Commit f240bdd2 authored by Linus Torvalds's avatar Linus Torvalds

Merge tag 'powerpc-4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix 32-bit TCE table init in kdump kernel from Nish

 - Fix kdump with non-power-of-2 crashkernel= from Nish

 - Abort cxl_pci_enable_device_hook() if PCI channel is offline from
   Andrew

 - Fix to release DRC when configure_connector() fails from Bharata

 - Wire up sys_userfaultfd()

 - Fix race condition in tearing down MSI interrupts from Paul

 - Fix unbalanced pci_dev_get() in cxl_probe() from Daniel

 - Fix cxl build failure due to -Wunused-variable gcc behaviour change
   from Ian

 - Tell the toolchain to use ABI v2 when building an LE boot wrapper
   from Benh

 - Fix THP to recompute hash value after a failed update from Aneesh

 - 32-bit memcpy/memset: only use dcbz once cache is enabled from
   Christophe

* tag 'powerpc-4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc32: memset: only use dcbz once cache is enabled
  powerpc32: memcpy: only use dcbz once cache is enabled
  powerpc/mm: Recompute hash value after a failed update
  powerpc/boot: Specify ABI v2 when building an LE boot wrapper
  cxl: Fix build failure due to -Wunused-variable behaviour change
  cxl: Fix unbalanced pci_dev_get in cxl_probe
  powerpc/MSI: Fix race condition in tearing down MSI interrupts
  powerpc: Wire up sys_userfaultfd()
  powerpc/pseries: Release DRC when configure_connector fails
  cxl: abort cxl_pci_enable_device_hook() if PCI channel is offline
  powerpc/powernv/pci-ioda: fix kdump with non-power-of-2 crashkernel=
  powerpc/powernv/pci-ioda: fix 32-bit TCE table init in kdump kernel
parents a7d5c189 400c47d8
...@@ -28,6 +28,9 @@ BOOTCFLAGS += -m64 ...@@ -28,6 +28,9 @@ BOOTCFLAGS += -m64
endif endif
ifdef CONFIG_CPU_BIG_ENDIAN ifdef CONFIG_CPU_BIG_ENDIAN
BOOTCFLAGS += -mbig-endian BOOTCFLAGS += -mbig-endian
else
BOOTCFLAGS += -mlittle-endian
BOOTCFLAGS += $(call cc-option,-mabi=elfv2)
endif endif
BOOTAFLAGS := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional -nostdinc BOOTAFLAGS := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional -nostdinc
......
...@@ -368,3 +368,4 @@ SYSCALL_SPU(memfd_create) ...@@ -368,3 +368,4 @@ SYSCALL_SPU(memfd_create)
SYSCALL_SPU(bpf) SYSCALL_SPU(bpf)
COMPAT_SYS(execveat) COMPAT_SYS(execveat)
PPC64ONLY(switch_endian) PPC64ONLY(switch_endian)
SYSCALL_SPU(userfaultfd)
...@@ -12,7 +12,7 @@ ...@@ -12,7 +12,7 @@
#include <uapi/asm/unistd.h> #include <uapi/asm/unistd.h>
#define __NR_syscalls 364 #define __NR_syscalls 365
#define __NR__exit __NR_exit #define __NR__exit __NR_exit
#define NR_syscalls __NR_syscalls #define NR_syscalls __NR_syscalls
......
...@@ -386,5 +386,6 @@ ...@@ -386,5 +386,6 @@
#define __NR_bpf 361 #define __NR_bpf 361
#define __NR_execveat 362 #define __NR_execveat 362
#define __NR_switch_endian 363 #define __NR_switch_endian 363
#define __NR_userfaultfd 364
#endif /* _UAPI_ASM_POWERPC_UNISTD_H_ */ #endif /* _UAPI_ASM_POWERPC_UNISTD_H_ */
...@@ -38,6 +38,7 @@ ...@@ -38,6 +38,7 @@
#include <asm/udbg.h> #include <asm/udbg.h>
#include <asm/mmu_context.h> #include <asm/mmu_context.h>
