at doing that :) The patch also allocates per cpu memory node local on NUMA machines. This patch from Rusty has been sitting in my queue _forever_ but stalled when I hit the compiler bug. Sorry about that. Finally I also only allocate per cpu data for possible cpus, which comes straight out of the x86-64 port. On a pseries kernel (with NR_CPUS == 128) and 4 possible cpus we see some nice gains: total used free shared buffers cached Mem: 4012228 212860 3799368 0 0 162424 total used free shared buffers cached Mem: 4016200 212984 3803216 0 0 162424 A saving of 3.75MB. Quite nice for smaller machines. Note: we now have to be careful of per cpu users that touch data for !possible cpus. At this stage it might be worth making the NUMA and possible cpu optimisations generic, but per cpu init is done so early we have to be careful that all architectures have their possible map setup correctly. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-11 10:16:44 +08:00
{
memblock: replace __alloc_bootmem_node with appropriate memblock_ API Use memblock_alloc_try_nid whenever goal (i.e. minimal address is specified) and memblock_alloc_node otherwise. Link: http://lkml.kernel.org/r/1536927045-23536-17-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-31 06:08:45 +08:00
return memblock_alloc_try_nid(size, align, __pa(MAX_DMA_ADDRESS),
memblock: replace BOOTMEM_ALLOC_* with MEMBLOCK variants Drop BOOTMEM_ALLOC_ACCESSIBLE and BOOTMEM_ALLOC_ANYWHERE in favor of identical MEMBLOCK definitions. Link: http://lkml.kernel.org/r/1536927045-23536-29-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-31 06:09:44 +08:00
MEMBLOCK_ALLOC_ACCESSIBLE,
memblock: replace __alloc_bootmem_node with appropriate memblock_ API Use memblock_alloc_try_nid whenever goal (i.e. minimal address is specified) and memblock_alloc_node otherwise. Link: http://lkml.kernel.org/r/1536927045-23536-17-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-31 06:08:45 +08:00
early_cpu_to_node(cpu));
}
[PATCH] powerpc/64: per cpu data optimisations The current ppc64 per cpu data implementation is quite slow. eg: lhz 11,18(13) /* smp_processor_id() */ ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */ ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */ sldi 11,11,3 /* form index into __per_cpu_offset */ mr 10,9 ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */ ldx 0,10,9 /* load per cpu data */ 5 loads for something that is supposed to be fast, pretty awful. One reason for the large number of loads is that we have to synthesize 2 64bit constants (per_cpu__variable_name and __per_cpu_offset). By putting __per_cpu_offset into the paca we can avoid the 2 loads associated with it: ld 11,56(13) /* paca->data_offset */ ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */ ldx 0,9,11 /* load per cpu data Longer term we can should be able to do even better than 3 loads. If per_cpu__variable_name wasnt a 64bit constant and paca->data_offset was in a register we could cut it down to one load. A suggestion from Rusty is to use gcc's __thread extension here. In order to do this we would need to free up r13 (the __thread register and where the paca currently is). So far Ive had a few unsuccessful attempts at doing that :) The patch also allocates per cpu memory node local on NUMA machines. This patch from Rusty has been sitting in my queue _forever_ but stalled when I hit the compiler bug. Sorry about that. Finally I also only allocate per cpu data for possible cpus, which comes straight out of the x86-64 port. On a pseries kernel (with NR_CPUS == 128) and 4 possible cpus we see some nice gains: total used free shared buffers cached Mem: 4012228 212860 3799368 0 0 162424 total used free shared buffers cached Mem: 4016200 212984 3803216 0 0 162424 A saving of 3.75MB. Quite nice for smaller machines. Note: we now have to be careful of per cpu users that touch data for !possible cpus. At this stage it might be worth making the NUMA and possible cpu optimisations generic, but per cpu init is done so early we have to be careful that all architectures have their possible map setup correctly. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-11 10:16:44 +08:00
static void __init pcpu_fc_free(void *ptr, size_t size)
{
