diff options
Diffstat (limited to 'arch/powerpc/platforms')
60 files changed, 2309 insertions, 780 deletions
diff --git a/arch/powerpc/platforms/44x/gpio.c b/arch/powerpc/platforms/44x/gpio.c index e5f2319e5cbe..d540e261d85a 100644 --- a/arch/powerpc/platforms/44x/gpio.c +++ b/arch/powerpc/platforms/44x/gpio.c @@ -75,8 +75,7 @@ __ppc4xx_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) clrbits32(®s->or, GPIO_MASK(gpio)); } -static void -ppc4xx_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) +static int ppc4xx_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) { struct ppc4xx_gpio_chip *chip = gpiochip_get_data(gc); unsigned long flags; @@ -88,6 +87,8 @@ ppc4xx_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) spin_unlock_irqrestore(&chip->lock, flags); pr_debug("%s: gpio: %d val: %d\n", __func__, gpio, val); + + return 0; } static int ppc4xx_gpio_dir_in(struct gpio_chip *gc, unsigned int gpio) @@ -179,7 +180,7 @@ static int __init ppc4xx_add_gpiochips(void) gc->direction_input = ppc4xx_gpio_dir_in; gc->direction_output = ppc4xx_gpio_dir_out; gc->get = ppc4xx_gpio_get; - gc->set = ppc4xx_gpio_set; + gc->set_rv = ppc4xx_gpio_set; ret = of_mm_gpiochip_add_data(np, mm_gc, ppc4xx_gc); if (ret) diff --git a/arch/powerpc/platforms/44x/uic.c b/arch/powerpc/platforms/44x/uic.c index 8b03ae4cb3f6..85daf841fd3f 100644 --- a/arch/powerpc/platforms/44x/uic.c +++ b/arch/powerpc/platforms/44x/uic.c @@ -254,8 +254,9 @@ static struct uic * __init uic_init_one(struct device_node *node) } uic->dcrbase = *dcrreg; - uic->irqhost = irq_domain_add_linear(node, NR_UIC_INTS, &uic_host_ops, - uic); + uic->irqhost = irq_domain_create_linear(of_fwnode_handle(node), + NR_UIC_INTS, &uic_host_ops, + uic); if (! uic->irqhost) return NULL; /* FIXME: panic? */ @@ -291,7 +292,7 @@ void __init uic_init_tree(void) if (!primary_uic) panic("Unable to initialize primary UIC %pOF\n", np); - irq_set_default_host(primary_uic->irqhost); + irq_set_default_domain(primary_uic->irqhost); of_node_put(np); /* The scan again for cascaded UICs */ @@ -327,5 +328,5 @@ unsigned int uic_get_irq(void) msr = mfdcr(primary_uic->dcrbase + UIC_MSR); src = 32 - ffs(msr); - return irq_linear_revmap(primary_uic->irqhost, src); + return irq_find_mapping(primary_uic->irqhost, src); } diff --git a/arch/powerpc/platforms/512x/mpc5121_ads_cpld.c b/arch/powerpc/platforms/512x/mpc5121_ads_cpld.c index e995eb30bf09..2cf3c6237337 100644 --- a/arch/powerpc/platforms/512x/mpc5121_ads_cpld.c +++ b/arch/powerpc/platforms/512x/mpc5121_ads_cpld.c @@ -188,7 +188,8 @@ mpc5121_ads_cpld_pic_init(void) cpld_pic_node = of_node_get(np); - cpld_pic_host = irq_domain_add_linear(np, 16, &cpld_pic_host_ops, NULL); + cpld_pic_host = irq_domain_create_linear(of_fwnode_handle(np), 16, + &cpld_pic_host_ops, NULL); if (!cpld_pic_host) { printk(KERN_ERR "CPLD PIC: failed to allocate irq host!\n"); goto end; diff --git a/arch/powerpc/platforms/512x/mpc512x_lpbfifo.c b/arch/powerpc/platforms/512x/mpc512x_lpbfifo.c index 9668b052cd4b..f251e0f68262 100644 --- a/arch/powerpc/platforms/512x/mpc512x_lpbfifo.c +++ b/arch/powerpc/platforms/512x/mpc512x_lpbfifo.c @@ -240,10 +240,8 @@ static int mpc512x_lpbfifo_kick(void) dma_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES; /* Make DMA channel work with LPB FIFO data register */ - if (dma_dev->device_config(lpbfifo.chan, &dma_conf)) { - ret = -EINVAL; - goto err_dma_prep; - } + if (dma_dev->device_config(lpbfifo.chan, &dma_conf)) + return -EINVAL; sg_init_table(&sg, 1); diff --git a/arch/powerpc/platforms/52xx/media5200.c b/arch/powerpc/platforms/52xx/media5200.c index 19626cd42406..bc7f83cfec1d 100644 --- a/arch/powerpc/platforms/52xx/media5200.c +++ b/arch/powerpc/platforms/52xx/media5200.c @@ -168,7 +168,7 @@ static void __init media5200_init_irq(void) spin_lock_init(&media5200_irq.lock); - media5200_irq.irqhost = irq_domain_add_linear(fpga_np, + media5200_irq.irqhost = irq_domain_create_linear(of_fwnode_handle(fpga_np), MEDIA5200_NUM_IRQS, &media5200_irq_ops, &media5200_irq); if (!media5200_irq.irqhost) goto out; diff --git a/arch/powerpc/platforms/52xx/mpc52xx_gpt.c b/arch/powerpc/platforms/52xx/mpc52xx_gpt.c index 1ea591ec6083..bda707d848a6 100644 --- a/arch/powerpc/platforms/52xx/mpc52xx_gpt.c +++ b/arch/powerpc/platforms/52xx/mpc52xx_gpt.c @@ -247,9 +247,9 @@ mpc52xx_gpt_irq_setup(struct mpc52xx_gpt_priv *gpt, struct device_node *node) if (!cascade_virq) return; - gpt->irqhost = irq_domain_add_linear(node, 1, &mpc52xx_gpt_irq_ops, gpt); + gpt->irqhost = irq_domain_create_linear(of_fwnode_handle(node), 1, &mpc52xx_gpt_irq_ops, gpt); if (!gpt->irqhost) { - dev_err(gpt->dev, "irq_domain_add_linear() failed\n"); + dev_err(gpt->dev, "irq_domain_create_linear() failed\n"); return; } @@ -280,7 +280,7 @@ static int mpc52xx_gpt_gpio_get(struct gpio_chip *gc, unsigned int gpio) return (in_be32(&gpt->regs->status) >> 8) & 1; } -static void +static int mpc52xx_gpt_gpio_set(struct gpio_chip *gc, unsigned int gpio, int v) { struct mpc52xx_gpt_priv *gpt = gpiochip_get_data(gc); @@ -293,6 +293,8 @@ mpc52xx_gpt_gpio_set(struct gpio_chip *gc, unsigned int gpio, int v) raw_spin_lock_irqsave(&gpt->lock, flags); clrsetbits_be32(&gpt->regs->mode, MPC52xx_GPT_MODE_GPIO_MASK, r); raw_spin_unlock_irqrestore(&gpt->lock, flags); + + return 0; } static int mpc52xx_gpt_gpio_dir_in(struct gpio_chip *gc, unsigned int gpio) @@ -334,7 +336,7 @@ static void mpc52xx_gpt_gpio_setup(struct mpc52xx_gpt_priv *gpt) gpt->gc.direction_input = mpc52xx_gpt_gpio_dir_in; gpt->gc.direction_output = mpc52xx_gpt_gpio_dir_out; gpt->gc.get = mpc52xx_gpt_gpio_get; - gpt->gc.set = mpc52xx_gpt_gpio_set; + gpt->gc.set_rv = mpc52xx_gpt_gpio_set; gpt->gc.base = -1; gpt->gc.parent = gpt->dev; @@ -369,7 +371,7 @@ struct mpc52xx_gpt_priv *mpc52xx_gpt_from_irq(int irq) mutex_lock(&mpc52xx_gpt_list_mutex); list_for_each(pos, &mpc52xx_gpt_list) { gpt = container_of(pos, struct mpc52xx_gpt_priv, list); - if (gpt->irqhost && irq == irq_linear_revmap(gpt->irqhost, 0)) { + if (gpt->irqhost && irq == irq_find_mapping(gpt->irqhost, 0)) { mutex_unlock(&mpc52xx_gpt_list_mutex); return gpt; } diff --git a/arch/powerpc/platforms/52xx/mpc52xx_pic.c b/arch/powerpc/platforms/52xx/mpc52xx_pic.c index 1e0a5e9644dc..eb6a4e745c08 100644 --- a/arch/powerpc/platforms/52xx/mpc52xx_pic.c +++ b/arch/powerpc/platforms/52xx/mpc52xx_pic.c @@ -446,14 +446,14 @@ void __init mpc52xx_init_irq(void) * As last step, add an irq host to translate the real * hw irq information provided by the ofw to linux virq */ - mpc52xx_irqhost = irq_domain_add_linear(picnode, + mpc52xx_irqhost = irq_domain_create_linear(of_fwnode_handle(picnode), MPC52xx_IRQ_HIGHTESTHWIRQ, &mpc52xx_irqhost_ops, NULL); if (!mpc52xx_irqhost) panic(__FILE__ ": Cannot allocate the IRQ host\n"); - irq_set_default_host(mpc52xx_irqhost); + irq_set_default_domain(mpc52xx_irqhost); pr_info("MPC52xx PIC is up and running!\n"); } @@ -515,5 +515,5 @@ unsigned int mpc52xx_get_irq(void) return 0; } - return irq_linear_revmap(mpc52xx_irqhost, irq); + return irq_find_mapping(mpc52xx_irqhost, irq); } diff --git a/arch/powerpc/platforms/83xx/mcu_mpc8349emitx.c b/arch/powerpc/platforms/83xx/mcu_mpc8349emitx.c index 4d8fa9ed1a67..6e37dfc6c5c9 100644 --- a/arch/powerpc/platforms/83xx/mcu_mpc8349emitx.c +++ b/arch/powerpc/platforms/83xx/mcu_mpc8349emitx.c @@ -92,10 +92,11 @@ static void mcu_power_off(void) mutex_unlock(&mcu->lock); } -static void mcu_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) +static int mcu_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) { struct mcu *mcu = gpiochip_get_data(gc); u8 bit = 1 << (4 + gpio); + int ret; mutex_lock(&mcu->lock); if (val) @@ -103,14 +104,16 @@ static void mcu_gpio_set(struct gpio_chip *gc, unsigned int gpio, int val) else mcu->reg_ctrl |= bit; - i2c_smbus_write_byte_data(mcu->client, MCU_REG_CTRL, mcu->reg_ctrl); + ret = i2c_smbus_write_byte_data(mcu->client, MCU_REG_CTRL, + mcu->reg_ctrl); mutex_unlock(&mcu->lock); + + return ret; } static int mcu_gpio_dir_out(struct gpio_chip *gc, unsigned int gpio, int val) { - mcu_gpio_set(gc, gpio, val); - return 0; + return mcu_gpio_set(gc, gpio, val); } static int mcu_gpiochip_add(struct mcu *mcu) @@ -123,7 +126,7 @@ static int mcu_gpiochip_add(struct mcu *mcu) gc->can_sleep = 1; gc->ngpio = MCU_NUM_GPIO; gc->base = -1; - gc->set = mcu_gpio_set; + gc->set_rv = mcu_gpio_set; gc->direction_output = mcu_gpio_dir_out; gc->parent = dev; diff --git a/arch/powerpc/platforms/85xx/socrates_fpga_pic.c b/arch/powerpc/platforms/85xx/socrates_fpga_pic.c index 60e0b8947ce6..4b69fb321a68 100644 --- a/arch/powerpc/platforms/85xx/socrates_fpga_pic.c +++ b/arch/powerpc/platforms/85xx/socrates_fpga_pic.c @@ -83,7 +83,7 @@ static inline unsigned int socrates_fpga_pic_get_irq(unsigned int irq) if (cause >> (i + 16)) break; } - return irq_linear_revmap(socrates_fpga_pic_irq_host, + return irq_find_mapping(socrates_fpga_pic_irq_host, (irq_hw_number_t)i); } @@ -278,7 +278,7 @@ void __init socrates_fpga_pic_init(struct device_node *pic) int i; /* Setup an irq_domain structure */ - socrates_fpga_pic_irq_host = irq_domain_add_linear(pic, + socrates_fpga_pic_irq_host = irq_domain_create_linear(of_fwnode_handle(pic), SOCRATES_FPGA_NUM_IRQS, &socrates_fpga_pic_host_ops, NULL); if (socrates_fpga_pic_irq_host == NULL) { pr_err("FPGA PIC: Unable to allocate host\n"); diff --git a/arch/powerpc/platforms/8xx/cpm1-ic.c b/arch/powerpc/platforms/8xx/cpm1-ic.c index a18fc7c99f83..a49d4a9ab3bc 100644 --- a/arch/powerpc/platforms/8xx/cpm1-ic.c +++ b/arch/powerpc/platforms/8xx/cpm1-ic.c @@ -59,7 +59,7 @@ static int cpm_get_irq(struct irq_desc *desc) cpm_vec = in_be16(&data->reg->cpic_civr); cpm_vec >>= 11; - return irq_linear_revmap(data->host, cpm_vec); + return irq_find_mapping(data->host, cpm_vec); } static void cpm_cascade(struct irq_desc *desc) @@ -110,7 +110,8 @@ static int cpm_pic_probe(struct platform_device *pdev) out_be32(&data->reg->cpic_cimr, 0); - data->host = irq_domain_add_linear(dev->of_node, 64, &cpm_pic_host_ops, data); + data->host = irq_domain_create_linear(of_fwnode_handle(dev->of_node), + 64, &cpm_pic_host_ops, data); if (!data->host) return -ENODEV; diff --git a/arch/powerpc/platforms/8xx/cpm1.c b/arch/powerpc/platforms/8xx/cpm1.c index 1dc095ad48fc..7462c221115c 100644 --- a/arch/powerpc/platforms/8xx/cpm1.c +++ b/arch/powerpc/platforms/8xx/cpm1.c @@ -417,7 +417,7 @@ static void __cpm1_gpio16_set(struct cpm1_gpio16_chip *cpm1_gc, u16 pin_mask, in out_be16(&iop->dat, cpm1_gc->cpdata); } -static void cpm1_gpio16_set(struct gpio_chip *gc, unsigned int gpio, int value) +static int cpm1_gpio16_set(struct gpio_chip *gc, unsigned int gpio, int value) { struct cpm1_gpio16_chip *cpm1_gc = gpiochip_get_data(gc); unsigned long flags; @@ -428,6 +428,8 @@ static void cpm1_gpio16_set(struct gpio_chip *gc, unsigned int gpio, int value) __cpm1_gpio16_set(cpm1_gc, pin_mask, value); spin_unlock_irqrestore(&cpm1_gc->lock, flags); + + return 0; } static int cpm1_gpio16_to_irq(struct gpio_chip *gc, unsigned int gpio) @@ -497,7 +499,7 @@ int cpm1_gpiochip_add16(struct device *dev) gc->direction_input = cpm1_gpio16_dir_in; gc->direction_output = cpm1_gpio16_dir_out; gc->get = cpm1_gpio16_get; - gc->set = cpm1_gpio16_set; + gc->set_rv = cpm1_gpio16_set; gc->to_irq = cpm1_gpio16_to_irq; gc->parent = dev; gc->owner = THIS_MODULE; @@ -554,7 +556,7 @@ static void __cpm1_gpio32_set(struct cpm1_gpio32_chip *cpm1_gc, u32 pin_mask, in out_be32(&iop->dat, cpm1_gc->cpdata); } -static void cpm1_gpio32_set(struct gpio_chip *gc, unsigned int gpio, int value) +static int cpm1_gpio32_set(struct gpio_chip *gc, unsigned int gpio, int value) { struct cpm1_gpio32_chip *cpm1_gc = gpiochip_get_data(gc); unsigned long flags; @@ -565,6 +567,8 @@ static void cpm1_gpio32_set(struct gpio_chip *gc, unsigned int gpio, int value) __cpm1_gpio32_set(cpm1_gc, pin_mask, value); spin_unlock_irqrestore(&cpm1_gc->lock, flags); + + return 0; } static int cpm1_gpio32_dir_out(struct gpio_chip *gc, unsigned int gpio, int val) @@ -618,7 +622,7 @@ int cpm1_gpiochip_add32(struct device *dev) gc->direction_input = cpm1_gpio32_dir_in; gc->direction_output = cpm1_gpio32_dir_out; gc->get = cpm1_gpio32_get; - gc->set = cpm1_gpio32_set; + gc->set_rv = cpm1_gpio32_set; gc->parent = dev; gc->owner = THIS_MODULE; diff --git a/arch/powerpc/platforms/8xx/pic.c b/arch/powerpc/platforms/8xx/pic.c index ea6b0e523c60..933d6ab7f512 100644 --- a/arch/powerpc/platforms/8xx/pic.c +++ b/arch/powerpc/platforms/8xx/pic.c @@ -80,7 +80,7 @@ unsigned int mpc8xx_get_irq(void) if (irq == PIC_VEC_SPURRIOUS) return 0; - return irq_linear_revmap(mpc8xx_pic_host, irq); + return irq_find_mapping(mpc8xx_pic_host, irq); } @@ -146,7 +146,8 @@ void __init mpc8xx_pic_init(void) if (!siu_reg) goto out; - mpc8xx_pic_host = irq_domain_add_linear(np, 64, &mpc8xx_pic_host_ops, NULL); + mpc8xx_pic_host = irq_domain_create_linear(of_fwnode_handle(np), 64, + &mpc8xx_pic_host_ops, NULL); if (!mpc8xx_pic_host) printk(KERN_ERR "MPC8xx PIC: failed to allocate irq host!\n"); diff --git a/arch/powerpc/platforms/amigaone/setup.c b/arch/powerpc/platforms/amigaone/setup.c index 2c8dc0886912..33f852a7625f 100644 --- a/arch/powerpc/platforms/amigaone/setup.c +++ b/arch/powerpc/platforms/amigaone/setup.c @@ -109,7 +109,7 @@ static void __init amigaone_init_IRQ(void) i8259_init(pic, int_ack); ppc_md.get_irq = i8259_irq; - irq_set_default_host(i8259_get_host()); + irq_set_default_domain(i8259_get_host()); } static int __init request_isa_regions(void) diff --git a/arch/powerpc/platforms/book3s/vas-api.c b/arch/powerpc/platforms/book3s/vas-api.c index 0b6365d85d11..49b15e7a8265 100644 --- a/arch/powerpc/platforms/book3s/vas-api.c +++ b/arch/powerpc/platforms/book3s/vas-api.c @@ -425,23 +425,22 @@ static vm_fault_t vas_mmap_fault(struct vm_fault *vmf) return VM_FAULT_SIGBUS; } - mutex_lock(&txwin->task_ref.mmap_mutex); /* * The window may be inactive due to lost credit (Ex: core * removal with DLPAR). If the window is active again when * the credit is available, map the new paste address at the * window virtual address. */ - if (txwin->status == VAS_WIN_ACTIVE) { - paste_addr = cp_inst->coproc->vops->paste_addr(txwin); - if (paste_addr) { - fault = vmf_insert_pfn(vma, vma->vm_start, - (paste_addr >> PAGE_SHIFT)); - mutex_unlock(&txwin->task_ref.mmap_mutex); - return fault; + scoped_guard(mutex, &txwin->task_ref.mmap_mutex) { + if (txwin->status == VAS_WIN_ACTIVE) { + paste_addr = cp_inst->coproc->vops->paste_addr(txwin); + if (paste_addr) { + fault = vmf_insert_pfn(vma, vma->vm_start, + (paste_addr >> PAGE_SHIFT)); + return fault; + } } } - mutex_unlock(&txwin->task_ref.mmap_mutex); /* * Received this fault due to closing the actual window. @@ -494,9 +493,8 @@ static void vas_mmap_close(struct vm_area_struct *vma) return; } - mutex_lock(&txwin->task_ref.mmap_mutex); - txwin->task_ref.vma = NULL; - mutex_unlock(&txwin->task_ref.mmap_mutex); + scoped_guard(mutex, &txwin->task_ref.mmap_mutex) + txwin->task_ref.vma = NULL; } static const struct vm_operations_struct vas_vm_ops = { @@ -521,6 +519,15 @@ static int coproc_mmap(struct file *fp, struct vm_area_struct *vma) return -EINVAL; } + /* + * Map complete page to the paste address. So the user + * space should pass 0ULL to the offset parameter. + */ + if (vma->vm_pgoff) { + pr_debug("Page offset unsupported to map paste address\n"); + return -EINVAL; + } + /* Ensure instance has an open send window */ if (!txwin) { pr_err("No send window open?