#include <asm/epapr_hcalls.h> #include <asm/epapr_hcalls.h>
#include <asm/code-patching.h>
#define DBG(fmt...) #define DBG(fmt...)
...@@ -109,6 +110,8 @@ notrace unsigned long __init early_init(unsigned long dt_ptr) ...@@ -109,6 +110,8 @@ notrace unsigned long __init early_init(unsigned long dt_ptr)
* This is called very early on the boot process, after a minimal * This is called very early on the boot process, after a minimal
* MMU environment has been set up but before MMU_init is called. * MMU environment has been set up but before MMU_init is called.
*/ */
extern unsigned int memset_nocache_branch; /* Insn to be replaced by NOP */
notrace void __init machine_init(u64 dt_ptr) notrace void __init machine_init(u64 dt_ptr)
{ {
lockdep_init(); lockdep_init();
...@@ -116,6 +119,9 @@ notrace void __init machine_init(u64 dt_ptr) ...@@ -116,6 +119,9 @@ notrace void __init machine_init(u64 dt_ptr)
/* Enable early debugging if any specified (see udbg.h) */ /* Enable early debugging if any specified (see udbg.h) */
udbg_early_init(); udbg_early_init();
patch_instruction((unsigned int *)&memcpy, PPC_INST_NOP);
patch_instruction(&memset_nocache_branch, PPC_INST_NOP);
/* Do some early initialization based on the flat device tree */ /* Do some early initialization based on the flat device tree */
early_init_devtree(__va(dt_ptr)); early_init_devtree(__va(dt_ptr));
......
...@@ -73,6 +73,10 @@ CACHELINE_MASK = (L1_CACHE_BYTES-1) ...@@ -73,6 +73,10 @@ CACHELINE_MASK = (L1_CACHE_BYTES-1)
* Use dcbz on the complete cache lines in the destination * Use dcbz on the complete cache lines in the destination
* to set them to zero. This requires that the destination * to set them to zero. This requires that the destination
* area is cacheable. -- paulus * area is cacheable. -- paulus
*
* During early init, cache might not be active yet, so dcbz cannot be used.
* We therefore skip the optimised bloc that uses dcbz. This jump is
* replaced by a nop once cache is active. This is done in machine_init()
*/ */
_GLOBAL(memset) _GLOBAL(memset)
rlwimi r4,r4,8,16,23 rlwimi r4,r4,8,16,23
...@@ -88,6 +92,8 @@ _GLOBAL(memset) ...@@ -88,6 +92,8 @@ _GLOBAL(memset)
subf r6,r0,r6 subf r6,r0,r6
cmplwi 0,r4,0 cmplwi 0,r4,0
bne 2f /* Use normal procedure if r4 is not zero */ bne 2f /* Use normal procedure if r4 is not zero */
_GLOBAL(memset_nocache_branch)
b 2f /* Skip optimised bloc until cache is enabled */
clrlwi r7,r6,32-LG_CACHELINE_BYTES clrlwi r7,r6,32-LG_CACHELINE_BYTES
add r8,r7,r5 add r8,r7,r5
...@@ -128,6 +134,10 @@ _GLOBAL(memset) ...@@ -128,6 +134,10 @@ _GLOBAL(memset)
* the destination area is cacheable. * the destination area is cacheable.
* We only use this version if the source and dest don't overlap. * We only use this version if the source and dest don't overlap.
* -- paulus. * -- paulus.
*
* During early init, cache might not be active yet, so dcbz cannot be used.
* We therefore jump to generic_memcpy which doesn't use dcbz. This jump is
* replaced by a nop once cache is active. This is done in machine_init()
*/ */
_GLOBAL(memmove) _GLOBAL(memmove)
cmplw 0,r3,r4 cmplw 0,r3,r4
...@@ -135,6 +145,7 @@ _GLOBAL(memmove) ...@@ -135,6 +145,7 @@ _GLOBAL(memmove)