memblock: replace free_bootmem{_node} with memblock_free The free_bootmem and free_bootmem_node are merely wrappers for memblock_free. Replace their usage with a call to memblock_free using the following semantic patch: @@ expression e1, e2, e3; @@ ( - free_bootmem(e1, e2) + memblock_free(e1, e2) | - free_bootmem_node(e1, e2, e3) + memblock_free(e2, e3) ) Link: http://lkml.kernel.org/r/1536927045-23536-24-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-31 06:09:21 +08:00
memblock_free(__pa(ptr), size);
}
[PATCH] powerpc/64: per cpu data optimisations The current ppc64 per cpu data implementation is quite slow. eg: lhz 11,18(13) /* smp_processor_id() */ ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */ ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */ sldi 11,11,3 /* form index into __per_cpu_offset */ mr 10,9 ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */ ldx 0,10,9 /* load per cpu data */ 5 loads for something that is supposed to be fast, pretty awful. One reason for the large number of loads is that we have to synthesize 2 64bit constants (per_cpu__variable_name and __per_cpu_offset). By putting __per_cpu_offset into the paca we can avoid the 2 loads associated with it: ld 11,56(13) /* paca->data_offset */ ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */ ldx 0,9,11 /* load per cpu data Longer term we can should be able to do even better than 3 loads. If per_cpu__variable_name wasnt a 64bit constant and paca->data_offset was in a register we could cut it down to one load. A suggestion from Rusty is to use gcc's __thread extension here. In order to do this we would need to free up r13 (the __thread register and where the paca currently is). So far Ive had a few unsuccessful attempts at doing that :) The patch also allocates per cpu memory node local on NUMA machines. This patch from Rusty has been sitting in my queue _forever_ but stalled when I hit the compiler bug. Sorry about that. Finally I also only allocate per cpu data for possible cpus, which comes straight out of the x86-64 port. On a pseries kernel (with NR_CPUS == 128) and 4 possible cpus we see some nice gains: total used free shared buffers cached Mem: 4012228 212860 3799368 0 0 162424 total used free shared buffers cached Mem: 4016200 212984 3803216 0 0 162424 A saving of 3.75MB. Quite nice for smaller machines. Note: we now have to be careful of per cpu users that touch data for !possible cpus. At this stage it might be worth making the NUMA and possible cpu optimisations generic, but per cpu init is done so early we have to be careful that all architectures have their possible map setup correctly. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-11 10:16:44 +08:00
static int pcpu_cpu_distance(unsigned int from, unsigned int to)
{
powerpc/numa: Fix percpu allocations to be NUMA aware In commit 8c272261194d ("powerpc/numa: Enable USE_PERCPU_NUMA_NODE_ID"), we switched to the generic implementation of cpu_to_node(), which uses a percpu variable to hold the NUMA node for each CPU. Unfortunately we neglected to notice that we use cpu_to_node() in the allocation of our percpu areas, leading to a chicken and egg problem. In practice what happens is when we are setting up the percpu areas, cpu_to_node() reports that all CPUs are on node 0, so we allocate all percpu areas on node 0. This is visible in the dmesg output, as all pcpu allocs being in group 0: pcpu-alloc: [0] 00 01 02 03 [0] 04 05 06 07 pcpu-alloc: [0] 08 09 10 11 [0] 12 13 14 15 pcpu-alloc: [0] 16 17 18 19 [0] 20 21 22 23 pcpu-alloc: [0] 24 25 26 27 [0] 28 29 30 31 pcpu-alloc: [0] 32 33 34 35 [0] 36 37 38 39 pcpu-alloc: [0] 40 41 42 43 [0] 44 45 46 47 To fix it we need an early_cpu_to_node() which can run prior to percpu being setup. We already have the numa_cpu_lookup_table we can use, so just plumb it in. With the patch dmesg output shows two groups, 0 and 1: pcpu-alloc: [0] 00 01 02 03 [0] 04 05 06 07 pcpu-alloc: [0] 08 09 10 11 [0] 12 13 14 15 pcpu-alloc: [0] 16 17 18 19 [0] 20 21 22 23 pcpu-alloc: [1] 24 25 26 27 [1] 28 29 30 31 pcpu-alloc: [1] 32 33 34 35 [1] 36 37 38 39 pcpu-alloc: [1] 40 41 42 43 [1] 44 45 46 47 We can also check the data_offset in the paca of various CPUs, with the fix we see: CPU 0: data_offset = 0x0ffe8b0000 CPU 24: data_offset = 0x1ffe5b0000 And we can see from dmesg that CPU 24 has an allocation on node 1: node 0: [mem 0x0000000000000000-0x0000000fffffffff] node 1: [mem 0x0000001000000000-0x0000001fffffffff] Cc: stable@vger.kernel.org # v3.16+ Fixes: 8c272261194d ("powerpc/numa: Enable USE_PERCPU_NUMA_NODE_ID") Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Reviewed-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-06-06 18:23:57 +08:00