\n"); @@ -543,18 +550,16 @@ static int coproc_mmap(struct file *fp, struct vm_area_struct *vma) * close/open event and allows mmap() only when the window is * active. */ - mutex_lock(&txwin->task_ref.mmap_mutex); + guard(mutex)(&txwin->task_ref.mmap_mutex); if (txwin->status != VAS_WIN_ACTIVE) { pr_err("Window is not active\n"); - rc = -EACCES; - goto out; + return -EACCES; } paste_addr = cp_inst->coproc->vops->paste_addr(txwin); if (!paste_addr) { pr_err("Window paste address failed\n"); - rc = -EINVAL; - goto out; + return -EINVAL; } pfn = paste_addr >> PAGE_SHIFT; @@ -574,8 +579,6 @@ static int coproc_mmap(struct file *fp, struct vm_area_struct *vma) txwin->task_ref.vma = vma; vma->vm_ops = &vas_vm_ops; -out: - mutex_unlock(&txwin->task_ref.mmap_mutex); return rc; } diff --git a/arch/powerpc/platforms/cell/spufs/gang.c b/arch/powerpc/platforms/cell/spufs/gang.c index 827d338deaf4..2c2999de6bfa 100644 --- a/arch/powerpc/platforms/cell/spufs/gang.c +++ b/arch/powerpc/platforms/cell/spufs/gang.c @@ -25,6 +25,7 @@ struct spu_gang *alloc_spu_gang(void) mutex_init(&gang->aff_mutex); INIT_LIST_HEAD(&gang->list); INIT_LIST_HEAD(&gang->aff_list_head); + gang->alive = 1; out: return gang; diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 70236d1df3d3..7ec60290abe6 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -143,42 +143,13 @@ spufs_evict_inode(struct inode *inode) put_spu_gang(ei->i_gang); } -static void spufs_prune_dir(struct dentry *dir) -{ - struct dentry *dentry; - struct hlist_node *n; - - inode_lock(d_inode(dir)); - hlist_for_each_entry_safe(dentry, n, &dir->d_children, d_sib) { - spin_lock(&dentry->d_lock); - if (simple_positive(dentry)) { - dget_dlock(dentry); - __d_drop(dentry); - spin_unlock(&dentry->d_lock); - simple_unlink(d_inode(dir), dentry); - /* XXX: what was dcache_lock protecting here? Other - * filesystems (IB, configfs) release dcache_lock - * before unlink */ - dput(dentry); - } else { - spin_unlock(&dentry->d_lock); - } - } - shrink_dcache_parent(dir); - inode_unlock(d_inode(dir)); -} - /* Caller must hold parent->i_mutex */ -static int spufs_rmdir(struct inode *parent, struct dentry *dir) +static void spufs_rmdir(struct inode *parent, struct dentry *dir) { - /* remove all entries */ - int res; - spufs_prune_dir(dir); - d_drop(dir); - res = simple_rmdir(parent, dir); - /* We have to give up the mm_struct */ - spu_forget(SPUFS_I(d_inode(dir))->i_ctx); - return res; + struct spu_context *ctx = SPUFS_I(d_inode(dir))->i_ctx; + + locked_recursive_removal(dir, NULL); + spu_forget(ctx); } static int spufs_fill_dir(struct dentry *dir, @@ -192,27 +163,45 @@ static int spufs_fill_dir(struct dentry *dir, return -ENOMEM; ret = spufs_new_file(dir->d_sb, dentry, files->ops, files->mode & mode, files->size, ctx); - if (ret) + if (ret) { + dput(dentry); return ret; + } files++; } return 0; } +static void unuse_gang(struct dentry *dir) +{ + struct inode *inode = dir->d_inode; + struct spu_gang *gang = SPUFS_I(inode)->i_gang; + + if (gang) { + bool dead; + + inode_lock(inode); // exclusion with spufs_create_context() + dead = !--gang->alive; + inode_unlock(inode); + + if (dead) + simple_recursive_removal(dir, NULL); + } +} + static int spufs_dir_close(struct inode *inode, struct file *file) { struct inode *parent; struct dentry *dir; - int ret; dir = file->f_path.dentry; parent = d_inode(dir->d_parent); inode_lock_nested(parent, I_MUTEX_PARENT); - ret = spufs_rmdir(parent, dir); + spufs_rmdir(parent, dir); inode_unlock(parent); - WARN_ON(ret); + unuse_gang(dir->d_parent); return dcache_dir_close(inode, file); } @@ -268,11 +257,11 @@ spufs_mkdir(struct inode *dir, struct dentry *dentry, unsigned int flags, ret = spufs_fill_dir(dentry, spufs_dir_debug_contents, mode, ctx); + inode_unlock(inode); + if (ret) spufs_rmdir(dir, dentry); - inode_unlock(inode); - return ret; } @@ -405,7 +394,7 @@ spufs_create_context(struct inode *inode, struct dentry *dentry, { int ret; int affinity; - struct spu_gang *gang; + struct spu_gang *gang = SPUFS_I(inode)->i_gang; struct spu_context *neighbor; struct path path = {.mnt = mnt, .dentry = dentry}; @@ -420,11 +409,15 @@ spufs_create_context(struct inode *inode, struct dentry *dentry, if ((flags & SPU_CREATE_ISOLATE) && !isolated_loader) return -ENODEV; - gang = NULL; + if (gang) { + if (!gang->alive) + return -ENOENT; + gang->alive++; + } + neighbor = NULL; affinity = flags & (SPU_CREATE_AFFINITY_MEM | SPU_CREATE_AFFINITY_SPU); if (affinity) { - gang = SPUFS_I(inode)->i_gang; if (!gang) return -EINVAL; mutex_lock(&gang->aff_mutex); @@ -436,8 +429,11 @@ spufs_create_context(struct inode *inode, struct dentry *dentry, } ret = spufs_mkdir(inode, dentry, flags, mode & 0777); - if (ret) + if (ret) { + if (neighbor) + put_spu_context(neighbor); goto out_aff_unlock; + } if (affinity) { spufs_set_affinity(flags, SPUFS_I(d_inode(dentry))->i_ctx, @@ -448,11 +444,13 @@ spufs_create_context(struct inode *inode, struct dentry *dentry, ret = spufs_context_open(&path); if (ret < 0) - WARN_ON(spufs_rmdir(inode, dentry)); + spufs_rmdir(inode, dentry); out_aff_unlock: if (affinity) mutex_unlock(&gang->aff_mutex); + if (ret && gang) + gang->alive--; // can't reach 0 return ret; } @@ -482,6 +480,7 @@ spufs_mkgang(struct inode *dir, struct dentry *dentry, umode_t mode) inode->i_fop = &simple_dir_operations; d_instantiate(dentry, inode); + dget(dentry); inc_nlink(dir); inc_nlink(d_inode(dentry)); return ret; @@ -492,6 +491,21 @@ out: return ret; } +static int spufs_gang_close(struct inode *inode, struct file *file) +{ + unuse_gang(file->f_path.dentry); + return dcache_dir_close(inode, file); +} + +static const struct file_operations spufs_gang_fops = { + .open = dcache_dir_open, + .release = spufs_gang_close, + .llseek = dcache_dir_lseek, + .read = generic_read_dir, + .iterate_shared = dcache_readdir, + .fsync = noop_fsync, +}; + static int spufs_gang_open(const struct path *path) { int ret; @@ -511,7 +525,7 @@ static int spufs_gang_open(const struct path *path) return PTR_ERR(filp); } - filp->f_op = &simple_dir_operations; + filp->f_op = &spufs_gang_fops; fd_install(ret, filp); return ret; } @@ -526,10 +540,8 @@ static int spufs_create_gang(struct inode *inode, ret = spufs_mkgang(inode, dentry, mode & 0777); if (!ret) { ret = spufs_gang_open(&path); - if (ret < 0) { - int err = simple_rmdir(inode, dentry); - WARN_ON(err); - } + if (ret < 0) + unuse_gang(dentry); } return ret; } diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c index 610ca8570682..8e7ed010bfde 100644 --- a/arch/powerpc/platforms/cell/spufs/sched.c +++ b/arch/powerpc/platforms/cell/spufs/sched.c @@ -508,7 +508,7 @@ static void __spu_del_from_rq(struct spu_context *ctx) if (!list_empty(&ctx->rq)) { if (!--spu_prio->nr_waiting) - del_timer(&spusched_timer); + timer_delete(&spusched_timer); list_del_init(&ctx->rq); if (list_empty(&spu_prio->runq[prio])) @@ -1126,8 +1126,8 @@ void spu_sched_exit(void) remove_proc_entry("spu_loadavg", NULL); - del_timer_sync(&spusched_timer); - del_timer_sync(&spuloadavg_timer); + timer_delete_sync(&spusched_timer); + timer_delete_sync(&spuloadavg_timer); kthread_stop(spusched_task); for (node = 0; node < MAX_NUMNODES; node++) { diff --git a/arch/powerpc/platforms/cell/spufs/spufs.h b/arch/powerpc/platforms/cell/spufs/spufs.h index 84958487f696..d33787c57c39 100644 --- a/arch/powerpc/platforms/cell/spufs/spufs.h +++ b/arch/powerpc/platforms/cell/spufs/spufs.h @@ -151,6 +151,8 @@ struct spu_gang { int aff_flags; struct spu *aff_ref_spu; atomic_t aff_sched_count; + + int alive; }; /* Flag bits for spu_gang aff_flags */ diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c index 36ee3a5056a1..c1bfa4c3444c 100644 --- a/arch/powerpc/platforms/chrp/setup.c +++ b/arch/powerpc/platforms/chrp/setup.c @@ -486,7 +486,7 @@ static void __init chrp_find_8259(void) i8259_init(pic, chrp_int_ack); if (ppc_md.get_irq == NULL) { ppc_md.get_irq = i8259_irq; - irq_set_default_host(i8259_get_host()); + irq_set_default_domain(i8259_get_host()); } if (chrp_mpic != NULL) { cascade_irq = irq_of_parse_and_map(pic, 0); diff --git a/arch/powerpc/platforms/embedded6xx/flipper-pic.c b/arch/powerpc/platforms/embedded6xx/flipper-pic.c index 4d9200bdba78..91a8f0a7086e 100644 --- a/arch/powerpc/platforms/embedded6xx/flipper-pic.c +++ b/arch/powerpc/platforms/embedded6xx/flipper-pic.c @@ -149,8 +149,9 @@ static struct irq_domain * __init flipper_pic_init(struct device_node *np) __flipper_quiesce(io_base); - irq_domain = irq_domain_add_linear(np, FLIPPER_NR_IRQS, - &flipper_irq_domain_ops, io_base); + irq_domain = irq_domain_create_linear(of_fwnode_handle(np), + FLIPPER_NR_IRQS, + &flipper_irq_domain_ops, io_base); if (!irq_domain) { pr_err("failed to allocate irq_domain\n"); return NULL; @@ -172,7 +173,7 @@ unsigned int flipper_pic_get_irq(void) return 0; /* no more IRQs pending */ irq = __ffs(irq_status); - return irq_linear_revmap(flipper_irq_host, irq); + return irq_find_mapping(flipper_irq_host, irq); } /* @@ -190,7 +191,7 @@ void __init flipper_pic_probe(void) flipper_irq_host = flipper_pic_init(np); BUG_ON(!flipper_irq_host); - irq_set_default_host(flipper_irq_host); + irq_set_default_domain(flipper_irq_host); of_node_put(np); } diff --git a/arch/powerpc/platforms/embedded6xx/hlwd-pic.c b/arch/powerpc/platforms/embedded6xx/hlwd-pic.c index 4d2d92de30af..b57e87b0b3ce 100644 --- a/arch/powerpc/platforms/embedded6xx/hlwd-pic.c +++ b/arch/powerpc/platforms/embedded6xx/hlwd-pic.c @@ -175,8 +175,9 @@ static struct irq_domain *__init hlwd_pic_init(struct device_node *np) __hlwd_quiesce(io_base); - irq_domain = irq_domain_add_linear(np, HLWD_NR_IRQS, - &hlwd_irq_domain_ops, io_base); + irq_domain = irq_domain_create_linear(of_fwnode_handle(np), + HLWD_NR_IRQS, + &hlwd_irq_domain_ops, io_base); if (!irq_domain) { pr_err("failed to allocate irq_domain\n"); iounmap(io_base); @@ -189,7 +190,7 @@ static struct irq_domain *__init hlwd_pic_init(struct device_node *np) unsigned int hlwd_pic_get_irq(void) { unsigned int hwirq = __hlwd_pic_get_irq(hlwd_irq_host); - return hwirq ? irq_linear_revmap(hlwd_irq_host, hwirq) : 0; + return hwirq ? irq_find_mapping(hlwd_irq_host, hwirq) : 0; } /* diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c index 0761d98e5be3..d03b41336901 100644 --- a/arch/powerpc/platforms/pasemi/setup.c +++ b/arch/powerpc/platforms/pasemi/setup.c @@ -228,7 +228,7 @@ static void __init nemo_init_IRQ(struct mpic *mpic) irq_set_chained_handler(gpio_virq, sb600_8259_cascade); mpic_unmask_irq(irq_get_irq_data(gpio_virq)); - irq_set_default_host(mpic->irqhost); + irq_set_default_domain(mpic->irqhost); } #else diff --git a/arch/powerpc/platforms/powermac/low_i2c.c b/arch/powerpc/platforms/powermac/low_i2c.c index c097d591670e..02474e27df9b 100644 --- a/arch/powerpc/platforms/powermac/low_i2c.c +++ b/arch/powerpc/platforms/powermac/low_i2c.c @@ -347,7 +347,7 @@ static irqreturn_t kw_i2c_irq(int irq, void *dev_id) unsigned long flags; spin_lock_irqsave(&host->lock, flags); - del_timer(&host->timeout_timer); + timer_delete(&host->timeout_timer); kw_i2c_handle_interrupt(host, kw_read_reg(reg_isr)); if (host->state != state_idle) { host->timeout_timer.expires = jiffies + KW_POLL_TIMEOUT; @@ -359,7 +359,8 @@ static irqreturn_t kw_i2c_irq(int irq, void *dev_id) static void kw_i2c_timeout(struct timer_list *t) { - struct pmac_i2c_host_kw *host = from_timer(host, t, timeout_timer); + struct pmac_i2c_host_kw *host = timer_container_of(host, t, + timeout_timer); unsigned long flags; spin_lock_irqsave(&host->lock, flags); diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c index 2202bf77c7a3..c37783a03d25 100644 --- a/arch/powerpc/platforms/powermac/pic.c +++ b/arch/powerpc/platforms/powermac/pic.c @@ -250,7 +250,7 @@ static unsigned int pmac_pic_get_irq(void) raw_spin_unlock_irqrestore(&pmac_pic_lock, flags); if (unlikely(irq < 0)) return 0; - return irq_linear_revmap(pmac_pic_host, irq); + return irq_find_mapping(pmac_pic_host, irq); } static int pmac_pic_host_match(struct irq_domain *h, struct device_node *node, @@ -327,10 +327,11 @@ static void __init pmac_pic_probe_oldstyle(void) /* * Allocate an irq host */ - pmac_pic_host = irq_domain_add_linear(master, max_irqs, - &pmac_pic_host_ops, NULL); + pmac_pic_host = irq_domain_create_linear(of_fwnode_handle(master), + max_irqs, + &pmac_pic_host_ops, NULL); BUG_ON(pmac_pic_host == NULL); - irq_set_default_host(pmac_pic_host); + irq_set_default_domain(pmac_pic_host); /* Get addresses of first controller if we have a node for it */ BUG_ON(of_address_to_resource(master, 0, &r)); diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c index 6de1cd5d8a58..eb092f293113 100644 --- a/arch/powerpc/platforms/powermac/setup.c +++ b/arch/powerpc/platforms/powermac/setup.c @@ -28,13 +28,11 @@ #include <linux/ptrace.h> #include <linux/export.h> #include <linux/user.h> -#include <linux/tty.h> #include <linux/string.h> #include <linux/delay.h> #include <linux/ioport.h> #include <linux/major.h> #include <linux/initrd.h> -#include <linux/vt_kern.h> #include <linux/console.h> #include <linux/pci.h> #include <linux/adb.h> @@ -45,6 +43,7 @@ #include <linux/root_dev.h> #include <linux/bitops.h> #include <linux/suspend.h> +#include <linux/string_choices.h> #include <linux/of.h> #include <linux/of_platform.h> @@ -238,8 +237,7 @@ static void __init l2cr_init(void) _set_L2CR(0); _set_L2CR(*l2cr); pr_info("L2CR overridden (0x%x), backside cache is %s\n", - *l2cr, ((*l2cr) & 0x80000000) ? - "enabled" : "disabled"); + *l2cr, str_enabled_disabled((*l2cr) & 0x80000000)); } of_node_put(np); break; diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c index 09e7fe24fac1..88e92af8acf9 100644 --- a/arch/powerpc/platforms/powermac/smp.c +++ b/arch/powerpc/platforms/powermac/smp.c @@ -190,7 +190,7 @@ static int __init psurge_secondary_ipi_init(void) { int rc = -ENOMEM; - psurge_host = irq_domain_add_nomap(NULL, ~0, &psurge_host_ops, NULL); + psurge_host = irq_domain_create_nomap(NULL, ~0, &psurge_host_ops, NULL); if (psurge_host) psurge_secondary_virq = irq_create_direct_mapping(psurge_host); diff --git a/arch/powerpc/platforms/powermac/time.c b/arch/powerpc/platforms/powermac/time.c index 8633891b7aa5..b4426a35aca3 100644 --- a/arch/powerpc/platforms/powermac/time.c +++ b/arch/powerpc/platforms/powermac/time.c @@ -15,6 +15,7 @@ #include <linux/kernel.h> #include <linux/param.h> #include <linux/string.h> +#include <linux/string_choices.h> #include <linux/mm.h> #include <linux/init.h> #include <linux/time.h> @@ -77,7 +78,7 @@ long __init pmac_time_init(void) delta |= 0xFF000000UL; dst = ((pmac_xpram_read(PMAC_XPRAM_MACHINE_LOC + 0x8) & 0x80) != 0); printk("GMT Delta read from XPRAM: %d minutes, DST: %s\n", delta/60, - dst ? "on" : "off"); + str_on_off(dst)); #endif return delta; } diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/powernv/Kconfig index 3fbe0295ce14..95d7ba73d43d 100644 --- a/arch/powerpc/platforms/powernv/Kconfig +++ b/arch/powerpc/platforms/powernv/Kconfig @@ -17,7 +17,7 @@ config PPC_POWERNV select MMU_NOTIFIER select FORCE_SMP select ARCH_SUPPORTS_PER_VMA_LOCK - select PPC_RADIX_BROADCAST_TLBIE + select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU default y config OPAL_PRD diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile index 19f0fc5c6f1b..9e5d0c847ee2 100644 --- a/arch/powerpc/platforms/powernv/Makefile +++ b/arch/powerpc/platforms/powernv/Makefile @@ -21,7 +21,6 @@ obj-$(CONFIG_PRESERVE_FA_DUMP) += opal-fadump.o obj-$(CONFIG_OPAL_CORE) += opal-core.o obj-$(CONFIG_PCI) += pci.o pci-ioda.o pci-ioda-tce.o obj-$(CONFIG_PCI_IOV) += pci-sriov.o -obj-$(CONFIG_CXL_BASE) += pci-cxl.o obj-$(CONFIG_EEH) += eeh-powernv.o obj-$(CONFIG_MEMORY_FAILURE) += opal-memory-errors.o obj-$(CONFIG_OPAL_PRD) += opal-prd.o diff --git a/arch/powerpc/platforms/powernv/memtrace.c b/arch/powerpc/platforms/powernv/memtrace.c index 4ac9808e55a4..2ea30b343354 100644 --- a/arch/powerpc/platforms/powernv/memtrace.c +++ b/arch/powerpc/platforms/powernv/memtrace.c @@ -48,11 +48,15 @@ static ssize_t memtrace_read(struct file *filp, char __user *ubuf, static int memtrace_mmap(struct file *filp, struct vm_area_struct *vma) { struct memtrace_entry *ent = filp->private_data; + unsigned long ent_nrpages = ent->size >> PAGE_SHIFT; + unsigned long vma_nrpages = vma_pages(vma); - if (ent->size < vma->vm_end - vma->vm_start) + /* The requested page offset should be within object's page count */ + if (vma->vm_pgoff >= ent_nrpages) return -EINVAL; - if (vma->vm_pgoff << PAGE_SHIFT >= ent->size) + /* The requested mapping range should remain within the bounds */ + if (vma_nrpages > ent_nrpages - vma->vm_pgoff) return -EINVAL; vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); diff --git a/arch/powerpc/platforms/powernv/ocxl.c b/arch/powerpc/platforms/powernv/ocxl.c index 64a9c7125c29..f8139948348e 100644 --- a/arch/powerpc/platforms/powernv/ocxl.c +++ b/arch/powerpc/platforms/powernv/ocxl.c @@ -172,12 +172,11 @@ static void pnv_ocxl_fixup_actag(struct pci_dev *dev) if (phb->type != PNV_PHB_NPU_OCAPI) return; - mutex_lock(&links_list_lock); + guard(mutex)(&links_list_lock); link = find_link(dev); if (!link) { dev_warn(&dev->dev, "couldn't update actag information\n"); - mutex_unlock(&links_list_lock); return; } @@ -206,7 +205,6 @@ static void pnv_ocxl_fixup_actag(struct pci_dev *dev) dev_dbg(&dev->dev, "total actags for function: %d\n", link->fn_desired_actags[PCI_FUNC(dev->devfn)]); - mutex_unlock(&links_list_lock); } DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pnv_ocxl_fixup_actag); @@ -253,12 +251,11 @@ int pnv_ocxl_get_actag(struct pci_dev *dev, u16 *base, u16 *enabled, { struct npu_link *link; - mutex_lock(&links_list_lock); + guard(mutex)(&links_list_lock); link = find_link(dev); if (!link) { dev_err(&dev->dev, "actag information not found\n"); - mutex_unlock(&links_list_lock); return -ENODEV; } /* @@ -274,7 +271,6 @@ int pnv_ocxl_get_actag(struct pci_dev *dev, u16 *base, u16 *enabled, *enabled = link->fn_actags[PCI_FUNC(dev->devfn)].count; *supported = link->fn_desired_actags[PCI_FUNC(dev->devfn)]; - mutex_unlock(&links_list_lock); return 0; } EXPORT_SYMBOL_GPL(pnv_ocxl_get_actag); @@ -293,12 +289,11 @@ int pnv_ocxl_get_pasid_count(struct pci_dev *dev, int *count) * * We only support one AFU-carrying function for now. */ - mutex_lock(&links_list_lock); + guard(mutex)(&links_list_lock); link = find_link(dev); if (!link) { dev_err(&dev->dev, "actag information not found\n"); - mutex_unlock(&links_list_lock); return -ENODEV; } @@ -309,7 +304,6 @@ int pnv_ocxl_get_pasid_count(struct pci_dev *dev, int *count) break; } - mutex_unlock(&links_list_lock); dev_dbg(&dev->dev, "%d PASIDs available for function\n", rc ? 