/* fall through */ /* fall through */
_GLOBAL(memcpy) _GLOBAL(memcpy)
b generic_memcpy
add r7,r3,r5 /* test if the src & dst overlap */ add r7,r3,r5 /* test if the src & dst overlap */
add r8,r4,r5 add r8,r4,r5
cmplw 0,r4,r7 cmplw 0,r4,r7
......
...@@ -85,7 +85,6 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid, ...@@ -85,7 +85,6 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid,
BUG_ON(index >= 4096); BUG_ON(index >= 4096);
vpn = hpt_vpn(ea, vsid, ssize); vpn = hpt_vpn(ea, vsid, ssize);
hash = hpt_hash(vpn, shift, ssize);
hpte_slot_array = get_hpte_slot_array(pmdp); hpte_slot_array = get_hpte_slot_array(pmdp);
if (psize == MMU_PAGE_4K) { if (psize == MMU_PAGE_4K) {
/* /*
...@@ -101,6 +100,7 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid, ...@@ -101,6 +100,7 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid,
valid = hpte_valid(hpte_slot_array, index); valid = hpte_valid(hpte_slot_array, index);
if (valid) { if (valid) {
/* update the hpte bits */ /* update the hpte bits */
hash = hpt_hash(vpn, shift, ssize);
hidx = hpte_hash_index(hpte_slot_array, index); hidx = hpte_hash_index(hpte_slot_array, index);
if (hidx & _PTEIDX_SECONDARY) if (hidx & _PTEIDX_SECONDARY)
hash = ~hash; hash = ~hash;
...@@ -126,6 +126,7 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid, ...@@ -126,6 +126,7 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid,
if (!valid) { if (!valid) {
unsigned long hpte_group; unsigned long hpte_group;
hash = hpt_hash(vpn, shift, ssize);
/* insert new entry */ /* insert new entry */
pa = pmd_pfn(__pmd(old_pmd)) << PAGE_SHIFT; pa = pmd_pfn(__pmd(old_pmd)) << PAGE_SHIFT;
new_pmd |= _PAGE_HASHPTE; new_pmd |= _PAGE_HASHPTE;
......
...@@ -63,6 +63,7 @@ static struct irq_chip mpic_pasemi_msi_chip = { ...@@ -63,6 +63,7 @@ static struct irq_chip mpic_pasemi_msi_chip = {
static void pasemi_msi_teardown_msi_irqs(struct pci_dev *pdev) static void pasemi_msi_teardown_msi_irqs(struct pci_dev *pdev)
{ {
struct msi_desc *entry; struct msi_desc *entry;
irq_hw_number_t hwirq;
pr_debug("pasemi_msi_teardown_msi_irqs, pdev %p\n", pdev); pr_debug("pasemi_msi_teardown_msi_irqs, pdev %p\n", pdev);
...@@ -70,10 +71,10 @@ static void pasemi_msi_teardown_msi_irqs(struct pci_dev *pdev) ...@@ -70,10 +71,10 @@ static void pasemi_msi_teardown_msi_irqs(struct pci_dev *pdev)
if (entry->irq == NO_IRQ) if (entry->irq == NO_IRQ)
continue; continue;
hwirq = virq_to_hw(entry->irq);
irq_set_msi_desc(entry->irq, NULL); irq_set_msi_desc(entry->irq, NULL);
msi_bitmap_free_hwirqs(&msi_mpic->msi_bitmap,
virq_to_hw(entry->irq), ALLOC_CHUNK);
irq_dispose_mapping(entry->irq); irq_dispose_mapping(entry->irq);
msi_bitmap_free_hwirqs(&msi_mpic->msi_bitmap, hwirq, ALLOC_CHUNK);
} }
return; return;
......
...@@ -2049,9 +2049,23 @@ static long pnv_pci_ioda2_setup_default_config(struct pnv_ioda_pe *pe) ...@@ -2049,9 +2049,23 @@ static long pnv_pci_ioda2_setup_default_config(struct pnv_ioda_pe *pe)
struct iommu_table *tbl = NULL; struct iommu_table *tbl = NULL;
long rc; long rc;
/*
* crashkernel= specifies the kdump kernel's maximum memory at
* some offset and there is no guaranteed the result is a power