if (early_cpu_to_node(from) == early_cpu_to_node(to))
return LOCAL_DISTANCE;
else
return REMOTE_DISTANCE;
}
powerpc: Optimise per cpu accesses on 64bit Now we dynamically allocate the paca array, it takes an extra load whenever we want to access another cpu's paca. One place we do that a lot is per cpu variables. A simple example: DEFINE_PER_CPU(unsigned long, vara); unsigned long test4(int cpu) { return per_cpu(vara, cpu); } This takes 4 loads, 5 if you include the actual load of the per cpu variable: ld r11,-32760(r30) # load address of paca pointer ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,9 # get offset into paca (each entry is 512 bytes) ld r0,0(r11) # load paca pointer add r3,r0,r3 # paca + offset ld r11,64(r3) # load paca[cpu].data_offset ldx r3,r9,r11 # load per cpu variable If we remove the ppc64 specific per_cpu_offset(), we get the generic one which indexes into a statically allocated array. This removes one load and one add: ld r11,-32760(r30) # load address of __per_cpu_offset ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,3 # get offset into __per_cpu_offset (each entry 8 bytes) ldx r11,r11,r3 # load __per_cpu_offset[cpu] ldx r3,r9,r11 # load per cpu variable Having all the offsets in one array also helps when iterating over a per cpu variable across a number of cpus, such as in the scheduler. Before we would need to load one paca cacheline when calculating each per cpu offset. Now we have 16 (128 / sizeof(long)) per cpu offsets in each cacheline. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-06-01 02:45:11 +08:00
unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
EXPORT_SYMBOL(__per_cpu_offset);
void __init setup_per_cpu_areas(void)
{
const size_t dyn_size = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE;
size_t atom_size;
unsigned long delta;
unsigned int cpu;
int rc;
/*
* Linear mapping is one of 4K, 1M and 16M. For 4K, no need
* to group units. For larger mappings, use 1M atom which
* should be large enough to contain a number of units.
*/
if (mmu_linear_psize == MMU_PAGE_4K)
atom_size = PAGE_SIZE;
else
atom_size = 1 << 20;
rc = pcpu_embed_first_chunk(0, dyn_size, atom_size, pcpu_cpu_distance,
pcpu_fc_alloc, pcpu_fc_free);
if (rc < 0)
panic("cannot initialize percpu area (err=%d)", rc);
delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
powerpc: Optimise per cpu accesses on 64bit Now we dynamically allocate the paca array, it takes an extra load whenever we want to access another cpu's paca. One place we do that a lot is per cpu variables. A simple example: DEFINE_PER_CPU(unsigned long, vara); unsigned long test4(int cpu) { return per_cpu(vara, cpu); } This takes 4 loads, 5 if you include the actual load of the per cpu variable: ld r11,-32760(r30) # load address of paca pointer ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,9 # get offset into paca (each entry is 512 bytes) ld r0,0(r11) # load paca pointer add r3,r0,r3 # paca + offset ld r11,64(r3) # load paca[cpu].data_offset ldx r3,r9,r11 # load per cpu variable If we remove the ppc64 specific per_cpu_offset(), we get the generic one which indexes into a statically allocated array. This removes one load and one add: ld r11,-32760(r30) # load address of __per_cpu_offset ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,3 # get offset into __per_cpu_offset (each entry 8 bytes) ldx r11,r11,r3 # load __per_cpu_offset[cpu] ldx r3,r9,r11 # load per cpu variable Having all the offsets in one array also helps when iterating over a per cpu variable across a number of cpus, such as in the scheduler. Before we would need to load one paca cacheline when calculating each per cpu offset. Now we have 16 (128 / sizeof(long)) per cpu offsets in each cacheline. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-06-01 02:45:11 +08:00
for_each_possible_cpu(cpu) {
__per_cpu_offset[cpu] = delta + pcpu_unit_offsets[cpu];
paca_ptrs[cpu]->data_offset = __per_cpu_offset[cpu];