0 : *count); return rc; diff --git a/arch/powerpc/platforms/powernv/opal-core.c b/arch/powerpc/platforms/powernv/opal-core.c index e652da8f986f..784602a48afb 100644 --- a/arch/powerpc/platforms/powernv/opal-core.c +++ b/arch/powerpc/platforms/powernv/opal-core.c @@ -208,7 +208,7 @@ static ssize_t read_opalcore(struct file *file, struct kobject *kobj, static struct bin_attribute opal_core_attr __ro_after_init = { .attr = {.name = "core", .mode = 0400}, - .read_new = read_opalcore + .read = read_opalcore }; /* @@ -607,7 +607,7 @@ static const struct bin_attribute *const mpipl_bin_attr[] = { static const struct attribute_group mpipl_group = { .attrs = mpipl_attr, - .bin_attrs_new = mpipl_bin_attr, + .bin_attrs = mpipl_bin_attr, }; static int __init opalcore_init(void) diff --git a/arch/powerpc/platforms/powernv/opal-dump.c b/arch/powerpc/platforms/powernv/opal-dump.c index 27e25693cf39..cc3cc9ddf9d1 100644 --- a/arch/powerpc/platforms/powernv/opal-dump.c +++ b/arch/powerpc/platforms/powernv/opal-dump.c @@ -342,7 +342,7 @@ static void create_dump_obj(uint32_t id, size_t size, uint32_t type) dump->dump_attr.attr.name = "dump"; dump->dump_attr.attr.mode = 0400; dump->dump_attr.size = size; - dump->dump_attr.read_new = dump_attr_read; + dump->dump_attr.read = dump_attr_read; dump->id = id; dump->size = size; diff --git a/arch/powerpc/platforms/powernv/opal-elog.c b/arch/powerpc/platforms/powernv/opal-elog.c index de33f354e9fd..c3fc5d258146 100644 --- a/arch/powerpc/platforms/powernv/opal-elog.c +++ b/arch/powerpc/platforms/powernv/opal-elog.c @@ -203,7 +203,7 @@ static void create_elog_obj(uint64_t id, size_t size, uint64_t type) elog->raw_attr.attr.name = "raw"; elog->raw_attr.attr.mode = 0400; elog->raw_attr.size = size; - elog->raw_attr.read_new = raw_attr_read; + elog->raw_attr.read = raw_attr_read; elog->id = id; elog->size = size; diff --git a/arch/powerpc/platforms/powernv/opal-flash.c b/arch/powerpc/platforms/powernv/opal-flash.c index fd8c8621e973..a3f7a2928767 100644 --- a/arch/powerpc/platforms/powernv/opal-flash.c +++ b/arch/powerpc/platforms/powernv/opal-flash.c @@ -493,7 +493,7 @@ out: static const struct bin_attribute image_data_attr = { .attr = {.name = "image", .mode = 0200}, .size = MAX_IMAGE_SIZE, /* Limit image size */ - .write_new = image_data_write, + .write = image_data_write, }; static struct kobj_attribute validate_attribute = diff --git a/arch/powerpc/platforms/powernv/opal-irqchip.c b/arch/powerpc/platforms/powernv/opal-irqchip.c index d92759c21fae..e180bd8e1400 100644 --- a/arch/powerpc/platforms/powernv/opal-irqchip.c +++ b/arch/powerpc/platforms/powernv/opal-irqchip.c @@ -191,7 +191,8 @@ int __init opal_event_init(void) * fall back to the legacy method (opal_event_request(...)) * anyway. */ dn = of_find_compatible_node(NULL, NULL, "ibm,opal-event"); - opal_event_irqchip.domain = irq_domain_add_linear(dn, MAX_NUM_EVENTS, + opal_event_irqchip.domain = irq_domain_create_linear(of_fwnode_handle(dn), + MAX_NUM_EVENTS, &opal_event_domain_ops, &opal_event_irqchip); of_node_put(dn); if (!opal_event_irqchip.domain) { diff --git a/arch/powerpc/platforms/powernv/opal-msglog.c b/arch/powerpc/platforms/powernv/opal-msglog.c index f1988d0ab45c..992a6b379a66 100644 --- a/arch/powerpc/platforms/powernv/opal-msglog.c +++ b/arch/powerpc/platforms/powernv/opal-msglog.c @@ -102,7 +102,7 @@ static ssize_t opal_msglog_read(struct file *file, struct kobject *kobj, static struct bin_attribute opal_msglog_attr __ro_after_init = { .attr = {.name = "msglog", .mode = 0400}, - .read_new = opal_msglog_read + .read = opal_msglog_read }; struct memcons *__init memcons_init(struct device_node *node, const char *mc_prop_name) diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c index 9ec265fcaff4..09bd93464b4f 100644 --- a/arch/powerpc/platforms/powernv/opal.c +++ b/arch/powerpc/platforms/powernv/opal.c @@ -815,7 +815,7 @@ static int opal_add_one_export(struct kobject *parent, const char *export_name, sysfs_bin_attr_init(attr); attr->attr.name = name; attr->attr.mode = 0400; - attr->read_new = sysfs_bin_attr_simple_read; + attr->read = sysfs_bin_attr_simple_read; attr->private = __va(vals[0]); attr->size = vals[1]; diff --git a/arch/powerpc/platforms/powernv/pci-cxl.c b/arch/powerpc/platforms/powernv/pci-cxl.c deleted file mode 100644 index 7e419de71db8..000000000000 --- a/arch/powerpc/platforms/powernv/pci-cxl.c +++ /dev/null @@ -1,153 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * Copyright 2014-2016 IBM Corp. - */ - -#include <linux/module.h> -#include <misc/cxl-base.h> -#include <asm/pnv-pci.h> -#include <asm/opal.h> - -#include "pci.h" - -int pnv_phb_to_cxl_mode(struct pci_dev *dev, uint64_t mode) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - struct pnv_ioda_pe *pe; - int rc; - - pe = pnv_ioda_get_pe(dev); - if (!pe) - return -ENODEV; - - pe_info(pe, "Switching PHB to CXL\n"); - - rc = opal_pci_set_phb_cxl_mode(phb->opal_id, mode, pe->pe_number); - if (rc == OPAL_UNSUPPORTED) - dev_err(&dev->dev, "Required cxl mode not supported by firmware - update skiboot\n"); - else if (rc) - dev_err(&dev->dev, "opal_pci_set_phb_cxl_mode failed: %i\n", rc); - - return rc; -} -EXPORT_SYMBOL(pnv_phb_to_cxl_mode); - -/* Find PHB for cxl dev and allocate MSI hwirqs? - * Returns the absolute hardware IRQ number - */ -int pnv_cxl_alloc_hwirqs(struct pci_dev *dev, int num) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - int hwirq = msi_bitmap_alloc_hwirqs(&phb->msi_bmp, num); - - if (hwirq < 0) { - dev_warn(&dev->dev, "Failed to find a free MSI\n"); - return -ENOSPC; - } - - return phb->msi_base + hwirq; -} -EXPORT_SYMBOL(pnv_cxl_alloc_hwirqs); - -void pnv_cxl_release_hwirqs(struct pci_dev *dev, int hwirq, int num) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - - msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq - phb->msi_base, num); -} -EXPORT_SYMBOL(pnv_cxl_release_hwirqs); - -void pnv_cxl_release_hwirq_ranges(struct cxl_irq_ranges *irqs, - struct pci_dev *dev) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - int i, hwirq; - - for (i = 1; i < CXL_IRQ_RANGES; i++) { - if (!irqs->range[i]) - continue; - pr_devel("cxl release irq range 0x%x: offset: 0x%lx limit: %ld\n", - i, irqs->offset[i], - irqs->range[i]); - hwirq = irqs->offset[i] - phb->msi_base; - msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq, - irqs->range[i]); - } -} -EXPORT_SYMBOL(pnv_cxl_release_hwirq_ranges); - -int pnv_cxl_alloc_hwirq_ranges(struct cxl_irq_ranges *irqs, - struct pci_dev *dev, int num) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - int i, hwirq, try; - - memset(irqs, 0, sizeof(struct cxl_irq_ranges)); - - /* 0 is reserved for the multiplexed PSL DSI interrupt */ - for (i = 1; i < CXL_IRQ_RANGES && num; i++) { - try = num; - while (try) { - hwirq = msi_bitmap_alloc_hwirqs(&phb->msi_bmp, try); - if (hwirq >= 0) - break; - try /= 2; - } - if (!try) - goto fail; - - irqs->offset[i] = phb->msi_base + hwirq; - irqs->range[i] = try; - pr_devel("cxl alloc irq range 0x%x: offset: 0x%lx limit: %li\n", - i, irqs->offset[i], irqs->range[i]); - num -= try; - } - if (num) - goto fail; - - return 0; -fail: - pnv_cxl_release_hwirq_ranges(irqs, dev); - return -ENOSPC; -} -EXPORT_SYMBOL(pnv_cxl_alloc_hwirq_ranges); - -int pnv_cxl_get_irq_count(struct pci_dev *dev) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - - return phb->msi_bmp.irq_count; -} -EXPORT_SYMBOL(pnv_cxl_get_irq_count); - -int pnv_cxl_ioda_msi_setup(struct pci_dev *dev, unsigned int hwirq, - unsigned int virq) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - struct pnv_phb *phb = hose->private_data; - unsigned int xive_num = hwirq - phb->msi_base; - struct pnv_ioda_pe *pe; - int rc; - - if (!(pe = pnv_ioda_get_pe(dev))) - return -ENODEV; - - /* Assign XIVE to PE */ - rc = opal_pci_set_xive_pe(phb->opal_id, pe->pe_number, xive_num); - if (rc) { - pe_warn(pe, "%s: OPAL error %d setting msi_base 0x%x " - "hwirq 0x%x XIVE 0x%x PE\n", - pci_name(dev), rc, phb->msi_base, hwirq, xive_num); - return -EIO; - } - pnv_set_msi_irq_chip(phb, virq); - - return 0; -} -EXPORT_SYMBOL(pnv_cxl_ioda_msi_setup); diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index b0a14e48175c..d8ccf2c9b98a 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -39,8 +39,6 @@ #include <asm/mmzone.h> #include <asm/xive.h> -#include <misc/cxl-base.h> - #include "powernv.h" #include "pci.h" #include "../../../../drivers/pci/pci.h" @@ -1636,47 +1634,6 @@ int64_t pnv_opal_pci_msi_eoi(struct irq_data *d) return opal_pci_msi_eoi(phb->opal_id, d->parent_data->hwirq); } -/* - * The IRQ data is mapped in the XICS domain, with OPAL HW IRQ numbers - */ -static void pnv_ioda2_msi_eoi(struct irq_data *d) -{ - int64_t rc; - unsigned int hw_irq = (unsigned int)irqd_to_hwirq(d); - struct pci_controller *hose = irq_data_get_irq_chip_data(d); - struct pnv_phb *phb = hose->private_data; - - rc = opal_pci_msi_eoi(phb->opal_id, hw_irq); - WARN_ON_ONCE(rc); - - icp_native_eoi(d); -} - -/* P8/CXL only */ -void pnv_set_msi_irq_chip(struct pnv_phb *phb, unsigned int virq) -{ - struct irq_data *idata; - struct irq_chip *ichip; - - /* The MSI EOI OPAL call is only needed on PHB3 */ - if (phb->model != PNV_PHB_MODEL_PHB3) - return; - - if (!phb->ioda.irq_chip_init) { - /* - * First time we setup an MSI IRQ, we need to setup the - * corresponding IRQ chip to route correctly. - */ - idata = irq_get_irq_data(virq); - ichip = irq_data_get_irq_chip(idata); - phb->ioda.irq_chip_init = 1; - phb->ioda.irq_chip = *ichip; - phb->ioda.irq_chip.irq_eoi = pnv_ioda2_msi_eoi; - } - irq_set_chip(virq, &phb->ioda.irq_chip); - irq_set_chip_data(virq, phb->hose); -} - static struct irq_chip pnv_pci_msi_irq_chip; /* @@ -1924,7 +1881,7 @@ static const struct irq_domain_ops pnv_irq_domain_ops = { static int __init pnv_msi_allocate_domains(struct pci_controller *hose, unsigned int count) { struct pnv_phb *phb = hose->private_data; - struct irq_domain *parent = irq_get_default_host(); + struct irq_domain *parent = irq_get_default_domain(); hose->fwnode = irq_domain_alloc_named_id_fwnode("PNV-MSI", phb->opal_id); if (!hose->fwnode) @@ -1940,7 +1897,7 @@ static int __init pnv_msi_allocate_domains(struct pci_controller *hose, unsigned return -ENOMEM; } - hose->msi_domain = pci_msi_create_irq_domain(of_node_to_fwnode(hose->dn), + hose->msi_domain = pci_msi_create_irq_domain(of_fwnode_handle(hose->dn), &pnv_msi_domain_info, hose->dev_domain); if (!hose->msi_domain) { diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c index 35f566aa0424..b2c1da025410 100644 --- a/arch/powerpc/platforms/powernv/pci.c +++ b/arch/powerpc/platforms/powernv/pci.c @@ -14,7 +14,6 @@ #include <linux/io.h> #include <linux/msi.h> #include <linux/iommu.h> -#include <linux/sched/mm.h> #include <asm/sections.h> #include <asm/io.h> @@ -33,8 +32,6 @@ #include "powernv.h" #include "pci.h" -static DEFINE_MUTEX(tunnel_mutex); - int pnv_pci_get_slot_id(struct device_node *np, uint64_t *id) { struct device_node *node = np; @@ -744,64 +741,6 @@ struct iommu_table *pnv_pci_table_alloc(int nid) return tbl; } -struct device_node *pnv_pci_get_phb_node(struct pci_dev *dev) -{ - struct pci_controller *hose = pci_bus_to_host(dev->bus); - - return of_node_get(hose->dn); -} -EXPORT_SYMBOL(pnv_pci_get_phb_node); - -int pnv_pci_set_tunnel_bar(struct pci_dev *dev, u64 addr, int enable) -{ - struct pnv_phb *phb = pci_bus_to_pnvhb(dev->bus); - u64 tunnel_bar; - __be64 val; - int rc; - - if (!opal_check_token(OPAL_PCI_GET_PBCQ_TUNNEL_BAR)) - return -ENXIO; - if (!opal_check_token(OPAL_PCI_SET_PBCQ_TUNNEL_BAR)) - return -ENXIO; - - mutex_lock(&tunnel_mutex); - rc = opal_pci_get_pbcq_tunnel_bar(phb->opal_id, &val); - if (rc != OPAL_SUCCESS) { - rc = -EIO; - goto out; - } - tunnel_bar = be64_to_cpu(val); - if (enable) { - /* - * Only one device per PHB can use atomics. - * Our policy is first-come, first-served. - */ - if (tunnel_bar) { - if (tunnel_bar != addr) - rc = -EBUSY; - else - rc = 0; /* Setting same address twice is ok */ - goto out; - } - } else { - /* - * The device that owns atomics and wants to release - * them must pass the same address with enable == 0. - */ - if (tunnel_bar != addr) { - rc = -EPERM; - goto out; - } - addr = 0x0ULL; - } - rc = opal_pci_set_pbcq_tunnel_bar(phb->opal_id, addr); - rc = opal_error_code(rc); -out: - mutex_unlock(&tunnel_mutex); - return rc; -} -EXPORT_SYMBOL_GPL(pnv_pci_set_tunnel_bar); - void pnv_pci_shutdown(void) { struct pci_controller *hose; diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h index 93fba1f8661f..42075501663b 100644 --- a/arch/powerpc/platforms/powernv/pci.h +++ b/arch/powerpc/platforms/powernv/pci.h @@ -163,7 +163,6 @@ struct pnv_phb { unsigned int *io_segmap; /* IRQ chip */ - int irq_chip_init; struct irq_chip irq_chip; /* Sorted list of used PE's based @@ -281,7 +280,6 @@ extern int pnv_eeh_phb_reset(struct pci_controller *hose, int option); extern struct pnv_ioda_pe *pnv_pci_bdfn_to_pe(struct pnv_phb *phb, u16 bdfn); extern struct pnv_ioda_pe *pnv_ioda_get_pe(struct pci_dev *dev); -extern void pnv_set_msi_irq_chip(struct pnv_phb *phb, unsigned int virq); extern unsigned long pnv_pci_ioda2_get_table_size(__u32 page_shift, __u64 window_size, __u32 levels); extern int pnv_eeh_post_init(void); diff --git a/arch/powerpc/platforms/powernv/ultravisor.c b/arch/powerpc/platforms/powernv/ultravisor.c index 157d9a8134e4..c526871a1229 100644 --- a/arch/powerpc/platforms/powernv/ultravisor.c +++ b/arch/powerpc/platforms/powernv/ultravisor.c @@ -40,7 +40,7 @@ static ssize_t uv_msglog_read(struct file *file, struct kobject *kobj, static struct bin_attribute uv_msglog_attr __ro_after_init = { .attr = {.name = "msglog", .mode = 0400}, - .read_new = uv_msglog_read + .read = uv_msglog_read }; static int __init uv_init(void) diff --git a/arch/powerpc/platforms/ps3/device-init.c b/arch/powerpc/platforms/ps3/device-init.c index 61722133eb2d..22d91ac424dd 100644 --- a/arch/powerpc/platforms/ps3/device-init.c +++ b/arch/powerpc/platforms/ps3/device-init.c @@ -14,6 +14,7 @@ #include <linux/slab.h> #include <linux/reboot.h> #include <linux/rcuwait.h> +#include <linux/string_choices.h> #include <asm/firmware.h> #include <asm/lv1call.h> @@ -724,7 +725,7 @@ static irqreturn_t ps3_notification_interrupt(int irq, void *data) static int ps3_notification_read_write(struct ps3_notification_device *dev, u64 lpar, int write) { - const char *op = write ? "write" : "read"; + const char *op = str_write_read(write); unsigned long flags; int res; diff --git a/arch/powerpc/platforms/ps3/interrupt.c b/arch/powerpc/platforms/ps3/interrupt.c