* of 2, which will cause errors later.
*/
const u64 max_memory = __rounddown_pow_of_two(memory_hotplug_max());
/*
* In memory constrained environments, e.g. kdump kernel, the
* DMA window can be larger than available memory, which will
* cause errors later.
*/
const u64 window_size = min((u64)pe->table_group.tce32_size, max_memory);
rc = pnv_pci_ioda2_create_table(&pe->table_group, 0, rc = pnv_pci_ioda2_create_table(&pe->table_group, 0,
IOMMU_PAGE_SHIFT_4K, IOMMU_PAGE_SHIFT_4K,
pe->table_group.tce32_size, window_size,
POWERNV_IOMMU_DEFAULT_LEVELS, &tbl); POWERNV_IOMMU_DEFAULT_LEVELS, &tbl);
if (rc) { if (rc) {
pe_err(pe, "Failed to create 32-bit TCE table, err %ld", pe_err(pe, "Failed to create 32-bit TCE table, err %ld",
......
...@@ -99,6 +99,7 @@ void pnv_teardown_msi_irqs(struct pci_dev *pdev) ...@@ -99,6 +99,7 @@ void pnv_teardown_msi_irqs(struct pci_dev *pdev)
struct pci_controller *hose = pci_bus_to_host(pdev->bus); struct pci_controller *hose = pci_bus_to_host(pdev->bus);
struct pnv_phb *phb = hose->private_data; struct pnv_phb *phb = hose->private_data;
struct msi_desc *entry; struct msi_desc *entry;
irq_hw_number_t hwirq;
if (WARN_ON(!phb)) if (WARN_ON(!phb))
return; return;
...@@ -106,10 +107,10 @@ void pnv_teardown_msi_irqs(struct pci_dev *pdev) ...@@ -106,10 +107,10 @@ void pnv_teardown_msi_irqs(struct pci_dev *pdev)
for_each_pci_msi_entry(entry, pdev) { for_each_pci_msi_entry(entry, pdev) {
if (entry->irq == NO_IRQ) if (entry->irq == NO_IRQ)
continue; continue;
hwirq = virq_to_hw(entry->irq);
irq_set_msi_desc(entry->irq, NULL); irq_set_msi_desc(entry->irq, NULL);
msi_bitmap_free_hwirqs(&phb->msi_bmp,
virq_to_hw(entry->irq) - phb->msi_base, 1);
irq_dispose_mapping(entry->irq); irq_dispose_mapping(entry->irq);
msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq - phb->msi_base, 1);
} }
} }
#endif /* CONFIG_PCI_MSI */ #endif /* CONFIG_PCI_MSI */
......
...@@ -422,8 +422,10 @@ static ssize_t dlpar_cpu_probe(const char *buf, size_t count) ...@@ -422,8 +422,10 @@ static ssize_t dlpar_cpu_probe(const char *buf, size_t count)
dn = dlpar_configure_connector(cpu_to_be32(drc_index), parent); dn = dlpar_configure_connector(cpu_to_be32(drc_index), parent);
of_node_put(parent); of_node_put(parent);
if (!dn) if (!dn) {
dlpar_release_drc(drc_index);
return -EINVAL; return -EINVAL;
}
rc = dlpar_attach_node(dn); rc = dlpar_attach_node(dn);
if (rc) { if (rc) {
......
...@@ -128,15 +128,16 @@ static void fsl_teardown_msi_irqs(struct pci_dev *pdev) ...@@ -128,15 +128,16 @@ static void fsl_teardown_msi_irqs(struct pci_dev *pdev)
{ {
struct msi_desc *entry; struct msi_desc *entry;
struct fsl_msi *msi_data; struct fsl_msi *msi_data;
irq_hw_number_t hwirq;
for_each_pci_msi_entry(entry, pdev) { for_each_pci_msi_entry(entry, pdev) {
if (entry->irq == NO_IRQ) if (entry->irq == NO_IRQ)
continue; continue;
hwirq = virq_to_hw(entry->irq);
msi_data = irq_get_chip_data(entry->irq); msi_data = irq_get_chip_data(entry->irq);
irq_set_msi_desc(entry->irq, NULL); irq_set_msi_desc(entry->irq, NULL);
msi_bitmap_free_hwirqs(&msi_data->bitmap,
virq_to_hw(entry->irq), 1);