powerpc: Optimise per cpu accesses on 64bit Now we dynamically allocate the paca array, it takes an extra load whenever we want to access another cpu's paca. One place we do that a lot is per cpu variables. A simple example: DEFINE_PER_CPU(unsigned long, vara); unsigned long test4(int cpu) { return per_cpu(vara, cpu); } This takes 4 loads, 5 if you include the actual load of the per cpu variable: ld r11,-32760(r30) # load address of paca pointer ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,9 # get offset into paca (each entry is 512 bytes) ld r0,0(r11) # load paca pointer add r3,r0,r3 # paca + offset ld r11,64(r3) # load paca[cpu].data_offset ldx r3,r9,r11 # load per cpu variable If we remove the ppc64 specific per_cpu_offset(), we get the generic one which indexes into a statically allocated array. This removes one load and one add: ld r11,-32760(r30) # load address of __per_cpu_offset ld r9,-32768(r30) # load link address of percpu variable sldi r3,r29,3 # get offset into __per_cpu_offset (each entry 8 bytes) ldx r11,r11,r3 # load __per_cpu_offset[cpu] ldx r3,r9,r11 # load per cpu variable Having all the offsets in one array also helps when iterating over a per cpu variable across a number of cpus, such as in the scheduler. Before we would need to load one paca cacheline when calculating each per cpu offset. Now we have 16 (128 / sizeof(long)) per cpu offsets in each cacheline. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-06-01 02:45:11 +08:00
}
[PATCH] powerpc/64: per cpu data optimisations The current ppc64 per cpu data implementation is quite slow. eg: lhz 11,18(13) /* smp_processor_id() */ ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */ ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */ sldi 11,11,3 /* form index into __per_cpu_offset */ mr 10,9 ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */ ldx 0,10,9 /* load per cpu data */ 5 loads for something that is supposed to be fast, pretty awful. One reason for the large number of loads is that we have to synthesize 2 64bit constants (per_cpu__variable_name and __per_cpu_offset). By putting __per_cpu_offset into the paca we can avoid the 2 loads associated with it: ld 11,56(13) /* paca->data_offset */ ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */ ldx 0,9,11 /* load per cpu data Longer term we can should be able to do even better than 3 loads. If per_cpu__variable_name wasnt a 64bit constant and paca->data_offset was in a register we could cut it down to one load. A suggestion from Rusty is to use gcc's __thread extension here. In order to do this we would need to free up r13 (the __thread register and where the paca currently is). So far Ive had a few unsuccessful attempts at doing that :) The patch also allocates per cpu memory node local on NUMA machines. This patch from Rusty has been sitting in my queue _forever_ but stalled when I hit the compiler bug. Sorry about that. Finally I also only allocate per cpu data for possible cpus, which comes straight out of the x86-64 port. On a pseries kernel (with NR_CPUS == 128) and 4 possible cpus we see some nice gains: total used free shared buffers cached Mem: 4012228 212860 3799368 0 0 162424 total used free shared buffers cached Mem: 4016200 212984 3803216 0 0 162424 A saving of 3.75MB. Quite nice for smaller machines. Note: we now have to be careful of per cpu users that touch data for !possible cpus. At this stage it might be worth making the NUMA and possible cpu optimisations generic, but per cpu init is done so early we have to be careful that all architectures have their possible map setup correctly. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-11 10:16:44 +08:00
}
#endif
[POWERPC] Allow hooking of PCI MMIO & PIO accessors on 64 bits This patch reworks the way iSeries hooks on PCI IO operations (both MMIO and PIO) and provides a generic way for other platforms to do so (we have need to do that for various other platforms). While reworking the IO ops, I ended up doing some spring cleaning in io.h and eeh.h which I might want to split into 2 or 3 patches (among others, eeh.h had a lot of useless stuff in it). A side effect is that EEH for PIO should work now (it used to pass IO ports down to the eeh address check functions which is bogus). Also, new are MMIO "repeat" ops, which other archs like ARM already had, and that we have too now: readsb, readsw, readsl, writesb, writesw, writesl. In the long run, I might also make EEH use the hooks instead of wrapping at the toplevel, which would make things even cleaner and relegate EEH completely in platforms/iseries, but we have to measure the performance impact there (though it's really only on MMIO reads) Since I also need to hook on ioremap, I shuffled the functions a bit there. I introduced ioremap_flags() to use by drivers who want to pass explicit flags to ioremap (and it can be hooked). The old __ioremap() is still there as a low level and cannot be hooked, thus drivers who use it should migrate unless they know they want the low level version. The patch "arch provides generic iomap missing accessors" (should be number 4 in this series) is a pre-requisite to provide full iomap API support with this patch. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-11-11 14:25:10 +08:00
#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
unsigned long memory_block_size_bytes(void)
{
if (ppc_md.memory_block_size)
return ppc_md.memory_block_size();
return MIN_MEMORY_BLOCK_SIZE;
}
#endif
[POWERPC] Allow hooking of PCI MMIO & PIO accessors on 64 bits This patch reworks the way iSeries hooks on PCI IO operations (both MMIO and PIO) and provides a generic way for other platforms to do so (we have need to do that for various other platforms). While reworking the IO ops, I ended up doing some spring cleaning in io.h and eeh.h which I might want to split into 2 or 3 patches (among others, eeh.h had a lot of useless stuff in it). A side effect is that EEH for PIO should work now (it used to pass IO ports down to the eeh address check functions which is bogus). Also, new are MMIO "repeat" ops, which other archs like ARM already had, and that we have too now: readsb, readsw, readsl, writesb, writesw, writesl. In the long run, I might also make EEH use the hooks instead of wrapping at the toplevel, which would make things even cleaner and relegate EEH completely in platforms/iseries, but we have to measure the performance impact there (though it's really only on MMIO reads) Since I also need to hook on ioremap, I shuffled the functions a bit there. I introduced ioremap_flags() to use by drivers who want to pass explicit flags to ioremap (and it can be hooked). The old __ioremap() is still there as a low level and cannot be hooked, thus drivers who use it should migrate unless they know they want the low level version. The patch "arch provides generic iomap missing accessors" (should be number 4 in this series) is a pre-requisite to provide full iomap API support with this patch. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-11-11 14:25:10 +08:00
#if defined(CONFIG_PPC_INDIRECT_PIO) || defined(CONFIG_PPC_INDIRECT_MMIO)
[POWERPC] Allow hooking of PCI MMIO & PIO accessors on 64 bits This patch reworks the way iSeries hooks on PCI IO operations (both MMIO and PIO) and provides a generic way for other platforms to do so (we have need to do that for various other platforms). While reworking the IO ops, I ended up doing some spring cleaning in io.h and eeh.h which I might want to split into 2 or 3 patches (among others, eeh.h had a lot of useless stuff in it). A side effect is that EEH for PIO should work now (it used to pass IO ports down to the eeh address check functions which is bogus). Also, new are MMIO "repeat" ops, which other archs like ARM already had, and that we have too now: readsb, readsw, readsl, writesb, writesw, writesl. In the long run, I might also make EEH use the hooks instead of wrapping at the toplevel, which would make things even cleaner and relegate EEH completely in platforms/iseries, but we have to measure the performance impact there (though it's really only on MMIO reads) Since I also need to hook on ioremap, I shuffled the functions a bit there. I introduced ioremap_flags() to use by drivers who want to pass explicit flags to ioremap (and it can be hooked). The old __ioremap() is still there as a low level and cannot be hooked, thus drivers who use it should migrate unless they know they want the low level version. The patch "arch provides generic iomap missing accessors" (should be number 4 in this series) is a pre-requisite to provide full iomap API support with this patch. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-11-11 14:25:10 +08:00
struct ppc_pci_io ppc_pci_io;
EXPORT_SYMBOL(ppc_pci_io);
#endif
#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF
u64 hw_nmi_get_sample_period(int watchdog_thresh)
{
return ppc_proc_freq * watchdog_thresh;
}
#endif
/*
* The perf based hardlockup detector breaks PMU event based branches, so
* disable it by default. Book3S has a soft-nmi hardlockup detector based
* on the decrementer interrupt, so it does not suffer from this problem.
*
* It is likely to get false positives in VM guests, so disable it there
* by default too.
*/
static int __init disable_hardlockup_detector(void)
{
#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF
hardlockup_detector_disable();
#else
if (firmware_has_feature(FW_FEATURE_LPAR))
hardlockup_detector_disable();
#endif
return 0;
}
early_initcall(disable_hardlockup_detector);
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
#ifdef CONFIG_PPC_BOOK3S_64
static enum l1d_flush_type enabled_flush_types;
static void *l1d_flush_fallback_area;
static bool no_rfi_flush;
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
bool rfi_flush;
static int __init handle_no_rfi_flush(char *p)
{
pr_info("rfi-flush: disabled on command line.");
no_rfi_flush = true;
return 0;
}
early_param("no_rfi_flush", handle_no_rfi_flush);
/*
* The RFI flush is not KPTI, but because users will see doco that says to use
* nopti we hijack that option here to also disable the RFI flush.
*/
static int __init handle_no_pti(char *p)
{
pr_info("rfi-flush: disabling due to 'nopti' on command line.\n");
handle_no_rfi_flush(NULL);
return 0;
}
early_param("nopti", handle_no_pti);
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
static void do_nothing(void *unused)
{
/*
* We don't need to do the flush explicitly, just enter+exit kernel is
* sufficient, the RFI exit handlers will do the right thing.
*/
}
void rfi_flush_enable(bool enable)
{
if (enable) {
do_rfi_flush_fixups(enabled_flush_types);
on_each_cpu(do_nothing, NULL, 1);
} else
do_rfi_flush_fixups(L1D_FLUSH_NONE);
rfi_flush = enable;
}
static void __ref init_fallback_flush(void)
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
{
u64 l1d_size, limit;
int cpu;
/* Only allocate the fallback flush area once (at boot time). */
if (l1d_flush_fallback_area)
return;
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
l1d_size = ppc64_caches.l1d.size;
/*
* If there is no d-cache-size property in the device tree, l1d_size
* could be zero. That leads to the loop in the asm wrapping around to
* 2^64-1, and then walking off the end of the fallback area and
* eventually causing a page fault which is fatal. Just default to
* something vaguely sane.
*/
if (!l1d_size)
l1d_size = (64 * 1024);
limit = min(ppc64_bolted_size(), ppc64_rma_size);
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
/*
* Align to L1d size, and size it at 2x L1d size, to catch possible
* hardware prefetch runoff. We don't have a recipe for load patterns to
* reliably avoid the prefetcher.
*/
powerpc: prefer memblock APIs returning virtual address Patch series "memblock: simplify several early memory allocation", v4. These patches simplify some of the early memory allocations by replacing usage of older memblock APIs with newer and shinier ones. Quite a few places in the arch/ code allocated memory using a memblock API that returns a physical address of the allocated area, then converted this physical address to a virtual one and then used memset(0) to clear the allocated range. More recent memblock APIs do all the three steps in one call and their usage simplifies the code. It's important to note that regardless of API used, the core allocation is nearly identical for any set of memblock allocators: first it tries to find a free memory with all the constraints specified by the caller and then falls back to the allocation with some or all constraints disabled. The first three patches perform the conversion of call sites that have exact requirements for the node and the possible memory range. The fourth patch is a bit one-off as it simplifies openrisc's implementation of pte_alloc_one_kernel(), and not only the memblock usage. The fifth patch takes care of simpler cases when the allocation can be satisfied with a simple call to memblock_alloc(). The sixth patch removes one-liner wrappers for memblock_alloc on arm and unicore32, as suggested by Christoph. This patch (of 6): There are a several places that allocate memory using memblock APIs that return a physical address, convert the returned address to the virtual address and frequently also memset(0) the allocated range. Update these places to use memblock allocators already returning a virtual address. Use memblock functions that clear the allocated memory instead of calling memset(0) where appropriate. The calls to memblock_alloc_base() that were not followed by memset(0) are replaced with memblock_alloc_try_nid_raw(). Since the latter does not panic() when the allocation fails, the appropriate panic() calls are added to the call sites. Link: http://lkml.kernel.org/r/1546248566-14910-2-git-send-email-rppt@linux.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Greentime Hu <green.hu@gmail.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Hocko <mhocko@suse.com> Cc: Michal Simek <monstr@monstr.eu> Cc: Mark Salter <msalter@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Stefan Kristiansson <stefan.kristiansson@saunalahti.fi> Cc: Stafford Horne <shorne@gmail.com> Cc: Vincent Chen <deanbo422@gmail.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Christoph Hellwig <hch@infradead.org> Cc: Michal Simek <michal.simek@xilinx.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-08 08:30:48 +08:00
l1d_flush_fallback_area = memblock_alloc_try_nid(l1d_size * 2,
l1d_size, MEMBLOCK_LOW_LIMIT,
limit, NUMA_NO_NODE);
treewide: add checks for the return value of memblock_alloc*() Add check for the return value of memblock_alloc*() functions and call panic() in case of error. The panic message repeats the one used by panicing memblock allocators with adjustment of parameters to include only relevant ones. The replacement was mostly automated with semantic patches like the one below with manual massaging of format strings. @@ expression ptr, size, align; @@ ptr = memblock_alloc(size, align); + if (!ptr) + panic("%s: Failed to allocate %lu bytes align=0x%lx\n", __func__, size, align); [anders.roxell@linaro.org: use '%pa' with 'phys_addr_t' type] Link: http://lkml.kernel.org/r/20190131161046.21886-1-anders.roxell@linaro.org [rppt@linux.ibm.com: fix format strings for panics after memblock_alloc] Link: http://lkml.kernel.org/r/1548950940-15145-1-git-send-email-rppt@linux.ibm.com [rppt@linux.ibm.com: don't panic if the allocation in sparse_buffer_init fails] Link: http://lkml.kernel.org/r/20190131074018.GD28876@rapoport-lnx [akpm@linux-foundation.org: fix xtensa printk warning] Link: http://lkml.kernel.org/r/1548057848-15136-20-git-send-email-rppt@linux.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Reviewed-by: Guo Ren <ren_guo@c-sky.com> [c-sky] Acked-by: Paul Burton <paul.burton@mips.com> [MIPS] Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com> [s390] Reviewed-by: Juergen Gross <jgross@suse.com> [Xen] Reviewed-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Max Filippov <jcmvbkbc@gmail.com> [xtensa] Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Christophe Leroy <christophe.leroy@c-s.fr> Cc: Christoph Hellwig <hch@lst.de> Cc: "David S. Miller" <davem@davemloft.net> Cc: Dennis Zhou <dennis@kernel.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Guo Ren <guoren@kernel.org> Cc: Mark Salter <msalter@redhat.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Petr Mladek <pmladek@suse.com> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Rob Herring <robh+dt@kernel.org> Cc: Rob Herring <robh@kernel.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Stafford Horne <shorne@gmail.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-12 14:30:31 +08:00
if (!l1d_flush_fallback_area)
panic("%s: Failed to allocate %llu bytes align=0x%llx max_addr=%pa\n",
__func__, l1d_size * 2, l1d_size, &limit);
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
for_each_possible_cpu(cpu) {
struct paca_struct *paca = paca_ptrs[cpu];
paca->rfi_flush_fallback_area = l1d_flush_fallback_area;
paca->l1d_flush_size = l1d_size;
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
}
}
void setup_rfi_flush(enum l1d_flush_type types, bool enable)
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
{
if (types & L1D_FLUSH_FALLBACK) {
pr_info("rfi-flush: fallback displacement flush available\n");
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
init_fallback_flush();
}
if (types & L1D_FLUSH_ORI)
pr_info("rfi-flush: ori type flush available\n");
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
if (types & L1D_FLUSH_MTTRIG)
pr_info("rfi-flush: mttrig type flush available\n");
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
enabled_flush_types = types;
powerpc/speculation: Support 'mitigations=' cmdline option Configure powerpc CPU runtime speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v1, Spectre v2, and Speculative Store Bypass. The default behavior is unchanged. Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Jiri Kosina <jkosina@suse.cz> (on x86) Reviewed-by: Jiri Kosina <jkosina@suse.cz> Cc: Borislav Petkov <bp@alien8.de> Cc: "H . Peter Anvin" <hpa@zytor.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Jiri Kosina <jikos@kernel.org> Cc: Waiman Long <longman@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Jon Masters <jcm@redhat.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: linuxppc-dev@lists.ozlabs.org Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux-s390@vger.kernel.org Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will.deacon@arm.com> Cc: linux-arm-kernel@lists.infradead.org Cc: linux-arch@vger.kernel.org Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Tyler Hicks <tyhicks@canonical.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Steven Price <steven.price@arm.com> Cc: Phil Auld <pauld@redhat.com> Link: https://lkml.kernel.org/r/245a606e1a42a558a310220312d9b6adb9159df6.1555085500.git.jpoimboe@redhat.com
2019-04-13 04:39:30 +08:00
if (!no_rfi_flush && !cpu_mitigations_off())
rfi_flush_enable(enable);
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
}
#ifdef CONFIG_DEBUG_FS
static int rfi_flush_set(void *data, u64 val)
{
bool enable;
if (val == 1)
enable = true;
else if (val == 0)
enable = false;
else
return -EINVAL;
/* Only do anything if we're changing state */
if (enable != rfi_flush)
rfi_flush_enable(enable);
return 0;
}
static int rfi_flush_get(void *data, u64 *val)
{
*val = rfi_flush ? 1 : 0;
return 0;
}
DEFINE_SIMPLE_ATTRIBUTE(fops_rfi_flush, rfi_flush_get, rfi_flush_set, "%llu\n");
static __init int rfi_flush_debugfs_init(void)
{
debugfs_create_file("rfi_flush", 0600, powerpc_debugfs_root, NULL, &fops_rfi_flush);
return 0;
}
device_initcall(rfi_flush_debugfs_init);
#endif
powerpc/64s: Add support for RFI flush of L1-D cache On some CPUs we can prevent the Meltdown vulnerability by flushing the L1-D cache on exit from kernel to user mode, and from hypervisor to guest. This is known to be the case on at least Power7, Power8 and Power9. At this time we do not know the status of the vulnerability on other CPUs such as the 970 (Apple G5), pasemi CPUs (AmigaOne X1000) or Freescale CPUs. As more information comes to light we can enable this, or other mechanisms on those CPUs. The vulnerability occurs when the load of an architecturally inaccessible memory region (eg. userspace load of kernel memory) is speculatively executed to the point where its result can influence the address of a subsequent speculatively executed load. In order for that to happen, the first load must hit in the L1, because before the load is sent to the L2 the permission check is performed. Therefore if no kernel addresses hit in the L1 the vulnerability can not occur. We can ensure that is the case by flushing the L1 whenever we return to userspace. Similarly for hypervisor vs guest. In order to flush the L1-D cache on exit, we add a section of nops at each (h)rfi location that returns to a lower privileged context, and patch that with some sequence. Newer firmwares are able to advertise to us that there is a special nop instruction that flushes the L1-D. If we do not see that advertised, we fall back to doing a displacement flush in software. For guest kernels we support migration between some CPU versions, and different CPUs may use different flush instructions. So that we are prepared to migrate to a machine with a different flush instruction activated, we may have to patch more than one flush instruction at boot if the hypervisor tells us to. In the end this patch is mostly the work of Nicholas Piggin and Michael Ellerman. However a cast of thousands contributed to analysis of the issue, earlier versions of the patch, back ports testing etc. Many thanks to all of them. Tested-by: Jon Masters <jcm@redhat.com> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-01-10 00:07:15 +08:00
#endif /* CONFIG_PPC_BOOK3S_64 */