index af3fe9f04f24..a4ad4b49eef7 100644 --- a/arch/powerpc/platforms/ps3/interrupt.c +++ b/arch/powerpc/platforms/ps3/interrupt.c @@ -743,8 +743,8 @@ void __init ps3_init_IRQ(void) unsigned cpu; struct irq_domain *host; - host = irq_domain_add_nomap(NULL, PS3_PLUG_MAX + 1, &ps3_host_ops, NULL); - irq_set_default_host(host); + host = irq_domain_create_nomap(NULL, PS3_PLUG_MAX + 1, &ps3_host_ops, NULL); + irq_set_default_domain(host); for_each_possible_cpu(cpu) { struct ps3_private *pd = &per_cpu(ps3_private, cpu); diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig index a934c2a262f6..fa3c2fff082a 100644 --- a/arch/powerpc/platforms/pseries/Kconfig +++ b/arch/powerpc/platforms/pseries/Kconfig @@ -23,7 +23,7 @@ config PPC_PSERIES select FORCE_SMP select SWIOTLB select ARCH_SUPPORTS_PER_VMA_LOCK - select PPC_RADIX_BROADCAST_TLBIE + select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU default y config PARAVIRT diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index 3f3e3492e436..57222678bb3f 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -3,7 +3,8 @@ ccflags-$(CONFIG_PPC_PSERIES_DEBUG) += -DDEBUG obj-y := lpar.o hvCall.o nvram.o reconfig.o \ of_helpers.o rtas-work-area.o papr-sysparm.o \ - papr-vpd.o \ + papr-rtas-common.o papr-vpd.o papr-indices.o \ + papr-platform-dump.o papr-phy-attest.o \ setup.o iommu.o event_sources.o ras.o \ firmware.o power.o dlpar.o mobility.o rng.o \ pci.o pci_dlpar.o eeh_pseries.o msi.o \ diff --git a/arch/powerpc/platforms/pseries/cmm.c b/arch/powerpc/platforms/pseries/cmm.c index 5f4037c1d7fe..5e0a718d1be7 100644 --- a/arch/powerpc/platforms/pseries/cmm.c +++ b/arch/powerpc/platforms/pseries/cmm.c @@ -532,7 +532,6 @@ static int cmm_migratepage(struct balloon_dev_info *b_dev_info, spin_lock_irqsave(&b_dev_info->pages_lock, flags); balloon_page_insert(b_dev_info, newpage); - balloon_page_delete(page); b_dev_info->isolated_pages--; spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); @@ -542,6 +541,7 @@ static int cmm_migratepage(struct balloon_dev_info *b_dev_info, */ plpar_page_set_active(page); + balloon_page_finalize(page); /* balloon page list reference */ put_page(page); diff --git a/arch/powerpc/platforms/pseries/dlpar.c b/arch/powerpc/platforms/pseries/dlpar.c index 213aa26dc8b3..979487da6522 100644 --- a/arch/powerpc/platforms/pseries/dlpar.c +++ b/arch/powerpc/platforms/pseries/dlpar.c @@ -404,6 +404,45 @@ get_device_node_with_drc_info(u32 index) return NULL; } +static struct device_node * +get_device_node_with_drc_indexes(u32 drc_index) +{ + struct device_node *np = NULL; + u32 nr_indexes, index; + int i, rc; + + for_each_node_with_property(np, "ibm,drc-indexes") { + /* + * First element in the array is the total number of + * DRC indexes returned. + */ + rc = of_property_read_u32_index(np, "ibm,drc-indexes", + 0, &nr_indexes); + if (rc) + goto out_put_np; + + /* + * Retrieve DRC index from the list and return the + * device node if matched with the specified index. + */ + for (i = 0; i < nr_indexes; i++) { + rc = of_property_read_u32_index(np, "ibm,drc-indexes", + i+1, &index); + if (rc) + goto out_put_np; + + if (drc_index == index) + return np; + } + } + + return NULL; + +out_put_np: + of_node_put(np); + return NULL; +} + static int dlpar_hp_dt_add(u32 index) { struct device_node *np, *nodes; @@ -423,10 +462,19 @@ static int dlpar_hp_dt_add(u32 index) goto out; } + /* + * Recent FW provides ibm,drc-info property. So search + * for the user specified DRC index from ibm,drc-info + * property. If this property is not available, search + * in the indexes array from ibm,drc-indexes property. + */ np = get_device_node_with_drc_info(index); - if (!np) - return -EIO; + if (!np) { + np = get_device_node_with_drc_indexes(index); + if (!np) + return -EIO; + } /* Next, configure the connector. */ nodes = dlpar_configure_connector(cpu_to_be32(index), np); diff --git a/arch/powerpc/platforms/pseries/htmdump.c b/arch/powerpc/platforms/pseries/htmdump.c index 57fc1700f604..742ec52c9d4d 100644 --- a/arch/powerpc/platforms/pseries/htmdump.c +++ b/arch/powerpc/platforms/pseries/htmdump.c @@ -10,28 +10,40 @@ #include <asm/io.h> #include <asm/machdep.h> #include <asm/plpar_wrappers.h> +#include <asm/kvm_guest.h> static void *htm_buf; +static void *htm_status_buf; +static void *htm_info_buf; +static void *htm_caps_buf; static u32 nodeindex; static u32 nodalchipindex; static u32 coreindexonchip; static u32 htmtype; +static u32 htmconfigure; +static u32 htmstart; +static u32 htmsetup; +static u64 htmflags; + static struct dentry *htmdump_debugfs_dir; +#define HTM_ENABLE 1 +#define HTM_DISABLE 0 +#define HTM_NOWRAP 1 +#define HTM_WRAP 0 -static ssize_t htmdump_read(struct file *filp, char __user *ubuf, - size_t count, loff_t *ppos) +/* + * Check the return code for H_HTM hcall. + * Return non-zero value (1) if either H_PARTIAL or H_SUCCESS + * is returned. For other return codes: + * Return zero if H_NOT_AVAILABLE. + * Return -EBUSY if hcall return busy. + * Return -EINVAL if any parameter or operation is not valid. + * Return -EPERM if HTM Virtualization Engine Technology code + * is not applied. + * Return -EIO if the HTM state is not valid. + */ +static ssize_t htm_return_check(long rc) { - void *htm_buf = filp->private_data; - unsigned long page, read_size, available; - loff_t offset; - long rc; - - page = ALIGN_DOWN(*ppos, PAGE_SIZE); - offset = (*ppos) % PAGE_SIZE; - - rc = htm_get_dump_hardware(nodeindex, nodalchipindex, coreindexonchip, - htmtype, virt_to_phys(htm_buf), PAGE_SIZE, page); - switch (rc) { case H_SUCCESS: /* H_PARTIAL for the case where all available data can't be @@ -65,6 +77,38 @@ static ssize_t htmdump_read(struct file *filp, char __user *ubuf, return -EPERM; } + /* + * Return 1 for H_SUCCESS/H_PARTIAL + */ + return 1; +} + +static ssize_t htmdump_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + void *htm_buf = filp->private_data; + unsigned long page, read_size, available; + loff_t offset; + long rc, ret; + + page = ALIGN_DOWN(*ppos, PAGE_SIZE); + offset = (*ppos) % PAGE_SIZE; + + /* + * Invoke H_HTM call with: + * - operation as htm dump (H_HTM_OP_DUMP_DATA) + * - last three values are address, size and offset + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_DUMP_DATA, virt_to_phys(htm_buf), + PAGE_SIZE, page); + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed for op: H_HTM_OP_DUMP_DATA, returning %ld\n", ret); + return ret; + } + available = PAGE_SIZE; read_size = min(count, available); *ppos += read_size; @@ -77,6 +121,292 @@ static const struct file_operations htmdump_fops = { .open = simple_open, }; +static int htmconfigure_set(void *data, u64 val) +{ + long rc, ret; + unsigned long param1 = -1, param2 = -1; + + /* + * value as 1 : configure HTM. + * value as 0 : deconfigure HTM. Return -EINVAL for + * other values. + */ + if (val == HTM_ENABLE) { + /* + * Invoke H_HTM call with: + * - operation as htm configure (H_HTM_OP_CONFIGURE) + * - If htmflags is set, param1 and param2 will be -1 + * which is an indicator to use default htm mode reg mask + * and htm mode reg value. + * - last three values are unused, hence set to zero + */ + if (!htmflags) { + param1 = 0; + param2 = 0; + } + + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_CONFIGURE, param1, param2, 0); + } else if (val == HTM_DISABLE) { + /* + * Invoke H_HTM call with: + * - operation as htm deconfigure (H_HTM_OP_DECONFIGURE) + * - last three values are unused, hence set to zero + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_DECONFIGURE, 0, 0, 0); + } else + return -EINVAL; + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed, returning %ld\n", ret); + return ret; + } + + /* Set htmconfigure if operation succeeds */ + htmconfigure = val; + + return 0; +} + +static int htmconfigure_get(void *data, u64 *val) +{ + *val = htmconfigure; + return 0; +} + +static int htmstart_set(void *data, u64 val) +{ + long rc, ret; + + /* + * value as 1: start HTM + * value as 0: stop HTM + * Return -EINVAL for other values. + */ + if (val == HTM_ENABLE) { + /* + * Invoke H_HTM call with: + * - operation as htm start (H_HTM_OP_START) + * - last three values are unused, hence set to zero + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_START, 0, 0, 0); + + } else if (val == HTM_DISABLE) { + /* + * Invoke H_HTM call with: + * - operation as htm stop (H_HTM_OP_STOP) + * - last three values are unused, hence set to zero + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_STOP, 0, 0, 0); + } else + return -EINVAL; + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed, returning %ld\n", ret); + return ret; + } + + /* Set htmstart if H_HTM_OP_START/H_HTM_OP_STOP operation succeeds */ + htmstart = val; + + return 0; +} + +static int htmstart_get(void *data, u64 *val) +{ + *val = htmstart; + return 0; +} + +static ssize_t htmstatus_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + void *htm_status_buf = filp->private_data; + long rc, ret; + u64 *num_entries; + u64 to_copy; + int htmstatus_flag; + + /* + * Invoke H_HTM call with: + * - operation as htm status (H_HTM_OP_STATUS) + * - last three values as addr, size and offset + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_STATUS, virt_to_phys(htm_status_buf), + PAGE_SIZE, 0); + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed for op: H_HTM_OP_STATUS, returning %ld\n", ret); + return ret; + } + + /* + * HTM status buffer, start of buffer + 0x10 gives the + * number of HTM entries in the buffer. Each nest htm status + * entry is 0x6 bytes where each core htm status entry is + * 0x8 bytes. + * So total count to copy is: + * 32 bytes (for first 7 fields) + (number of HTM entries * entry size) + */ + num_entries = htm_status_buf + 0x10; + if (htmtype == 0x2) + htmstatus_flag = 0x8; + else + htmstatus_flag = 0x6; + to_copy = 32 + (be64_to_cpu(*num_entries) * htmstatus_flag); + return simple_read_from_buffer(ubuf, count, ppos, htm_status_buf, to_copy); +} + +static const struct file_operations htmstatus_fops = { + .llseek = NULL, + .read = htmstatus_read, + .open = simple_open, +}; + +static ssize_t htminfo_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + void *htm_info_buf = filp->private_data; + long rc, ret; + u64 *num_entries; + u64 to_copy; + + /* + * Invoke H_HTM call with: + * - operation as htm status (H_HTM_OP_STATUS) + * - last three values as addr, size and offset + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_DUMP_SYSPROC_CONF, virt_to_phys(htm_info_buf), + PAGE_SIZE, 0); + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed for op: H_HTM_OP_DUMP_SYSPROC_CONF, returning %ld\n", ret); + return ret; + } + + /* + * HTM status buffer, start of buffer + 0x10 gives the + * number of HTM entries in the buffer. Each entry of processor + * is 16 bytes. + * + * So total count to copy is: + * 32 bytes (for first 5 fields) + (number of HTM entries * entry size) + */ + num_entries = htm_info_buf + 0x10; + to_copy = 32 + (be64_to_cpu(*num_entries) * 16); + return simple_read_from_buffer(ubuf, count, ppos, htm_info_buf, to_copy); +} + +static ssize_t htmcaps_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + void *htm_caps_buf = filp->private_data; + long rc, ret; + + /* + * Invoke H_HTM call with: + * - operation as htm capabilities (H_HTM_OP_CAPABILITIES) + * - last three values as addr, size (0x80 for Capabilities Output Buffer + * and zero + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_CAPABILITIES, virt_to_phys(htm_caps_buf), + 0x80, 0); + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed for op: H_HTM_OP_CAPABILITIES, returning %ld\n", ret); + return ret; + } + + return simple_read_from_buffer(ubuf, count, ppos, htm_caps_buf, 0x80); +} + +static const struct file_operations htminfo_fops = { + .llseek = NULL, + .read = htminfo_read, + .open = simple_open, +}; + +static const struct file_operations htmcaps_fops = { + .llseek = NULL, + .read = htmcaps_read, + .open = simple_open, +}; + +static int htmsetup_set(void *data, u64 val) +{ + long rc, ret; + + /* + * Input value: HTM buffer size in the power of 2 + * example: hex value 0x21 ( decimal: 33 ) is for + * 8GB + * Invoke H_HTM call with: + * - operation as htm start (H_HTM_OP_SETUP) + * - parameter 1 set to input value. + * - last two values are unused, hence set to zero + */ + rc = htm_hcall_wrapper(htmflags, nodeindex, nodalchipindex, coreindexonchip, + htmtype, H_HTM_OP_SETUP, val, 0, 0); + + ret = htm_return_check(rc); + if (ret <= 0) { + pr_debug("H_HTM hcall failed for op: H_HTM_OP_SETUP, returning %ld\n", ret); + return ret; + } + + /* Set htmsetup if H_HTM_OP_SETUP operation succeeds */ + htmsetup = val; + + return 0; +} + +static int htmsetup_get(void *data, u64 *val) +{ + *val = htmsetup; + return 0; +} + +static int htmflags_set(void *data, u64 val) +{ + /* + * Input value: + * Currently supported flag value is to enable/disable + * HTM buffer wrap. wrap is used along with "configure" + * to prevent HTM buffer from wrapping. + * Writing 1 will set noWrap while configuring HTM + */ + if (val == HTM_NOWRAP) + htmflags = H_HTM_FLAGS_NOWRAP; + else if (val == HTM_WRAP) + htmflags = 0; + else + return -EINVAL; + + return 0; +} + +static int htmflags_get(void *data, u64 *val) +{ + *val = htmflags; + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(htmconfigure_fops, htmconfigure_get, htmconfigure_set, "%llu\n"); +DEFINE_SIMPLE_ATTRIBUTE(htmstart_fops, htmstart_get, htmstart_set, "%llu\n"); +DEFINE_SIMPLE_ATTRIBUTE(htmsetup_fops, htmsetup_get, htmsetup_set, "%llu\n"); +DEFINE_SIMPLE_ATTRIBUTE(htmflags_fops, htmflags_get, htmflags_set, "%llu\n"); + static int htmdump_init_debugfs(void) { htm_buf = kmalloc(PAGE_SIZE, GFP_KERNEL); @@ -98,11 +428,50 @@ static int htmdump_init_debugfs(void) htmdump_debugfs_dir, &htmtype); debugfs_create_file("trace", 0400, htmdump_debugfs_dir, htm_buf, &htmdump_fops); + /* + * Debugfs interface files to control HTM operations: + */ + debugfs_create_file("htmconfigure", 0600, htmdump_debugfs_dir, NULL, &htmconfigure_fops); + debugfs_create_file("htmstart", 0600, htmdump_debugfs_dir, NULL, &htmstart_fops); + debugfs_create_file("htmsetup", 0600, htmdump_debugfs_dir, NULL, &htmsetup_fops); + debugfs_create_file("htmflags", 0600, htmdump_debugfs_dir, NULL, &htmflags_fops); + + /* Debugfs interface file to present status of HTM */ + htm_status_buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!htm_status_buf) { + pr_err("Failed to allocate htmstatus buf\n"); + return -ENOMEM; + } + + /* Debugfs interface file to present System Processor Configuration */ + htm_info_buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!htm_info_buf) { + pr_err("Failed to allocate htm info buf\n"); + return -ENOMEM; + } + + /* Debugfs interface file to present HTM capabilities */ + htm_caps_buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!htm_caps_buf) { + pr_err("Failed to allocate htm caps buf\n"); + return -ENOMEM; + } + + debugfs_create_file("htmstatus", 0400, htmdump_debugfs_dir, htm_status_buf, &htmstatus_fops); + debugfs_create_file("htminfo", 0400, htmdump_debugfs_dir, htm_info_buf, &htminfo_fops); + debugfs_create_file("htmcaps", 0400, htmdump_debugfs_dir, htm_caps_buf, &htmcaps_fops); + return 0; } static int __init htmdump_init(void) { + /* Disable on kvm guest */ + if (is_kvm_guest()) { + pr_info("htmdump not supported inside KVM guest\n"); + return -EOPNOTSUPP; + } + if (htmdump_init_debugfs()) return -ENOMEM; diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index d6ebc19fb99c..eec333dd2e59 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -197,7 +197,7 @@ static void tce_iommu_userspace_view_free(struct iommu_table *tbl) static void tce_free_pSeries(struct iommu_table *tbl) { - if (!tbl->it_userspace) + if (tbl->it_userspace) tce_iommu_userspace_view_free(tbl); } diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c index fdc2f7f38dc9..ee1c8c6898a3 100644 --- a/arch/powerpc/platforms/pseries/msi.c +++ b/arch/powerpc/platforms/pseries/msi.c @@ -525,7 +525,12 @@ static struct msi_domain_info pseries_msi_domain_info = { static void pseries_msi_compose_msg(struct irq_data *data, struct msi_msg *msg) { - __pci_read_msi_msg(irq_data_get_msi_desc(data), msg); + struct pci_dev *dev = msi_desc_to_pci_dev(irq_data_get_msi_desc(data)); + + if (dev->current_state == PCI_D0) + __pci_read_msi_msg(irq_data_get_msi_desc(data), msg); + else + get_cached_msi_msg(data->irq, msg); } static struct irq_chip pseries_msi_irq_chip = { @@ -611,7 +616,7 @@ static const struct irq_domain_ops pseries_irq_domain_ops = { static int __pseries_msi_allocate_domains(struct pci_controller *phb, unsigned int count) { - struct irq_domain *parent = irq_get_default_host(); + struct irq_domain *parent = irq_get_default_domain(); phb->fwnode = irq_domain_alloc_named_id_fwnode("pSeries-MSI", phb->global_number); @@ -628,7 +633,7 @@ static int __pseries_msi_allocate_domains(struct pci_controller *phb, return -ENOMEM; } - phb->msi_domain = pci_msi_create_irq_domain(of_node_to_fwnode(phb->dn), + phb->msi_domain = pci_msi_create_irq_domain(of_fwnode_handle(phb->dn), &pseries_msi_domain_info, phb->dev_domain); if (!phb->msi_domain) { diff --git a/arch/powerpc/platforms/pseries/papr-indices.c b/arch/powerpc/platforms/pseries/papr-indices.c new file mode 100644 index 000000000000..3c7545591c45 --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr-indices.c @@ -0,0 +1,488 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#define pr_fmt(fmt) "papr-indices: " fmt + +#include <linux/build_bug.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/lockdep.h> +#include <linux/kernel.h> +#include <linux/miscdevice.h> +#include <linux/signal.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/string_helpers.h> +#include <linux/uaccess.h> +#include <asm/machdep.h> +#include <asm/rtas-work-area.h> +#include <asm/rtas.h> +#include <uapi/asm/papr-indices.h> +#include "papr-rtas-common.h" + +/* + * Function-specific return values for ibm,set-dynamic-indicator and + * ibm,get-dynamic-sensor-state RTAS calls. + * PAPR+ v2.13 7.3.18 and 7.3.19. + */ +#define RTAS_IBM_DYNAMIC_INDICE_NO_INDICATOR -3 + +/** + * struct rtas_get_indices_params - Parameters (in and out) for + * ibm,get-indices. + * @is_sensor: In: Caller-provided whether sensor or indicator. + * @indice_type:In: Caller-provided indice (sensor or indicator) token + * @work_area: In: Caller-provided work area buffer for results. + * @next: In: Sequence number. Out: Next sequence number. + * @status: Out: RTAS call status. + */ +struct rtas_get_indices_params { + u8 is_sensor; + u32 indice_type; + struct rtas_work_area *work_area; + u32 next; + s32 status; +}; + +/* + * rtas_ibm_get_indices() - Call ibm,get-indices to fill a work area buffer. + * @params: See &struct rtas_ibm_get_indices_params. + * + * Calls ibm,get-indices until it errors or successfully deposits data + * into the supplied work area. Handles RTAS retry statuses. Maps RTAS + * error statuses to reasonable errno values. + * + * The caller is expected to invoke rtas_ibm_get_indices() multiple times + * to retrieve all indices data for the provided indice type. Only one + * sequence should be in progress at any time; starting a new sequence + * will disrupt any sequence already in progress. Serialization of + * indices retrieval sequences is the responsibility of the caller. + * + * The caller should inspect @params.status to determine whether more + * calls are needed to complete the sequence. + * + * Context: May sleep. + * Return: -ve on error, 0 otherwise. + */ +static int rtas_ibm_get_indices(struct rtas_get_indices_params *params) +{ + struct rtas_work_area *work_area = params->work_area; + const s32 token = rtas_function_token(RTAS_FN_IBM_GET_INDICES); + u32 rets; + s32 fwrc; + int ret; + + if (token == RTAS_UNKNOWN_SERVICE) + return -ENOENT; + + lockdep_assert_held(&rtas_ibm_get_indices_lock); + + do { + fwrc = rtas_call(token, 5, 2, &rets, params->is_sensor, + params->indice_type, + rtas_work_area_phys(work_area), + rtas_work_area_size(work_area), + params->next); + } while (rtas_busy_delay(fwrc)); + + switch (fwrc) { + case RTAS_HARDWARE_ERROR: + ret = -EIO; + break; + case RTAS_INVALID_PARAMETER: /* Indicator type is not supported */ + ret = -EINVAL; + break; + case RTAS_SEQ_START_OVER: + ret = -EAGAIN; + pr_info_ratelimited("Indices changed during retrieval, retrying\n"); + params->next = 1; + break; + case RTAS_SEQ_MORE_DATA: + params->next = rets; + ret = 0; + break; + case RTAS_SEQ_COMPLETE: + params->next = 0; + ret = 0; + break; + default: + ret = -EIO; + pr_err_ratelimited("unexpected ibm,get-indices status %d\n", fwrc); + break; + } + + params->status = fwrc; + return ret; +} + +/* + * Internal indices sequence APIs. A sequence is a series of calls to + * ibm,get-indices for a given location code. The sequence ends when + * an error is encountered or all indices for the input has been + * returned. + */ + +/* + * indices_sequence_begin() - Begin a indices retrieval sequence. + * + * Context: May sleep. + */ +static void indices_sequence_begin(struct papr_rtas_sequence *seq) +{ + struct rtas_get_indices_params *param; + + param = (struct rtas_get_indices_params *)seq->params; + /* + * We could allocate the work area before acquiring the + * function lock, but that would allow concurrent requests to + * exhaust the limited work area pool for no benefit. So + * allocate the work area under the lock. + */ + mutex_lock(&rtas_ibm_get_indices_lock); + param->work_area = rtas_work_area_alloc(RTAS_GET_INDICES_BUF_SIZE); + param->next = 1; + param->status = 0; +} + +/* + * indices_sequence_end() - Finalize a indices retrieval sequence. + * + * Releases resources obtained by indices_sequence_begin(). + */ +static void indices_sequence_end(struct papr_rtas_sequence *seq) +{ + struct rtas_get_indices_params *param; + + param = (struct rtas_get_indices_params *)seq->params; + rtas_work_area_free(param->work_area); + mutex_unlock(&rtas_ibm_get_indices_lock); +} + +/* + * Work function to be passed to papr_rtas_blob_generate(). + * + * ibm,get-indices RTAS call fills the work area with the certain + * format but does not return the bytes written in the buffer. So + * instead of kernel parsing this work area to determine the buffer + * length, copy the complete work area (RTAS_GET_INDICES_BUF_SIZE) + * to the blob and let the user space to obtain the data. + * Means RTAS_GET_INDICES_BUF_SIZE data will be returned for each + * read(). + */ + +static const char *indices_sequence_fill_work_area(struct papr_rtas_sequence *seq, + size_t *len) +{ + struct rtas_get_indices_params *p; + bool init_state; + + p = (struct rtas_get_indices_params *)seq->params; + init_state = (p->next == 1) ? true : false; + + if (papr_rtas_sequence_should_stop(seq, p->status, init_state)) + return NULL; + if (papr_rtas_sequence_set_err(seq, rtas_ibm_get_indices(p))) + return NULL; + + *len = RTAS_GET_INDICES_BUF_SIZE; + return rtas_work_area_raw_buf(p->work_area); +} + +/* + * papr_indices_handle_read - returns indices blob data to the user space + * + * ibm,get-indices RTAS call fills the work area with the certian + * format but does not return the bytes written in the buffer and + * copied RTAS_GET_INDICES_BUF_SIZE data to the blob for each RTAS + * call. So send RTAS_GET_INDICES_BUF_SIZE buffer to the user space + * for each read(). + */ +static ssize_t papr_indices_handle_read(struct file *file, + char __user *buf, size_t size, loff_t *off) +{ + const struct papr_rtas_blob *blob = file->private_data; + + /* we should not instantiate a handle without any data attached. */ + if (!papr_rtas_blob_has_data(blob)) { + pr_err_once("handle without data\n"); + return -EIO; + } + + if (size < RTAS_GET_INDICES_BUF_SIZE) { + pr_err_once("Invalid buffer length %ld, expect %d\n", + size, RTAS_GET_INDICES_BUF_SIZE); + return -EINVAL; + } else if (size > RTAS_GET_INDICES_BUF_SIZE) + size = RTAS_GET_INDICES_BUF_SIZE; + + return simple_read_from_buffer(buf, size, off, blob->data, blob->len); +} + +static const struct file_operations papr_indices_handle_ops = { + .read = papr_indices_handle_read, + .llseek = papr_rtas_common_handle_seek, + .release = papr_rtas_common_handle_release, +}; + +/* + * papr_indices_create_handle() - Create a fd-based handle for reading + * indices data + * @ubuf: Input parameters to RTAS call such as whether sensor or indicator + * and indice type in user memory + * + * Handler for PAPR_INDICES_IOC_GET ioctl command. Validates @ubuf + * and instantiates an immutable indices "blob" for it. The blob is + * attached to a file descriptor for reading by user space. The memory + * backing the blob is freed when the file is released. + * + * The entire requested indices is retrieved by this call and all + * necessary RTAS interactions are performed before returning the fd + * to user space. This keeps the read handler simple and ensures that + * the kernel can prevent interleaving of ibm,get-indices call sequences. + * + * Return: The installed fd number if successful, -ve errno otherwise. + */ +static long papr_indices_create_handle(struct papr_indices_io_block __user *ubuf) +{ + struct papr_rtas_sequence seq = {}; + struct rtas_get_indices_params params = {}; + int fd; + + if (get_user(params.is_sensor, &ubuf->indices.is_sensor)) + return -EFAULT; + + if (get_user(params.indice_type, &ubuf->indices.indice_type)) + return -EFAULT; + + seq = (struct papr_rtas_sequence) { + .begin = indices_sequence_begin, + .end = indices_sequence_end, + .work = indices_sequence_fill_work_area, + }; + + seq.params = ¶ms; + fd = papr_rtas_setup_file_interface(&seq, + &papr_indices_handle_ops, "[papr-indices]"); + + return fd; +} + +/* + * Create work area with the input parameters. This function is used + * for both ibm,set-dynamic-indicator and ibm,get-dynamic-sensor-state + * RTAS Calls. + */ +static struct rtas_work_area * +papr_dynamic_indice_buf_from_user(struct papr_indices_io_block __user *ubuf, + struct papr_indices_io_block *kbuf) +{ + struct rtas_work_area *work_area; + u32 length; + __be32 len_be; + + if (copy_from_user(kbuf, ubuf, sizeof(*kbuf))) + return ERR_PTR(-EFAULT); + + + if (!string_is_terminated(kbuf->dynamic_param.location_code_str, + ARRAY_SIZE(kbuf->dynamic_param.location_code_str))) + return ERR_PTR(-EINVAL); + + /* + * The input data in the work area should be as follows: + * - 32-bit integer length of the location code string, + * including NULL. + * - Location code string, NULL terminated, identifying the + * token (sensor or indicator). + * PAPR 2.13 - R1–7.3.18–5 ibm,set-dynamic-indicator + * - R1–7.3.19–5 ibm,get-dynamic-sensor-state + */ + /* + * Length that user space passed should also include NULL + * terminator. + */ + length = strlen(kbuf->dynamic_param.location_code_str) + 1; + if (length > LOC_CODE_SIZE) + return ERR_PTR(-EINVAL); + + len_be = cpu_to_be32(length); + + work_area = rtas_work_area_alloc(LOC_CODE_SIZE + sizeof(u32)); + memcpy(rtas_work_area_raw_buf(work_area), &len_be, sizeof(u32)); + memcpy((rtas_work_area_raw_buf(work_area) + sizeof(u32)), + &kbuf->dynamic_param.location_code_str, length); + + return work_area; +} + +/** + * papr_dynamic_indicator_ioc_set - ibm,set-dynamic-indicator RTAS Call + * PAPR 2.13 7.3.18 + * + * @ubuf: Input parameters to RTAS call such as indicator token and + * new state. + * + * Returns success or -errno. + */ +static long papr_dynamic_indicator_ioc_set(struct papr_indices_io_block __user *ubuf) +{ + struct papr_indices_io_block kbuf; + struct rtas_work_area *work_area; + s32 fwrc, token, ret; + + token = rtas_function_token(RTAS_FN_IBM_SET_DYNAMIC_INDICATOR); + if (token == RTAS_UNKNOWN_SERVICE) + return -ENOENT; + + mutex_lock(&rtas_ibm_set_dynamic_indicator_lock); + work_area = papr_dynamic_indice_buf_from_user(ubuf, &kbuf); + if (IS_ERR(work_area)) { + ret = PTR_ERR(work_area); + goto out; + } + + do { + fwrc = rtas_call(token, 3, 1, NULL, + kbuf.dynamic_param.token, + kbuf.dynamic_param.state, + rtas_work_area_phys(work_area)); + } while (rtas_busy_delay(fwrc)); + + rtas_work_area_free(work_area); + + switch (fwrc) { + case RTAS_SUCCESS: + ret = 0; + break; + case RTAS_IBM_DYNAMIC_INDICE_NO_INDICATOR: /* No such indicator */ + ret = -EOPNOTSUPP; + break; + default: + pr_err("unexpected ibm,set-dynamic-indicator result %d\n", + fwrc); + fallthrough; + case RTAS_HARDWARE_ERROR: /* Hardware/platform error */ + ret = -EIO; + break; + } + +out: + mutex_unlock(&rtas_ibm_set_dynamic_indicator_lock); + return ret; +} + +/** + * papr_dynamic_sensor_ioc_get - ibm,get-dynamic-sensor-state RTAS Call + * PAPR 2.13 7.3.19 + * + * @ubuf: Input parameters to RTAS call such as sensor token + * Copies the state in user space buffer. + * + * + * Returns success or -errno. + */ + +static long papr_dynamic_sensor_ioc_get(struct papr_indices_io_block __user *ubuf) +{ + struct papr_indices_io_block kbuf; + struct rtas_work_area *work_area; + s32 fwrc, token, ret; + u32 rets; + + token = rtas_function_token(RTAS_FN_IBM_GET_DYNAMIC_SENSOR_STATE); + if (token == RTAS_UNKNOWN_SERVICE) + return -ENOENT; + + mutex_lock(&rtas_ibm_get_dynamic_sensor_state_lock); + work_area = papr_dynamic_indice_buf_from_user(ubuf, &kbuf); + if (IS_ERR(work_area)) { + ret = PTR_ERR(work_area); + goto out; + } + + do { + fwrc = rtas_call(token, 2, 2, &rets, + kbuf.dynamic_param.token, + rtas_work_area_phys(work_area)); + } while (rtas_busy_delay(fwrc)); + + rtas_work_area_free(work_area); + + switch (fwrc) { + case RTAS_SUCCESS: + if (put_user(rets, &ubuf->dynamic_param.state)) + ret = -EFAULT; + else + ret = 0; + break; + case RTAS_IBM_DYNAMIC_INDICE_NO_INDICATOR: /* No such indicator */ + ret = -EOPNOTSUPP; + break; + default: + pr_err("unexpected ibm,get-dynamic-sensor result %d\n", + fwrc); + fallthrough; + case RTAS_HARDWARE_ERROR: /* Hardware/platform error */ + ret = -EIO; + break; + } + +out: + mutex_unlock(&rtas_ibm_get_dynamic_sensor_state_lock); + return ret; +} + +/* + * Top-level ioctl handler for /dev/papr-indices. + */ +static long papr_indices_dev_ioctl(struct file *filp, unsigned int ioctl, + unsigned long arg) +{ + void __user *argp = (__force void __user *)arg; + long ret; + + switch (ioctl) { + case PAPR_INDICES_IOC_GET: + ret = papr_indices_create_handle(argp); + break; + case PAPR_DYNAMIC_SENSOR_IOC_GET: + ret = papr_dynamic_sensor_ioc_get(argp); + break; + case PAPR_DYNAMIC_INDICATOR_IOC_SET: + if (filp->f_mode & FMODE_WRITE) + ret = papr_dynamic_indicator_ioc_set(argp); + else + ret = -EBADF; + break; + default: + ret = -ENOIOCTLCMD; + break; + } + + return ret; +} + +static const struct file_operations papr_indices_ops = { + .unlocked_ioctl = papr_indices_dev_ioctl, +}; + +static struct miscdevice papr_indices_dev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "papr-indices", + .fops = &papr_indices_ops, +}; + +static __init int papr_indices_init(void) +{ + if (!rtas_function_implemented(RTAS_FN_IBM_GET_INDICES)) + return -ENODEV; + + if (!rtas_function_implemented(RTAS_FN_IBM_SET_DYNAMIC_INDICATOR)) + return -ENODEV; + + if (!rtas_function_implemented(RTAS_FN_IBM_GET_DYNAMIC_SENSOR_STATE)) + return -ENODEV; + + return misc_register(&papr_indices_dev); +} +machine_device_initcall(pseries, papr_indices_init); diff --git a/arch/powerpc/platforms/pseries/papr-phy-attest.c b/arch/powerpc/platforms/pseries/papr-phy-attest.c new file mode 100644 index 000000000000..1907f2411567 --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr-phy-attest.c @@ -0,0 +1,288 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#define pr_fmt(fmt) "papr-phy-attest: " fmt + +#include <linux/build_bug.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/lockdep.h> +#include <linux/kernel.h> +#include <linux/miscdevice.h> +#include <linux/signal.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/string_helpers.h> +#include <linux/uaccess.h> +#include <asm/machdep.h> +#include <asm/rtas-work-area.h> +#include <asm/rtas.h> +#include <uapi/asm/papr-physical-attestation.h> +#include "papr-rtas-common.h" + +/** + * struct rtas_phy_attest_params - Parameters (in and out) for + * ibm,physical-attestation. + * + * @cmd: In: Caller-provided attestation command buffer. Must be + * RTAS-addressable. + * @work_area: In: Caller-provided work area buffer for attestation + * command structure + * Out: Caller-provided work area buffer for the response + * @cmd_len: In: Caller-provided attestation command structure + * length + * @sequence: In: Sequence number. Out: Next sequence number. + * @written: Out: Bytes written by ibm,physical-attestation to + * @work_area. + * @status: Out: RTAS call status. + */ +struct rtas_phy_attest_params { + struct papr_phy_attest_io_block cmd; + struct rtas_work_area *work_area; + u32 cmd_len; + u32 sequence; + u32 written; + s32 status; +}; + +/** + * rtas_physical_attestation() - Call ibm,physical-attestation to + * fill a work area buffer. + * @params: See &struct rtas_phy_attest_params. + * + * Calls ibm,physical-attestation until it errors or successfully + * deposits data into the supplied work area. Handles RTAS retry + * statuses. Maps RTAS error statuses to reasonable errno values. + * + * The caller is expected to invoke rtas_physical_attestation() + * multiple times to retrieve all the data for the provided + * attestation command. Only one sequence should be in progress at + * any time; starting a new sequence will disrupt any sequence + * already in progress. Serialization of attestation retrieval + * sequences is the responsibility of the caller. + * + * The caller should inspect @params.status to determine whether more + * calls are needed to complete the sequence. + * + * Context: May sleep. + * Return: -ve on error, 0 otherwise. + */ +static int rtas_physical_attestation(struct rtas_phy_attest_params *params) +{ + struct rtas_work_area *work_area; + s32 fwrc, token; + u32 rets[2]; + int ret; + + work_area = params->work_area; + token = rtas_function_token(RTAS_FN_IBM_PHYSICAL_ATTESTATION); + if (token == RTAS_UNKNOWN_SERVICE) + return -ENOENT; + + lockdep_assert_held(&rtas_ibm_physical_attestation_lock); + + do { + fwrc = rtas_call(token, 3, 3, rets, + rtas_work_area_phys(work_area), + params->cmd_len, + params->sequence); + } while (rtas_busy_delay(fwrc)); + + switch (fwrc) { + case RTAS_HARDWARE_ERROR: + ret = -EIO; + break; + case RTAS_INVALID_PARAMETER: + ret = -EINVAL; + break; + case RTAS_SEQ_MORE_DATA: + params->sequence = rets[0]; + fallthrough; + case RTAS_SEQ_COMPLETE: + params->written = rets[1]; + /* + * Kernel or firmware bug, do not continue. + */ + if (WARN(params->written > rtas_work_area_size(work_area), + "possible write beyond end of work area")) + ret = -EFAULT; + else + ret = 0; + break; + default: + ret = -EIO; + pr_err_ratelimited("unexpected ibm,get-phy_attest status %d\n", fwrc); + break; + } + + params->status = fwrc; + return ret; +} + +/* + * Internal physical-attestation sequence APIs. A physical-attestation + * sequence is a series of calls to get ibm,physical-attestation + * for a given attestation command. The sequence ends when an error + * is encountered or all data for the attestation command has been + * returned. + */ + +/** + * phy_attest_sequence_begin() - Begin a response data for attestation + * command retrieval sequence. + * @seq: user specified parameters for RTAS call from seq struct. + * + * Context: May sleep. + */ +static void phy_attest_sequence_begin(struct papr_rtas_sequence *seq) +{ + struct rtas_phy_attest_params *param; + + /* + * We could allocate the work area before acquiring the + * function lock, but that would allow concurrent requests to + * exhaust the limited work area pool for no benefit. So + * allocate the work area under the lock. + */ + mutex_lock(&rtas_ibm_physical_attestation_lock); + param = (struct rtas_phy_attest_params *)seq->params; + param->work_area = rtas_work_area_alloc(SZ_4K); + memcpy(rtas_work_area_raw_buf(param->work_area), ¶m->cmd, + param->cmd_len); + param->sequence = 1; + param->status = 0; +} + +/** + * phy_attest_sequence_end() - Finalize a attestation command + * response retrieval sequence. + * @seq: Sequence state. + * + * Releases resources obtained by phy_attest_sequence_begin(). + */ +static void phy_attest_sequence_end(struct papr_rtas_sequence *seq) +{ + struct rtas_phy_attest_params *param; + + param = (struct rtas_phy_attest_params *)seq->params; + rtas_work_area_free(param->work_area); + mutex_unlock(&rtas_ibm_physical_attestation_lock); + kfree(param); +} + +/* + * Generator function to be passed to papr_rtas_blob_generate(). + */ +static const char *phy_attest_sequence_fill_work_area(struct papr_rtas_sequence *seq, + size_t *len) +{ + struct rtas_phy_attest_params *p; + bool init_state; + + p = (struct rtas_phy_attest_params *)seq->params; + init_state = (p->written == 0) ? true : false; + + if (papr_rtas_sequence_should_stop(seq, p->status, init_state)) + return NULL; + if (papr_rtas_sequence_set_err(seq, rtas_physical_attestation(p))) + return NULL; + *len = p->written; + return rtas_work_area_raw_buf(p->work_area); +} + +static const struct file_operations papr_phy_attest_handle_ops = { + .read = papr_rtas_common_handle_read, + .llseek = papr_rtas_common_handle_seek, + .release = papr_rtas_common_handle_release, +}; + +/** + * papr_phy_attest_create_handle() - Create a fd-based handle for + * reading the response for the given attestation command. + * @ulc: Attestation command in user memory; defines the scope of + * data for the attestation command to retrieve. + * + * Handler for PAPR_PHYSICAL_ATTESTATION_IOC_CREATE_HANDLE ioctl + * command. Validates @ulc and instantiates an immutable response + * "blob" for attestation command. The blob is attached to a file + * descriptor for reading by user space. The memory backing the blob + * is freed when the file is released. + * + * The entire requested response buffer for the attestation command + * retrieved by this call and all necessary RTAS interactions are + * performed before returning the fd to user space. This keeps the + * read handler simple and ensures that kernel can prevent + * interleaving ibm,physical-attestation call sequences. + * + * Return: The installed fd number if successful, -ve errno otherwise. + */ +static long papr_phy_attest_create_handle(struct papr_phy_attest_io_block __user *ulc) +{ + struct rtas_phy_attest_params *params; + struct papr_rtas_sequence seq = {}; + int fd; + + /* + * Freed in phy_attest_sequence_end(). + */ + params = kzalloc(sizeof(*params), GFP_KERNEL_ACCOUNT); + if (!params) + return -ENOMEM; + + if (copy_from_user(¶ms->cmd, ulc, + sizeof(struct papr_phy_attest_io_block))) + return -EFAULT; + + params->cmd_len = be32_to_cpu(params->cmd.length); + seq = (struct papr_rtas_sequence) { + .begin = phy_attest_sequence_begin, + .end = phy_attest_sequence_end, + .work = phy_attest_sequence_fill_work_area, + }; + + seq.params = (void *)params; + + fd = papr_rtas_setup_file_interface(&seq, + &papr_phy_attest_handle_ops, + "[papr-physical-attestation]"); + + return fd; +} + +/* + * Top-level ioctl handler for /dev/papr-physical-attestation. + */ +static long papr_phy_attest_dev_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) +{ + void __user *argp = (__force void __user *)arg; + long ret; + + switch (ioctl) { + case PAPR_PHY_ATTEST_IOC_HANDLE: + ret = papr_phy_attest_create_handle(argp); + break; + default: + ret = -ENOIOCTLCMD; + break; + } + return ret; +} + +static const struct file_operations papr_phy_attest_ops = { + .unlocked_ioctl = papr_phy_attest_dev_ioctl, +}; + +static struct miscdevice papr_phy_attest_dev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "papr-physical-attestation", + .fops = &papr_phy_attest_ops, +}; + +static __init int papr_phy_attest_init(void) +{ + if (!rtas_function_implemented(RTAS_FN_IBM_PHYSICAL_ATTESTATION)) + return -ENODEV; + + return misc_register(&papr_phy_attest_dev); +} +machine_device_initcall(pseries, papr_phy_attest_init); diff --git a/arch/powerpc/platforms/pseries/papr-platform-dump.c b/arch/powerpc/platforms/pseries/papr-platform-dump.c new file mode 100644 index 000000000000..f8d55eccdb6b --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr-platform-dump.c @@ -0,0 +1,411 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#define pr_fmt(fmt) "papr-platform-dump: " fmt + +#include <linux/anon_inodes.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/miscdevice.h> +#include <asm/machdep.h> +#include <asm/rtas-work-area.h> +#include <asm/rtas.h> +#include <uapi/asm/papr-platform-dump.h> + +/* + * Function-specific return values for ibm,platform-dump, derived from + * PAPR+ v2.13 7.3.3.4.1 "ibm,platform-dump RTAS Call". + */ +#define RTAS_IBM_PLATFORM_DUMP_COMPLETE 0 /* Complete dump retrieved. */ +#define RTAS_IBM_PLATFORM_DUMP_CONTINUE 1 /* Continue dump */ +#define RTAS_NOT_AUTHORIZED -9002 /* Not Authorized */ + +#define RTAS_IBM_PLATFORM_DUMP_START 2 /* Linux status to start dump */ + +/** + * struct ibm_platform_dump_params - Parameters (in and out) for + * ibm,platform-dump + * @work_area: In: work area buffer for results. + * @buf_length: In: work area buffer length in bytes + * @dump_tag_hi: In: Most-significant 32 bits of a Dump_Tag representing + * an id of the dump being processed. + * @dump_tag_lo: In: Least-significant 32 bits of a Dump_Tag representing + * an id of the dump being processed. + * @sequence_hi: In: Sequence number in most-significant 32 bits. + * Out: Next sequence number in most-significant 32 bits. + * @sequence_lo: In: Sequence number in Least-significant 32 bits + * Out: Next sequence number in Least-significant 32 bits. + * @bytes_ret_hi: Out: Bytes written in most-significant 32 bits. + * @bytes_ret_lo: Out: Bytes written in Least-significant 32 bits. + * @status: Out: RTAS call status. + * @list: Maintain the list of dumps are in progress. Can + * retrieve multiple dumps with different dump IDs at + * the same time but not with the same dump ID. This list + * is used to determine whether the dump for the same ID + * is in progress. + */ +struct ibm_platform_dump_params { + struct rtas_work_area *work_area; + u32 buf_length; + u32 dump_tag_hi; + u32 dump_tag_lo; + u32 sequence_hi; + u32 sequence_lo; + u32 bytes_ret_hi; + u32 bytes_ret_lo; + s32 status; + struct list_head list; +}; + +/* + * Multiple dumps with different dump IDs can be retrieved at the same + * time, but not with dame dump ID. platform_dump_list_mutex and + * platform_dump_list are used to prevent this behavior. + */ +static DEFINE_MUTEX(platform_dump_list_mutex); +static LIST_HEAD(platform_dump_list); + +/** + * rtas_ibm_platform_dump() - Call ibm,platform-dump to fill a work area + * buffer. + * @params: See &struct ibm_platform_dump_params. + * @buf_addr: Address of dump buffer (work_area) + * @buf_length: Length of the buffer in bytes (min. 1024) + * + * Calls ibm,platform-dump until it errors or successfully deposits data + * into the supplied work area. Handles RTAS retry statuses. Maps RTAS + * error statuses to reasonable errno values. + * + * Can request multiple dumps with different dump IDs at the same time, + * but not with the same dump ID which is prevented with the check in + * the ioctl code (papr_platform_dump_create_handle()). + * + * The caller should inspect @params.status to determine whether more + * calls are needed to complete the sequence. + * + * Context: May sleep. + * Return: -ve on error, 0 for dump complete and 1 for continue dump + */ +static int rtas_ibm_platform_dump(struct ibm_platform_dump_params *params, + phys_addr_t buf_addr, u32 buf_length) +{ + u32 rets[4]; + s32 fwrc; + int ret = 0; + + do { + fwrc = rtas_call(rtas_function_token(RTAS_FN_IBM_PLATFORM_DUMP), + 6, 5, + rets, + params->dump_tag_hi, + params->dump_tag_lo, + params->sequence_hi, + params->sequence_lo, + buf_addr, + buf_length); + } while (rtas_busy_delay(fwrc)); + + switch (fwrc) { + case RTAS_HARDWARE_ERROR: + ret = -EIO; + break; + case RTAS_NOT_AUTHORIZED: + ret = -EPERM; + break; + case RTAS_IBM_PLATFORM_DUMP_CONTINUE: + case RTAS_IBM_PLATFORM_DUMP_COMPLETE: + params->sequence_hi = rets[0]; + params->sequence_lo = rets[1]; + params->bytes_ret_hi = rets[2]; + params->bytes_ret_lo = rets[3]; + break; + default: + ret = -EIO; + pr_err_ratelimited("unexpected ibm,platform-dump status %d\n", + fwrc); + break; + } + + params->status = fwrc; + return ret; +} + +/* + * Platform dump is used with multiple RTAS calls to retrieve the + * complete dump for the provided dump ID. Once the complete dump is + * retrieved, the hypervisor returns dump complete status (0) for the + * last RTAS call and expects the caller issues one more call with + * NULL buffer to invalidate the dump so that the hypervisor can remove + * the dump. + * + * After the specific dump is invalidated in the hypervisor, expect the + * dump complete status for the new sequence - the user space initiates + * new request for the same dump ID. + */ +static ssize_t papr_platform_dump_handle_read(struct file *file, + char __user *buf, size_t size, loff_t *off) +{ + struct ibm_platform_dump_params *params = file->private_data; + u64 total_bytes; + s32 fwrc; + + /* + * Dump already completed with the previous read calls. + * In case if the user space issues further reads, returns + * -EINVAL. + */ + if (!params->buf_length) { + pr_warn_once("Platform dump completed for dump ID %llu\n", + (u64) (((u64)params->dump_tag_hi << 32) | + params->dump_tag_lo)); + return -EINVAL; + } + + /* + * The hypervisor returns status 0 if no more data available to + * download. The dump will be invalidated with ioctl (see below). + */ + if (params->status == RTAS_IBM_PLATFORM_DUMP_COMPLETE) { + params->buf_length = 0; + /* + * Returns 0 to the user space so that user + * space read stops. + */ + return 0; + } + + if (size < SZ_1K) { + pr_err_once("Buffer length should be minimum 1024 bytes\n"); + return -EINVAL; + } else if (size > params->buf_length) { + /* + * Allocate 4K work area. So if the user requests > 4K, + * resize the buffer length. + */ + size = params->buf_length; + } + + fwrc = rtas_ibm_platform_dump(params, + rtas_work_area_phys(params->work_area), + size); + if (fwrc < 0) + return fwrc; + + total_bytes = (u64) (((u64)params->bytes_ret_hi << 32) | + params->bytes_ret_lo); + + /* + * Kernel or firmware bug, do not continue. + */ + if (WARN(total_bytes > size, "possible write beyond end of work area")) + return -EFAULT; + + if (copy_to_user(buf, rtas_work_area_raw_buf(params->work_area), + total_bytes)) + return -EFAULT; + + return total_bytes; +} + +static int papr_platform_dump_handle_release(struct inode *inode, + struct file *file) +{ + struct ibm_platform_dump_params *params = file->private_data; + + if (params->work_area) + rtas_work_area_free(params->work_area); + + mutex_lock(&platform_dump_list_mutex); + list_del(¶ms->list); + mutex_unlock(&platform_dump_list_mutex); + + kfree(params); + file->private_data = NULL; + return 0; +} + +/* + * This ioctl is used to invalidate the dump assuming the user space + * issue this ioctl after obtain the complete dump. + * Issue the last RTAS call with NULL buffer to invalidate the dump + * which means dump will be freed in the hypervisor. + */ +static long papr_platform_dump_invalidate_ioctl(struct file *file, + unsigned int ioctl, unsigned long arg) +{ + struct ibm_platform_dump_params *params; + u64 __user *argp = (void __user *)arg; + u64 param_dump_tag, dump_tag; + + if (ioctl != PAPR_PLATFORM_DUMP_IOC_INVALIDATE) + return -ENOIOCTLCMD; + + if (get_user(dump_tag, argp)) + return -EFAULT; + + /* + * private_data is freeded during release(), so should not + * happen. + */ + if (!file->private_data) { + pr_err("No valid FD to invalidate dump for the ID(%llu)\n", + dump_tag); + return -EINVAL; + } + + params = file->private_data; + param_dump_tag = (u64) (((u64)params->dump_tag_hi << 32) | + params->dump_tag_lo); + if (dump_tag != param_dump_tag) { + pr_err("Invalid dump ID(%llu) to invalidate dump\n", + dump_tag); + return -EINVAL; + } + + if (params->status != RTAS_IBM_PLATFORM_DUMP_COMPLETE) { + pr_err("Platform dump is not complete, but requested " + "to invalidate dump for ID(%llu)\n", + dump_tag); + return -EINPROGRESS; + } + + return rtas_ibm_platform_dump(params, 0, 0); +} + +static const struct file_operations papr_platform_dump_handle_ops = { + .read = papr_platform_dump_handle_read, + .release = papr_platform_dump_handle_release, + .unlocked_ioctl = papr_platform_dump_invalidate_ioctl, +}; + +/** + * papr_platform_dump_create_handle() - Create a fd-based handle for + * reading platform dump + * + * Handler for PAPR_PLATFORM_DUMP_IOC_CREATE_HANDLE ioctl command + * Allocates RTAS parameter struct and work area and attached to the + * file descriptor for reading by user space with the multiple RTAS + * calls until the dump is completed. This memory allocation is freed + * when the file is released. + * + * Multiple dump requests with different IDs are allowed at the same + * time, but not with the same dump ID. So if the user space is + * already opened file descriptor for the specific dump ID, return + * -EALREADY for the next request. + * + * @dump_tag: Dump ID for the dump requested to retrieve from the + * hypervisor + * + * Return: The installed fd number if successful, -ve errno otherwise. + */ +static long papr_platform_dump_create_handle(u64 dump_tag) +{ + struct ibm_platform_dump_params *params; + u64 param_dump_tag; + struct file *file; + long err; + int fd; + + /* + * Return failure if the user space is already opened FD for + * the specific dump ID. This check will prevent multiple dump + * requests for the same dump ID at the same time. Generally + * should not expect this, but in case. + */ + list_for_each_entry(params, &platform_dump_list, list) { + param_dump_tag = (u64) (((u64)params->dump_tag_hi << 32) | + params->dump_tag_lo); + if (dump_tag == param_dump_tag) { + pr_err("Platform dump for ID(%llu) is already in progress\n", + dump_tag); + return -EALREADY; + } + } + + params = kzalloc(sizeof(struct ibm_platform_dump_params), + GFP_KERNEL_ACCOUNT); + if (!params) + return -ENOMEM; + + params->work_area = rtas_work_area_alloc(SZ_4K); + params->buf_length = SZ_4K; + params->dump_tag_hi = (u32)(dump_tag >> 32); + params->dump_tag_lo = (u32)(dump_tag & 0x00000000ffffffffULL); + params->status = RTAS_IBM_PLATFORM_DUMP_START; + + fd = get_unused_fd_flags(O_RDONLY | O_CLOEXEC); + if (fd < 0) { + err = fd; + goto free_area; + } + + file = anon_inode_getfile_fmode("[papr-platform-dump]", + &papr_platform_dump_handle_ops, + (void *)params, O_RDONLY, + FMODE_LSEEK | FMODE_PREAD); + if (IS_ERR(file)) { + err = PTR_ERR(file); + goto put_fd; + } + + fd_install(fd, file); + + list_add(¶ms->list, &platform_dump_list); + + pr_info("%s (%d) initiated platform dump for dump tag %llu\n", + current->comm, current->pid, dump_tag); + return fd; +put_fd: + put_unused_fd(fd); +free_area: + rtas_work_area_free(params->work_area); + kfree(params); + return err; +} + +/* + * Top-level ioctl handler for /dev/papr-platform-dump. + */ +static long papr_platform_dump_dev_ioctl(struct file *filp, + unsigned int ioctl, + unsigned long arg) +{ + u64 __user *argp = (void __user *)arg; + u64 dump_tag; + long ret; + + if (get_user(dump_tag, argp)) + return -EFAULT; + + switch (ioctl) { + case PAPR_PLATFORM_DUMP_IOC_CREATE_HANDLE: + mutex_lock(&platform_dump_list_mutex); + ret = papr_platform_dump_create_handle(dump_tag); + mutex_unlock(&platform_dump_list_mutex); + break; + default: + ret = -ENOIOCTLCMD; + break; + } + return ret; +} + +static const struct file_operations papr_platform_dump_ops = { + .unlocked_ioctl = papr_platform_dump_dev_ioctl, +}; + +static struct miscdevice papr_platform_dump_dev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "papr-platform-dump", + .fops = &papr_platform_dump_ops, +}; + +static __init int papr_platform_dump_init(void) +{ + if (!rtas_function_implemented(RTAS_FN_IBM_PLATFORM_DUMP)) + return -ENODEV; + + return misc_register(&papr_platform_dump_dev); +} +machine_device_initcall(pseries, papr_platform_dump_init); diff --git a/arch/powerpc/platforms/pseries/papr-rtas-common.c b/arch/powerpc/platforms/pseries/papr-rtas-common.c new file mode 100644 index 000000000000..33c606e3378a --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr-rtas-common.c @@ -0,0 +1,311 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#define pr_fmt(fmt) "papr-common: " fmt + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/signal.h> +#include <linux/slab.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/anon_inodes.h> +#include <linux/sched/signal.h> +#include "papr-rtas-common.h" + +/* + * Sequence based RTAS HCALL has to issue multiple times to retrieve + * complete data from the hypervisor. For some of these RTAS calls, + * the OS should not interleave calls with different input until the + * sequence is completed. So data is collected for these calls during + * ioctl handle and export to user space with read() handle. + * This file provides common functions needed for such sequence based + * RTAS calls Ex: ibm,get-vpd and ibm,get-indices. + */ + +bool papr_rtas_blob_has_data(const struct papr_rtas_blob *blob) +{ + return blob->data && blob->len; +} + +void papr_rtas_blob_free(const struct papr_rtas_blob *blob) +{ + if (blob) { + kvfree(blob->data); + kfree(blob); + } +} + +/** + * papr_rtas_blob_extend() - Append data to a &struct papr_rtas_blob. + * @blob: The blob to extend. + * @data: The new data to append to @blob. + * @len: The length of @data. + * + * Context: May sleep. + * Return: -ENOMEM on allocation failure, 0 otherwise. + */ +static int papr_rtas_blob_extend(struct papr_rtas_blob *blob, + const char *data, size_t len) +{ + const size_t new_len = blob->len + len; + const size_t old_len = blob->len; + const char *old_ptr = blob->data; + char *new_ptr; + + new_ptr = kvrealloc(old_ptr, new_len, GFP_KERNEL_ACCOUNT); + if (!new_ptr) + return -ENOMEM; + + memcpy(&new_ptr[old_len], data, len); + blob->data = new_ptr; + blob->len = new_len; + return 0; +} + +/** + * papr_rtas_blob_generate() - Construct a new &struct papr_rtas_blob. + * @seq: work function of the caller that is called to obtain + * data with the caller RTAS call. + * + * The @work callback is invoked until it returns NULL. @seq is + * passed to @work in its first argument on each call. When + * @work returns data, it should store the data length in its + * second argument. + * + * Context: May sleep. + * Return: A completely populated &struct papr_rtas_blob, or NULL on error. + */ +static const struct papr_rtas_blob * +papr_rtas_blob_generate(struct papr_rtas_sequence *seq) +{ + struct papr_rtas_blob *blob; + const char *buf; + size_t len; + int err = 0; + + blob = kzalloc(sizeof(*blob), GFP_KERNEL_ACCOUNT); + if (!blob) + return NULL; + + if (!seq->work) + return ERR_PTR(-EINVAL); + + + while (err == 0 && (buf = seq->work(seq, &len))) + err = papr_rtas_blob_extend(blob, buf, len); + + if (err != 0 || !papr_rtas_blob_has_data(blob)) + goto free_blob; + + return blob; +free_blob: + papr_rtas_blob_free(blob); + return NULL; +} + +int papr_rtas_sequence_set_err(struct papr_rtas_sequence *seq, int err) +{ + /* Preserve the first error recorded. */ + if (seq->error == 0) + seq->error = err; + + return seq->error; +} + +/* + * Higher-level retrieval code below. These functions use the + * papr_rtas_blob_* and sequence_* APIs defined above to create fd-based + * handles for consumption by user space. + */ + +/** + * papr_rtas_run_sequence() - Run a single retrieval sequence. + * @seq: Functions of the caller to complete the sequence + * + * Context: May sleep. Holds a mutex and an RTAS work area for its + * duration. Typically performs multiple sleepable slab + * allocations. + * + * Return: A populated &struct papr_rtas_blob on success. Encoded error + * pointer otherwise. + */ +static const struct papr_rtas_blob *papr_rtas_run_sequence(struct papr_rtas_sequence *seq) +{ + const struct papr_rtas_blob *blob; + + if (seq->begin) + seq->begin(seq); + + blob = papr_rtas_blob_generate(seq); + if (!blob) + papr_rtas_sequence_set_err(seq, -ENOMEM); + + if (seq->end) + seq->end(seq); + + + if (seq->error) { + papr_rtas_blob_free(blob); + return ERR_PTR(seq->error); + } + + return blob; +} + +/** + * papr_rtas_retrieve() - Return the data blob that is exposed to + * user space. + * @seq: RTAS call specific functions to be invoked until the + * sequence is completed. + * + * Run sequences against @param until a blob is successfully + * instantiated, or a hard error is encountered, or a fatal signal is + * pending. + * + * Context: May sleep. + * Return: A fully populated data blob when successful. Encoded error + * pointer otherwise. + */ +const struct papr_rtas_blob *papr_rtas_retrieve(struct papr_rtas_sequence *seq) +{ + const struct papr_rtas_blob *blob; + + /* + * EAGAIN means the sequence returns error with a -4 (data + * changed and need to start the sequence) status from RTAS calls + * and we should attempt a new sequence. PAPR+ (v2.13 R1–7.3.20–5 + * - ibm,get-vpd, R1–7.3.17–6 - ibm,get-indices) indicates that + * this should be a transient condition, not something that + * happens continuously. But we'll stop trying on a fatal signal. + */ + do { + blob = papr_rtas_run_sequence(seq); + if (!IS_ERR(blob)) /* Success. */ + break; + if (PTR_ERR(blob) != -EAGAIN) /* Hard error. */ + break; + cond_resched(); + } while (!fatal_signal_pending(current)); + + return blob; +} + +/** + * papr_rtas_setup_file_interface - Complete the sequence and obtain + * the data and export to user space with fd-based handles. Then the + * user spave gets the data with read() handle. + * @seq: RTAS call specific functions to get the data. + * @fops: RTAS call specific file operations such as read(). + * @name: RTAS call specific char device node. + * + * Return: FD handle for consumption by user space + */ +long papr_rtas_setup_file_interface(struct papr_rtas_sequence *seq, + const struct file_operations *fops, + char *name) +{ + const struct papr_rtas_blob *blob; + struct file *file; + long ret; + int fd; + + blob = papr_rtas_retrieve(seq); + if (IS_ERR(blob)) + return PTR_ERR(blob); + + fd = get_unused_fd_flags(O_RDONLY | O_CLOEXEC); + if (fd < 0) { + ret = fd; + goto free_blob; + } + + file = anon_inode_getfile_fmode(name, fops, (void *)blob, + O_RDONLY, FMODE_LSEEK | FMODE_PREAD); + if (IS_ERR(file)) { + ret = PTR_ERR(file); + goto put_fd; + } + + fd_install(fd, file); + return fd; + +put_fd: + put_unused_fd(fd); +free_blob: + papr_rtas_blob_free(blob); + return ret; +} + +/* + * papr_rtas_sequence_should_stop() - Determine whether RTAS retrieval + * sequence should continue. + * + * Examines the sequence error state and outputs of the last call to + * the specific RTAS to determine whether the sequence in progress + * should continue or stop. + * + * Return: True if the sequence has encountered an error or if all data + * for this sequence has been retrieved. False otherwise. + */ +bool papr_rtas_sequence_should_stop(const struct papr_rtas_sequence *seq, + s32 status, bool init_state) +{ + bool done; + + if (seq->error) + return true; + + switch (status) { + case RTAS_SEQ_COMPLETE: + if (init_state) + done = false; /* Initial state. */ + else + done = true; /* All data consumed. */ + break; + case RTAS_SEQ_MORE_DATA: + done = false; /* More data available. */ + break; + default: + done = true; /* Error encountered. */ + break; + } + + return done; +} + +/* + * User space read to retrieve data for the corresponding RTAS call. + * papr_rtas_blob is filled with the data using the corresponding RTAS + * call sequence API. + */ +ssize_t papr_rtas_common_handle_read(struct file *file, + char __user *buf, size_t size, loff_t *off) +{ + const struct papr_rtas_blob *blob = file->private_data; + + /* We should not instantiate a handle without any data attached. */ + if (!papr_rtas_blob_has_data(blob)) { + pr_err_once("handle without data\n"); + return -EIO; + } + + return simple_read_from_buffer(buf, size, off, blob->data, blob->len); +} + +int papr_rtas_common_handle_release(struct inode *inode, + struct file *file) +{ + const struct papr_rtas_blob *blob = file->private_data; + + papr_rtas_blob_free(blob); + + return 0; +} + +loff_t papr_rtas_common_handle_seek(struct file *file, loff_t off, + int whence) +{ + const struct papr_rtas_blob *blob = file->private_data; + + return fixed_size_llseek(file, off, whence, blob->len); +} diff --git a/arch/powerpc/platforms/pseries/papr-rtas-common.h b/arch/powerpc/platforms/pseries/papr-rtas-common.h new file mode 100644 index 000000000000..4ceabcaf4905 --- /dev/null +++ b/arch/powerpc/platforms/pseries/papr-rtas-common.h @@ -0,0 +1,61 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +#ifndef _ASM_POWERPC_PAPR_RTAS_COMMON_H +#define _ASM_POWERPC_PAPR_RTAS_COMMON_H + +#include <linux/types.h> + +/* + * Return codes for sequence based RTAS calls. + * Not listed under PAPR+ v2.13 7.2.8: "Return Codes". + * But defined in the specific section of each RTAS call. + */ +#define RTAS_SEQ_COMPLETE 0 /* All data has been retrieved. */ +#define RTAS_SEQ_MORE_DATA 1 /* More data is available */ +#define RTAS_SEQ_START_OVER -4 /* Data changed, restart call sequence. */ + +/* + * Internal "blob" APIs for accumulating RTAS call results into + * an immutable buffer to be attached to a file descriptor. + */ +struct papr_rtas_blob { + const char *data; + size_t len; +}; + +/** + * struct papr_sequence - State for managing a sequence of RTAS calls. + * @error: Shall be zero as long as the sequence has not encountered an error, + * -ve errno otherwise. Use papr_rtas_sequence_set_err() to update. + * @params: Parameter block to pass to rtas_*() calls. + * @begin: Work area allocation and initialize the needed parameter + * values passed to RTAS call + * @end: Free the allocated work area + * @work: Obtain data with RTAS call and invoke it until the sequence is + * completed. + * + */ +struct papr_rtas_sequence { + int error; + void *params; + void (*begin)(struct papr_rtas_sequence *seq); + void (*end)(struct papr_rtas_sequence *seq); + const char *(*work)(struct papr_rtas_sequence *seq, size_t *len); +}; + +extern bool papr_rtas_blob_has_data(const struct papr_rtas_blob *blob); +extern void papr_rtas_blob_free(const struct papr_rtas_blob *blob); +extern int papr_rtas_sequence_set_err(struct papr_rtas_sequence *seq, + int err); +extern const struct papr_rtas_blob *papr_rtas_retrieve(struct papr_rtas_sequence *seq); +extern long papr_rtas_setup_file_interface(struct papr_rtas_sequence *seq, + const struct file_operations *fops, char *name); +extern bool papr_rtas_sequence_should_stop(const struct papr_rtas_sequence *seq, + s32 status, bool init_state); +extern ssize_t papr_rtas_common_handle_read(struct file *file, + char __user *buf, size_t size, loff_t *off); +extern int papr_rtas_common_handle_release(struct inode *inode, + struct file *file); +extern loff_t papr_rtas_common_handle_seek(struct file *file, loff_t off, + int whence); +#endif /* _ASM_POWERPC_PAPR_RTAS_COMMON_H */ + diff --git a/arch/powerpc/platforms/pseries/papr-vpd.c b/arch/powerpc/platforms/pseries/papr-vpd.c index c86950d7105a..f38c188fc4a1 100644 --- a/arch/powerpc/platforms/pseries/papr-vpd.c +++ b/arch/powerpc/platforms/pseries/papr-vpd.c @@ -2,7 +2,6 @@ #define pr_fmt(fmt) "papr-vpd: " fmt -#include <linux/anon_inodes.h> #include <linux/build_bug.h> #include <linux/file.h> #include <linux/fs.h> @@ -20,14 +19,7 @@ #include <asm/rtas-work-area.h> #include <asm/rtas.h> #include <uapi/asm/papr-vpd.h> - -/* - * Function-specific return values for ibm,get-vpd, derived from PAPR+ - * v2.13 7.3.20 "ibm,get-vpd RTAS Call". - */ -#define RTAS_IBM_GET_VPD_COMPLETE 0 /* All VPD has been retrieved. */ -#define RTAS_IBM_GET_VPD_MORE_DATA 1 /* More VPD is available. */ -#define RTAS_IBM_GET_VPD_START_OVER -4 /* VPD changed, restart call sequence. */ +#include "papr-rtas-common.h" /** * struct rtas_ibm_get_vpd_params - Parameters (in and out) for ibm,get-vpd. @@ -91,13 +83,14 @@ static int rtas_ibm_get_vpd(struct rtas_ibm_get_vpd_params *params) case RTAS_INVALID_PARAMETER: ret = -EINVAL; break; - case RTAS_IBM_GET_VPD_START_OVER: + case RTAS_SEQ_START_OVER: ret = -EAGAIN; + pr_info_ratelimited("VPD changed during retrieval, retrying\n"); break; - case RTAS_IBM_GET_VPD_MORE_DATA: + case RTAS_SEQ_MORE_DATA: params->sequence = rets[0]; fallthrough; - case RTAS_IBM_GET_VPD_COMPLETE: + case RTAS_SEQ_COMPLETE: params->written = rets[1]; /* * Kernel or firmware bug, do not continue. @@ -119,91 +112,6 @@ static int rtas_ibm_get_vpd(struct rtas_ibm_get_vpd_params *params) } /* - * Internal VPD "blob" APIs for accumulating ibm,get-vpd results into - * an immutable buffer to be attached to a file descriptor. - */ -struct vpd_blob { - const char *data; - size_t len; -}; - -static bool vpd_blob_has_data(const struct vpd_blob *blob) -{ - return blob->data && blob->len; -} - -static void vpd_blob_free(const struct vpd_blob *blob) -{ - if (blob) { - kvfree(blob->data); - kfree(blob); - } -} - -/** - * vpd_blob_extend() - Append data to a &struct vpd_blob. - * @blob: The blob to extend. - * @data: The new data to append to @blob. - * @len: The length of @data. - * - * Context: May sleep. - * Return: -ENOMEM on allocation failure, 0 otherwise. - */ -static int vpd_blob_extend(struct vpd_blob *blob, const char *data, size_t len) -{ - const size_t new_len = blob->len + len; - const size_t old_len = blob->len; - const char *old_ptr = blob->data; - char *new_ptr; - - new_ptr = kvrealloc(old_ptr, new_len, GFP_KERNEL_ACCOUNT); - if (!new_ptr) - return -ENOMEM; - - memcpy(&new_ptr[old_len], data, len); - blob->data = new_ptr; - blob->len = new_len; - return 0; -} - -/** - * vpd_blob_generate() - Construct a new &struct vpd_blob. - * @generator: Function that supplies the blob data. - * @arg: Context pointer supplied by caller, passed to @generator. - * - * The @generator callback is invoked until it returns NULL. @arg is - * passed to @generator in its first argument on each call. When - * @generator returns data, it should store the data length in its - * second argument. - * - * Context: May sleep. - * Return: A completely populated &struct vpd_blob, or NULL on error. - */ -static const struct vpd_blob * -vpd_blob_generate(const char * (*generator)(void *, size_t *), void *arg) -{ - struct vpd_blob *blob; - const char *buf; - size_t len; - int err = 0; - - blob = kzalloc(sizeof(*blob), GFP_KERNEL_ACCOUNT); - if (!blob) - return NULL; - - while (err == 0 && (buf = generator(arg, &len))) - err = vpd_blob_extend(blob, buf, len); - - if (err != 0 || !vpd_blob_has_data(blob)) - goto free_blob; - - return blob; -free_blob: - vpd_blob_free(blob); - return NULL; -} - -/* * Internal VPD sequence APIs. A VPD sequence is a series of calls to * ibm,get-vpd for a given location code. The sequence ends when an * error is encountered or all VPD for the location code has been @@ -211,30 +119,14 @@ free_blob: */ /** - * struct vpd_sequence - State for managing a VPD sequence. - * @error: Shall be zero as long as the sequence has not encountered an error, - * -ve errno otherwise. Use vpd_sequence_set_err() to update this. - * @params: Parameter block to pass to rtas_ibm_get_vpd(). - */ -struct vpd_sequence { - int error; - struct rtas_ibm_get_vpd_params params; -}; - -/** * vpd_sequence_begin() - Begin a VPD retrieval sequence. - * @seq: Uninitialized sequence state. - * @loc_code: Location code that defines the scope of the VPD to return. - * - * Initializes @seq with the resources necessary to carry out a VPD - * sequence. Callers must pass @seq to vpd_sequence_end() regardless - * of whether the sequence succeeds. + * @seq: vpd call parameters from sequence struct * * Context: May sleep. */ -static void vpd_sequence_begin(struct vpd_sequence *seq, - const struct papr_location_code *loc_code) +static void vpd_sequence_begin(struct papr_rtas_sequence *seq) { + struct rtas_ibm_get_vpd_params *vpd_params; /* * Use a static data structure for the location code passed to * RTAS to ensure it's in the RMA and avoid a separate work @@ -242,6 +134,7 @@ static void vpd_sequence_begin(struct vpd_sequence *seq, */ static struct papr_location_code static_loc_code; + vpd_params = (struct rtas_ibm_get_vpd_params *)seq->params; /* * We could allocate the work area before acquiring the * function lock, but that would allow concurrent requests to @@ -249,14 +142,12 @@ static void vpd_sequence_begin(struct vpd_sequence *seq, * allocate the work area under the lock. */ mutex_lock(&rtas_ibm_get_vpd_lock); - static_loc_code = *loc_code; - *seq = (struct vpd_sequence) { - .params = { - .work_area = rtas_work_area_alloc(SZ_4K), - .loc_code = &static_loc_code, - .sequence = 1, - }, - }; + static_loc_code = *(struct papr_location_code *)vpd_params->loc_code; + vpd_params = (struct rtas_ibm_get_vpd_params *)seq->params; + vpd_params->work_area = rtas_work_area_alloc(SZ_4K); + vpd_params->loc_code = &static_loc_code; + vpd_params->sequence = 1; + vpd_params->status = 0; } /** @@ -265,180 +156,39 @@ static void vpd_sequence_begin(struct vpd_sequence *seq, * * Releases resources obtained by vpd_sequence_begin(). */ -static void vpd_sequence_end(struct vpd_sequence *seq) +static void vpd_sequence_end(struct papr_rtas_sequence *seq) { - rtas_work_area_free(seq->params.work_area); - mutex_unlock(&rtas_ibm_get_vpd_lock); -} - -/** - * vpd_sequence_should_stop() - Determine whether a VPD retrieval sequence - * should continue. - * @seq: VPD sequence state. - * - * Examines the sequence error state and outputs of the last call to - * ibm,get-vpd to determine whether the sequence in progress should - * continue or stop. - * - * Return: True if the sequence has encountered an error or if all VPD for - * this sequence has been retrieved. False otherwise. - */ -static bool vpd_sequence_should_stop(const struct vpd_sequence *seq) -{ - bool done; - - if (seq->error) - return true; + struct rtas_ibm_get_vpd_params *vpd_params; - switch (seq->params.status) { - case 0: - if (seq->params.written == 0) - done = false; /* Initial state. */ - else - done = true; /* All data consumed. */ - break; - case 1: - done = false; /* More data available. */ - break; - default: - done = true; /* Error encountered. */ - break; - } - - return done; -} - -static int vpd_sequence_set_err(struct vpd_sequence *seq, int err) -{ - /* Preserve the first error recorded. */ - if (seq->error == 0) - seq->error = err; - - return seq->error; + vpd_params = (struct rtas_ibm_get_vpd_params *)seq->params; + rtas_work_area_free(vpd_params->work_area); + mutex_unlock(&rtas_ibm_get_vpd_lock); } /* - * Generator function to be passed to vpd_blob_generate(). + * Generator function to be passed to papr_rtas_blob_generate(). */ -static const char *vpd_sequence_fill_work_area(void *arg, size_t *len) +static const char *vpd_sequence_fill_work_area(struct papr_rtas_sequence *seq, + size_t *len) { - struct vpd_sequence *seq = arg; - struct rtas_ibm_get_vpd_params *p = &seq->params; + struct rtas_ibm_get_vpd_params *p; + bool init_state; - if (vpd_sequence_should_stop(seq)) + p = (struct rtas_ibm_get_vpd_params *)seq->params; + init_state = (p->written == 0) ? true : false; + + if (papr_rtas_sequence_should_stop(seq, p->status, init_state)) return NULL; - if (vpd_sequence_set_err(seq, rtas_ibm_get_vpd(p))) + if (papr_rtas_sequence_set_err(seq, rtas_ibm_get_vpd(p))) return NULL; *len = p->written; return rtas_work_area_raw_buf(p->work_area); } -/* - * Higher-level VPD retrieval code below. These functions use the - * vpd_blob_* and vpd_sequence_* APIs defined above to create fd-based - * VPD handles for consumption by user space. - */ - -/** - * papr_vpd_run_sequence() - Run a single VPD retrieval sequence. - * @loc_code: Location code that defines the scope of VPD to return. - * - * Context: May sleep. Holds a mutex and an RTAS work area for its - * duration. Typically performs multiple sleepable slab - * allocations. - * - * Return: A populated &struct vpd_blob on success. Encoded error - * pointer otherwise. - */ -static const struct vpd_blob *papr_vpd_run_sequence(const struct papr_location_code *loc_code) -{ - const struct vpd_blob *blob; - struct vpd_sequence seq; - - vpd_sequence_begin(&seq, loc_code); - blob = vpd_blob_generate(vpd_sequence_fill_work_area, &seq); - if (!blob) - vpd_sequence_set_err(&seq, -ENOMEM); - vpd_sequence_end(&seq); - - if (seq.error) { - vpd_blob_free(blob); - return ERR_PTR(seq.error); - } - - return blob; -} - -/** - * papr_vpd_retrieve() - Return the VPD for a location code. - * @loc_code: Location code that defines the scope of VPD to return. - * - * Run VPD sequences against @loc_code until a blob is successfully - * instantiated, or a hard error is encountered, or a fatal signal is - * pending. - * - * Context: May sleep. - * Return: A fully populated VPD blob when successful. Encoded error - * pointer otherwise. - */ -static const struct vpd_blob *papr_vpd_retrieve(const struct papr_location_code *loc_code) -{ - const struct vpd_blob *blob; - - /* - * EAGAIN means the sequence errored with a -4 (VPD changed) - * status from ibm,get-vpd, and we should attempt a new - * sequence. PAPR+ v2.13 R1–7.3.20–5 indicates that this - * should be a transient condition, not something that happens - * continuously. But we'll stop trying on a fatal signal. - */ - do { - blob = papr_vpd_run_sequence(loc_code); - if (!IS_ERR(blob)) /* Success. */ - break; - if (PTR_ERR(blob) != -EAGAIN) /* Hard error. */ - break; - pr_info_ratelimited("VPD changed during retrieval, retrying\n"); - cond_resched(); - } while (!fatal_signal_pending(current)); - - return blob; -} - -static ssize_t papr_vpd_handle_read(struct file *file, char __user *buf, size_t size, loff_t *off) -{ - const struct vpd_blob *blob = file->private_data; - - /* bug: we should not instantiate a handle without any data attached. */ - if (!vpd_blob_has_data(blob)) { - pr_err_once("handle without data\n"); - return -EIO; - } - - return simple_read_from_buffer(buf, size, off, blob->data, blob->len); -} - -static int papr_vpd_handle_release(struct inode *inode, struct file *file) -{ - const struct vpd_blob *blob = file->private_data; - - vpd_blob_free(blob); - - return 0; -} - -static loff_t papr_vpd_handle_seek(struct file *file, loff_t off, int whence) -{ - const struct vpd_blob *blob = file->private_data; - - return fixed_size_llseek(file, off, whence, blob->len); -} - - static const struct file_operations papr_vpd_handle_ops = { - .read = papr_vpd_handle_read, - .llseek = papr_vpd_handle_seek, - .release = papr_vpd_handle_release, + .read = papr_rtas_common_handle_read, + .llseek = papr_rtas_common_handle_seek, + .release = papr_rtas_common_handle_release, }; /** @@ -460,10 +210,9 @@ static const struct file_operations papr_vpd_handle_ops = { */ static long papr_vpd_create_handle(struct papr_location_code __user *ulc) { + struct rtas_ibm_get_vpd_params vpd_params = {}; + struct papr_rtas_sequence seq = {}; struct papr_location_code klc; - const struct vpd_blob *blob; - struct file *file; - long err; int fd; if (copy_from_user(&klc, ulc, sizeof(klc))) @@ -472,30 +221,19 @@ static long papr_vpd_create_handle(struct papr_location_code __user *ulc) if (!string_is_terminated(klc.str, ARRAY_SIZE(klc.str))) return -EINVAL; - blob = papr_vpd_retrieve(&klc); - if (IS_ERR(blob)) - return PTR_ERR(blob); + seq = (struct papr_rtas_sequence) { + .begin = vpd_sequence_begin, + .end = vpd_sequence_end, + .work = vpd_sequence_fill_work_area, + }; - fd = get_unused_fd_flags(O_RDONLY | O_CLOEXEC); - if (fd < 0) { - err = fd; - goto free_blob; - } + vpd_params.loc_code = &klc; + seq.params = (void *)&vpd_params; + + fd = papr_rtas_setup_file_interface(&seq, &papr_vpd_handle_ops, + "[papr-vpd]"); - file = anon_inode_getfile_fmode("[papr-vpd]", &papr_vpd_handle_ops, - (void *)blob, O_RDONLY, - FMODE_LSEEK | FMODE_PREAD); - if (IS_ERR(file)) { - err = PTR_ERR(file); - goto put_fd; - } - fd_install(fd, file); return fd; -put_fd: - put_unused_fd(fd); -free_blob: - vpd_blob_free(blob); - return err; } /* diff --git a/arch/powerpc/platforms/pseries/pci_dlpar.c b/arch/powerpc/platforms/pseries/pci_dlpar.c index 52e2623a741d..aeb8633a3d00 100644 --- a/arch/powerpc/platforms/pseries/pci_dlpar.c +++ b/arch/powerpc/platforms/pseries/pci_dlpar.c @@ -29,7 +29,7 @@ struct pci_controller *init_phb_dynamic(struct device_node *dn) nid = of_node_to_nid(dn); if (likely((nid) >= 0)) { if (!node_online(nid)) { - if (__register_one_node(nid)) { + if (register_one_node(nid)) { pr_err("PCI: Failed to register node %d\n", nid); } else { update_numa_distance(dn); diff --git a/arch/powerpc/platforms/pseries/plpks-secvar.c b/arch/powerpc/platforms/pseries/plpks-secvar.c index 257fd1f8bc19..f9e9cc40c9d0 100644 --- a/arch/powerpc/platforms/pseries/plpks-secvar.c +++ b/arch/powerpc/platforms/pseries/plpks-secvar.c @@ -59,7 +59,14 @@ static u32 get_policy(const char *name) return PLPKS_SIGNEDUPDATE; } -static const char * const plpks_var_names[] = { +static const char * const plpks_var_names_static[] = { + "PK", + "moduledb", + "trustedcadb", + NULL, +}; + +static const char * const plpks_var_names_dynamic[] = { "PK", "KEK", "db", @@ -152,39 +159,55 @@ err: return rc; } -// PLPKS dynamic secure boot doesn't give us a format string in the same way OPAL does. -// Instead, report the format using the SB_VERSION variable in the keystore. -// The string is made up by us, and takes the form "ibm,plpks-sb-v<n>" (or "ibm,plpks-sb-unknown" -// if the SB_VERSION variable doesn't exist). Hypervisor defines the SB_VERSION variable as a -// "1 byte unsigned integer value". -static ssize_t plpks_secvar_format(char *buf, size_t bufsize) +/* + * Return the key management mode. + * + * SB_VERSION is defined as a "1 byte unsigned integer value", taking values + * starting from 1. It is owned by the Partition Firmware and its presence + * indicates that the key management mode is dynamic. Any failure in + * reading SB_VERSION defaults the key management mode to static. The error + * codes -ENOENT or -EPERM are expected in static key management mode. An + * unexpected error code will have to be investigated. Only signed variables + * have null bytes in their names, SB_VERSION does not. + * + * Return 0 to indicate that the key management mode is static. Otherwise + * return the SB_VERSION value to indicate that the key management mode is + * dynamic. + */ +static u8 plpks_get_sb_keymgmt_mode(void) { - struct plpks_var var = {0}; - ssize_t ret; - u8 version; - - var.component = NULL; - // Only the signed variables have null bytes in their names, this one doesn't - var.name = "SB_VERSION"; - var.namelen = strlen(var.name); - var.datalen = 1; - var.data = &version; - - // Unlike the other vars, SB_VERSION is owned by firmware instead of the OS - ret = plpks_read_fw_var(&var); - if (ret) { - if (ret == -ENOENT) { - ret = snprintf(buf, bufsize, "ibm,plpks-sb-unknown"); - } else { - pr_err("Error %ld reading SB_VERSION from firmware\n", ret); - ret = -EIO; - } - goto err; + u8 mode; + ssize_t rc; + struct plpks_var var = { + .component = NULL, + .name = "SB_VERSION", + .namelen = 10, + .datalen = 1, + .data = &mode, + }; + + rc = plpks_read_fw_var(&var); + if (rc) { + if (rc != -ENOENT && rc != -EPERM) + pr_info("Error %ld reading SB_VERSION from firmware\n", rc); + mode = 0; } + return mode; +} - ret = snprintf(buf, bufsize, "ibm,plpks-sb-v%hhu", version); -err: - return ret; +/* + * PLPKS dynamic secure boot doesn't give us a format string in the same way + * OPAL does. Instead, report the format using the SB_VERSION variable in the + * keystore. The string, made up by us, takes the form of either + * "ibm,plpks-sb-v<n>" or "ibm,plpks-sb-v0", based on the key management mode, + * and return the length of the secvar format property. + */ +static ssize_t plpks_secvar_format(char *buf, size_t bufsize) +{ + u8 mode; + + mode = plpks_get_sb_keymgmt_mode(); + return snprintf(buf, bufsize, "ibm,plpks-sb-v%hhu", mode); } static int plpks_max_size(u64 *max_size) @@ -197,21 +220,34 @@ static int plpks_max_size(u64 *max_size) return 0; } +static const struct secvar_operations plpks_secvar_ops_static = { + .get = plpks_get_variable, + .set = plpks_set_variable, + .format = plpks_secvar_format, + .max_size = plpks_max_size, + .config_attrs = config_attrs, + .var_names = plpks_var_names_static, +}; -static const struct secvar_operations plpks_secvar_ops = { +static const struct secvar_operations plpks_secvar_ops_dynamic = { .get = plpks_get_variable, .set = plpks_set_variable, .format = plpks_secvar_format, .max_size = plpks_max_size, .config_attrs = config_attrs, - .var_names = plpks_var_names, + .var_names = plpks_var_names_dynamic, }; static int plpks_secvar_init(void) { + u8 mode; + if (!plpks_is_available()) return -ENODEV; - return set_secvar_ops(&plpks_secvar_ops); + mode = plpks_get_sb_keymgmt_mode(); + if (mode) + return set_secvar_ops(&plpks_secvar_ops_dynamic); + return set_secvar_ops(&plpks_secvar_ops_static); } machine_device_initcall(pseries, plpks_secvar_init); |