irq_dispose_mapping(entry->irq); irq_dispose_mapping(entry->irq);
msi_bitmap_free_hwirqs(&msi_data->bitmap, hwirq, 1);
} }
return; return;
......
...@@ -107,15 +107,16 @@ static u64 find_u4_magic_addr(struct pci_dev *pdev, unsigned int hwirq) ...@@ -107,15 +107,16 @@ static u64 find_u4_magic_addr(struct pci_dev *pdev, unsigned int hwirq)
static void u3msi_teardown_msi_irqs(struct pci_dev *pdev) static void u3msi_teardown_msi_irqs(struct pci_dev *pdev)
{ {
struct msi_desc *entry; struct msi_desc *entry;
irq_hw_number_t hwirq;
for_each_pci_msi_entry(entry, pdev) { for_each_pci_msi_entry(entry, pdev) {
if (entry->irq == NO_IRQ) if (entry->irq == NO_IRQ)
continue; continue;
hwirq = virq_to_hw(entry->irq);
irq_set_msi_desc(entry->irq, NULL); irq_set_msi_desc(entry->irq, NULL);
msi_bitmap_free_hwirqs(&msi_mpic->msi_bitmap,
virq_to_hw(entry->irq), 1);
irq_dispose_mapping(entry->irq); irq_dispose_mapping(entry->irq);
msi_bitmap_free_hwirqs(&msi_mpic->msi_bitmap, hwirq, 1);
} }
return; return;
......
...@@ -124,16 +124,17 @@ void ppc4xx_teardown_msi_irqs(struct pci_dev *dev) ...@@ -124,16 +124,17 @@ void ppc4xx_teardown_msi_irqs(struct pci_dev *dev)
{ {
struct msi_desc *entry; struct msi_desc *entry;
struct ppc4xx_msi *msi_data = &ppc4xx_msi; struct ppc4xx_msi *msi_data = &ppc4xx_msi;
irq_hw_number_t hwirq;
dev_dbg(&dev->dev, "PCIE-MSI: tearing down msi irqs\n"); dev_dbg(&dev->dev, "PCIE-MSI: tearing down msi irqs\n");
for_each_pci_msi_entry(entry, dev) { for_each_pci_msi_entry(entry, dev) {
if (entry->irq == NO_IRQ) if (entry->irq == NO_IRQ)
continue; continue;
hwirq = virq_to_hw(entry->irq);
irq_set_msi_desc(entry->irq, NULL); irq_set_msi_desc(entry->irq, NULL);
msi_bitmap_free_hwirqs(&msi_data->bitmap,
virq_to_hw(entry->irq), 1);
irq_dispose_mapping(entry->irq); irq_dispose_mapping(entry->irq);
msi_bitmap_free_hwirqs(&msi_data->bitmap, hwirq, 1);
} }
} }
......
ccflags-y := -Werror ccflags-y := -Werror -Wno-unused-const-variable
cxl-y += main.o file.o irq.o fault.o native.o cxl-y += main.o file.o irq.o fault.o native.o
cxl-y += context.o sysfs.o debugfs.o pci.o trace.o cxl-y += context.o sysfs.o debugfs.o pci.o trace.o
......
...@@ -1249,8 +1249,6 @@ static int cxl_probe(struct pci_dev *dev, const struct pci_device_id *id) ...@@ -1249,8 +1249,6 @@ static int cxl_probe(struct pci_dev *dev, const struct pci_device_id *id)
int slice; int slice;
int rc; int rc;
pci_dev_get(dev);
if (cxl_verbose) if (cxl_verbose)
dump_cxl_config_space(dev); dump_cxl_config_space(dev);
......
...@@ -48,6 +48,12 @@ static bool cxl_pci_enable_device_hook(struct pci_dev *dev) ...@@ -48,6 +48,12 @@ static bool cxl_pci_enable_device_hook(struct pci_dev *dev)
phb = pci_bus_to_host(dev->bus); phb = pci_bus_to_host(dev->bus);
afu = (struct cxl_afu *)phb->private_data; afu = (struct cxl_afu *)phb->private_data;
if (!cxl_adapter_link_ok(afu->adapter)) {
dev_warn(&dev->dev, "%s: Device link is down, refusing to enable AFU\n", __func__);
return false;
}
set_dma_ops(&dev->dev, &dma_direct_ops); set_dma_ops(&dev->dev, &dma_direct_ops);
set_dma_offset(&dev->dev, PAGE_OFFSET); set_dma_offset(&dev->dev, PAGE_OFFSET);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment