diff options
Diffstat (limited to 'drivers')
64 files changed, 1550 insertions, 821 deletions
diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index d3a59c688fe..8a267c42762 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -17,7 +17,7 @@ #include <linux/bitops.h> #include <linux/mutex.h> #include <linux/kthread.h> - +#include <linux/highmem.h> #include <linux/firmware.h> #include "base.h" @@ -45,7 +45,10 @@ struct firmware_priv { struct bin_attribute attr_data; struct firmware *fw; unsigned long status; - int alloc_size; + struct page **pages; + int nr_pages; + int page_array_size; + const char *vdata; struct timer_list timeout; }; @@ -122,6 +125,10 @@ static ssize_t firmware_loading_show(struct device *dev, return sprintf(buf, "%d\n", loading); } +/* Some architectures don't have PAGE_KERNEL_RO */ +#ifndef PAGE_KERNEL_RO +#define PAGE_KERNEL_RO PAGE_KERNEL +#endif /** * firmware_loading_store - set value in the 'loading' control file * @dev: device pointer @@ -141,6 +148,7 @@ static ssize_t firmware_loading_store(struct device *dev, { struct firmware_priv *fw_priv = dev_get_drvdata(dev); int loading = simple_strtol(buf, NULL, 10); + int i; switch (loading) { case 1: @@ -151,13 +159,30 @@ static ssize_t firmware_loading_store(struct device *dev, } vfree(fw_priv->fw->data); fw_priv->fw->data = NULL; + for (i = 0; i < fw_priv->nr_pages; i++) + __free_page(fw_priv->pages[i]); + kfree(fw_priv->pages); + fw_priv->pages = NULL; + fw_priv->page_array_size = 0; + fw_priv->nr_pages = 0; fw_priv->fw->size = 0; - fw_priv->alloc_size = 0; set_bit(FW_STATUS_LOADING, &fw_priv->status); mutex_unlock(&fw_lock); break; case 0: if (test_bit(FW_STATUS_LOADING, &fw_priv->status)) { + vfree(fw_priv->fw->data); + fw_priv->fw->data = vmap(fw_priv->pages, + fw_priv->nr_pages, + 0, PAGE_KERNEL_RO); + if (!fw_priv->fw->data) { + dev_err(dev, "%s: vmap() failed\n", __func__); + goto err; + } + /* Pages will be freed by vfree() */ + fw_priv->pages = NULL; + fw_priv->page_array_size = 0; + fw_priv->nr_pages = 0; complete(&fw_priv->completion); clear_bit(FW_STATUS_LOADING, &fw_priv->status); break; @@ -167,6 +192,7 @@ static ssize_t firmware_loading_store(struct device *dev, dev_err(dev, "%s: unexpected value (%d)\n", __func__, loading); /* fallthrough */ case -1: + err: fw_load_abort(fw_priv); break; } @@ -191,8 +217,28 @@ firmware_data_read(struct kobject *kobj, struct bin_attribute *bin_attr, ret_count = -ENODEV; goto out; } - ret_count = memory_read_from_buffer(buffer, count, &offset, - fw->data, fw->size); + if (offset > fw->size) + return 0; + if (count > fw->size - offset) + count = fw->size - offset; + + ret_count = count; + + while (count) { + void *page_data; + int page_nr = offset >> PAGE_SHIFT; + int page_ofs = offset & (PAGE_SIZE-1); + int page_cnt = min_t(size_t, PAGE_SIZE - page_ofs, count); + + page_data = kmap(fw_priv->pages[page_nr]); + + memcpy(buffer, page_data + page_ofs, page_cnt); + + kunmap(fw_priv->pages[page_nr]); + buffer += page_cnt; + offset += page_cnt; + count -= page_cnt; + } out: mutex_unlock(&fw_lock); return ret_count; @@ -201,27 +247,39 @@ out: static int fw_realloc_buffer(struct firmware_priv *fw_priv, int min_size) { - u8 *new_data; - int new_size = fw_priv->alloc_size; + int pages_needed = ALIGN(min_size, PAGE_SIZE) >> PAGE_SHIFT; + + /* If the array of pages is too small, grow it... */ + if (fw_priv->page_array_size < pages_needed) { + int new_array_size = max(pages_needed, + fw_priv->page_array_size * 2); + struct page **new_pages; + + new_pages = kmalloc(new_array_size * sizeof(void *), + GFP_KERNEL); + if (!new_pages) { + fw_load_abort(fw_priv); + return -ENOMEM; + } + memcpy(new_pages, fw_priv->pages, + fw_priv->page_array_size * sizeof(void *)); + memset(&new_pages[fw_priv->page_array_size], 0, sizeof(void *) * + (new_array_size - fw_priv->page_array_size)); + kfree(fw_priv->pages); + fw_priv->pages = new_pages; + fw_priv->page_array_size = new_array_size; + } - if (min_size <= fw_priv->alloc_size) - return 0; + while (fw_priv->nr_pages < pages_needed) { + fw_priv->pages[fw_priv->nr_pages] = + alloc_page(GFP_KERNEL | __GFP_HIGHMEM); - new_size = ALIGN(min_size, PAGE_SIZE); - new_data = vmalloc(new_size); - if (!new_data) { - printk(KERN_ERR "%s: unable to alloc buffer\n", __func__); - /* Make sure that we don't keep incomplete data */ - fw_load_abort(fw_priv); - return -ENOMEM; - } - fw_priv->alloc_size = new_size; - if (fw_priv->fw->data) { - memcpy(new_data, fw_priv->fw->data, fw_priv->fw->size); - vfree(fw_priv->fw->data); + if (!fw_priv->pages[fw_priv->nr_pages]) { + fw_load_abort(fw_priv); + return -ENOMEM; + } + fw_priv->nr_pages++; } - fw_priv->fw->data = new_data; - BUG_ON(min_size > fw_priv->alloc_size); return 0; } @@ -258,10 +316,25 @@ firmware_data_write(struct kobject *kobj, struct bin_attribute *bin_attr, if (retval) goto out; - memcpy((u8 *)fw->data + offset, buffer, count); - - fw->size = max_t(size_t, offset + count, fw->size); retval = count; + + while (count) { + void *page_data; + int page_nr = offset >> PAGE_SHIFT; + int page_ofs = offset & (PAGE_SIZE - 1); + int page_cnt = min_t(size_t, PAGE_SIZE - page_ofs, count); + + page_data = kmap(fw_priv->pages[page_nr]); + + memcpy(page_data + page_ofs, buffer, page_cnt); + + kunmap(fw_priv->pages[page_nr]); + buffer += page_cnt; + offset += page_cnt; + count -= page_cnt; + } + + fw->size = max_t(size_t, offset, fw->size); out: mutex_unlock(&fw_lock); return retval; @@ -277,7 +350,11 @@ static struct bin_attribute firmware_attr_data_tmpl = { static void fw_dev_release(struct device *dev) { struct firmware_priv *fw_priv = dev_get_drvdata(dev); + int i; + for (i = 0; i < fw_priv->nr_pages; i++) + __free_page(fw_priv->pages[i]); + kfree(fw_priv->pages); kfree(fw_priv); kfree(dev); diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index c0facaa55cf..43db3ea15b5 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -254,7 +254,7 @@ static int index_to_minor(int index) return index << PART_BITS; } -static int virtblk_probe(struct virtio_device *vdev) +static int __devinit virtblk_probe(struct virtio_device *vdev) { struct virtio_blk *vblk; int err; @@ -288,7 +288,7 @@ static int virtblk_probe(struct virtio_device *vdev) sg_init_table(vblk->sg, vblk->sg_elems); /* We expect one virtqueue, for output. */ - vblk->vq = vdev->config->find_vq(vdev, 0, blk_done); + vblk->vq = virtio_find_single_vq(vdev, blk_done, "requests"); if (IS_ERR(vblk->vq)) { err = PTR_ERR(vblk->vq); goto out_free_vblk; @@ -388,14 +388,14 @@ out_put_disk: out_mempool: mempool_destroy(vblk->pool); out_free_vq: - vdev->config->del_vq(vblk->vq); + vdev->config->del_vqs(vdev); out_free_vblk: kfree(vblk); out: return err; } -static void virtblk_remove(struct virtio_device *vdev) +static void __devexit virtblk_remove(struct virtio_device *vdev) { struct virtio_blk *vblk = vdev->priv; @@ -409,7 +409,7 @@ static void virtblk_remove(struct virtio_device *vdev) blk_cleanup_queue(vblk->disk->queue); put_disk(vblk->disk); mempool_destroy(vblk->pool); - vdev->config->del_vq(vblk->vq); + vdev->config->del_vqs(vdev); kfree(vblk); } diff --git a/drivers/char/hw_random/virtio-rng.c b/drivers/char/hw_random/virtio-rng.c index 86e83f88313..32216b62324 100644 --- a/drivers/char/hw_random/virtio-rng.c +++ b/drivers/char/hw_random/virtio-rng.c @@ -35,13 +35,13 @@ static DECLARE_COMPLETION(have_data); static void random_recv_done(struct virtqueue *vq) { - int len; + unsigned int len; /* We can get spurious callbacks, e.g. shared IRQs + virtio_pci. */ if (!vq->vq_ops->get_buf(vq, &len)) return; - data_left = len / sizeof(random_data[0]); + data_left += len; complete(&have_data); } @@ -49,7 +49,7 @@ static void register_buffer(void) { struct scatterlist sg; - sg_init_one(&sg, random_data, RANDOM_DATA_SIZE); + sg_init_one(&sg, random_data+data_left, RANDOM_DATA_SIZE-data_left); /* There should always be room for one buffer. */ if (vq->vq_ops->add_buf(vq, &sg, 0, 1, random_data) != 0) BUG(); @@ -59,24 +59,32 @@ static void register_buffer(void) /* At least we don't udelay() in a loop like some other drivers. */ static int virtio_data_present(struct hwrng *rng, int wait) { - if (data_left) + if (data_left >= sizeof(u32)) return 1; +again: if (!wait) return 0; wait_for_completion(&have_data); + + /* Not enough? Re-register. */ + if (unlikely(data_left < sizeof(u32))) { + register_buffer(); + goto again; + } + return 1; } /* virtio_data_present() must have succeeded before this is called. */ static int virtio_data_read(struct hwrng *rng, u32 *data) { - BUG_ON(!data_left); - - *data = random_data[--data_left]; + BUG_ON(data_left < sizeof(u32)); + data_left -= sizeof(u32); + *data = random_data[data_left / 4]; - if (!data_left) { + if (data_left < sizeof(u32)) { init_completion(&have_data); register_buffer(); } @@ -94,13 +102,13 @@ static int virtrng_probe(struct virtio_device *vdev) int err; /* We expect a single virtqueue. */ - vq = vdev->config->find_vq(vdev, 0, random_recv_done); + vq = virtio_find_single_vq(vdev, random_recv_done, "input"); if (IS_ERR(vq)) return PTR_ERR(vq); err = hwrng_register(&virtio_hwrng); if (err) { - vdev->config->del_vq(vq); + vdev->config->del_vqs(vdev); return err; } @@ -112,7 +120,7 @@ static void virtrng_remove(struct virtio_device *vdev) { vdev->config->reset(vdev); hwrng_unregister(&virtio_hwrng); - vdev->config->del_vq(vq); + vdev->config->del_vqs(vdev); } static struct virtio_device_id id_table[] = { diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c index ff6f5a4b58f..c74dacfa679 100644 --- a/drivers/char/virtio_console.c +++ b/drivers/char/virtio_console.c @@ -188,6 +188,9 @@ static void hvc_handle_input(struct virtqueue *vq) * Finally we put our input buffer in the input queue, ready to receive. */ static int __devinit virtcons_probe(struct virtio_device *dev) { + vq_callback_t *callbacks[] = { hvc_handle_input, NULL}; + const char *names[] = { "input", "output" }; + struct virtqueue *vqs[2]; int err; vdev = dev; @@ -199,20 +202,15 @@ static int __devinit virtcons_probe(struct virtio_device *dev) goto fail; } - /* Find the input queue. */ + /* Find the queues. */ /* FIXME: This is why we want to wean off hvc: we do nothing * when input comes in. */ - in_vq = vdev->config->find_vq(vdev, 0, hvc_handle_input); - if (IS_ERR(in_vq)) { - err = PTR_ERR(in_vq); + err = vdev->config->find_vqs(vdev, 2, vqs, callbacks, names); + if (err) goto free; - } - out_vq = vdev->config->find_vq(vdev, 1, NULL); - if (IS_ERR(out_vq)) { - err = PTR_ERR(out_vq); - goto free_in_vq; - } + in_vq = vqs[0]; + out_vq = vqs[1]; /* Start using the new console output. */ virtio_cons.get_chars = get_chars; @@ -233,17 +231,15 @@ static int __devinit virtcons_probe(struct virtio_device *dev) hvc = hvc_alloc(0, 0, &virtio_cons, PAGE_SIZE); if (IS_ERR(hvc)) { err = PTR_ERR(hvc); - goto free_out_vq; + goto free_vqs; } /* Register the input buffer the first time. */ add_inbuf(); return 0; -free_out_vq: - vdev->config->del_vq(out_vq); -free_in_vq: - vdev->config->del_vq(in_vq); +free_vqs: + vdev->config->del_vqs(vdev); free: kfree(inbuf); fail: diff --git a/drivers/ide/at91_ide.c b/drivers/ide/at91_ide.c index 403d0e4265d..fc0949a8cfd 100644 --- a/drivers/ide/at91_ide.c +++ b/drivers/ide/at91_ide.c @@ -216,6 +216,7 @@ static const struct ide_port_info at91_ide_port_info __initdata = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA | IDE_HFLAG_SINGLE | IDE_HFLAG_NO_IO_32BIT | IDE_HFLAG_UNMASK_IRQS, .pio_mask = ATA_PIO6, + .chipset = ide_generic, }; /* @@ -246,8 +247,7 @@ irqreturn_t at91_irq_handler(int irq, void *dev_id) static int __init at91_ide_probe(struct platform_device *pdev) { int ret; - hw_regs_t hw; - hw_regs_t *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; struct ide_host *host; struct resource *res; unsigned long tf_base = 0, ctl_base = 0; @@ -304,10 +304,9 @@ static int __init at91_ide_probe(struct platform_device *pdev) ide_std_init_ports(&hw, tf_base, ctl_base + 6); hw.irq = board->irq_pin; - hw.chipset = ide_generic; hw.dev = &pdev->dev; - host = ide_host_alloc(&at91_ide_port_info, hws); + host = ide_host_alloc(&at91_ide_port_info, hws, 1); if (!host) { perr("failed to allocate ide host\n"); return -ENOMEM; diff --git a/drivers/ide/au1xxx-ide.c b/drivers/ide/au1xxx-ide.c index 46013644c96..58121bd6c11 100644 --- a/drivers/ide/au1xxx-ide.c +++ b/drivers/ide/au1xxx-ide.c @@ -449,7 +449,7 @@ static int auide_ddma_init(ide_hwif_t *hwif, const struct ide_port_info *d) } #endif -static void auide_setup_ports(hw_regs_t *hw, _auide_hwif *ahwif) +static void auide_setup_ports(struct ide_hw *hw, _auide_hwif *ahwif) { int i; unsigned long *ata_regs = hw->io_ports_array; @@ -499,6 +499,7 @@ static const struct ide_port_info au1xxx_port_info = { #ifdef CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA .mwdma_mask = ATA_MWDMA2, #endif + .chipset = ide_au1xxx, }; static int au_ide_probe(struct platform_device *dev) @@ -507,7 +508,7 @@ static int au_ide_probe(struct platform_device *dev) struct resource *res; struct ide_host *host; int ret = 0; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; #if defined(CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA) char *mode = "MWDMA2"; @@ -548,9 +549,8 @@ static int au_ide_probe(struct platform_device *dev) auide_setup_ports(&hw, ahwif); hw.irq = ahwif->irq; hw.dev = &dev->dev; - hw.chipset = ide_au1xxx; - ret = ide_host_add(&au1xxx_port_info, hws, &host); + ret = ide_host_add(&au1xxx_port_info, hws, 1, &host); if (ret) goto out; diff --git a/drivers/ide/buddha.c b/drivers/ide/buddha.c index d028f8864bc..e3c6a591330 100644 --- a/drivers/ide/buddha.c +++ b/drivers/ide/buddha.c @@ -121,7 +121,7 @@ static int xsurf_ack_intr(ide_hwif_t *hwif) return 1; } -static void __init buddha_setup_ports(hw_regs_t *hw, unsigned long base, +static void __init buddha_setup_ports(struct ide_hw *hw, unsigned long base, unsigned long ctl, unsigned long irq_port, ide_ack_intr_t *ack_intr) { @@ -139,13 +139,12 @@ static void __init buddha_setup_ports(hw_regs_t *hw, unsigned long base, hw->irq = IRQ_AMIGA_PORTS; hw->ack_intr = ack_intr; - - hw->chipset = ide_generic; } static const struct ide_port_info buddha_port_info = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_generic, }; /* @@ -161,7 +160,7 @@ static int __init buddha_init(void) while ((z = zorro_find_device(ZORRO_WILDCARD, z))) { unsigned long board; - hw_regs_t hw[MAX_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[MAX_NUM_HWIFS], *hws[MAX_NUM_HWIFS]; if (z->id == ZORRO_PROD_INDIVIDUAL_COMPUTERS_BUDDHA) { buddha_num_hwifs = BUDDHA_NUM_HWIFS; @@ -225,7 +224,7 @@ fail_base2: hws[i] = &hw[i]; } - ide_host_add(&buddha_port_info, hws, NULL); + ide_host_add(&buddha_port_info, hws, i, NULL); } return 0; diff --git a/drivers/ide/cmd640.c b/drivers/ide/cmd640.c index 8890276fef7..1683ed5c732 100644 --- a/drivers/ide/cmd640.c +++ b/drivers/ide/cmd640.c @@ -708,7 +708,7 @@ static int __init cmd640x_init(void) int second_port_cmd640 = 0, rc; const char *bus_type, *port2; u8 b, cfr; - hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[2], *hws[2]; if (cmd640_vlb && probe_for_cmd640_vlb()) { bus_type = "VLB"; @@ -762,11 +762,9 @@ static int __init cmd640x_init(void) ide_std_init_ports(&hw[0], 0x1f0, 0x3f6); hw[0].irq = 14; - hw[0].chipset = ide_cmd640; ide_std_init_ports(&hw[1], 0x170, 0x376); hw[1].irq = 15; - hw[1].chipset = ide_cmd640; printk(KERN_INFO "cmd640: buggy cmd640%c interface on %s, config=0x%02x" "\n", 'a' + cmd640_chip_version - 1, bus_type, cfr); @@ -824,7 +822,8 @@ static int __init cmd640x_init(void) cmd640_dump_regs(); #endif - return ide_host_add(&cmd640_port_info, hws, NULL); + return ide_host_add(&cmd640_port_info, hws, second_port_cmd640 ? 2 : 1, + NULL); } module_param_named(probe_vlb, cmd640_vlb, bool, 0); diff --git a/drivers/ide/cs5520.c b/drivers/ide/cs5520.c index 87987a7d36c..bd066bb9d61 100644 --- a/drivers/ide/cs5520.c +++ b/drivers/ide/cs5520.c @@ -110,7 +110,7 @@ static const struct ide_port_info cyrix_chipset __devinitdata = { static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id) { const struct ide_port_info *d = &cyrix_chipset; - hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[2], *hws[] = { NULL, NULL }; ide_setup_pci_noise(dev, d); @@ -136,7 +136,7 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic ide_pci_setup_ports(dev, d, &hw[0], &hws[0]); hw[0].irq = 14; - return ide_host_add(d, hws, NULL); + return ide_host_add(d, hws, 2, NULL); } static const struct pci_device_id cs5520_pci_tbl[] = { diff --git a/drivers/ide/delkin_cb.c b/drivers/ide/delkin_cb.c index f153b95619b..1e10eba62ce 100644 --- a/drivers/ide/delkin_cb.c +++ b/drivers/ide/delkin_cb.c @@ -68,6 +68,7 @@ static const struct ide_port_info delkin_cb_port_info = { IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, .init_chipset = delkin_cb_init_chipset, + .chipset = ide_pci, }; static int __devinit @@ -76,7 +77,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id) struct ide_host *host; unsigned long base; int rc; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; rc = pci_enable_device(dev); if (rc) { @@ -97,9 +98,8 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id) ide_std_init_ports(&hw, base + 0x10, base + 0x1e); hw.irq = dev->irq; hw.dev = &dev->dev; - hw.chipset = ide_pci; /* this enables IRQ sharing */ - rc = ide_host_add(&delkin_cb_port_info, hws, &host); + rc = ide_host_add(&delkin_cb_port_info, hws, 1, &host); if (rc) goto out_disable; diff --git a/drivers/ide/falconide.c b/drivers/ide/falconide.c index 0e2df6755ec..22fa27389c3 100644 --- a/drivers/ide/falconide.c +++ b/drivers/ide/falconide.c @@ -111,9 +111,10 @@ static const struct ide_port_info falconide_port_info = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_SERIALIZE | IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_generic, }; -static void __init falconide_setup_ports(hw_regs_t *hw) +static void __init falconide_setup_ports(struct ide_hw *hw) { int i; @@ -128,8 +129,6 @@ static void __init falconide_setup_ports(hw_regs_t *hw) hw->irq = IRQ_MFP_IDE; hw->ack_intr = NULL; - - hw->chipset = ide_generic; } /* @@ -139,7 +138,7 @@ static void __init falconide_setup_ports(hw_regs_t *hw) static int __init falconide_init(void) { struct ide_host *host; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; int rc; if (!MACH_IS_ATARI || !ATARIHW_PRESENT(IDE)) @@ -154,7 +153,7 @@ static int __init falconide_init(void) falconide_setup_ports(&hw); - host = ide_host_alloc(&falconide_port_info, hws); + host = ide_host_alloc(&falconide_port_info, hws, 1); if (host == NULL) { rc = -ENOMEM; goto err; diff --git a/drivers/ide/gayle.c b/drivers/ide/gayle.c index c7119516c5a..4451a6a5dfe 100644 --- a/drivers/ide/gayle.c +++ b/drivers/ide/gayle.c @@ -88,7 +88,7 @@ static int gayle_ack_intr_a1200(ide_hwif_t *hwif) return 1; } -static void __init gayle_setup_ports(hw_regs_t *hw, unsigned long base, +static void __init gayle_setup_ports(struct ide_hw *hw, unsigned long base, unsigned long ctl, unsigned long irq_port, ide_ack_intr_t *ack_intr) { @@ -106,14 +106,13 @@ static void __init gayle_setup_ports(hw_regs_t *hw, unsigned long base, hw->irq = IRQ_AMIGA_PORTS; hw->ack_intr = ack_intr; - - hw->chipset = ide_generic; } static const struct ide_port_info gayle_port_info = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_SERIALIZE | IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_generic, }; /* @@ -126,7 +125,7 @@ static int __init gayle_init(void) unsigned long base, ctrlport, irqport; ide_ack_intr_t *ack_intr; int a4000, i, rc; - hw_regs_t hw[GAYLE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[GAYLE_NUM_HWIFS], *hws[GAYLE_NUM_HWIFS]; if (!MACH_IS_AMIGA) return -ENODEV; @@ -171,7 +170,7 @@ found: hws[i] = &hw[i]; } - rc = ide_host_add(&gayle_port_info, hws, NULL); + rc = ide_host_add(&gayle_port_info, hws, i, NULL); if (rc) release_mem_region(res_start, res_n); diff --git a/drivers/ide/hpt366.c b/drivers/ide/hpt366.c index 0feb66c720e..7ce68ef6b90 100644 --- a/drivers/ide/hpt366.c +++ b/drivers/ide/hpt366.c @@ -138,14 +138,6 @@ #undef HPT_RESET_STATE_ENGINE #undef HPT_DELAY_INTERRUPT -static const char *quirk_drives[] = { - "QUANTUM FIREBALLlct08 08", - "QUANTUM FIREBALLP KA6.4", - "QUANTUM FIREBALLP LM20.4", - "QUANTUM FIREBALLP LM20.5", - NULL -}; - static const char *bad_ata100_5[] = { "IBM-DTLA-307075", "IBM-DTLA-307060", @@ -729,27 +721,13 @@ static void hpt3xx_set_pio_mode(ide_drive_t *drive, const u8 pio) hpt3xx_set_mode(drive, XFER_PIO_0 + pio); } -static void hpt3xx_quirkproc(ide_drive_t *drive) -{ - char *m = (char *)&drive->id[ATA_ID_PROD]; - const char **list = quirk_drives; - - while (*list) - if (strstr(m, *list++)) { - drive->quirk_list = 1; - return; - } - - drive->quirk_list = 0; -} - static void hpt3xx_maskproc(ide_drive_t *drive, int mask) { ide_hwif_t *hwif = drive->hwif; struct pci_dev *dev = to_pci_dev(hwif->dev); struct hpt_info *info = hpt3xx_get_info(hwif->dev); - if (drive->quirk_list == 0) + if ((drive->dev_flags & IDE_DFLAG_NIEN_QUIRK) == 0) return; if (info->chip_type >= HPT370) { @@ -1404,7 +1382,6 @@ static int __devinit hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2) static const struct ide_port_ops hpt3xx_port_ops = { .set_pio_mode = hpt3xx_set_pio_mode, .set_dma_mode = hpt3xx_set_mode, - .quirkproc = hpt3xx_quirkproc, .maskproc = hpt3xx_maskproc, .mdma_filter = hpt3xx_mdma_filter, .udma_filter = hpt3xx_udma_filter, diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c index 36da913cc55..5af3d0ffaf0 100644 --- a/drivers/ide/icside.c +++ b/drivers/ide/icside.c @@ -65,8 +65,6 @@ static struct cardinfo icside_cardinfo_v6_2 = { }; struct icside_state { - unsigned int channel; - unsigned int enabled; void __iomem *irq_port; void __iomem *ioc_base; unsigned int sel; @@ -116,18 +114,11 @@ static void icside_irqenable_arcin_v6 (struct expansion_card *ec, int irqnr) struct icside_state *state = ec->irq_data; void __iomem *base = state->irq_port; - state->enabled = 1; + writeb(0, base + ICS_ARCIN_V6_INTROFFSET_1); + readb(base + ICS_ARCIN_V6_INTROFFSET_2); - switch (state->channel) { - case 0: - writeb(0, base + ICS_ARCIN_V6_INTROFFSET_1); - readb(base + ICS_ARCIN_V6_INTROFFSET_2); - break; - case 1: - writeb(0, base + ICS_ARCIN_V6_INTROFFSET_2); - readb(base + ICS_ARCIN_V6_INTROFFSET_1); - break; - } + writeb(0, base + ICS_ARCIN_V6_INTROFFSET_2); + readb(base + ICS_ARCIN_V6_INTROFFSET_1); } /* Prototype: icside_irqdisable_arcin_v6 (struct expansion_card *ec, int irqnr) @@ -137,8 +128,6 @@ static void icside_irqdisable_arcin_v6 (struct expansion_card *ec, int irqnr) { struct icside_state *state = ec->irq_data; - state->enabled = 0; - readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_1); readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_2); } @@ -160,44 +149,6 @@ static const expansioncard_ops_t icside_ops_arcin_v6 = { .irqpending = icside_irqpending_arcin_v6, }; -/* - * Handle routing of interrupts. This is called before - * we write the command to the drive. - */ -static void icside_maskproc(ide_drive_t *drive, int mask) -{ - ide_hwif_t *hwif = drive->hwif; - struct expansion_card *ec = ECARD_DEV(hwif->dev); - struct icside_state *state = ecard_get_drvdata(ec); - unsigned long flags; - - local_irq_save(flags); - - state->channel = hwif->channel; - - if (state->enabled && !mask) { - switch (hwif->channel) { - case 0: - writeb(0, state->irq_port + ICS_ARCIN_V6_INTROFFSET_1); - readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_2); - break; - case 1: - writeb(0, state->irq_port + ICS_ARCIN_V6_INTROFFSET_2); - readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_1); - break; - } - } else { - readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_2); - readb(state->irq_port + ICS_ARCIN_V6_INTROFFSET_1); - } - - local_irq_restore(flags); -} - -static const struct ide_port_ops icside_v6_no_dma_port_ops = { - .maskproc = icside_maskproc, -}; - #ifdef CONFIG_BLK_DEV_IDEDMA_ICS /* * SG-DMA support. @@ -275,7 +226,6 @@ static void icside_set_dma_mode(ide_drive_t *drive, const u8 xfer_mode) static const struct ide_port_ops icside_v6_port_ops = { .set_dma_mode = icside_set_dma_mode, - .maskproc = icside_maskproc, }; static void icside_dma_host_set(ide_drive_t *drive, int on) @@ -320,11 +270,6 @@ static int icside_dma_setup(ide_drive_t *drive, struct ide_cmd *cmd) BUG_ON(dma_channel_active(ec->dma)); /* - * Ensure that we have the right interrupt routed. - */ - icside_maskproc(drive, 0); - - /* * Route the DMA signals to the correct interface. */ writeb(state->sel | hwif->channel, state->ioc_base); @@ -381,7 +326,7 @@ static int icside_dma_off_init(ide_hwif_t *hwif, const struct ide_port_info *d) return -EOPNOTSUPP; } -static void icside_setup_ports(hw_regs_t *hw, void __iomem *base, +static void icside_setup_ports(struct ide_hw *hw, void __iomem *base, struct cardinfo *info, struct expansion_card *ec) { unsigned long port = (unsigned long)base + info->dataoffset; @@ -398,11 +343,11 @@ static void icside_setup_ports(hw_regs_t *hw, void __iomem *base, hw->irq = ec->irq; hw->dev = &ec->dev; - hw->chipset = ide_acorn; } static const struct ide_port_info icside_v5_port_info = { .host_flags = IDE_HFLAG_NO_DMA, + .chipset = ide_acorn, }; static int __devinit @@ -410,7 +355,7 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec) { void __iomem *base; struct ide_host *host; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; int ret; base = ecardm_iomap(ec, ECARD_RES_MEMC, 0, 0); @@ -431,7 +376,7 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec) icside_setup_ports(&hw, base, &icside_cardinfo_v5, ec); - host = ide_host_alloc(&icside_v5_port_info, hws); + host = ide_host_alloc(&icside_v5_port_info, hws, 1); if (host == NULL) return -ENODEV; @@ -452,11 +397,11 @@ err_free: static const struct ide_port_info icside_v6_port_info __initdata = { .init_dma = icside_dma_off_init, - .port_ops = &icside_v6_no_dma_port_ops, .dma_ops = &icside_v6_dma_ops, .host_flags = IDE_HFLAG_SERIALIZE | IDE_HFLAG_MMIO, .mwdma_mask = ATA_MWDMA2, .swdma_mask = ATA_SWDMA2, + .chipset = ide_acorn, }; static int __devinit @@ -466,7 +411,7 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec) struct ide_host *host; unsigned int sel = 0; int ret; - hw_regs_t hw[2], *hws[] = { &hw[0], &hw[1], NULL, NULL }; + struct ide_hw hw[2], *hws[] = { &hw[0], &hw[1] }; struct ide_port_info d = icside_v6_port_info; ioc_base = ecardm_iomap(ec, ECARD_RES_IOCFAST, 0, 0); @@ -506,7 +451,7 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec) icside_setup_ports(&hw[0], easi_base, &icside_cardinfo_v6_1, ec); icside_setup_ports(&hw[1], easi_base, &icside_cardinfo_v6_2, ec); - host = ide_host_alloc(&d, hws); + host = ide_host_alloc(&d, hws, 2); if (host == NULL) return -ENODEV; diff --git a/drivers/ide/ide-4drives.c b/drivers/ide/ide-4drives.c index 78aca75a2c4..979d342c338 100644 --- a/drivers/ide/ide-4drives.c +++ b/drivers/ide/ide-4drives.c @@ -25,12 +25,13 @@ static const struct ide_port_info ide_4drives_port_info = { .port_ops = &ide_4drives_port_ops, .host_flags = IDE_HFLAG_SERIALIZE | IDE_HFLAG_NO_DMA | IDE_HFLAG_4DRIVES, + .chipset = ide_4drives, }; static int __init ide_4drives_init(void) { unsigned long base = 0x1f0, ctl = 0x3f6; - hw_regs_t hw, *hws[] = { &hw, &hw, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw, &hw }; if (probe_4drives == 0) return -ENODEV; @@ -52,9 +53,8 @@ static int __init ide_4drives_init(void) ide_std_init_ports(&hw, base, ctl); hw.irq = 14; - hw.chipset = ide_4drives; - return ide_host_add(&ide_4drives_port_info, hws, NULL); + return ide_host_add(&ide_4drives_port_info, hws, 2, NULL); } module_init(ide_4drives_init); diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c index 757e5956b13..bbdd2547f12 100644 --- a/drivers/ide/ide-atapi.c +++ b/drivers/ide/ide-atapi.c @@ -259,7 +259,7 @@ void ide_retry_pc(ide_drive_t *drive) pc->req_xfer = blk_rq_bytes(sense_rq); if (drive->media == ide_tape) - set_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags); + drive->atapi_flags |= IDE_AFLAG_IGNORE_DSC; /* * Push back the failed request and put request sense on top diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c index 9e47f3529d5..527908ff298 100644 --- a/drivers/ide/ide-cs.c +++ b/drivers/ide/ide-cs.c @@ -155,6 +155,7 @@ static const struct ide_port_info idecs_port_info = { .port_ops = &idecs_port_ops, .host_flags = IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_pci, }; static struct ide_host *idecs_register(unsigned long io, unsigned long ctl, @@ -163,7 +164,7 @@ static struct ide_host *idecs_register(unsigned long io, unsigned long ctl, struct ide_host *host; ide_hwif_t *hwif; int i, rc; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; if (!request_region(io, 8, DRV_NAME)) { printk(KERN_ERR "%s: I/O resource 0x%lX-0x%lX not free.\n", @@ -181,10 +182,9 @@ static struct ide_host *idecs_register(unsigned long io, unsigned long ctl, memset(&hw, 0, sizeof(hw)); ide_std_init_ports(&hw, io, ctl); hw.irq = irq; - hw.chipset = ide_pci; hw.dev = &handle->dev; - rc = ide_host_add(&idecs_port_info, hws, &host); + rc = ide_host_add(&idecs_port_info, hws, 1, &host); if (rc) goto out_release; diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c index c6f7fcfb9d6..6a1de216970 100644 --- a/drivers/ide/ide-disk.c +++ b/drivers/ide/ide-disk.c @@ -302,14 +302,12 @@ static const struct drive_list_entry hpa_list[] = { { NULL, NULL } }; -static void idedisk_check_hpa(ide_drive_t *drive) +static u64 ide_disk_hpa_get_native_capacity(ide_drive_t *drive, int lba48) { - unsigned long long capacity, set_max; - int lba48 = ata_id_lba48_enabled(drive->id); + u64 capacity, set_max; capacity = drive->capacity64; - - set_max = idedisk_read_native_max_address(drive, lba48); + set_max = idedisk_read_native_max_address(drive, lba48); if (ide_in_drive_list(drive->id, hpa_list)) { /* @@ -320,9 +318,31 @@ static void idedisk_check_hpa(ide_drive_t *drive) set_max--; } + return set_max; +} + +static u64 ide_disk_hpa_set_capacity(ide_drive_t *drive, u64 set_max, int lba48) +{ + set_max = idedisk_set_max_address(drive, set_max, lba48); + if (set_max) + drive->capacity64 = set_max; + + return set_max; +} + +static void idedisk_check_hpa(ide_drive_t *drive) +{ + u64 capacity, set_max; + int lba48 = ata_id_lba48_enabled(drive->id); + + capacity = drive->capacity64; + set_max = ide_disk_hpa_get_native_capacity(drive, lba48); + if (set_max <= capacity) return; + drive->probed_capacity = set_max; + printk(KERN_INFO "%s: Host Protected Area detected.\n" "\tcurrent capacity is %llu sectors (%llu MB)\n" "\tnative capacity is %llu sectors (%llu MB)\n", @@ -330,13 +350,13 @@ static void idedisk_check_hpa(ide_drive_t *drive) capacity, sectors_to_MB(capacity), set_max, sectors_to_MB(set_max)); - set_max = idedisk_set_max_address(drive, set_max, lba48); + if ((drive->dev_flags & IDE_DFLAG_NOHPA) == 0) + return; - if (set_max) { - drive->capacity64 = set_max; + set_max = ide_disk_hpa_set_capacity(drive, set_max, lba48); + if (set_max) printk(KERN_INFO "%s: Host Protected Area disabled.\n", drive->name); - } } static int ide_disk_get_capacity(ide_drive_t *drive) @@ -358,6 +378,8 @@ static int ide_disk_get_capacity(ide_drive_t *drive) drive->capacity64 = drive->cyl * drive->head * drive->sect; } + drive->probed_capacity = drive->capacity64; + if (lba) { drive->dev_flags |= IDE_DFLAG_LBA; @@ -376,7 +398,7 @@ static int ide_disk_get_capacity(ide_drive_t *drive) "%llu sectors (%llu MB)\n", drive->name, (unsigned long long)drive->capacity64, sectors_to_MB(drive->capacity64)); - drive->capacity64 = 1ULL << 28; + drive->probed_capacity = drive->capacity64 = 1ULL << 28; } if ((drive->hwif->host_flags & IDE_HFLAG_NO_LBA48_DMA) && @@ -392,6 +414,34 @@ static int ide_disk_get_capacity(ide_drive_t *drive) return 0; } +static u64 ide_disk_set_capacity(ide_drive_t *drive, u64 capacity) +{ + u64 set = min(capacity, drive->probed_capacity); + u16 *id = drive->id; + int lba48 = ata_id_lba48_enabled(id); + + if ((drive->dev_flags & IDE_DFLAG_LBA) == 0 || + ata_id_hpa_enabled(id) == 0) + goto out; + + /* + * according to the spec the SET MAX ADDRESS command shall be + * immediately preceded by a READ NATIVE MAX ADDRESS command + */ + capacity = ide_disk_hpa_get_native_capacity(drive, lba48); + if (capacity == 0) + goto out; + + set = ide_disk_hpa_set_capacity(drive, set, lba48); + if (set) { + /* needed for ->resume to disable HPA */ + drive->dev_flags |= IDE_DFLAG_NOHPA; + return set; + } +out: + return drive->capacity64; +} + static void idedisk_prepare_flush(struct request_queue *q, struct request *rq) { ide_drive_t *drive = q->queuedata; @@ -428,14 +478,14 @@ static int set_multcount(ide_drive_t *drive, int arg) if (arg < 0 || arg > (drive->id[ATA_ID_MAX_MULTSECT] & 0xff)) return -EINVAL; - if (drive->special.b.set_multmode) + if (drive->special_flags & IDE_SFLAG_SET_MULTMODE) return -EBUSY; rq = blk_get_request(drive->queue, READ, __GFP_WAIT); rq->cmd_type = REQ_TYPE_ATA_TASKFILE; drive->mult_req = arg; - drive->special.b.set_multmode = 1; + drive->special_flags |= IDE_SFLAG_SET_MULTMODE; error = blk_execute_rq(drive->queue, NULL, rq, 0); blk_put_request(rq); @@ -740,6 +790,7 @@ static int ide_disk_set_doorlock(ide_drive_t *drive, struct gendisk *disk, const struct ide_disk_ops ide_ata_disk_ops = { .check = ide_disk_check, + .set_capacity = ide_disk_set_capacity, .get_capacity = ide_disk_get_capacity, .setup = ide_disk_setup, .flush = ide_disk_flush, diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c index 001f68f0bb2..219e6fb78dc 100644 --- a/drivers/ide/ide-dma.c +++ b/drivers/ide/ide-dma.c @@ -347,7 +347,6 @@ u8 ide_find_dma_mode(ide_drive_t *drive, u8 req_mode) return mode; } -EXPORT_SYMBOL_GPL(ide_find_dma_mode); static int ide_tune_dma(ide_drive_t *drive) { diff --git a/drivers/ide/ide-eh.c b/drivers/ide/ide-eh.c index 5d5fb961b5c..2b914197961 100644 --- a/drivers/ide/ide-eh.c +++ b/drivers/ide/ide-eh.c @@ -52,7 +52,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, } if ((rq->errors & ERROR_RECAL) == ERROR_RECAL) - drive->special.b.recalibrate = 1; + drive->special_flags |= IDE_SFLAG_RECALIBRATE; ++rq->errors; @@ -268,9 +268,8 @@ static void ide_disk_pre_reset(ide_drive_t *drive) { int legacy = (drive->id[ATA_ID_CFS_ENABLE_2] & 0x0400) ? 0 : 1; - drive->special.all = 0; - drive->special.b.set_geometry = legacy; - drive->special.b.recalibrate = legacy; + drive->special_flags = + legacy ? (IDE_SFLAG_SET_GEOMETRY | IDE_SFLAG_RECALIBRATE) : 0; drive->mult_count = 0; drive->dev_flags &= ~IDE_DFLAG_PARKED; @@ -280,7 +279,7 @@ static void ide_disk_pre_reset(ide_drive_t *drive) drive->mult_req = 0; if (drive->mult_req != drive->mult_count) - drive->special.b.set_multmode = 1; + drive->special_flags |= IDE_SFLAG_SET_MULTMODE; } static void pre_reset(ide_drive_t *drive) @@ -408,8 +407,9 @@ static ide_startstop_t do_reset1(ide_drive_t *drive, int do_not_try_atapi) /* more than enough time */ udelay(10); /* clear SRST, leave nIEN (unless device is on the quirk list) */ - tp_ops->write_devctl(hwif, (drive->quirk_list == 2 ? 0 : ATA_NIEN) | - ATA_DEVCTL_OBS); + tp_ops->write_devctl(hwif, + ((drive->dev_flags & IDE_DFLAG_NIEN_QUIRK) ? 0 : ATA_NIEN) | + ATA_DEVCTL_OBS); /* more than enough time */ udelay(10); hwif->poll_timeout = jiffies + WAIT_WORSTCASE; diff --git a/drivers/ide/ide-gd.c b/drivers/ide/ide-gd.c index 4b6b71e2cdf..214119026b3 100644 --- a/drivers/ide/ide-gd.c +++ b/drivers/ide/ide-gd.c @@ -287,6 +287,19 @@ static int ide_gd_media_changed(struct gendisk *disk) return ret; } +static unsigned long long ide_gd_set_capacity(struct gendisk *disk, + unsigned long long capacity) +{ + struct ide_disk_obj *idkp = ide_drv_g(disk, ide_disk_obj); + ide_drive_t *drive = idkp->drive; + const struct ide_disk_ops *disk_ops = drive->disk_ops; + + if (disk_ops->set_capacity) + return disk_ops->set_capacity(drive, capacity); + + return drive->capacity64; +} + static int ide_gd_revalidate_disk(struct gendisk *disk) { struct ide_disk_obj *idkp = ide_drv_g(disk, ide_disk_obj); @@ -315,6 +328,7 @@ static struct block_device_operations ide_gd_ops = { .locked_ioctl = ide_gd_ioctl, .getgeo = ide_gd_getgeo, .media_changed = ide_gd_media_changed, + .set_capacity = ide_gd_set_capacity, .revalidate_disk = ide_gd_revalidate_disk }; diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c index 7812ca0be13..54d7c4685d2 100644 --- a/drivers/ide/ide-generic.c +++ b/drivers/ide/ide-generic.c @@ -29,6 +29,7 @@ MODULE_PARM_DESC(probe_mask, "probe mask for legacy ISA IDE ports"); static const struct ide_port_info ide_generic_port_info = { .host_flags = IDE_HFLAG_NO_DMA, + .chipset = ide_generic, }; #ifdef CONFIG_ARM @@ -85,7 +86,7 @@ static void ide_generic_check_pci_legacy_iobases(int *primary, int *secondary) static int __init ide_generic_init(void) { - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; unsigned long io_addr; int i, rc = 0, primary = 0, secondary = 0; @@ -132,9 +133,7 @@ static int __init ide_generic_init(void) #else hw.irq = legacy_irqs[i]; #endif - hw.chipset = ide_generic; - - rc = ide_host_add(&ide_generic_port_info, hws, NULL); + rc = ide_host_add(&ide_generic_port_info, hws, 1, NULL); if (rc) { release_region(io_addr + 0x206, 1); release_region(io_addr, 8); diff --git a/drivers/ide/ide-h8300.c b/drivers/ide/ide-h8300.c index c06ebdc4a13..520f42c5445 100644 --- a/drivers/ide/ide-h8300.c +++ b/drivers/ide/ide-h8300.c @@ -64,26 +64,26 @@ static const struct ide_tp_ops h8300_tp_ops = { #define H8300_IDE_GAP (2) -static inline void hw_setup(hw_regs_t *hw) +static inline void hw_setup(struct ide_hw *hw) { int i; - memset(hw, 0, sizeof(hw_regs_t)); + memset(hw, 0, sizeof(*hw)); for (i = 0; i <= 7; i++) hw->io_ports_array[i] = CONFIG_H8300_IDE_BASE + H8300_IDE_GAP*i; hw->io_ports.ctl_addr = CONFIG_H8300_IDE_ALT; hw->irq = EXT_IRQ0 + CONFIG_H8300_IDE_IRQ; - hw->chipset = ide_generic; } static const struct ide_port_info h8300_port_info = { .tp_ops = &h8300_tp_ops, .host_flags = IDE_HFLAG_NO_IO_32BIT | IDE_HFLAG_NO_DMA, + .chipset = ide_generic, }; static int __init h8300_ide_init(void) { - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; printk(KERN_INFO DRV_NAME ": H8/300 generic IDE interface\n"); @@ -96,7 +96,7 @@ static int __init h8300_ide_init(void) hw_setup(&hw); - return ide_host_add(&h8300_port_info, hws, NULL); + return ide_host_add(&h8300_port_info, hws, 1, NULL); out_busy: printk(KERN_ERR "ide-h8300: IDE I/F resource already used.\n"); diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index bba4297f2f0..272cc38f6db 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -184,29 +184,42 @@ static void ide_tf_set_setmult_cmd(ide_drive_t *drive, struct ide_taskfile *tf) tf->command = ATA_CMD_SET_MULTI; } -static ide_startstop_t ide_disk_special(ide_drive_t *drive) +/** + * do_special - issue some special commands + * @drive: drive the command is for + * + * do_special() is used to issue ATA_CMD_INIT_DEV_PARAMS, + * ATA_CMD_RESTORE and ATA_CMD_SET_MULTI commands to a drive. + */ + +static ide_startstop_t do_special(ide_drive_t *drive) { - special_t *s = &drive->special; struct ide_cmd cmd; +#ifdef DEBUG + printk(KERN_DEBUG "%s: %s: 0x%02x\n", drive->name, __func__, + drive->special_flags); +#endif + if (drive->media != ide_disk) { + drive->special_flags = 0; + drive->mult_req = 0; + return ide_stopped; + } + memset(&cmd, 0, sizeof(cmd)); cmd.protocol = ATA_PROT_NODATA; - if (s->b.set_geometry) { - s->b.set_geometry = 0; + if (drive->special_flags & IDE_SFLAG_SET_GEOMETRY) { + drive->special_flags &= ~IDE_SFLAG_SET_GEOMETRY; ide_tf_set_specify_cmd(drive, &cmd.tf); - } else if (s->b.recalibrate) { - s->b.recalibrate = 0; + } else if (drive->special_flags & IDE_SFLAG_RECALIBRATE) { + drive->special_flags &= ~IDE_SFLAG_RECALIBRATE; ide_tf_set_restore_cmd(drive, &cmd.tf); - } else if (s->b.set_multmode) { - s->b.set_multmode = 0; + } else if (drive->special_flags & IDE_SFLAG_SET_MULTMODE) { + drive->special_flags &= ~IDE_SFLAG_SET_MULTMODE; ide_tf_set_setmult_cmd(drive, &cmd.tf); - } else if (s->all) { - int special = s->all; - s->all = 0; - printk(KERN_ERR "%s: bad special flag: 0x%02x\n", drive->name, special); - return ide_stopped; - } + } else + BUG(); cmd.valid.out.tf = IDE_VALID_OUT_TF | IDE_VALID_DEVICE; cmd.valid.in.tf = IDE_VALID_IN_TF | IDE_VALID_DEVICE; @@ -217,31 +230,6 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive) return ide_started; } -/** - * do_special - issue some special commands - * @drive: drive the command is for - * - * do_special() is used to issue ATA_CMD_INIT_DEV_PARAMS, - * ATA_CMD_RESTORE and ATA_CMD_SET_MULTI commands to a drive. - * - * It used to do much more, but has been scaled back. - */ - -static ide_startstop_t do_special (ide_drive_t *drive) -{ - special_t *s = &drive->special; - -#ifdef DEBUG - printk("%s: do_special: 0x%02x\n", drive->name, s->all); -#endif - if (drive->media == ide_disk) - return ide_disk_special(drive); - - s->all = 0; - drive->mult_req = 0; - return ide_stopped; -} - void ide_map_sg(ide_drive_t *drive, struct ide_cmd *cmd) { ide_hwif_t *hwif = drive->hwif; @@ -351,7 +339,8 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq) printk(KERN_ERR "%s: drive not ready for command\n", drive->name); return startstop; } - if (!drive->special.all) { + + if (drive->special_flags == 0) { struct ide_driver *drv; /* @@ -499,11 +488,15 @@ repeat: if ((hwif->host->host_flags & IDE_HFLAG_SERIALIZE) && hwif != prev_port) { + ide_drive_t *cur_dev = + prev_port ? prev_port->cur_dev : NULL; + /* * set nIEN for previous port, drives in the - * quirk_list may not like intr setups/cleanups + * quirk list may not like intr setups/cleanups */ - if (prev_port && prev_port->cur_dev->quirk_list == 0) + if (cur_dev && + (cur_dev->dev_flags & IDE_DFLAG_NIEN_QUIRK) == 0) prev_port->tp_ops->write_devctl(prev_port, ATA_NIEN | ATA_DEVCTL_OBS); diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c index 06fe002116e..fa047150a1c 100644 --- a/drivers/ide/ide-iops.c +++ b/drivers/ide/ide-iops.c @@ -282,6 +282,29 @@ no_80w: return 0; } +static const char *nien_quirk_list[] = { + "QUANTUM FIREBALLlct08 08", + "QUANTUM FIREBALLP KA6.4", + "QUANTUM FIREBALLP KA9.1", + "QUANTUM FIREBALLP KX13.6", + "QUANTUM FIREBALLP KX20.5", + "QUANTUM FIREBALLP KX27.3", + "QUANTUM FIREBALLP LM20.4", + "QUANTUM FIREBALLP LM20.5", + NULL +}; + +void ide_check_nien_quirk_list(ide_drive_t *drive) +{ + const char **list, *m = (char *)&drive->id[ATA_ID_PROD]; + + for (list = nien_quirk_list; *list != NULL; list++) + if (strstr(m, *list) != NULL) { + drive->dev_flags |= IDE_DFLAG_NIEN_QUIRK; + return; + } +} + int ide_driveid_update(ide_drive_t *drive) { u16 *id; @@ -311,7 +334,6 @@ int ide_driveid_update(ide_drive_t *drive) return 1; out_err: - SELECT_MASK(drive, 0); if (rc == 2) printk(KERN_ERR "%s: %s: bad status\n", drive->name, __func__); kfree(id); @@ -365,7 +387,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed) tp_ops->exec_command(hwif, ATA_CMD_SET_FEATURES); - if (drive->quirk_list == 2) + if (drive->dev_flags & IDE_DFLAG_NIEN_QUIRK) tp_ops->write_devctl(hwif, ATA_DEVCTL_OBS); error = __ide_wait_stat(drive, drive->ready_stat, diff --git a/drivers/ide/ide-legacy.c b/drivers/ide/ide-legacy.c index 8c5dcbf2254..b9654a7bb7b 100644 --- a/drivers/ide/ide-legacy.c +++ b/drivers/ide/ide-legacy.c @@ -1,7 +1,7 @@ #include <linux/kernel.h> #include <linux/ide.h> -static void ide_legacy_init_one(hw_regs_t **hws, hw_regs_t *hw, +static void ide_legacy_init_one(struct ide_hw **hws, struct ide_hw *hw, u8 port_no, const struct ide_port_info *d, unsigned long config) { @@ -33,7 +33,6 @@ static void ide_legacy_init_one(hw_regs_t **hws, hw_regs_t *hw, ide_std_init_ports(hw, base, ctl); hw->irq = irq; - hw->chipset = d->chipset; hw->config = config; hws[port_no] = hw; @@ -41,7 +40,7 @@ static void ide_legacy_init_one(hw_regs_t **hws, hw_regs_t *hw, int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config) { - hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[2], *hws[] = { NULL, NULL }; memset(&hw, 0, sizeof(hw)); @@ -53,6 +52,6 @@ int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config) (d->host_flags & IDE_HFLAG_SINGLE)) return -ENOENT; - return ide_host_add(d, hws, NULL); + return ide_host_add(d, hws, 2, NULL); } EXPORT_SYMBOL_GPL(ide_legacy_device_add); diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c index 6e80b774e88..017b1df3b80 100644 --- a/drivers/ide/ide-pnp.c +++ b/drivers/ide/ide-pnp.c @@ -29,6 +29,7 @@ static struct pnp_device_id idepnp_devices[] = { static const struct ide_port_info ide_pnp_port_info = { .host_flags = IDE_HFLAG_NO_DMA, + .chipset = ide_generic, }; static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) @@ -36,7 +37,7 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) struct ide_host *host; unsigned long base, ctl; int rc; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; printk(KERN_INFO DRV_NAME ": generic PnP IDE interface\n"); @@ -62,9 +63,8 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) memset(&hw, 0, sizeof(hw)); ide_std_init_ports(&hw, base, ctl); hw.irq = pnp_irq(dev, 0); - hw.chipset = ide_generic; - rc = ide_host_add(&ide_pnp_port_info, hws, &host); + rc = ide_host_add(&ide_pnp_port_info, hws, 1, &host); if (rc) goto out; diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index c895ed52b2e..f371b0de314 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -97,7 +97,7 @@ static void ide_disk_init_mult_count(ide_drive_t *drive) drive->mult_req = id[ATA_ID_MULTSECT] & 0xff; if (drive->mult_req) - drive->special.b.set_multmode = 1; + drive->special_flags |= IDE_SFLAG_SET_MULTMODE; } } @@ -465,23 +465,8 @@ static u8 probe_for_drive(ide_drive_t *drive) int rc; u8 cmd; - /* - * In order to keep things simple we have an id - * block for all drives at all times. If the device - * is pre ATA or refuses ATA/ATAPI identify we - * will add faked data to this. - * - * Also note that 0 everywhere means "can't do X" - */ - drive->dev_flags &= ~IDE_DFLAG_ID_READ; - drive->id = kzalloc(SECTOR_SIZE, GFP_KERNEL); - if (drive->id == NULL) { - printk(KERN_ERR "ide: out of memory for id data.\n"); - return 0; - } - m = (char *)&drive->id[ATA_ID_PROD]; strcpy(m, "UNKNOWN"); @@ -497,7 +482,7 @@ static u8 probe_for_drive(ide_drive_t *drive) } if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0) - goto out_free; + return 0; /* identification failed? */ if ((drive->dev_flags & IDE_DFLAG_ID_READ) == 0) { @@ -521,7 +506,7 @@ static u8 probe_for_drive(ide_drive_t *drive) } if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0) - goto out_free; + return 0; /* The drive wasn't being helpful. Add generic info only */ if ((drive->dev_flags & IDE_DFLAG_ID_READ) == 0) { @@ -535,9 +520,6 @@ static u8 probe_for_drive(ide_drive_t *drive) } return 1; -out_free: - kfree(drive->id); - return 0; } static void hwif_release_dev(struct device *dev) @@ -702,8 +684,14 @@ static int ide_probe_port(ide_hwif_t *hwif) if (irqd) disable_irq(hwif->irq); - if (ide_port_wait_ready(hwif) == -EBUSY) - printk(KERN_DEBUG "%s: Wait for ready failed before probe !\n", hwif->name); + rc = ide_port_wait_ready(hwif); + if (rc == -ENODEV) { + printk(KERN_INFO "%s: no devices on the port\n", hwif->name); + goto out; + } else if (rc == -EBUSY) + printk(KERN_ERR "%s: not ready before the probe\n", hwif->name); + else + rc = -ENODEV; /* * Second drive should only exist if first drive was found, @@ -714,7 +702,7 @@ static int ide_probe_port(ide_hwif_t *hwif) if (drive->dev_flags & IDE_DFLAG_PRESENT) rc = 0; } - +out: /* * Use cached IRQ number. It might be (and is...) changed by probe * code above @@ -732,6 +720,8 @@ static void ide_port_tune_devices(ide_hwif_t *hwif) int i; ide_port_for_each_present_dev(i, drive, hwif) { + ide_check_nien_quirk_list(drive); + if (port_ops && port_ops->quirkproc) port_ops->quirkproc(drive); } @@ -817,8 +807,6 @@ static int ide_port_setup_devices(ide_hwif_t *hwif) if (ide_init_queue(drive)) { printk(KERN_ERR "ide: failed to init %s\n", drive->name); - kfree(drive->id); - drive->id = NULL; drive->dev_flags &= ~IDE_DFLAG_PRESENT; continue; } @@ -947,9 +935,6 @@ static void drive_release_dev (struct device *dev) blk_cleanup_queue(drive->queue); drive->queue = NULL; - kfree(drive->id); - drive->id = NULL; - drive->dev_flags &= ~IDE_DFLAG_PRESENT; complete(&drive->gendev_rel_comp); @@ -1035,6 +1020,15 @@ static void ide_port_init_devices(ide_hwif_t *hwif) if (port_ops && port_ops->init_dev) port_ops->init_dev(drive); } + + ide_port_for_each_dev(i, drive, hwif) { + /* + * default to PIO Mode 0 before we figure out + * the most suited mode for the attached device + */ + if (port_ops && port_ops->set_pio_mode) + port_ops->set_pio_mode(drive, 0); + } } static void ide_init_port(ide_hwif_t *hwif, unsigned int port, @@ -1042,8 +1036,7 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port, { hwif->channel = port; - if (d->chipset) - hwif->chipset = d->chipset; + hwif->chipset = d->chipset ? d->chipset : ide_pci; if (d->init_iops) d->init_iops(hwif); @@ -1124,16 +1117,19 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif) ide_port_for_each_dev(i, drive, hwif) { u8 j = (hwif->index * MAX_DRIVES) + i; + u16 *saved_id = drive->id; memset(drive, 0, sizeof(*drive)); + memset(saved_id, 0, SECTOR_SIZE); + drive->id = saved_id; drive->media = ide_disk; drive->select = (i << 4) | ATA_DEVICE_OBS; drive->hwif = hwif; drive->ready_stat = ATA_DRDY; drive->bad_wstat = BAD_W_STAT; - drive->special.b.recalibrate = 1; - drive->special.b.set_geometry = 1; + drive->special_flags = IDE_SFLAG_RECALIBRATE | + IDE_SFLAG_SET_GEOMETRY; drive->name[0] = 'h'; drive->name[1] = 'd'; drive->name[2] = 'a' + j; @@ -1168,11 +1164,10 @@ static void ide_init_port_data(ide_hwif_t *hwif, unsigned int index) ide_port_init_devices_data(hwif); } -static void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw) +static void ide_init_port_hw(ide_hwif_t *hwif, struct ide_hw *hw) { memcpy(&hwif->io_ports, &hw->io_ports, sizeof(hwif->io_ports)); hwif->irq = hw->irq; - hwif->chipset = hw->chipset; hwif->dev = hw->dev; hwif->gendev.parent = hw->parent ? hw->parent : hw->dev; hwif->ack_intr = hw->ack_intr; @@ -1233,8 +1228,10 @@ static void ide_port_free_devices(ide_hwif_t *hwif) ide_drive_t *drive; int i; - ide_port_for_each_dev(i, drive, hwif) + ide_port_for_each_dev(i, drive, hwif) { + kfree(drive->id); kfree(drive); + } } static int ide_port_alloc_devices(ide_hwif_t *hwif, int node) @@ -1248,6 +1245,18 @@ static int ide_port_alloc_devices(ide_hwif_t *hwif, int node) if (drive == NULL) goto out_nomem; + /* + * In order to keep things simple we have an id + * block for all drives at all times. If the device + * is pre ATA or refuses ATA/ATAPI identify we + * will add faked data to this. + * + * Also note that 0 everywhere means "can't do X" + */ + drive->id = kzalloc_node(SECTOR_SIZE, GFP_KERNEL, node); + if (drive->id == NULL) + goto out_nomem; + hwif->devices[i] = drive; } return 0; @@ -1257,7 +1266,8 @@ out_nomem: return -ENOMEM; } -struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws) +struct ide_host *ide_host_alloc(const struct ide_port_info *d, + struct ide_hw **hws, unsigned int n_ports) { struct ide_host *host; struct device *dev = hws[0] ? hws[0]->dev : NULL; @@ -1268,7 +1278,7 @@ struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws) if (host == NULL) return NULL; - for (i = 0; i < MAX_HOST_PORTS; i++) { + for (i = 0; i < n_ports; i++) { ide_hwif_t *hwif; int idx; @@ -1288,6 +1298,7 @@ struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws) if (idx < 0) { printk(KERN_ERR "%s: no free slot for interface\n", d ? d->name : "ide"); + ide_port_free_devices(hwif); kfree(hwif); continue; } @@ -1344,7 +1355,7 @@ static void ide_disable_port(ide_hwif_t *hwif) } int ide_host_register(struct ide_host *host, const struct ide_port_info *d, - hw_regs_t **hws) + struct ide_hw **hws) { ide_hwif_t *hwif, *mate = NULL; int i, j = 0; @@ -1438,13 +1449,13 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d, } EXPORT_SYMBOL_GPL(ide_host_register); -int ide_host_add(const struct ide_port_info *d, hw_regs_t **hws, - struct ide_host **hostp) +int ide_host_add(const struct ide_port_info *d, struct ide_hw **hws, + unsigned int n_ports, struct ide_host **hostp) { struct ide_host *host; int rc; - host = ide_host_alloc(d, hws); + host = ide_host_alloc(d, hws, n_ports); if (host == NULL) return -ENOMEM; diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index d9764f0bc82..4b447a8a49d 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -240,18 +240,27 @@ static struct class *idetape_sysfs_class; static void ide_tape_release(struct device *); -static struct ide_tape_obj *ide_tape_get(struct gendisk *disk) +static struct ide_tape_obj *idetape_devs[MAX_HWIFS * MAX_DRIVES]; + +static struct ide_tape_obj *ide_tape_get(struct gendisk *disk, bool cdev, + unsigned int i) { struct ide_tape_obj *tape = NULL; mutex_lock(&idetape_ref_mutex); - tape = ide_drv_g(disk, ide_tape_obj); + + if (cdev) + tape = idetape_devs[i]; + else + tape = ide_drv_g(disk, ide_tape_obj); + if (tape) { if (ide_device_get(tape->drive)) tape = NULL; else get_device(&tape->dev); } + mutex_unlock(&idetape_ref_mutex); return tape; } @@ -267,24 +276,6 @@ static void ide_tape_put(struct ide_tape_obj *tape) } /* - * The variables below are used for the character device interface. Additional - * state variables are defined in our ide_drive_t structure. - */ -static struct ide_tape_obj *idetape_devs[MAX_HWIFS * MAX_DRIVES]; - -static struct ide_tape_obj *ide_tape_chrdev_get(unsigned int i) -{ - struct ide_tape_obj *tape = NULL; - - mutex_lock(&idetape_ref_mutex); - tape = idetape_devs[i]; - if (tape) - get_device(&tape->dev); - mutex_unlock(&idetape_ref_mutex); - return tape; -} - -/* * called on each failed packet command retry to analyze the request sense. We * currently do not utilize this information. */ @@ -397,7 +388,8 @@ static int ide_tape_callback(ide_drive_t *drive, int dsc) if (readpos[0] & 0x4) { printk(KERN_INFO "ide-tape: Block location is unknown" "to the tape\n"); - clear_bit(IDE_AFLAG_ADDRESS_VALID, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_ADDRESS_VALID), + &drive->atapi_flags); uptodate = 0; err = IDE_DRV_ERROR_GENERAL; } else { @@ -406,7 +398,8 @@ static int ide_tape_callback(ide_drive_t *drive, int dsc) tape->partition = readpos[1]; tape->first_frame = be32_to_cpup((__be32 *)&readpos[4]); - set_bit(IDE_AFLAG_ADDRESS_VALID, &drive->atapi_flags); + set_bit(ilog2(IDE_AFLAG_ADDRESS_VALID), + &drive->atapi_flags); } } @@ -656,15 +649,15 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive, if ((drive->dev_flags & IDE_DFLAG_DSC_OVERLAP) == 0 && (rq->cmd[13] & REQ_IDETAPE_PC2) == 0) - set_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags); + drive->atapi_flags |= IDE_AFLAG_IGNORE_DSC; if (drive->dev_flags & IDE_DFLAG_POST_RESET) { - set_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags); + drive->atapi_flags |= IDE_AFLAG_IGNORE_DSC; drive->dev_flags &= ~IDE_DFLAG_POST_RESET; } - if (!test_and_clear_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags) && - (stat & ATA_DSC) == 0) { + if (!(drive->atapi_flags & IDE_AFLAG_IGNORE_DSC) && + !(stat & ATA_DSC)) { if (postponed_rq == NULL) { tape->dsc_polling_start = jiffies; tape->dsc_poll_freq = tape->best_dsc_rw_freq; @@ -684,7 +677,9 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive, tape->dsc_poll_freq = IDETAPE_DSC_MA_SLOW; idetape_postpone_request(drive); return ide_stopped; - } + } else + drive->atapi_flags &= ~IDE_AFLAG_IGNORE_DSC; + if (rq->cmd[13] & REQ_IDETAPE_READ) { pc = &tape->queued_pc; ide_tape_create_rw_cmd(tape, pc, rq, READ_6); @@ -744,7 +739,7 @@ static int idetape_wait_ready(ide_drive_t *drive, unsigned long timeout) int load_attempted = 0; /* Wait for the tape to become ready */ - set_bit(IDE_AFLAG_MEDIUM_PRESENT, &drive->atapi_flags); + set_bit(ilog2(IDE_AFLAG_MEDIUM_PRESENT), &drive->atapi_flags); timeout += jiffies; while (time_before(jiffies, timeout)) { if (ide_do_test_unit_ready(drive, disk) == 0) @@ -820,7 +815,7 @@ static void __ide_tape_discard_merge_buffer(ide_drive_t *drive) if (tape->chrdev_dir != IDETAPE_DIR_READ) return; - clear_bit(IDE_AFLAG_FILEMARK, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_FILEMARK), &drive->atapi_flags); tape->valid = 0; if (tape->buf != NULL) { kfree(tape->buf); @@ -1113,7 +1108,8 @@ static int idetape_space_over_filemarks(ide_drive_t *drive, short mt_op, if (tape->chrdev_dir == IDETAPE_DIR_READ) { tape->valid = 0; - if (test_and_clear_bit(IDE_AFLAG_FILEMARK, &drive->atapi_flags)) + if (test_and_clear_bit(ilog2(IDE_AFLAG_FILEMARK), + &drive->atapi_flags)) ++count; ide_tape_discard_merge_buffer(drive, 0); } @@ -1168,7 +1164,7 @@ static ssize_t idetape_chrdev_read(struct file *file, char __user *buf, debug_log(DBG_CHRDEV, "Enter %s, count %Zd\n", __func__, count); if (tape->chrdev_dir != IDETAPE_DIR_READ) { - if (test_bit(IDE_AFLAG_DETECT_BS, &drive->atapi_flags)) + if (test_bit(ilog2(IDE_AFLAG_DETECT_BS), &drive->atapi_flags)) if (count > tape->blk_size && (count % tape->blk_size) == 0) tape->user_bs_factor = count / tape->blk_size; @@ -1184,7 +1180,8 @@ static ssize_t idetape_chrdev_read(struct file *file, char __user *buf, /* refill if staging buffer is empty */ if (!tape->valid) { /* If we are at a filemark, nothing more to read */ - if (test_bit(IDE_AFLAG_FILEMARK, &drive->atapi_flags)) + if (test_bit(ilog2(IDE_AFLAG_FILEMARK), + &drive->atapi_flags)) break; /* read */ if (idetape_queue_rw_tail(drive, REQ_IDETAPE_READ, @@ -1202,7 +1199,7 @@ static ssize_t idetape_chrdev_read(struct file *file, char __user *buf, done += todo; } - if (!done && test_bit(IDE_AFLAG_FILEMARK, &drive->atapi_flags)) { + if (!done && test_bit(ilog2(IDE_AFLAG_FILEMARK), &drive->atapi_flags)) { debug_log(DBG_SENSE, "%s: spacing over filemark\n", tape->name); idetape_space_over_filemarks(drive, MTFSF, 1); @@ -1336,7 +1333,8 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count) ide_tape_discard_merge_buffer(drive, 0); retval = ide_do_start_stop(drive, disk, !IDETAPE_LU_LOAD_MASK); if (!retval) - clear_bit(IDE_AFLAG_MEDIUM_PRESENT, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_MEDIUM_PRESENT), + &drive->atapi_flags); return retval; case MTNOP: ide_tape_discard_merge_buffer(drive, 0); @@ -1358,9 +1356,11 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count) mt_count % tape->blk_size) return -EIO; tape->user_bs_factor = mt_count / tape->blk_size; - clear_bit(IDE_AFLAG_DETECT_BS, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_DETECT_BS), + &drive->atapi_flags); } else - set_bit(IDE_AFLAG_DETECT_BS, &drive->atapi_flags); + set_bit(ilog2(IDE_AFLAG_DETECT_BS), + &drive->atapi_flags); return 0; case MTSEEK: ide_tape_discard_merge_buffer(drive, 0); @@ -1486,7 +1486,7 @@ static int idetape_chrdev_open(struct inode *inode, struct file *filp) return -ENXIO; lock_kernel(); - tape = ide_tape_chrdev_get(i); + tape = ide_tape_get(NULL, true, i); if (!tape) { unlock_kernel(); return -ENXIO; @@ -1505,20 +1505,20 @@ static int idetape_chrdev_open(struct inode *inode, struct file *filp) filp->private_data = tape; - if (test_and_set_bit(IDE_AFLAG_BUSY, &drive->atapi_flags)) { + if (test_and_set_bit(ilog2(IDE_AFLAG_BUSY), &drive->atapi_flags)) { retval = -EBUSY; goto out_put_tape; } retval = idetape_wait_ready(drive, 60 * HZ); if (retval) { - clear_bit(IDE_AFLAG_BUSY, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_BUSY), &drive->atapi_flags); printk(KERN_ERR "ide-tape: %s: drive not ready\n", tape->name); goto out_put_tape; } idetape_read_position(drive); - if (!test_bit(IDE_AFLAG_ADDRESS_VALID, &drive->atapi_flags)) + if (!test_bit(ilog2(IDE_AFLAG_ADDRESS_VALID), &drive->atapi_flags)) (void)idetape_rewind_tape(drive); /* Read block size and write protect status from drive. */ @@ -1534,7 +1534,7 @@ static int idetape_chrdev_open(struct inode *inode, struct file *filp) if (tape->write_prot) { if ((filp->f_flags & O_ACCMODE) == O_WRONLY || (filp->f_flags & O_ACCMODE) == O_RDWR) { - clear_bit(IDE_AFLAG_BUSY, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_BUSY), &drive->atapi_flags); retval = -EROFS; goto out_put_tape; } @@ -1591,15 +1591,17 @@ static int idetape_chrdev_release(struct inode *inode, struct file *filp) ide_tape_discard_merge_buffer(drive, 1); } - if (minor < 128 && test_bit(IDE_AFLAG_MEDIUM_PRESENT, &drive->atapi_flags)) + if (minor < 128 && test_bit(ilog2(IDE_AFLAG_MEDIUM_PRESENT), + &drive->atapi_flags)) (void) idetape_rewind_tape(drive); + if (tape->chrdev_dir == IDETAPE_DIR_NONE) { if (tape->door_locked == DOOR_LOCKED) { if (!ide_set_media_lock(drive, tape->disk, 0)) tape->door_locked = DOOR_UNLOCKED; } } - clear_bit(IDE_AFLAG_BUSY, &drive->atapi_flags); + clear_bit(ilog2(IDE_AFLAG_BUSY), &drive->atapi_flags); ide_tape_put(tape); unlock_kernel(); return 0; @@ -1905,7 +1907,7 @@ static const struct file_operations idetape_fops = { static int idetape_open(struct block_device *bdev, fmode_t mode) { - struct ide_tape_obj *tape = ide_tape_get(bdev->bd_disk); + struct ide_tape_obj *tape = ide_tape_get(bdev->bd_disk, false, 0); if (!tape) return -ENXIO; diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c index a0c3e1b2f73..75b85a8cd2d 100644 --- a/drivers/ide/ide-taskfile.c +++ b/drivers/ide/ide-taskfile.c @@ -98,7 +98,6 @@ ide_startstop_t do_rw_taskfile(ide_drive_t *drive, struct ide_cmd *orig_cmd) if ((cmd->tf_flags & IDE_TFLAG_DMA_PIO_FALLBACK) == 0) { ide_tf_dump(drive->name, cmd); tp_ops->write_devctl(hwif, ATA_DEVCTL_OBS); - SELECT_MASK(drive, 0); if (cmd->ftf_flags & IDE_FTFLAG_OUT_DATA) { u8 data[2] = { cmd->tf.data, cmd->hob.data }; @@ -166,7 +165,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive) if (!OK_STAT(stat, ATA_DRDY, BAD_STAT)) { if (custom && tf->command == ATA_CMD_SET_MULTI) { drive->mult_req = drive->mult_count = 0; - drive->special.b.recalibrate = 1; + drive->special_flags |= IDE_SFLAG_RECALIBRATE; (void)ide_dump_status(drive, __func__, stat); return ide_stopped; } else if (custom && tf->command == ATA_CMD_INIT_DEV_PARAMS) { diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 92c9b90931e..16d056939f9 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -211,6 +211,11 @@ static unsigned int ide_noflush; module_param_call(noflush, ide_set_dev_param_mask, NULL, &ide_noflush, 0); MODULE_PARM_DESC(noflush, "disable flush requests for a device"); +static unsigned int ide_nohpa; + +module_param_call(nohpa, ide_set_dev_param_mask, NULL, &ide_nohpa, 0); +MODULE_PARM_DESC(nohpa, "disable Host Protected Area for a device"); + static unsigned int ide_noprobe; module_param_call(noprobe, ide_set_dev_param_mask, NULL, &ide_noprobe, 0); @@ -281,6 +286,11 @@ static void ide_dev_apply_params(ide_drive_t *drive, u8 unit) drive->name); drive->dev_flags |= IDE_DFLAG_NOFLUSH; } + if (ide_nohpa & (1 << i)) { + printk(KERN_INFO "ide: disabling Host Protected Area for %s\n", + drive->name); + drive->dev_flags |= IDE_DFLAG_NOHPA; + } if (ide_noprobe & (1 << i)) { printk(KERN_INFO "ide: skipping probe for %s\n", drive->name); drive->dev_flags |= IDE_DFLAG_NOPROBE; diff --git a/drivers/ide/ide_platform.c b/drivers/ide/ide_platform.c index 051b4ab0f35..ee9b55ecc62 100644 --- a/drivers/ide/ide_platform.c +++ b/drivers/ide/ide_platform.c @@ -21,7 +21,7 @@ #include <linux/platform_device.h> #include <linux/io.h> -static void __devinit plat_ide_setup_ports(hw_regs_t *hw, +static void __devinit plat_ide_setup_ports(struct ide_hw *hw, void __iomem *base, void __iomem *ctrl, struct pata_platform_info *pdata, @@ -40,12 +40,11 @@ static void __devinit plat_ide_setup_ports(hw_regs_t *hw, hw->io_ports.ctl_addr = (unsigned long)ctrl; hw->irq = irq; - - hw->chipset = ide_generic; } static const struct ide_port_info platform_ide_port_info = { .host_flags = IDE_HFLAG_NO_DMA, + .chipset = ide_generic, }; static int __devinit plat_ide_probe(struct platform_device *pdev) @@ -55,7 +54,7 @@ static int __devinit plat_ide_probe(struct platform_device *pdev) struct pata_platform_info *pdata; struct ide_host *host; int ret = 0, mmio = 0; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; struct ide_port_info d = platform_ide_port_info; pdata = pdev->dev.platform_data; @@ -99,7 +98,7 @@ static int __devinit plat_ide_probe(struct platform_device *pdev) if (mmio) d.host_flags |= IDE_HFLAG_MMIO; - ret = ide_host_add(&d, hws, &host); + ret = ide_host_add(&d, hws, 1, &host); if (ret) goto out; diff --git a/drivers/ide/macide.c b/drivers/ide/macide.c index 4b1718e8328..1447c8c9056 100644 --- a/drivers/ide/macide.c +++ b/drivers/ide/macide.c @@ -62,7 +62,7 @@ int macide_ack_intr(ide_hwif_t* hwif) return 0; } -static void __init macide_setup_ports(hw_regs_t *hw, unsigned long base, +static void __init macide_setup_ports(struct ide_hw *hw, unsigned long base, int irq, ide_ack_intr_t *ack_intr) { int i; @@ -76,13 +76,12 @@ static void __init macide_setup_ports(hw_regs_t *hw, unsigned long base, hw->irq = irq; hw->ack_intr = ack_intr; - - hw->chipset = ide_generic; } static const struct ide_port_info macide_port_info = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_generic, }; static const char *mac_ide_name[] = @@ -97,7 +96,7 @@ static int __init macide_init(void) ide_ack_intr_t *ack_intr; unsigned long base; int irq; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; if (!MACH_IS_MAC) return -ENODEV; @@ -127,7 +126,7 @@ static int __init macide_init(void) macide_setup_ports(&hw, base, irq, ack_intr); - return ide_host_add(&macide_port_info, hws, NULL); + return ide_host_add(&macide_port_info, hws, 1, NULL); } module_init(macide_init); diff --git a/drivers/ide/palm_bk3710.c b/drivers/ide/palm_bk3710.c index 09d813d313f..3c1dc015215 100644 --- a/drivers/ide/palm_bk3710.c +++ b/drivers/ide/palm_bk3710.c @@ -306,6 +306,7 @@ static struct ide_port_info __devinitdata palm_bk3710_port_info = { .host_flags = IDE_HFLAG_MMIO, .pio_mask = ATA_PIO4, .mwdma_mask = ATA_MWDMA2, + .chipset = ide_palm3710, }; static int __init palm_bk3710_probe(struct platform_device *pdev) @@ -315,7 +316,7 @@ static int __init palm_bk3710_probe(struct platform_device *pdev) void __iomem *base; unsigned long rate, mem_size; int i, rc; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; clk = clk_get(&pdev->dev, "IDECLK"); if (IS_ERR(clk)) @@ -363,13 +364,12 @@ static int __init palm_bk3710_probe(struct platform_device *pdev) (base + IDE_PALM_ATA_PRI_CTL_OFFSET); hw.irq = irq->start; hw.dev = &pdev->dev; - hw.chipset = ide_palm3710; palm_bk3710_port_info.udma_mask = rate < 100000000 ? ATA_UDMA4 : ATA_UDMA5; /* Register the IDE interface with Linux */ - rc = ide_host_add(&palm_bk3710_port_info, hws, NULL); + rc = ide_host_add(&palm_bk3710_port_info, hws, 1, NULL); if (rc) goto out; diff --git a/drivers/ide/pdc202xx_new.c b/drivers/ide/pdc202xx_new.c index b68906c3c17..65ba8239e7b 100644 --- a/drivers/ide/pdc202xx_new.c +++ b/drivers/ide/pdc202xx_new.c @@ -40,18 +40,6 @@ #define DBG(fmt, args...) #endif -static const char *pdc_quirk_drives[] = { - "QUANTUM FIREBALLlct08 08", - "QUANTUM FIREBALLP KA6.4", - "QUANTUM FIREBALLP KA9.1", - "QUANTUM FIREBALLP LM20.4", - "QUANTUM FIREBALLP KX13.6", - "QUANTUM FIREBALLP KX20.5", - "QUANTUM FIREBALLP KX27.3", - "QUANTUM FIREBALLP LM20.5", - NULL -}; - static u8 max_dma_rate(struct pci_dev *pdev) { u8 mode; @@ -200,19 +188,6 @@ static u8 pdcnew_cable_detect(ide_hwif_t *hwif) return ATA_CBL_PATA80; } -static void pdcnew_quirkproc(ide_drive_t *drive) -{ - const char **list, *m = (char *)&drive->id[ATA_ID_PROD]; - - for (list = pdc_quirk_drives; *list != NULL; list++) - if (strstr(m, *list) != NULL) { - drive->quirk_list = 2; - return; - } - - drive->quirk_list = 0; -} - static void pdcnew_reset(ide_drive_t *drive) { /* @@ -473,7 +448,6 @@ static struct pci_dev * __devinit pdc20270_get_dev2(struct pci_dev *dev) static const struct ide_port_ops pdcnew_port_ops = { .set_pio_mode = pdcnew_set_pio_mode, .set_dma_mode = pdcnew_set_dma_mode, - .quirkproc = pdcnew_quirkproc, .resetproc = pdcnew_reset, .cable_detect = pdcnew_cable_detect, }; diff --git a/drivers/ide/pdc202xx_old.c b/drivers/ide/pdc202xx_old.c index e24ecc87a9b..b6abf7e52ca 100644 --- a/drivers/ide/pdc202xx_old.c +++ b/drivers/ide/pdc202xx_old.c @@ -23,18 +23,6 @@ #define PDC202XX_DEBUG_DRIVE_INFO 0 -static const char *pdc_quirk_drives[] = { - "QUANTUM FIREBALLlct08 08", - "QUANTUM FIREBALLP KA6.4", - "QUANTUM FIREBALLP KA9.1", - "QUANTUM FIREBALLP LM20.4", - "QUANTUM FIREBALLP KX13.6", - "QUANTUM FIREBALLP KX20.5", - "QUANTUM FIREBALLP KX27.3", - "QUANTUM FIREBALLP LM20.5", - NULL -}; - static void pdc_old_disable_66MHz_clock(ide_hwif_t *); static void pdc202xx_set_mode(ide_drive_t *drive, const u8 speed) @@ -151,19 +139,6 @@ static void pdc_old_disable_66MHz_clock(ide_hwif_t *hwif) outb(clock & ~(hwif->channel ? 0x08 : 0x02), clock_reg); } -static void pdc202xx_quirkproc(ide_drive_t *drive) -{ - const char **list, *m = (char *)&drive->id[ATA_ID_PROD]; - - for (list = pdc_quirk_drives; *list != NULL; list++) - if (strstr(m, *list) != NULL) { - drive->quirk_list = 2; - return; - } - - drive->quirk_list = 0; -} - static void pdc202xx_dma_start(ide_drive_t *drive) { if (drive->current_speed > XFER_UDMA_2) @@ -203,52 +178,6 @@ static int pdc202xx_dma_end(ide_drive_t *drive) return ide_dma_end(drive); } -static int pdc202xx_dma_test_irq(ide_drive_t *drive) -{ - ide_hwif_t *hwif = drive->hwif; - unsigned long high_16 = hwif->extra_base - 16; - u8 dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS); - u8 sc1d = inb(high_16 + 0x001d); - - if (hwif->channel) { - /* bit7: Error, bit6: Interrupting, bit5: FIFO Full, bit4: FIFO Empty */ - if ((sc1d & 0x50) == 0x50) - goto somebody_else; - else if ((sc1d & 0x40) == 0x40) - return (dma_stat & 4) == 4; - } else { - /* bit3: Error, bit2: Interrupting, bit1: FIFO Full, bit0: FIFO Empty */ - if ((sc1d & 0x05) == 0x05) - goto somebody_else; - else if ((sc1d & 0x04) == 0x04) - return (dma_stat & 4) == 4; - } -somebody_else: - return (dma_stat & 4) == 4; /* return 1 if INTR asserted */ -} - -static void pdc202xx_reset(ide_drive_t *drive) -{ - ide_hwif_t *hwif = drive->hwif; - unsigned long high_16 = hwif->extra_base - 16; - u8 udma_speed_flag = inb(high_16 | 0x001f); - - printk(KERN_WARNING "PDC202xx: software reset...\n"); - - outb(udma_speed_flag | 0x10, high_16 | 0x001f); - mdelay(100); - outb(udma_speed_flag & ~0x10, high_16 | 0x001f); - mdelay(2000); /* 2 seconds ?! */ - - ide_set_max_pio(drive); -} - -static void pdc202xx_dma_lost_irq(ide_drive_t *drive) -{ - pdc202xx_reset(drive); - ide_dma_lost_irq(drive); -} - static int init_chipset_pdc202xx(struct pci_dev *dev) { unsigned long dmabase = pci_resource_start(dev, 4); @@ -302,37 +231,22 @@ static void __devinit pdc202ata4_fixup_irq(struct pci_dev *dev, static const struct ide_port_ops pdc20246_port_ops = { .set_pio_mode = pdc202xx_set_pio_mode, .set_dma_mode = pdc202xx_set_mode, - .quirkproc = pdc202xx_quirkproc, }; static const struct ide_port_ops pdc2026x_port_ops = { .set_pio_mode = pdc202xx_set_pio_mode, .set_dma_mode = pdc202xx_set_mode, - .quirkproc = pdc202xx_quirkproc, - .resetproc = pdc202xx_reset, .cable_detect = pdc2026x_cable_detect, }; -static const struct ide_dma_ops pdc20246_dma_ops = { - .dma_host_set = ide_dma_host_set, - .dma_setup = ide_dma_setup, - .dma_start = ide_dma_start, - .dma_end = ide_dma_end, - .dma_test_irq = pdc202xx_dma_test_irq, - .dma_lost_irq = ide_dma_lost_irq, - .dma_timer_expiry = ide_dma_sff_timer_expiry, - .dma_sff_read_status = ide_dma_sff_read_status, -}; - static const struct ide_dma_ops pdc2026x_dma_ops = { .dma_host_set = ide_dma_host_set, .dma_setup = ide_dma_setup, .dma_start = pdc202xx_dma_start, .dma_end = pdc202xx_dma_end, - .dma_test_irq = pdc202xx_dma_test_irq, - .dma_lost_irq = pdc202xx_dma_lost_irq, + .dma_test_irq = ide_dma_test_irq, + .dma_lost_irq = ide_dma_lost_irq, .dma_timer_expiry = ide_dma_sff_timer_expiry, - .dma_clear = pdc202xx_reset, .dma_sff_read_status = ide_dma_sff_read_status, }; @@ -354,7 +268,7 @@ static const struct ide_port_info pdc202xx_chipsets[] __devinitdata = { .name = DRV_NAME, .init_chipset = init_chipset_pdc202xx, .port_ops = &pdc20246_port_ops, - .dma_ops = &pdc20246_dma_ops, + .dma_ops = &sff_dma_ops, .host_flags = IDE_HFLAGS_PDC202XX, .pio_mask = ATA_PIO4, .mwdma_mask = ATA_MWDMA2, diff --git a/drivers/ide/pmac.c b/drivers/ide/pmac.c index f76e4e6b408..97642a7a79c 100644 --- a/drivers/ide/pmac.c +++ b/drivers/ide/pmac.c @@ -1023,13 +1023,14 @@ static const struct ide_port_info pmac_port_info = { * Setup, register & probe an IDE channel driven by this driver, this is * called by one of the 2 probe functions (macio or PCI). */ -static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw) +static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, + struct ide_hw *hw) { struct device_node *np = pmif->node; const int *bidp; struct ide_host *host; ide_hwif_t *hwif; - hw_regs_t *hws[] = { hw, NULL, NULL, NULL }; + struct ide_hw *hws[] = { hw }; struct ide_port_info d = pmac_port_info; int rc; @@ -1077,7 +1078,7 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw) /* Make sure we have sane timings */ sanitize_timings(pmif); - host = ide_host_alloc(&d, hws); + host = ide_host_alloc(&d, hws, 1); if (host == NULL) return -ENOMEM; hwif = host->ports[0]; @@ -1124,7 +1125,7 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw) return 0; } -static void __devinit pmac_ide_init_ports(hw_regs_t *hw, unsigned long base) +static void __devinit pmac_ide_init_ports(struct ide_hw *hw, unsigned long base) { int i; @@ -1144,7 +1145,7 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match) unsigned long regbase; pmac_ide_hwif_t *pmif; int irq, rc; - hw_regs_t hw; + struct ide_hw hw; pmif = kzalloc(sizeof(*pmif), GFP_KERNEL); if (pmif == NULL) @@ -1268,7 +1269,7 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id) void __iomem *base; unsigned long rbase, rlen; int rc; - hw_regs_t hw; + struct ide_hw hw; np = pci_device_to_OF_node(pdev); if (np == NULL) { diff --git a/drivers/ide/q40ide.c b/drivers/ide/q40ide.c index c7934667924..ab49a97023d 100644 --- a/drivers/ide/q40ide.c +++ b/drivers/ide/q40ide.c @@ -51,11 +51,11 @@ static int q40ide_default_irq(unsigned long base) /* * Addresses are pretranslated for Q40 ISA access. */ -static void q40_ide_setup_ports(hw_regs_t *hw, unsigned long base, +static void q40_ide_setup_ports(struct ide_hw *hw, unsigned long base, ide_ack_intr_t *ack_intr, int irq) { - memset(hw, 0, sizeof(hw_regs_t)); + memset(hw, 0, sizeof(*hw)); /* BIG FAT WARNING: assumption: only DATA port is ever used in 16 bit mode */ hw->io_ports.data_addr = Q40_ISA_IO_W(base); @@ -70,8 +70,6 @@ static void q40_ide_setup_ports(hw_regs_t *hw, unsigned long base, hw->irq = irq; hw->ack_intr = ack_intr; - - hw->chipset = ide_generic; } static void q40ide_input_data(ide_drive_t *drive, struct ide_cmd *cmd, @@ -119,6 +117,7 @@ static const struct ide_port_info q40ide_port_info = { .tp_ops = &q40ide_tp_ops, .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA, .irq_flags = IRQF_SHARED, + .chipset = ide_generic, }; /* @@ -136,7 +135,7 @@ static const char *q40_ide_names[Q40IDE_NUM_HWIFS]={ static int __init q40ide_init(void) { int i; - hw_regs_t hw[Q40IDE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL }; + struct ide_hw hw[Q40IDE_NUM_HWIFS], *hws[] = { NULL, NULL }; if (!MACH_IS_Q40) return -ENODEV; @@ -163,7 +162,7 @@ static int __init q40ide_init(void) hws[i] = &hw[i]; } - return ide_host_add(&q40ide_port_info, hws, NULL); + return ide_host_add(&q40ide_port_info, hws, Q40IDE_NUM_HWIFS, NULL); } module_init(q40ide_init); diff --git a/drivers/ide/rapide.c b/drivers/ide/rapide.c index d5003ca6980..00f54248f41 100644 --- a/drivers/ide/rapide.c +++ b/drivers/ide/rapide.c @@ -13,9 +13,10 @@ static const struct ide_port_info rapide_port_info = { .host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA, + .chipset = ide_generic, }; -static void rapide_setup_ports(hw_regs_t *hw, void __iomem *base, +static void rapide_setup_ports(struct ide_hw *hw, void __iomem *base, void __iomem *ctrl, unsigned int sz, int irq) { unsigned long port = (unsigned long)base; @@ -35,7 +36,7 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id) void __iomem *base; struct ide_host *host; int ret; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; ret = ecard_request_resources(ec); if (ret) @@ -49,10 +50,9 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id) memset(&hw, 0, sizeof(hw)); rapide_setup_ports(&hw, base, base + 0x818, 1 << 6, ec->irq); - hw.chipset = ide_generic; hw.dev = &ec->dev; - ret = ide_host_add(&rapide_port_info, hws, &host); + ret = ide_host_add(&rapide_port_info, hws, 1, &host); if (ret) goto release; diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c index 5be41f25204..1104bb301eb 100644 --- a/drivers/ide/scc_pata.c +++ b/drivers/ide/scc_pata.c @@ -559,7 +559,7 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev, { struct scc_ports *ports = pci_get_drvdata(dev); struct ide_host *host; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; int i, rc; memset(&hw, 0, sizeof(hw)); @@ -567,9 +567,8 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev, hw.io_ports_array[i] = ports->dma + 0x20 + i * 4; hw.irq = dev->irq; hw.dev = &dev->dev; - hw.chipset = ide_pci; - rc = ide_host_add(d, hws, &host); + rc = ide_host_add(d, hws, 1, &host); if (rc) return rc; @@ -823,6 +822,7 @@ static const struct ide_port_info scc_chipset __devinitdata = { .host_flags = IDE_HFLAG_SINGLE, .irq_flags = IRQF_SHARED, .pio_mask = ATA_PIO4, + .chipset = ide_pci, }; /** diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c index 7a3a12d6e63..ab3db61d2ba 100644 --- a/drivers/ide/setup-pci.c +++ b/drivers/ide/setup-pci.c @@ -1,7 +1,7 @@ /* * Copyright (C) 1998-2000 Andre Hedrick <andre@linux-ide.org> * Copyright (C) 1995-1998 Mark Lord - * Copyright (C) 2007 Bartlomiej Zolnierkiewicz + * Copyright (C) 2007-2009 Bartlomiej Zolnierkiewicz * * May be copied or modified under the terms of the GNU General Public License */ @@ -301,11 +301,11 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info * } /** - * ide_hw_configure - configure a hw_regs_t instance + * ide_hw_configure - configure a struct ide_hw instance * @dev: PCI device holding interface * @d: IDE port info * @port: port number - * @hw: hw_regs_t instance corresponding to this port + * @hw: struct ide_hw instance corresponding to this port * * Perform the initial set up for the hardware interface structure. This * is done per interface port rather than per PCI device. There may be @@ -315,7 +315,7 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info * */ static int ide_hw_configure(struct pci_dev *dev, const struct ide_port_info *d, - unsigned int port, hw_regs_t *hw) + unsigned int port, struct ide_hw *hw) { unsigned long ctl = 0, base = 0; @@ -344,7 +344,6 @@ static int ide_hw_configure(struct pci_dev *dev, const struct ide_port_info *d, memset(hw, 0, sizeof(*hw)); hw->dev = &dev->dev; - hw->chipset = d->chipset ? d->chipset : ide_pci; ide_std_init_ports(hw, base, ctl | 2); return 0; @@ -446,8 +445,8 @@ out: * ide_pci_setup_ports - configure ports/devices on PCI IDE * @dev: PCI device * @d: IDE port info - * @hw: hw_regs_t instances corresponding to this PCI IDE device - * @hws: hw_regs_t pointers table to update + * @hw: struct ide_hw instances corresponding to this PCI IDE device + * @hws: struct ide_hw pointers table to update * * Scan the interfaces attached to this device and do any * necessary per port setup. Attach the devices and ask the @@ -459,7 +458,7 @@ out: */ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d, - hw_regs_t *hw, hw_regs_t **hws) + struct ide_hw *hw, struct ide_hw **hws) { int channels = (d->host_flags & IDE_HFLAG_SINGLE) ? 1 : 2, port; u8 tmp; @@ -535,61 +534,15 @@ out: return ret; } -int ide_pci_init_one(struct pci_dev *dev, const struct ide_port_info *d, - void *priv) -{ - struct ide_host *host; - hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL }; - int ret; - - ret = ide_setup_pci_controller(dev, d, 1); - if (ret < 0) - goto out; - - ide_pci_setup_ports(dev, d, &hw[0], &hws[0]); - - host = ide_host_alloc(d, hws); - if (host == NULL) { - ret = -ENOMEM; - goto out; - } - - host->dev[0] = &dev->dev; - - host->host_priv = priv; - - host->irq_flags = IRQF_SHARED; - - pci_set_drvdata(dev, host); - - ret = do_ide_setup_pci_device(dev, d, 1); - if (ret < 0) - goto out; - - /* fixup IRQ */ - if (ide_pci_is_in_compatibility_mode(dev)) { - hw[0].irq = pci_get_legacy_ide_irq(dev, 0); - hw[1].irq = pci_get_legacy_ide_irq(dev, 1); - } else - hw[1].irq = hw[0].irq = ret; - - ret = ide_host_register(host, d, hws); - if (ret) - ide_host_free(host); -out: - return ret; -} -EXPORT_SYMBOL_GPL(ide_pci_init_one); - int ide_pci_init_two(struct pci_dev *dev1, struct pci_dev *dev2, const struct ide_port_info *d, void *priv) { struct pci_dev *pdev[] = { dev1, dev2 }; struct ide_host *host; - int ret, i; - hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL }; + int ret, i, n_ports = dev2 ? 4 : 2; + struct ide_hw hw[4], *hws[] = { NULL, NULL, NULL, NULL }; - for (i = 0; i < 2; i++) { + for (i = 0; i < n_ports / 2; i++) { ret = ide_setup_pci_controller(pdev[i], d, !i); if (ret < 0) goto out; @@ -597,23 +550,24 @@ int ide_pci_init_two(struct pci_dev *dev1, struct pci_dev *dev2, ide_pci_setup_ports(pdev[i], d, &hw[i*2], &hws[i*2]); } - host = ide_host_alloc(d, hws); + host = ide_host_alloc(d, hws, n_ports); if (host == NULL) { ret = -ENOMEM; goto out; } host->dev[0] = &dev1->dev; - host->dev[1] = &dev2->dev; + if (dev2) + host->dev[1] = &dev2->dev; host->host_priv = priv; - host->irq_flags = IRQF_SHARED; pci_set_drvdata(pdev[0], host); - pci_set_drvdata(pdev[1], host); + if (dev2) + pci_set_drvdata(pdev[1], host); - for (i = 0; i < 2; i++) { + for (i = 0; i < n_ports / 2; i++) { ret = do_ide_setup_pci_device(pdev[i], d, !i); /* @@ -639,6 +593,13 @@ out: } EXPORT_SYMBOL_GPL(ide_pci_init_two); +int ide_pci_init_one(struct pci_dev *dev, const struct ide_port_info *d, + void *priv) +{ + return ide_pci_init_two(dev, NULL, d, priv); +} +EXPORT_SYMBOL_GPL(ide_pci_init_one); + void ide_pci_remove(struct pci_dev *dev) { struct ide_host *host = pci_get_drvdata(dev); diff --git a/drivers/ide/sgiioc4.c b/drivers/ide/sgiioc4.c index e5d2a48a84d..5f37f168f94 100644 --- a/drivers/ide/sgiioc4.c +++ b/drivers/ide/sgiioc4.c @@ -91,7 +91,7 @@ typedef struct { static void -sgiioc4_init_hwif_ports(hw_regs_t * hw, unsigned long data_port, +sgiioc4_init_hwif_ports(struct ide_hw *hw, unsigned long data_port, unsigned long ctrl_port, unsigned long irq_port) { unsigned long reg = data_port; @@ -546,7 +546,7 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev) unsigned long cmd_base, irqport; unsigned long bar0, cmd_phys_base, ctl; void __iomem *virt_base; - hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; int rc; /* Get the CmdBlk and CtrlBlk Base Registers */ @@ -575,13 +575,12 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev) memset(&hw, 0, sizeof(hw)); sgiioc4_init_hwif_ports(&hw, cmd_base, ctl, irqport); hw.irq = dev->irq; - hw.chipset = ide_pci; hw.dev = &dev->dev; /* Initializing chipset IRQ Registers */ writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4)); - rc = ide_host_add(&sgiioc4_port_info, hws, NULL); + rc = ide_host_add(&sgiioc4_port_info, hws, 1, NULL); if (!rc) return 0; diff --git a/drivers/ide/siimage.c b/drivers/ide/siimage.c index e4973cd1fba..bd82d228608 100644 --- a/drivers/ide/siimage.c +++ b/drivers/ide/siimage.c @@ -451,8 +451,8 @@ static int sil_sata_reset_poll(ide_drive_t *drive) static void sil_sata_pre_reset(ide_drive_t *drive) { if (drive->media == ide_disk) { - drive->special.b.set_geometry = 0; - drive->special.b.recalibrate = 0; + drive->special_flags &= + ~(IDE_SFLAG_SET_GEOMETRY | IDE_SFLAG_RECALIBRATE); } } diff --git a/drivers/ide/sl82c105.c b/drivers/ide/sl82c105.c index b0a46062533..0924abff52f 100644 --- a/drivers/ide/sl82c105.c +++ b/drivers/ide/sl82c105.c @@ -10,7 +10,7 @@ * with the timing registers setup. * -- Benjamin Herrenschmidt (01/11/03) benh@kernel.crashing.org * - * Copyright (C) 2006-2007 MontaVista Software, Inc. <source@mvista.com> + * Copyright (C) 2006-2007,2009 MontaVista Software, Inc. <source@mvista.com> * Copyright (C) 2007 Bartlomiej Zolnierkiewicz */ @@ -146,14 +146,15 @@ static void sl82c105_dma_lost_irq(ide_drive_t *drive) u32 val, mask = hwif->channel ? CTRL_IDE_IRQB : CTRL_IDE_IRQA; u8 dma_cmd; - printk("sl82c105: lost IRQ, resetting host\n"); + printk(KERN_WARNING "sl82c105: lost IRQ, resetting host\n"); /* * Check the raw interrupt from the drive. */ pci_read_config_dword(dev, 0x40, &val); if (val & mask) - printk("sl82c105: drive was requesting IRQ, but host lost it\n"); + printk(KERN_INFO "sl82c105: drive was requesting IRQ, " + "but host lost it\n"); /* * Was DMA enabled? If so, disable it - we're resetting the @@ -162,7 +163,7 @@ static void sl82c105_dma_lost_irq(ide_drive_t *drive) dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD); if (dma_cmd & 1) { outb(dma_cmd & ~1, hwif->dma_base + ATA_DMA_CMD); - printk("sl82c105: DMA was enabled\n"); + printk(KERN_INFO "sl82c105: DMA was enabled\n"); } sl82c105_reset_host(dev); diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c index e33d764e294..ea89fddeed9 100644 --- a/drivers/ide/tx4938ide.c +++ b/drivers/ide/tx4938ide.c @@ -130,8 +130,7 @@ static const struct ide_port_info tx4938ide_port_info __initdata = { static int __init tx4938ide_probe(struct platform_device *pdev) { - hw_regs_t hw; - hw_regs_t *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; struct ide_host *host; struct resource *res; struct tx4938ide_platform_info *pdata = pdev->dev.platform_data; @@ -183,7 +182,7 @@ static int __init tx4938ide_probe(struct platform_device *pdev) tx4938ide_tune_ebusc(pdata->ebus_ch, pdata->gbus_clock, 0); else d.port_ops = NULL; - ret = ide_host_add(&d, hws, &host); + ret = ide_host_add(&d, hws, 1, &host); if (!ret) platform_set_drvdata(pdev, host); return ret; diff --git a/drivers/ide/tx4939ide.c b/drivers/ide/tx4939ide.c index 5ca76224f6d..64b58ecc3f0 100644 --- a/drivers/ide/tx4939ide.c +++ b/drivers/ide/tx4939ide.c @@ -537,8 +537,7 @@ static const struct ide_port_info tx4939ide_port_info __initdata = { static int __init tx4939ide_probe(struct platform_device *pdev) { - hw_regs_t hw; - hw_regs_t *hws[] = { &hw, NULL, NULL, NULL }; + struct ide_hw hw, *hws[] = { &hw }; struct ide_host *host; struct resource *res; int irq, ret; @@ -581,7 +580,7 @@ static int __init tx4939ide_probe(struct platform_device *pdev) hw.dev = &pdev->dev; pr_info("TX4939 IDE interface (base %#lx, irq %d)\n", mapbase, irq); - host = ide_host_alloc(&tx4939ide_port_info, hws); + host = ide_host_alloc(&tx4939ide_port_info, hws, 1); if (!host) return -ENOMEM; /* use extra_base for base address of the all registers */ diff --git a/drivers/lguest/Kconfig b/drivers/lguest/Kconfig index a3d3cbab359..0aaa0597a62 100644 --- a/drivers/lguest/Kconfig +++ b/drivers/lguest/Kconfig @@ -1,6 +1,6 @@ config LGUEST tristate "Linux hypervisor example code" - depends on X86_32 && EXPERIMENTAL && !X86_PAE && FUTEX + depends on X86_32 && EXPERIMENTAL && EVENTFD select HVC_DRIVER ---help--- This is a very simple module which allows you to run diff --git a/drivers/lguest/core.c b/drivers/lguest/core.c index 4845fb3cf74..a6974e9b8eb 100644 --- a/drivers/lguest/core.c +++ b/drivers/lguest/core.c @@ -95,7 +95,7 @@ static __init int map_switcher(void) * array of struct pages. It increments that pointer, but we don't * care. */ pagep = switcher_page; - err = map_vm_area(switcher_vma, PAGE_KERNEL, &pagep); + err = map_vm_area(switcher_vma, PAGE_KERNEL_EXEC, &pagep); if (err) { printk("lguest: map_vm_area failed: %i\n", err); goto free_vma; @@ -188,6 +188,9 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) { /* We stop running once the Guest is dead. */ while (!cpu->lg->dead) { + unsigned int irq; + bool more; + /* First we run any hypercalls the Guest wants done. */ if (cpu->hcall) do_hypercalls(cpu); @@ -195,23 +198,23 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) /* It's possible the Guest did a NOTIFY hypercall to the * Launcher, in which case we return from the read() now. */ if (cpu->pending_notify) { - if (put_user(cpu->pending_notify, user)) - return -EFAULT; - return sizeof(cpu->pending_notify); + if (!send_notify_to_eventfd(cpu)) { + if (put_user(cpu->pending_notify, user)) + return -EFAULT; + return sizeof(cpu->pending_notify); + } } /* Check for signals */ if (signal_pending(current)) return -ERESTARTSYS; - /* If Waker set break_out, return to Launcher. */ - if (cpu->break_out) - return -EAGAIN; - /* Check if there are any interrupts which can be delivered now: * if so, this sets up the hander to be executed when we next * run the Guest. */ - maybe_do_interrupt(cpu); + irq = interrupt_pending(cpu, &more); + if (irq < LGUEST_IRQS) + try_deliver_interrupt(cpu, irq, more); /* All long-lived kernel loops need to check with this horrible * thing called the freezer. If the Host is trying to suspend, @@ -224,10 +227,15 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) break; /* If the Guest asked to be stopped, we sleep. The Guest's - * clock timer or LHREQ_BREAK from the Waker will wake us. */ + * clock timer will wake us. */ if (cpu->halted) { set_current_state(TASK_INTERRUPTIBLE); - schedule(); + /* Just before we sleep, make sure no interrupt snuck in + * which we should be doing. */ + if (interrupt_pending(cpu, &more) < LGUEST_IRQS) + set_current_state(TASK_RUNNING); + else + schedule(); continue; } diff --git a/drivers/lguest/hypercalls.c b/drivers/lguest/hypercalls.c index 54d66f05fef..c29ffa19cb7 100644 --- a/drivers/lguest/hypercalls.c +++ b/drivers/lguest/hypercalls.c @@ -37,6 +37,10 @@ static void do_hcall(struct lg_cpu *cpu, struct hcall_args *args) /* This call does nothing, except by breaking out of the Guest * it makes us process all the asynchronous hypercalls. */ break; + case LHCALL_SEND_INTERRUPTS: + /* This call does nothing too, but by breaking out of the Guest + * it makes us process any pending interrupts. */ + break; case LHCALL_LGUEST_INIT: /* You can't get here unless you're already initialized. Don't * do that. */ @@ -73,11 +77,21 @@ static void do_hcall(struct lg_cpu *cpu, struct hcall_args *args) guest_set_stack(cpu, args->arg1, args->arg2, args->arg3); break; case LHCALL_SET_PTE: +#ifdef CONFIG_X86_PAE + guest_set_pte(cpu, args->arg1, args->arg2, + __pte(args->arg3 | (u64)args->arg4 << 32)); +#else guest_set_pte(cpu, args->arg1, args->arg2, __pte(args->arg3)); +#endif + break; + case LHCALL_SET_PGD: + guest_set_pgd(cpu->lg, args->arg1, args->arg2); break; +#ifdef CONFIG_X86_PAE case LHCALL_SET_PMD: guest_set_pmd(cpu->lg, args->arg1, args->arg2); break; +#endif case LHCALL_SET_CLOCKEVENT: guest_set_clockevent(cpu, args->arg1); break; diff --git a/drivers/lguest/interrupts_and_traps.c b/drivers/lguest/interrupts_and_traps.c index 6e99adbe194..0e9067b0d50 100644 --- a/drivers/lguest/interrupts_and_traps.c +++ b/drivers/lguest/interrupts_and_traps.c @@ -128,30 +128,39 @@ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, /*H:205 * Virtual Interrupts. * - * maybe_do_interrupt() gets called before every entry to the Guest, to see if - * we should divert the Guest to running an interrupt handler. */ -void maybe_do_interrupt(struct lg_cpu *cpu) + * interrupt_pending() returns the first pending interrupt which isn't blocked + * by the Guest. It is called before every entry to the Guest, and just before + * we go to sleep when the Guest has halted itself. */ +unsigned int interrupt_pending(struct lg_cpu *cpu, bool *more) { unsigned int irq; DECLARE_BITMAP(blk, LGUEST_IRQS); - struct desc_struct *idt; /* If the Guest hasn't even initialized yet, we can do nothing. */ if (!cpu->lg->lguest_data) - return; + return LGUEST_IRQS; /* Take our "irqs_pending" array and remove any interrupts the Guest * wants blocked: the result ends up in "blk". */ if (copy_from_user(&blk, cpu->lg->lguest_data->blocked_interrupts, sizeof(blk))) - return; + return LGUEST_IRQS; bitmap_andnot(blk, cpu->irqs_pending, blk, LGUEST_IRQS); /* Find the first interrupt. */ irq = find_first_bit(blk, LGUEST_IRQS); - /* None? Nothing to do */ - if (irq >= LGUEST_IRQS) - return; + *more = find_next_bit(blk, LGUEST_IRQS, irq+1); + + return irq; +} + +/* This actually diverts the Guest to running an interrupt handler, once an + * interrupt has been identified by interrupt_pending(). */ +void try_deliver_interrupt(struct lg_cpu *cpu, unsigned int irq, bool more) +{ + struct desc_struct *idt; + + BUG_ON(irq >= LGUEST_IRQS); /* They may be in the middle of an iret, where they asked us never to * deliver interrupts. */ @@ -170,8 +179,12 @@ void maybe_do_interrupt(struct lg_cpu *cpu) u32 irq_enabled; if (get_user(irq_enabled, &cpu->lg->lguest_data->irq_enabled)) irq_enabled = 0; - if (!irq_enabled) + if (!irq_enabled) { + /* Make sure they know an IRQ is pending. */ + put_user(X86_EFLAGS_IF, + &cpu->lg->lguest_data->irq_pending); return; + } } /* Look at the IDT entry the Guest gave us for this interrupt. The @@ -194,6 +207,25 @@ void maybe_do_interrupt(struct lg_cpu *cpu) * here is a compromise which means at least it gets updated every * timer interrupt. */ write_timestamp(cpu); + + /* If there are no other interrupts we want to deliver, clear + * the pending flag. */ + if (!more) + put_user(0, &cpu->lg->lguest_data->irq_pending); +} + +/* And this is the routine when we want to set an interrupt for the Guest. */ +void set_interrupt(struct lg_cpu *cpu, unsigned int irq) +{ + /* Next time the Guest runs, the core code will see if it can deliver + * this interrupt. */ + set_bit(irq, cpu->irqs_pending); + + /* Make sure it sees it; it might be asleep (eg. halted), or + * running the Guest right now, in which case kick_process() + * will knock it out. */ + if (!wake_up_process(cpu->tsk)) + kick_process(cpu->tsk); } /*:*/ @@ -510,10 +542,7 @@ static enum hrtimer_restart clockdev_fn(struct hrtimer *timer) struct lg_cpu *cpu = container_of(timer, struct lg_cpu, hrt); /* Remember the first interrupt is the timer interrupt. */ - set_bit(0, cpu->irqs_pending); - /* If the Guest is actually stopped, we need to wake it up. */ - if (cpu->halted) - wake_up_process(cpu->tsk); + set_interrupt(cpu, 0); return HRTIMER_NORESTART; } diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h index af92a176697..d4e8979735c 100644 --- a/drivers/lguest/lg.h +++ b/drivers/lguest/lg.h @@ -49,7 +49,7 @@ struct lg_cpu { u32 cr2; int ts; u32 esp1; - u8 ss1; + u16 ss1; /* Bitmap of what has changed: see CHANGED_* above. */ int changed; @@ -71,9 +71,7 @@ struct lg_cpu { /* Virtual clock device */ struct hrtimer hrt; - /* Do we need to stop what we're doing and return to userspace? */ - int break_out; - wait_queue_head_t break_wq; + /* Did the Guest tell us to halt? */ int halted; /* Pending virtual interrupts */ @@ -82,6 +80,16 @@ struct lg_cpu { struct lg_cpu_arch arch; }; +struct lg_eventfd { + unsigned long addr; + struct file *event; +}; + +struct lg_eventfd_map { + unsigned int num; + struct lg_eventfd map[]; +}; + /* The private info the thread maintains about the guest. */ struct lguest { @@ -102,6 +110,8 @@ struct lguest unsigned int stack_pages; u32 tsc_khz; + struct lg_eventfd_map *eventfds; + /* Dead? */ const char *dead; }; @@ -137,9 +147,13 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user); * in the kernel. */ #define pgd_flags(x) (pgd_val(x) & ~PAGE_MASK) #define pgd_pfn(x) (pgd_val(x) >> PAGE_SHIFT) +#define pmd_flags(x) (pmd_val(x) & ~PAGE_MASK) +#define pmd_pfn(x) (pmd_val(x) >> PAGE_SHIFT) /* interrupts_and_traps.c: */ -void maybe_do_interrupt(struct lg_cpu *cpu); +unsigned int interrupt_pending(struct lg_cpu *cpu, bool *more); +void try_deliver_interrupt(struct lg_cpu *cpu, unsigned int irq, bool more); +void set_interrupt(struct lg_cpu *cpu, unsigned int irq); bool deliver_trap(struct lg_cpu *cpu, unsigned int num); void load_guest_idt_entry(struct lg_cpu *cpu, unsigned int i, u32 low, u32 hi); @@ -150,6 +164,7 @@ void setup_default_idt_entries(struct lguest_ro_state *state, void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt, const unsigned long *def); void guest_set_clockevent(struct lg_cpu *cpu, unsigned long delta); +bool send_notify_to_eventfd(struct lg_cpu *cpu); void init_clockdev(struct lg_cpu *cpu); bool check_syscall_vector(struct lguest *lg); int init_interrupts(void); @@ -168,7 +183,10 @@ void copy_gdt_tls(const struct lg_cpu *cpu, struct desc_struct *gdt); int init_guest_pagetable(struct lguest *lg); void free_guest_pagetable(struct lguest *lg); void guest_new_pagetable(struct lg_cpu *cpu, unsigned long pgtable); +void guest_set_pgd(struct lguest *lg, unsigned long gpgdir, u32 i); +#ifdef CONFIG_X86_PAE void guest_set_pmd(struct lguest *lg, unsigned long gpgdir, u32 i); +#endif void guest_pagetable_clear_all(struct lg_cpu *cpu); void guest_pagetable_flush_user(struct lg_cpu *cpu); void guest_set_pte(struct lg_cpu *cpu, unsigned long gpgdir, diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c index df44d962626..e082cdac88b 100644 --- a/drivers/lguest/lguest_device.c +++ b/drivers/lguest/lguest_device.c @@ -228,7 +228,8 @@ extern void lguest_setup_irq(unsigned int irq); * function. */ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, unsigned index, - void (*callback)(struct virtqueue *vq)) + void (*callback)(struct virtqueue *vq), + const char *name) { struct lguest_device *ldev = to_lgdev(vdev); struct lguest_vq_info *lvq; @@ -263,7 +264,7 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, /* OK, tell virtio_ring.c to set up a virtqueue now we know its size * and we've got a pointer to its pages. */ vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN, - vdev, lvq->pages, lg_notify, callback); + vdev, lvq->pages, lg_notify, callback, name); if (!vq) { err = -ENOMEM; goto unmap; @@ -312,6 +313,38 @@ static void lg_del_vq(struct virtqueue *vq) kfree(lvq); } +static void lg_del_vqs(struct virtio_device *vdev) +{ + struct virtqueue *vq, *n; + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + lg_del_vq(vq); +} + +static int lg_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + struct lguest_device *ldev = to_lgdev(vdev); + int i; + + /* We must have this many virtqueues. */ + if (nvqs > ldev->desc->num_vq) + return -ENOENT; + + for (i = 0; i < nvqs; ++i) { + vqs[i] = lg_find_vq(vdev, i, callbacks[i], names[i]); + if (IS_ERR(vqs[i])) + goto error; + } + return 0; + +error: + lg_del_vqs(vdev); + return PTR_ERR(vqs[i]); +} + /* The ops structure which hooks everything together. */ static struct virtio_config_ops lguest_config_ops = { .get_features = lg_get_features, @@ -321,8 +354,8 @@ static struct virtio_config_ops lguest_config_ops = { .get_status = lg_get_status, .set_status = lg_set_status, .reset = lg_reset, - .find_vq = lg_find_vq, - .del_vq = lg_del_vq, + .find_vqs = lg_find_vqs, + .del_vqs = lg_del_vqs, }; /* The root device for the lguest virtio devices. This makes them appear as diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c index b8ee103eed5..32e29712105 100644 --- a/drivers/lguest/lguest_user.c +++ b/drivers/lguest/lguest_user.c @@ -7,32 +7,83 @@ #include <linux/miscdevice.h> #include <linux/fs.h> #include <linux/sched.h> +#include <linux/eventfd.h> +#include <linux/file.h> #include "lg.h" -/*L:055 When something happens, the Waker process needs a way to stop the - * kernel running the Guest and return to the Launcher. So the Waker writes - * LHREQ_BREAK and the value "1" to /dev/lguest to do this. Once the Launcher - * has done whatever needs attention, it writes LHREQ_BREAK and "0" to release - * the Waker. */ -static int break_guest_out(struct lg_cpu *cpu, const unsigned long __user*input) +bool send_notify_to_eventfd(struct lg_cpu *cpu) { - unsigned long on; + unsigned int i; + struct lg_eventfd_map *map; + + /* lg->eventfds is RCU-protected */ + rcu_read_lock(); + map = rcu_dereference(cpu->lg->eventfds); + for (i = 0; i < map->num; i++) { + if (map->map[i].addr == cpu->pending_notify) { + eventfd_signal(map->map[i].event, 1); + cpu->pending_notify = 0; + break; + } + } + rcu_read_unlock(); + return cpu->pending_notify == 0; +} - /* Fetch whether they're turning break on or off. */ - if (get_user(on, input) != 0) - return -EFAULT; +static int add_eventfd(struct lguest *lg, unsigned long addr, int fd) +{ + struct lg_eventfd_map *new, *old = lg->eventfds; - if (on) { - cpu->break_out = 1; - /* Pop it out of the Guest (may be running on different CPU) */ - wake_up_process(cpu->tsk); - /* Wait for them to reset it */ - return wait_event_interruptible(cpu->break_wq, !cpu->break_out); - } else { - cpu->break_out = 0; - wake_up(&cpu->break_wq); - return 0; + if (!addr) + return -EINVAL; + + /* Replace the old array with the new one, carefully: others can + * be accessing it at the same time */ + new = kmalloc(sizeof(*new) + sizeof(new->map[0]) * (old->num + 1), + GFP_KERNEL); + if (!new) + return -ENOMEM; + + /* First make identical copy. */ + memcpy(new->map, old->map, sizeof(old->map[0]) * old->num); + new->num = old->num; + + /* Now append new entry. */ + new->map[new->num].addr = addr; + new->map[new->num].event = eventfd_fget(fd); + if (IS_ERR(new->map[new->num].event)) { + kfree(new); + return PTR_ERR(new->map[new->num].event); } + new->num++; + + /* Now put new one in place. */ + rcu_assign_pointer(lg->eventfds, new); + + /* We're not in a big hurry. Wait until noone's looking at old + * version, then delete it. */ + synchronize_rcu(); + kfree(old); + + return 0; +} + +static int attach_eventfd(struct lguest *lg, const unsigned long __user *input) +{ + unsigned long addr, fd; + int err; + + if (get_user(addr, input) != 0) + return -EFAULT; + input++; + if (get_user(fd, input) != 0) + return -EFAULT; + + mutex_lock(&lguest_lock); + err = add_eventfd(lg, addr, fd); + mutex_unlock(&lguest_lock); + + return 0; } /*L:050 Sending an interrupt is done by writing LHREQ_IRQ and an interrupt @@ -45,9 +96,8 @@ static int user_send_irq(struct lg_cpu *cpu, const unsigned long __user *input) return -EFAULT; if (irq >= LGUEST_IRQS) return -EINVAL; - /* Next time the Guest runs, the core code will see if it can deliver - * this interrupt. */ - set_bit(irq, cpu->irqs_pending); + + set_interrupt(cpu, irq); return 0; } @@ -126,9 +176,6 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip) * address. */ lguest_arch_setup_regs(cpu, start_ip); - /* Initialize the queue for the Waker to wait on */ - init_waitqueue_head(&cpu->break_wq); - /* We keep a pointer to the Launcher task (ie. current task) for when * other Guests want to wake this one (eg. console input). */ cpu->tsk = current; @@ -185,6 +232,13 @@ static int initialize(struct file *file, const unsigned long __user *input) goto unlock; } + lg->eventfds = kmalloc(sizeof(*lg->eventfds), GFP_KERNEL); + if (!lg->eventfds) { + err = -ENOMEM; + goto free_lg; + } + lg->eventfds->num = 0; + /* Populate the easy fields of our "struct lguest" */ lg->mem_base = (void __user *)args[0]; lg->pfn_limit = args[1]; @@ -192,7 +246,7 @@ static int initialize(struct file *file, const unsigned long __user *input) /* This is the first cpu (cpu 0) and it will start booting at args[2] */ err = lg_cpu_start(&lg->cpus[0], 0, args[2]); if (err) - goto release_guest; + goto free_eventfds; /* Initialize the Guest's shadow page tables, using the toplevel * address the Launcher gave us. This allocates memory, so can fail. */ @@ -211,7 +265,9 @@ static int initialize(struct file *file, const unsigned long __user *input) free_regs: /* FIXME: This should be in free_vcpu */ free_page(lg->cpus[0].regs_page); -release_guest: +free_eventfds: + kfree(lg->eventfds); +free_lg: kfree(lg); unlock: mutex_unlock(&lguest_lock); @@ -252,11 +308,6 @@ static ssize_t write(struct file *file, const char __user *in, /* Once the Guest is dead, you can only read() why it died. */ if (lg->dead) return -ENOENT; - - /* If you're not the task which owns the Guest, all you can do - * is break the Launcher out of running the Guest. */ - if (current != cpu->tsk && req != LHREQ_BREAK) - return -EPERM; } switch (req) { @@ -264,8 +315,8 @@ static ssize_t write(struct file *file, const char __user *in, return initialize(file, input); case LHREQ_IRQ: return user_send_irq(cpu, input); - case LHREQ_BREAK: - return break_guest_out(cpu, input); + case LHREQ_EVENTFD: + return attach_eventfd(lg, input); default: return -EINVAL; } @@ -303,6 +354,12 @@ static int close(struct inode *inode, struct file *file) * the Launcher's memory management structure. */ mmput(lg->cpus[i].mm); } + + /* Release any eventfds they registered. */ + for (i = 0; i < lg->eventfds->num; i++) + fput(lg->eventfds->map[i].event); + kfree(lg->eventfds); + /* If lg->dead doesn't contain an error code it will be NULL or a * kmalloc()ed string, either of which is ok to hand to kfree(). */ if (!IS_ERR(lg->dead)) diff --git a/drivers/lguest/page_tables.c b/drivers/lguest/page_tables.c index a059cf9980f..a6fe1abda24 100644 --- a/drivers/lguest/page_tables.c +++ b/drivers/lguest/page_tables.c @@ -53,6 +53,17 @@ * page. */ #define SWITCHER_PGD_INDEX (PTRS_PER_PGD - 1) +/* For PAE we need the PMD index as well. We use the last 2MB, so we + * will need the last pmd entry of the last pmd page. */ +#ifdef CONFIG_X86_PAE +#define SWITCHER_PMD_INDEX (PTRS_PER_PMD - 1) +#define RESERVE_MEM 2U +#define CHECK_GPGD_MASK _PAGE_PRESENT +#else +#define RESERVE_MEM 4U +#define CHECK_GPGD_MASK _PAGE_TABLE +#endif + /* We actually need a separate PTE page for each CPU. Remember that after the * Switcher code itself comes two pages for each CPU, and we don't want this * CPU's guest to see the pages of any other CPU. */ @@ -73,24 +84,59 @@ static pgd_t *spgd_addr(struct lg_cpu *cpu, u32 i, unsigned long vaddr) { unsigned int index = pgd_index(vaddr); +#ifndef CONFIG_X86_PAE /* We kill any Guest trying to touch the Switcher addresses. */ if (index >= SWITCHER_PGD_INDEX) { kill_guest(cpu, "attempt to access switcher pages"); index = 0; } +#endif /* Return a pointer index'th pgd entry for the i'th page table. */ return &cpu->lg->pgdirs[i].pgdir[index]; } +#ifdef CONFIG_X86_PAE +/* This routine then takes the PGD entry given above, which contains the + * address of the PMD page. It then returns a pointer to the PMD entry for the + * given address. */ +static pmd_t *spmd_addr(struct lg_cpu *cpu, pgd_t spgd, unsigned long vaddr) +{ + unsigned int index = pmd_index(vaddr); + pmd_t *page; + + /* We kill any Guest trying to touch the Switcher addresses. */ + if (pgd_index(vaddr) == SWITCHER_PGD_INDEX && + index >= SWITCHER_PMD_INDEX) { + kill_guest(cpu, "attempt to access switcher pages"); + index = 0; + } + + /* You should never call this if the PGD entry wasn't valid */ + BUG_ON(!(pgd_flags(spgd) & _PAGE_PRESENT)); + page = __va(pgd_pfn(spgd) << PAGE_SHIFT); + + return &page[index]; +} +#endif + /* This routine then takes the page directory entry returned above, which * contains the address of the page table entry (PTE) page. It then returns a * pointer to the PTE entry for the given address. */ -static pte_t *spte_addr(pgd_t spgd, unsigned long vaddr) +static pte_t *spte_addr(struct lg_cpu *cpu, pgd_t spgd, unsigned long vaddr) { +#ifdef CONFIG_X86_PAE + pmd_t *pmd = spmd_addr(cpu, spgd, vaddr); + pte_t *page = __va(pmd_pfn(*pmd) << PAGE_SHIFT); + + /* You should never call this if the PMD entry wasn't valid */ + BUG_ON(!(pmd_flags(*pmd) & _PAGE_PRESENT)); +#else pte_t *page = __va(pgd_pfn(spgd) << PAGE_SHIFT); /* You should never call this if the PGD entry wasn't valid */ BUG_ON(!(pgd_flags(spgd) & _PAGE_PRESENT)); - return &page[(vaddr >> PAGE_SHIFT) % PTRS_PER_PTE]; +#endif + + return &page[pte_index(vaddr)]; } /* These two functions just like the above two, except they access the Guest @@ -101,12 +147,32 @@ static unsigned long gpgd_addr(struct lg_cpu *cpu, unsigned long vaddr) return cpu->lg->pgdirs[cpu->cpu_pgd].gpgdir + index * sizeof(pgd_t); } -static unsigned long gpte_addr(pgd_t gpgd, unsigned long vaddr) +#ifdef CONFIG_X86_PAE +static unsigned long gpmd_addr(pgd_t gpgd, unsigned long vaddr) +{ + unsigned long gpage = pgd_pfn(gpgd) << PAGE_SHIFT; + BUG_ON(!(pgd_flags(gpgd) & _PAGE_PRESENT)); + return gpage + pmd_index(vaddr) * sizeof(pmd_t); +} + +static unsigned long gpte_addr(struct lg_cpu *cpu, + pmd_t gpmd, unsigned long vaddr) +{ + unsigned long gpage = pmd_pfn(gpmd) << PAGE_SHIFT; + + BUG_ON(!(pmd_flags(gpmd) & _PAGE_PRESENT)); + return gpage + pte_index(vaddr) * sizeof(pte_t); +} +#else +static unsigned long gpte_addr(struct lg_cpu *cpu, + pgd_t gpgd, unsigned long vaddr) { unsigned long gpage = pgd_pfn(gpgd) << PAGE_SHIFT; + BUG_ON(!(pgd_flags(gpgd) & _PAGE_PRESENT)); - return gpage + ((vaddr>>PAGE_SHIFT) % PTRS_PER_PTE) * sizeof(pte_t); + return gpage + pte_index(vaddr) * sizeof(pte_t); } +#endif /*:*/ /*M:014 get_pfn is slow: we could probably try to grab batches of pages here as @@ -171,7 +237,7 @@ static void release_pte(pte_t pte) /* Remember that get_user_pages_fast() took a reference to the page, in * get_pfn()? We have to put it back now. */ if (pte_flags(pte) & _PAGE_PRESENT) - put_page(pfn_to_page(pte_pfn(pte))); + put_page(pte_page(pte)); } /*:*/ @@ -184,11 +250,20 @@ static void check_gpte(struct lg_cpu *cpu, pte_t gpte) static void check_gpgd(struct lg_cpu *cpu, pgd_t gpgd) { - if ((pgd_flags(gpgd) & ~_PAGE_TABLE) || + if ((pgd_flags(gpgd) & ~CHECK_GPGD_MASK) || (pgd_pfn(gpgd) >= cpu->lg->pfn_limit)) kill_guest(cpu, "bad page directory entry"); } +#ifdef CONFIG_X86_PAE +static void check_gpmd(struct lg_cpu *cpu, pmd_t gpmd) +{ + if ((pmd_flags(gpmd) & ~_PAGE_TABLE) || + (pmd_pfn(gpmd) >= cpu->lg->pfn_limit)) + kill_guest(cpu, "bad page middle directory entry"); +} +#endif + /*H:330 * (i) Looking up a page table entry when the Guest faults. * @@ -207,6 +282,11 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode) pte_t gpte; pte_t *spte; +#ifdef CONFIG_X86_PAE + pmd_t *spmd; + pmd_t gpmd; +#endif + /* First step: get the top-level Guest page table entry. */ gpgd = lgread(cpu, gpgd_addr(cpu, vaddr), pgd_t); /* Toplevel not present? We can't map it in. */ @@ -228,12 +308,45 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode) check_gpgd(cpu, gpgd); /* And we copy the flags to the shadow PGD entry. The page * number in the shadow PGD is the page we just allocated. */ - *spgd = __pgd(__pa(ptepage) | pgd_flags(gpgd)); + set_pgd(spgd, __pgd(__pa(ptepage) | pgd_flags(gpgd))); } +#ifdef CONFIG_X86_PAE + gpmd = lgread(cpu, gpmd_addr(gpgd, vaddr), pmd_t); + /* middle level not present? We can't map it in. */ + if (!(pmd_flags(gpmd) & _PAGE_PRESENT)) + return false; + + /* Now look at the matching shadow entry. */ + spmd = spmd_addr(cpu, *spgd, vaddr); + + if (!(pmd_flags(*spmd) & _PAGE_PRESENT)) { + /* No shadow entry: allocate a new shadow PTE page. */ + unsigned long ptepage = get_zeroed_page(GFP_KERNEL); + + /* This is not really the Guest's fault, but killing it is + * simple for this corner case. */ + if (!ptepage) { + kill_guest(cpu, "out of memory allocating pte page"); + return false; + } + + /* We check that the Guest pmd is OK. */ + check_gpmd(cpu, gpmd); + + /* And we copy the flags to the shadow PMD entry. The page + * number in the shadow PMD is the page we just allocated. */ + native_set_pmd(spmd, __pmd(__pa(ptepage) | pmd_flags(gpmd))); + } + + /* OK, now we look at the lower level in the Guest page table: keep its + * address, because we might update it later. */ + gpte_ptr = gpte_addr(cpu, gpmd, vaddr); +#else /* OK, now we look at the lower level in the Guest page table: keep its * address, because we might update it later. */ - gpte_ptr = gpte_addr(gpgd, vaddr); + gpte_ptr = gpte_addr(cpu, gpgd, vaddr); +#endif gpte = lgread(cpu, gpte_ptr, pte_t); /* If this page isn't in the Guest page tables, we can't page it in. */ @@ -259,7 +372,7 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode) gpte = pte_mkdirty(gpte); /* Get the pointer to the shadow PTE entry we're going to set. */ - spte = spte_addr(*spgd, vaddr); + spte = spte_addr(cpu, *spgd, vaddr); /* If there was a valid shadow PTE entry here before, we release it. * This can happen with a write to a previously read-only entry. */ release_pte(*spte); @@ -273,7 +386,7 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode) * table entry, even if the Guest says it's writable. That way * we will come back here when a write does actually occur, so * we can update the Guest's _PAGE_DIRTY flag. */ - *spte = gpte_to_spte(cpu, pte_wrprotect(gpte), 0); + native_set_pte(spte, gpte_to_spte(cpu, pte_wrprotect(gpte), 0)); /* Finally, we write the Guest PTE entry back: we've set the * _PAGE_ACCESSED and maybe the _PAGE_DIRTY flags. */ @@ -301,14 +414,23 @@ static bool page_writable(struct lg_cpu *cpu, unsigned long vaddr) pgd_t *spgd; unsigned long flags; +#ifdef CONFIG_X86_PAE + pmd_t *spmd; +#endif /* Look at the current top level entry: is it present? */ spgd = spgd_addr(cpu, cpu->cpu_pgd, vaddr); if (!(pgd_flags(*spgd) & _PAGE_PRESENT)) return false; +#ifdef CONFIG_X86_PAE + spmd = spmd_addr(cpu, *spgd, vaddr); + if (!(pmd_flags(*spmd) & _PAGE_PRESENT)) + return false; +#endif + /* Check the flags on the pte entry itself: it must be present and * writable. */ - flags = pte_flags(*(spte_addr(*spgd, vaddr))); + flags = pte_flags(*(spte_addr(cpu, *spgd, vaddr))); return (flags & (_PAGE_PRESENT|_PAGE_RW)) == (_PAGE_PRESENT|_PAGE_RW); } @@ -322,8 +444,43 @@ void pin_page(struct lg_cpu *cpu, unsigned long vaddr) kill_guest(cpu, "bad stack page %#lx", vaddr); } +#ifdef CONFIG_X86_PAE +static void release_pmd(pmd_t *spmd) +{ + /* If the entry's not present, there's nothing to release. */ + if (pmd_flags(*spmd) & _PAGE_PRESENT) { + unsigned int i; + pte_t *ptepage = __va(pmd_pfn(*spmd) << PAGE_SHIFT); + /* For each entry in the page, we might need to release it. */ + for (i = 0; i < PTRS_PER_PTE; i++) + release_pte(ptepage[i]); + /* Now we can free the page of PTEs */ + free_page((long)ptepage); + /* And zero out the PMD entry so we never release it twice. */ + native_set_pmd(spmd, __pmd(0)); + } +} + +static void release_pgd(pgd_t *spgd) +{ + /* If the entry's not present, there's nothing to release. */ + if (pgd_flags(*spgd) & _PAGE_PRESENT) { + unsigned int i; + pmd_t *pmdpage = __va(pgd_pfn(*spgd) << PAGE_SHIFT); + + for (i = 0; i < PTRS_PER_PMD; i++) + release_pmd(&pmdpage[i]); + + /* Now we can free the page of PMDs */ + free_page((long)pmdpage); + /* And zero out the PGD entry so we never release it twice. */ + set_pgd(spgd, __pgd(0)); + } +} + +#else /* !CONFIG_X86_PAE */ /*H:450 If we chase down the release_pgd() code, it looks like this: */ -static void release_pgd(struct lguest *lg, pgd_t *spgd) +static void release_pgd(pgd_t *spgd) { /* If the entry's not present, there's nothing to release. */ if (pgd_flags(*spgd) & _PAGE_PRESENT) { @@ -341,7 +498,7 @@ static void release_pgd(struct lguest *lg, pgd_t *spgd) *spgd = __pgd(0); } } - +#endif /*H:445 We saw flush_user_mappings() twice: once from the flush_user_mappings() * hypercall and once in new_pgdir() when we re-used a top-level pgdir page. * It simply releases every PTE page from 0 up to the Guest's kernel address. */ @@ -350,7 +507,7 @@ static void flush_user_mappings(struct lguest *lg, int idx) unsigned int i; /* Release every pgd entry up to the kernel's address. */ for (i = 0; i < pgd_index(lg->kernel_address); i++) - release_pgd(lg, lg->pgdirs[idx].pgdir + i); + release_pgd(lg->pgdirs[idx].pgdir + i); } /*H:440 (v) Flushing (throwing away) page tables, @@ -369,7 +526,9 @@ unsigned long guest_pa(struct lg_cpu *cpu, unsigned long vaddr) { pgd_t gpgd; pte_t gpte; - +#ifdef CONFIG_X86_PAE + pmd_t gpmd; +#endif /* First step: get the top-level Guest page table entry. */ gpgd = lgread(cpu, gpgd_addr(cpu, vaddr), pgd_t); /* Toplevel not present? We can't map it in. */ @@ -378,7 +537,14 @@ unsigned long guest_pa(struct lg_cpu *cpu, unsigned long vaddr) return -1UL; } - gpte = lgread(cpu, gpte_addr(gpgd, vaddr), pte_t); +#ifdef CONFIG_X86_PAE + gpmd = lgread(cpu, gpmd_addr(gpgd, vaddr), pmd_t); + if (!(pmd_flags(gpmd) & _PAGE_PRESENT)) + kill_guest(cpu, "Bad address %#lx", vaddr); + gpte = lgread(cpu, gpte_addr(cpu, gpmd, vaddr), pte_t); +#else + gpte = lgread(cpu, gpte_addr(cpu, gpgd, vaddr), pte_t); +#endif if (!(pte_flags(gpte) & _PAGE_PRESENT)) kill_guest(cpu, "Bad address %#lx", vaddr); @@ -405,6 +571,9 @@ static unsigned int new_pgdir(struct lg_cpu *cpu, int *blank_pgdir) { unsigned int next; +#ifdef CONFIG_X86_PAE + pmd_t *pmd_table; +#endif /* We pick one entry at random to throw out. Choosing the Least * Recently Used might be better, but this is easy. */ @@ -416,10 +585,27 @@ static unsigned int new_pgdir(struct lg_cpu *cpu, /* If the allocation fails, just keep using the one we have */ if (!cpu->lg->pgdirs[next].pgdir) next = cpu->cpu_pgd; - else - /* This is a blank page, so there are no kernel - * mappings: caller must map the stack! */ + else { +#ifdef CONFIG_X86_PAE + /* In PAE mode, allocate a pmd page and populate the + * last pgd entry. */ + pmd_table = (pmd_t *)get_zeroed_page(GFP_KERNEL); + if (!pmd_table) { + free_page((long)cpu->lg->pgdirs[next].pgdir); + set_pgd(cpu->lg->pgdirs[next].pgdir, __pgd(0)); + next = cpu->cpu_pgd; + } else { + set_pgd(cpu->lg->pgdirs[next].pgdir + + SWITCHER_PGD_INDEX, + __pgd(__pa(pmd_table) | _PAGE_PRESENT)); + /* This is a blank page, so there are no kernel + * mappings: caller must map the stack! */ + *blank_pgdir = 1; + } +#else *blank_pgdir = 1; +#endif + } } /* Record which Guest toplevel this shadows. */ cpu->lg->pgdirs[next].gpgdir = gpgdir; @@ -431,7 +617,7 @@ static unsigned int new_pgdir(struct lg_cpu *cpu, /*H:430 (iv) Switching page tables * - * Now we've seen all the page table setting and manipulation, let's see what + * Now we've seen all the page table setting and manipulation, let's see * what happens when the Guest changes page tables (ie. changes the top-level * pgdir). This occurs on almost every context switch. */ void guest_new_pagetable(struct lg_cpu *cpu, unsigned long pgtable) @@ -460,10 +646,25 @@ static void release_all_pagetables(struct lguest *lg) /* Every shadow pagetable this Guest has */ for (i = 0; i < ARRAY_SIZE(lg->pgdirs); i++) - if (lg->pgdirs[i].pgdir) + if (lg->pgdirs[i].pgdir) { +#ifdef CONFIG_X86_PAE + pgd_t *spgd; + pmd_t *pmdpage; + unsigned int k; + + /* Get the last pmd page. */ + spgd = lg->pgdirs[i].pgdir + SWITCHER_PGD_INDEX; + pmdpage = __va(pgd_pfn(*spgd) << PAGE_SHIFT); + + /* And release the pmd entries of that pmd page, + * except for the switcher pmd. */ + for (k = 0; k < SWITCHER_PMD_INDEX; k++) + release_pmd(&pmdpage[k]); +#endif /* Every PGD entry except the Switcher at the top */ for (j = 0; j < SWITCHER_PGD_INDEX; j++) - release_pgd(lg, lg->pgdirs[i].pgdir + j); + release_pgd(lg->pgdirs[i].pgdir + j); + } } /* We also throw away everything when a Guest tells us it's changed a kernel @@ -504,24 +705,37 @@ static void do_set_pte(struct lg_cpu *cpu, int idx, { /* Look up the matching shadow page directory entry. */ pgd_t *spgd = spgd_addr(cpu, idx, vaddr); +#ifdef CONFIG_X86_PAE + pmd_t *spmd; +#endif /* If the top level isn't present, there's no entry to update. */ if (pgd_flags(*spgd) & _PAGE_PRESENT) { - /* Otherwise, we start by releasing the existing entry. */ - pte_t *spte = spte_addr(*spgd, vaddr); - release_pte(*spte); - - /* If they're setting this entry as dirty or accessed, we might - * as well put that entry they've given us in now. This shaves - * 10% off a copy-on-write micro-benchmark. */ - if (pte_flags(gpte) & (_PAGE_DIRTY | _PAGE_ACCESSED)) { - check_gpte(cpu, gpte); - *spte = gpte_to_spte(cpu, gpte, - pte_flags(gpte) & _PAGE_DIRTY); - } else - /* Otherwise kill it and we can demand_page() it in - * later. */ - *spte = __pte(0); +#ifdef CONFIG_X86_PAE + spmd = spmd_addr(cpu, *spgd, vaddr); + if (pmd_flags(*spmd) & _PAGE_PRESENT) { +#endif + /* Otherwise, we start by releasing + * the existing entry. */ + pte_t *spte = spte_addr(cpu, *spgd, vaddr); + release_pte(*spte); + + /* If they're setting this entry as dirty or accessed, + * we might as well put that entry they've given us + * in now. This shaves 10% off a + * copy-on-write micro-benchmark. */ + if (pte_flags(gpte) & (_PAGE_DIRTY | _PAGE_ACCESSED)) { + check_gpte(cpu, gpte); + native_set_pte(spte, + gpte_to_spte(cpu, gpte, + pte_flags(gpte) & _PAGE_DIRTY)); + } else + /* Otherwise kill it and we can demand_page() + * it in later. */ + native_set_pte(spte, __pte(0)); +#ifdef CONFIG_X86_PAE + } +#endif } } @@ -568,12 +782,10 @@ void guest_set_pte(struct lg_cpu *cpu, * * So with that in mind here's our code to to update a (top-level) PGD entry: */ -void guest_set_pmd(struct lguest *lg, unsigned long gpgdir, u32 idx) +void guest_set_pgd(struct lguest *lg, unsigned long gpgdir, u32 idx) { int pgdir; - /* The kernel seems to try to initialize this early on: we ignore its - * attempts to map over the Switcher. */ if (idx >= SWITCHER_PGD_INDEX) return; @@ -581,8 +793,14 @@ void guest_set_pmd(struct lguest *lg, unsigned long gpgdir, u32 idx) pgdir = find_pgdir(lg, gpgdir); if (pgdir < ARRAY_SIZE(lg->pgdirs)) /* ... throw it away. */ - release_pgd(lg, lg->pgdirs[pgdir].pgdir + idx); + release_pgd(lg->pgdirs[pgdir].pgdir + idx); } +#ifdef CONFIG_X86_PAE +void guest_set_pmd(struct lguest *lg, unsigned long pmdp, u32 idx) +{ + guest_pagetable_clear_all(&lg->cpus[0]); +} +#endif /* Once we know how much memory we have we can construct simple identity * (which set virtual == physical) and linear mappings @@ -596,8 +814,16 @@ static unsigned long setup_pagetables(struct lguest *lg, { pgd_t __user *pgdir; pte_t __user *linear; - unsigned int mapped_pages, i, linear_pages, phys_linear; unsigned long mem_base = (unsigned long)lg->mem_base; + unsigned int mapped_pages, i, linear_pages; +#ifdef CONFIG_X86_PAE + pmd_t __user *pmds; + unsigned int j; + pgd_t pgd; + pmd_t pmd; +#else + unsigned int phys_linear; +#endif /* We have mapped_pages frames to map, so we need * linear_pages page tables to map them. */ @@ -610,6 +836,9 @@ static unsigned long setup_pagetables(struct lguest *lg, /* Now we use the next linear_pages pages as pte pages */ linear = (void *)pgdir - linear_pages * PAGE_SIZE; +#ifdef CONFIG_X86_PAE + pmds = (void *)linear - PAGE_SIZE; +#endif /* Linear mapping is easy: put every page's address into the * mapping in order. */ for (i = 0; i < mapped_pages; i++) { @@ -621,6 +850,22 @@ static unsigned long setup_pagetables(struct lguest *lg, /* The top level points to the linear page table pages above. * We setup the identity and linear mappings here. */ +#ifdef CONFIG_X86_PAE + for (i = j = 0; i < mapped_pages && j < PTRS_PER_PMD; + i += PTRS_PER_PTE, j++) { + native_set_pmd(&pmd, __pmd(((unsigned long)(linear + i) + - mem_base) | _PAGE_PRESENT | _PAGE_RW | _PAGE_USER)); + + if (copy_to_user(&pmds[j], &pmd, sizeof(pmd)) != 0) + return -EFAULT; + } + + set_pgd(&pgd, __pgd(((u32)pmds - mem_base) | _PAGE_PRESENT)); + if (copy_to_user(&pgdir[0], &pgd, sizeof(pgd)) != 0) + return -EFAULT; + if (copy_to_user(&pgdir[3], &pgd, sizeof(pgd)) != 0) + return -EFAULT; +#else phys_linear = (unsigned long)linear - mem_base; for (i = 0; i < mapped_pages; i += PTRS_PER_PTE) { pgd_t pgd; @@ -633,6 +878,7 @@ static unsigned long setup_pagetables(struct lguest *lg, &pgd, sizeof(pgd))) return -EFAULT; } +#endif /* We return the top level (guest-physical) address: remember where * this is. */ @@ -648,7 +894,10 @@ int init_guest_pagetable(struct lguest *lg) u64 mem; u32 initrd_size; struct boot_params __user *boot = (struct boot_params *)lg->mem_base; - +#ifdef CONFIG_X86_PAE + pgd_t *pgd; + pmd_t *pmd_table; +#endif /* Get the Guest memory size and the ramdisk size from the boot header * located at lg->mem_base (Guest address 0). */ if (copy_from_user(&mem, &boot->e820_map[0].size, sizeof(mem)) @@ -663,6 +912,15 @@ int init_guest_pagetable(struct lguest *lg) lg->pgdirs[0].pgdir = (pgd_t *)get_zeroed_page(GFP_KERNEL); if (!lg->pgdirs[0].pgdir) return -ENOMEM; +#ifdef CONFIG_X86_PAE + pgd = lg->pgdirs[0].pgdir; + pmd_table = (pmd_t *) get_zeroed_page(GFP_KERNEL); + if (!pmd_table) + return -ENOMEM; + + set_pgd(pgd + SWITCHER_PGD_INDEX, + __pgd(__pa(pmd_table) | _PAGE_PRESENT)); +#endif lg->cpus[0].cpu_pgd = 0; return 0; } @@ -672,17 +930,24 @@ void page_table_guest_data_init(struct lg_cpu *cpu) { /* We get the kernel address: above this is all kernel memory. */ if (get_user(cpu->lg->kernel_address, - &cpu->lg->lguest_data->kernel_address) - /* We tell the Guest that it can't use the top 4MB of virtual - * addresses used by the Switcher. */ - || put_user(4U*1024*1024, &cpu->lg->lguest_data->reserve_mem) - || put_user(cpu->lg->pgdirs[0].gpgdir, &cpu->lg->lguest_data->pgdir)) + &cpu->lg->lguest_data->kernel_address) + /* We tell the Guest that it can't use the top 2 or 4 MB + * of virtual addresses used by the Switcher. */ + || put_user(RESERVE_MEM * 1024 * 1024, + &cpu->lg->lguest_data->reserve_mem) + || put_user(cpu->lg->pgdirs[0].gpgdir, + &cpu->lg->lguest_data->pgdir)) kill_guest(cpu, "bad guest page %p", cpu->lg->lguest_data); /* In flush_user_mappings() we loop from 0 to * "pgd_index(lg->kernel_address)". This assumes it won't hit the * Switcher mappings, so check that now. */ +#ifdef CONFIG_X86_PAE + if (pgd_index(cpu->lg->kernel_address) == SWITCHER_PGD_INDEX && + pmd_index(cpu->lg->kernel_address) == SWITCHER_PMD_INDEX) +#else if (pgd_index(cpu->lg->kernel_address) >= SWITCHER_PGD_INDEX) +#endif kill_guest(cpu, "bad kernel address %#lx", cpu->lg->kernel_address); } @@ -708,16 +973,30 @@ void free_guest_pagetable(struct lguest *lg) void map_switcher_in_guest(struct lg_cpu *cpu, struct lguest_pages *pages) { pte_t *switcher_pte_page = __get_cpu_var(switcher_pte_pages); - pgd_t switcher_pgd; pte_t regs_pte; unsigned long pfn; +#ifdef CONFIG_X86_PAE + pmd_t switcher_pmd; + pmd_t *pmd_table; + + native_set_pmd(&switcher_pmd, pfn_pmd(__pa(switcher_pte_page) >> + PAGE_SHIFT, PAGE_KERNEL_EXEC)); + + pmd_table = __va(pgd_pfn(cpu->lg-> + pgdirs[cpu->cpu_pgd].pgdir[SWITCHER_PGD_INDEX]) + << PAGE_SHIFT); + native_set_pmd(&pmd_table[SWITCHER_PMD_INDEX], switcher_pmd); +#else + pgd_t switcher_pgd; + /* Make the last PGD entry for this Guest point to the Switcher's PTE * page for this CPU (with appropriate flags). */ - switcher_pgd = __pgd(__pa(switcher_pte_page) | __PAGE_KERNEL); + switcher_pgd = __pgd(__pa(switcher_pte_page) | __PAGE_KERNEL_EXEC); cpu->lg->pgdirs[cpu->cpu_pgd].pgdir[SWITCHER_PGD_INDEX] = switcher_pgd; +#endif /* We also change the Switcher PTE page. When we're running the Guest, * we want the Guest's "regs" page to appear where the first Switcher * page for this CPU is. This is an optimization: when the Switcher @@ -726,8 +1005,9 @@ void map_switcher_in_guest(struct lg_cpu *cpu, struct lguest_pages *pages) * page is already mapped there, we don't have to copy them out * again. */ pfn = __pa(cpu->regs_page) >> PAGE_SHIFT; - regs_pte = pfn_pte(pfn, __pgprot(__PAGE_KERNEL)); - switcher_pte_page[(unsigned long)pages/PAGE_SIZE%PTRS_PER_PTE] = regs_pte; + native_set_pte(®s_pte, pfn_pte(pfn, PAGE_KERNEL)); + native_set_pte(&switcher_pte_page[pte_index((unsigned long)pages)], + regs_pte); } /*:*/ @@ -752,21 +1032,21 @@ static __init void populate_switcher_pte_page(unsigned int cpu, /* The first entries are easy: they map the Switcher code. */ for (i = 0; i < pages; i++) { - pte[i] = mk_pte(switcher_page[i], - __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED)); + native_set_pte(&pte[i], mk_pte(switcher_page[i], + __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED))); } /* The only other thing we map is this CPU's pair of pages. */ i = pages + cpu*2; /* First page (Guest registers) is writable from the Guest */ - pte[i] = pfn_pte(page_to_pfn(switcher_page[i]), - __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED|_PAGE_RW)); + native_set_pte(&pte[i], pfn_pte(page_to_pfn(switcher_page[i]), + __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED|_PAGE_RW))); /* The second page contains the "struct lguest_ro_state", and is * read-only. */ - pte[i+1] = pfn_pte(page_to_pfn(switcher_page[i+1]), - __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED)); + native_set_pte(&pte[i+1], pfn_pte(page_to_pfn(switcher_page[i+1]), + __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED))); } /* We've made it through the page table code. Perhaps our tired brains are diff --git a/drivers/lguest/segments.c b/drivers/lguest/segments.c index 7ede64ffeef..482ed5a1875 100644 --- a/drivers/lguest/segments.c +++ b/drivers/lguest/segments.c @@ -150,7 +150,7 @@ void load_guest_gdt_entry(struct lg_cpu *cpu, u32 num, u32 lo, u32 hi) { /* We assume the Guest has the same number of GDT entries as the * Host, otherwise we'd have to dynamically allocate the Guest GDT. */ - if (num > ARRAY_SIZE(cpu->arch.gdt)) + if (num >= ARRAY_SIZE(cpu->arch.gdt)) kill_guest(cpu, "too many gdt entries %i", num); /* Set it up, then fix it. */ diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 4d1d47953fc..7fa620ddeb2 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -845,6 +845,10 @@ static int virtnet_probe(struct virtio_device *vdev) int err; struct net_device *dev; struct virtnet_info *vi; + struct virtqueue *vqs[3]; + vq_callback_t *callbacks[] = { skb_recv_done, skb_xmit_done, NULL}; + const char *names[] = { "input", "output", "control" }; + int nvqs; /* Allocate ourselves a network device with room for our info */ dev = alloc_etherdev(sizeof(struct virtnet_info)); @@ -905,25 +909,19 @@ static int virtnet_probe(struct virtio_device *vdev) if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF)) vi->mergeable_rx_bufs = true; - /* We expect two virtqueues, receive then send. */ - vi->rvq = vdev->config->find_vq(vdev, 0, skb_recv_done); - if (IS_ERR(vi->rvq)) { - err = PTR_ERR(vi->rvq); + /* We expect two virtqueues, receive then send, + * and optionally control. */ + nvqs = virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ? 3 : 2; + + err = vdev->config->find_vqs(vdev, nvqs, vqs, callbacks, names); + if (err) goto free; - } - vi->svq = vdev->config->find_vq(vdev, 1, skb_xmit_done); - if (IS_ERR(vi->svq)) { - err = PTR_ERR(vi->svq); - goto free_recv; - } + vi->rvq = vqs[0]; + vi->svq = vqs[1]; if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) { - vi->cvq = vdev->config->find_vq(vdev, 2, NULL); - if (IS_ERR(vi->cvq)) { - err = PTR_ERR(vi->svq); - goto free_send; - } + vi->cvq = vqs[2]; if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VLAN)) dev->features |= NETIF_F_HW_VLAN_FILTER; @@ -941,7 +939,7 @@ static int virtnet_probe(struct virtio_device *vdev) err = register_netdev(dev); if (err) { pr_debug("virtio_net: registering device failed\n"); - goto free_ctrl; + goto free_vqs; } /* Last of all, set up some receive buffers. */ @@ -962,13 +960,8 @@ static int virtnet_probe(struct virtio_device *vdev) unregister: unregister_netdev(dev); -free_ctrl: - if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) - vdev->config->del_vq(vi->cvq); -free_send: - vdev->config->del_vq(vi->svq); -free_recv: - vdev->config->del_vq(vi->rvq); +free_vqs: + vdev->config->del_vqs(vdev); free: free_netdev(dev); return err; @@ -994,12 +987,10 @@ static void virtnet_remove(struct virtio_device *vdev) BUG_ON(vi->num != 0); - vdev->config->del_vq(vi->svq); - vdev->config->del_vq(vi->rvq); - if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) - vdev->config->del_vq(vi->cvq); unregister_netdev(vi->dev); + vdev->config->del_vqs(vi->vdev); + while (vi->pages) __free_pages(get_a_page(vi, GFP_KERNEL), 0); diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c index cbc8566fab7..e38e5d306fa 100644 --- a/drivers/s390/kvm/kvm_virtio.c +++ b/drivers/s390/kvm/kvm_virtio.c @@ -173,8 +173,9 @@ static void kvm_notify(struct virtqueue *vq) * this device and sets it up. */ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, - unsigned index, - void (*callback)(struct virtqueue *vq)) + unsigned index, + void (*callback)(struct virtqueue *vq), + const char *name) { struct kvm_device *kdev = to_kvmdev(vdev); struct kvm_vqconfig *config; @@ -194,7 +195,7 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, vq = vring_new_virtqueue(config->num, KVM_S390_VIRTIO_RING_ALIGN, vdev, (void *) config->address, - kvm_notify, callback); + kvm_notify, callback, name); if (!vq) { err = -ENOMEM; goto unmap; @@ -226,6 +227,38 @@ static void kvm_del_vq(struct virtqueue *vq) KVM_S390_VIRTIO_RING_ALIGN)); } +static void kvm_del_vqs(struct virtio_device *vdev) +{ + struct virtqueue *vq, *n; + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + kvm_del_vq(vq); +} + +static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + struct kvm_device *kdev = to_kvmdev(vdev); + int i; + + /* We must have this many virtqueues. */ + if (nvqs > kdev->desc->num_vq) + return -ENOENT; + + for (i = 0; i < nvqs; ++i) { + vqs[i] = kvm_find_vq(vdev, i, callbacks[i], names[i]); + if (IS_ERR(vqs[i])) + goto error; + } + return 0; + +error: + kvm_del_vqs(vdev); + return PTR_ERR(vqs[i]); +} + /* * The config ops structure as defined by virtio config */ @@ -237,8 +270,8 @@ static struct virtio_config_ops kvm_vq_configspace_ops = { .get_status = kvm_get_status, .set_status = kvm_set_status, .reset = kvm_reset, - .find_vq = kvm_find_vq, - .del_vq = kvm_del_vq, + .find_vqs = kvm_find_vqs, + .del_vqs = kvm_del_vqs, }; /* diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c index 35e8eb02b9e..e4e4d433b00 100644 --- a/drivers/video/aty/aty128fb.c +++ b/drivers/video/aty/aty128fb.c @@ -354,7 +354,7 @@ static int default_crt_on __devinitdata = 0; static int default_lcd_on __devinitdata = 1; #ifdef CONFIG_MTRR -static int mtrr = 1; +static bool mtrr = true; #endif #ifdef CONFIG_PMAC_BACKLIGHT diff --git a/drivers/video/cyber2000fb.c b/drivers/video/cyber2000fb.c index 83c5cefc266..da7c01b39be 100644 --- a/drivers/video/cyber2000fb.c +++ b/drivers/video/cyber2000fb.c @@ -1736,10 +1736,8 @@ static int __init cyber2000fb_init(void) #ifdef CONFIG_ARCH_SHARK err = cyberpro_vl_probe(); - if (!err) { + if (!err) ret = 0; - __module_get(THIS_MODULE); - } #endif #ifdef CONFIG_PCI err = pci_register_driver(&cyberpro_driver); @@ -1749,14 +1747,15 @@ static int __init cyber2000fb_init(void) return ret ? err : 0; } +module_init(cyber2000fb_init); +#ifndef CONFIG_ARCH_SHARK static void __exit cyberpro_exit(void) { pci_unregister_driver(&cyberpro_driver); } - -module_init(cyber2000fb_init); module_exit(cyberpro_exit); +#endif MODULE_AUTHOR("Russell King"); MODULE_DESCRIPTION("CyberPro 2000, 2010 and 5000 framebuffer driver"); diff --git a/drivers/video/uvesafb.c b/drivers/video/uvesafb.c index 421770b5e6a..ca5b4643a40 100644 --- a/drivers/video/uvesafb.c +++ b/drivers/video/uvesafb.c @@ -45,7 +45,7 @@ static struct fb_fix_screeninfo uvesafb_fix __devinitdata = { static int mtrr __devinitdata = 3; /* enable mtrr by default */ static int blank = 1; /* enable blanking by default */ static int ypan = 1; /* 0: scroll, 1: ypan, 2: ywrap */ -static int pmi_setpal __devinitdata = 1; /* use PMI for palette changes */ +static bool pmi_setpal __devinitdata = true; /* use PMI for palette changes */ static int nocrtc __devinitdata; /* ignore CRTC settings */ static int noedid __devinitdata; /* don't try DDC transfers */ static int vram_remap __devinitdata; /* set amt. of memory to be used */ @@ -2002,11 +2002,7 @@ static void __devexit uvesafb_exit(void) module_exit(uvesafb_exit); -static int param_get_scroll(char *buffer, struct kernel_param *kp) -{ - return 0; -} - +#define param_get_scroll NULL static int param_set_scroll(const char *val, struct kernel_param *kp) { ypan = 0; @@ -2017,6 +2013,8 @@ static int param_set_scroll(const char *val, struct kernel_param *kp) ypan = 1; else if (!strcmp(val, "ywrap")) ypan = 2; + else + return -EINVAL; return 0; } diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c index 018c070a357..3a43ebf83a4 100644 --- a/drivers/virtio/virtio.c +++ b/drivers/virtio/virtio.c @@ -31,21 +31,37 @@ static ssize_t modalias_show(struct device *_d, return sprintf(buf, "virtio:d%08Xv%08X\n", dev->id.device, dev->id.vendor); } +static ssize_t features_show(struct device *_d, + struct device_attribute *attr, char *buf) +{ + struct virtio_device *dev = container_of(_d, struct virtio_device, dev); + unsigned int i; + ssize_t len = 0; + + /* We actually represent this as a bitstring, as it could be + * arbitrary length in future. */ + for (i = 0; i < ARRAY_SIZE(dev->features)*BITS_PER_LONG; i++) + len += sprintf(buf+len, "%c", + test_bit(i, dev->features) ? '1' : '0'); + len += sprintf(buf+len, "\n"); + return len; +} static struct device_attribute virtio_dev_attrs[] = { __ATTR_RO(device), __ATTR_RO(vendor), __ATTR_RO(status), __ATTR_RO(modalias), + __ATTR_RO(features), __ATTR_NULL }; static inline int virtio_id_match(const struct virtio_device *dev, const struct virtio_device_id *id) { - if (id->device != dev->id.device) + if (id->device != dev->id.device && id->device != VIRTIO_DEV_ANY_ID) return 0; - return id->vendor == VIRTIO_DEV_ANY_ID || id->vendor != dev->id.vendor; + return id->vendor == VIRTIO_DEV_ANY_ID || id->vendor == dev->id.vendor; } /* This looks through all the IDs a driver claims to support. If any of them @@ -118,13 +134,14 @@ static int virtio_dev_probe(struct device *_d) if (device_features & (1 << i)) set_bit(i, dev->features); + dev->config->finalize_features(dev); + err = drv->probe(dev); if (err) add_status(dev, VIRTIO_CONFIG_S_FAILED); - else { - dev->config->finalize_features(dev); + else add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK); - } + return err; } @@ -185,6 +202,8 @@ int register_virtio_device(struct virtio_device *dev) /* Acknowledge that we've seen the device. */ add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE); + INIT_LIST_HEAD(&dev->vqs); + /* device_register() causes the bus infrastructure to look for a * matching driver. */ err = device_register(&dev->dev); diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 9c76a061a04..26b27826479 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -204,6 +204,9 @@ static int balloon(void *_vballoon) static int virtballoon_probe(struct virtio_device *vdev) { struct virtio_balloon *vb; + struct virtqueue *vqs[2]; + vq_callback_t *callbacks[] = { balloon_ack, balloon_ack }; + const char *names[] = { "inflate", "deflate" }; int err; vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL); @@ -218,22 +221,17 @@ static int virtballoon_probe(struct virtio_device *vdev) vb->vdev = vdev; /* We expect two virtqueues. */ - vb->inflate_vq = vdev->config->find_vq(vdev, 0, balloon_ack); - if (IS_ERR(vb->inflate_vq)) { - err = PTR_ERR(vb->inflate_vq); + err = vdev->config->find_vqs(vdev, 2, vqs, callbacks, names); + if (err) goto out_free_vb; - } - vb->deflate_vq = vdev->config->find_vq(vdev, 1, balloon_ack); - if (IS_ERR(vb->deflate_vq)) { - err = PTR_ERR(vb->deflate_vq); - goto out_del_inflate_vq; - } + vb->inflate_vq = vqs[0]; + vb->deflate_vq = vqs[1]; vb->thread = kthread_run(balloon, vb, "vballoon"); if (IS_ERR(vb->thread)) { err = PTR_ERR(vb->thread); - goto out_del_deflate_vq; + goto out_del_vqs; } vb->tell_host_first @@ -241,10 +239,8 @@ static int virtballoon_probe(struct virtio_device *vdev) return 0; -out_del_deflate_vq: - vdev->config->del_vq(vb->deflate_vq); -out_del_inflate_vq: - vdev->config->del_vq(vb->inflate_vq); +out_del_vqs: + vdev->config->del_vqs(vdev); out_free_vb: kfree(vb); out: @@ -264,8 +260,7 @@ static void virtballoon_remove(struct virtio_device *vdev) /* Now we reset the device so we can clean up the queues. */ vdev->config->reset(vdev); - vdev->config->del_vq(vb->deflate_vq); - vdev->config->del_vq(vb->inflate_vq); + vdev->config->del_vqs(vdev); kfree(vb); } diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c index 330aacbdec1..193c8f0e5cc 100644 --- a/drivers/virtio/virtio_pci.c +++ b/drivers/virtio/virtio_pci.c @@ -42,6 +42,26 @@ struct virtio_pci_device /* a list of queues so we can dispatch IRQs */ spinlock_t lock; struct list_head virtqueues; + + /* MSI-X support */ + int msix_enabled; + int intx_enabled; + struct msix_entry *msix_entries; + /* Name strings for interrupts. This size should be enough, + * and I'm too lazy to allocate each name separately. */ + char (*msix_names)[256]; + /* Number of available vectors */ + unsigned msix_vectors; + /* Vectors allocated */ + unsigned msix_used_vectors; +}; + +/* Constants for MSI-X */ +/* Use first vector for configuration changes, second and the rest for + * virtqueues Thus, we need at least 2 vectors for MSI. */ +enum { + VP_MSIX_CONFIG_VECTOR = 0, + VP_MSIX_VQ_VECTOR = 1, }; struct virtio_pci_vq_info @@ -60,6 +80,9 @@ struct virtio_pci_vq_info /* the list node for the virtqueues list */ struct list_head node; + + /* MSI-X vector (or none) */ + unsigned vector; }; /* Qumranet donated their vendor ID for devices 0x1000 thru 0x10FF. */ @@ -109,7 +132,8 @@ static void vp_get(struct virtio_device *vdev, unsigned offset, void *buf, unsigned len) { struct virtio_pci_device *vp_dev = to_vp_device(vdev); - void __iomem *ioaddr = vp_dev->ioaddr + VIRTIO_PCI_CONFIG + offset; + void __iomem *ioaddr = vp_dev->ioaddr + + VIRTIO_PCI_CONFIG(vp_dev) + offset; u8 *ptr = buf; int i; @@ -123,7 +147,8 @@ static void vp_set(struct virtio_device *vdev, unsigned offset, const void *buf, unsigned len) { struct virtio_pci_device *vp_dev = to_vp_device(vdev); - void __iomem *ioaddr = vp_dev->ioaddr + VIRTIO_PCI_CONFIG + offset; + void __iomem *ioaddr = vp_dev->ioaddr + + VIRTIO_PCI_CONFIG(vp_dev) + offset; const u8 *ptr = buf; int i; @@ -164,6 +189,37 @@ static void vp_notify(struct virtqueue *vq) iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_NOTIFY); } +/* Handle a configuration change: Tell driver if it wants to know. */ +static irqreturn_t vp_config_changed(int irq, void *opaque) +{ + struct virtio_pci_device *vp_dev = opaque; + struct virtio_driver *drv; + drv = container_of(vp_dev->vdev.dev.driver, + struct virtio_driver, driver); + + if (drv && drv->config_changed) + drv->config_changed(&vp_dev->vdev); + return IRQ_HANDLED; +} + +/* Notify all virtqueues on an interrupt. */ +static irqreturn_t vp_vring_interrupt(int irq, void *opaque) +{ + struct virtio_pci_device *vp_dev = opaque; + struct virtio_pci_vq_info *info; + irqreturn_t ret = IRQ_NONE; + unsigned long flags; + + spin_lock_irqsave(&vp_dev->lock, flags); + list_for_each_entry(info, &vp_dev->virtqueues, node) { + if (vring_interrupt(irq, info->vq) == IRQ_HANDLED) + ret = IRQ_HANDLED; + } + spin_unlock_irqrestore(&vp_dev->lock, flags); + + return ret; +} + /* A small wrapper to also acknowledge the interrupt when it's handled. * I really need an EIO hook for the vring so I can ack the interrupt once we * know that we'll be handling the IRQ but before we invoke the callback since @@ -173,9 +229,6 @@ static void vp_notify(struct virtqueue *vq) static irqreturn_t vp_interrupt(int irq, void *opaque) { struct virtio_pci_device *vp_dev = opaque; - struct virtio_pci_vq_info *info; - irqreturn_t ret = IRQ_NONE; - unsigned long flags; u8 isr; /* reading the ISR has the effect of also clearing it so it's very @@ -187,34 +240,137 @@ static irqreturn_t vp_interrupt(int irq, void *opaque) return IRQ_NONE; /* Configuration change? Tell driver if it wants to know. */ - if (isr & VIRTIO_PCI_ISR_CONFIG) { - struct virtio_driver *drv; - drv = container_of(vp_dev->vdev.dev.driver, - struct virtio_driver, driver); + if (isr & VIRTIO_PCI_ISR_CONFIG) + vp_config_changed(irq, opaque); - if (drv && drv->config_changed) - drv->config_changed(&vp_dev->vdev); + return vp_vring_interrupt(irq, opaque); +} + +static void vp_free_vectors(struct virtio_device *vdev) +{ + struct virtio_pci_device *vp_dev = to_vp_device(vdev); + int i; + + if (vp_dev->intx_enabled) { + free_irq(vp_dev->pci_dev->irq, vp_dev); + vp_dev->intx_enabled = 0; } - spin_lock_irqsave(&vp_dev->lock, flags); - list_for_each_entry(info, &vp_dev->virtqueues, node) { - if (vring_interrupt(irq, info->vq) == IRQ_HANDLED) - ret = IRQ_HANDLED; + for (i = 0; i < vp_dev->msix_used_vectors; ++i) + free_irq(vp_dev->msix_entries[i].vector, vp_dev); + vp_dev->msix_used_vectors = 0; + + if (vp_dev->msix_enabled) { + /* Disable the vector used for configuration */ + iowrite16(VIRTIO_MSI_NO_VECTOR, + vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR); + /* Flush the write out to device */ + ioread16(vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR); + + vp_dev->msix_enabled = 0; + pci_disable_msix(vp_dev->pci_dev); } - spin_unlock_irqrestore(&vp_dev->lock, flags); +} - return ret; +static int vp_enable_msix(struct pci_dev *dev, struct msix_entry *entries, + int *options, int noptions) +{ + int i; + for (i = 0; i < noptions; ++i) + if (!pci_enable_msix(dev, entries, options[i])) + return options[i]; + return -EBUSY; +} + +static int vp_request_vectors(struct virtio_device *vdev, unsigned max_vqs) +{ + struct virtio_pci_device *vp_dev = to_vp_device(vdev); + const char *name = dev_name(&vp_dev->vdev.dev); + unsigned i, v; + int err = -ENOMEM; + /* We want at most one vector per queue and one for config changes. + * Fallback to separate vectors for config and a shared for queues. + * Finally fall back to regular interrupts. */ + int options[] = { max_vqs + 1, 2 }; + int nvectors = max(options[0], options[1]); + + vp_dev->msix_entries = kmalloc(nvectors * sizeof *vp_dev->msix_entries, + GFP_KERNEL); + if (!vp_dev->msix_entries) + goto error_entries; + vp_dev->msix_names = kmalloc(nvectors * sizeof *vp_dev->msix_names, + GFP_KERNEL); + if (!vp_dev->msix_names) + goto error_names; + + for (i = 0; i < nvectors; ++i) + vp_dev->msix_entries[i].entry = i; + + err = vp_enable_msix(vp_dev->pci_dev, vp_dev->msix_entries, + options, ARRAY_SIZE(options)); + if (err < 0) { + /* Can't allocate enough MSI-X vectors, use regular interrupt */ + vp_dev->msix_vectors = 0; + err = request_irq(vp_dev->pci_dev->irq, vp_interrupt, + IRQF_SHARED, name, vp_dev); + if (err) + goto error_irq; + vp_dev->intx_enabled = 1; + } else { + vp_dev->msix_vectors = err; + vp_dev->msix_enabled = 1; + + /* Set the vector used for configuration */ + v = vp_dev->msix_used_vectors; + snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names, + "%s-config", name); + err = request_irq(vp_dev->msix_entries[v].vector, + vp_config_changed, 0, vp_dev->msix_names[v], + vp_dev); + if (err) + goto error_irq; + ++vp_dev->msix_used_vectors; + + iowrite16(v, vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR); + /* Verify we had enough resources to assign the vector */ + v = ioread16(vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR); + if (v == VIRTIO_MSI_NO_VECTOR) { + err = -EBUSY; + goto error_irq; + } + } + + if (vp_dev->msix_vectors && vp_dev->msix_vectors != max_vqs + 1) { + /* Shared vector for all VQs */ + v = vp_dev->msix_used_vectors; + snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names, + "%s-virtqueues", name); + err = request_irq(vp_dev->msix_entries[v].vector, + vp_vring_interrupt, 0, vp_dev->msix_names[v], + vp_dev); + if (err) + goto error_irq; + ++vp_dev->msix_used_vectors; + } + return 0; +error_irq: + vp_free_vectors(vdev); + kfree(vp_dev->msix_names); +error_names: + kfree(vp_dev->msix_entries); +error_entries: + return err; } -/* the config->find_vq() implementation */ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index, - void (*callback)(struct virtqueue *vq)) + void (*callback)(struct virtqueue *vq), + const char *name) { struct virtio_pci_device *vp_dev = to_vp_device(vdev); struct virtio_pci_vq_info *info; struct virtqueue *vq; unsigned long flags, size; - u16 num; + u16 num, vector; int err; /* Select the queue we're interested in */ @@ -233,6 +389,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index, info->queue_index = index; info->num = num; + info->vector = VIRTIO_MSI_NO_VECTOR; size = PAGE_ALIGN(vring_size(num, VIRTIO_PCI_VRING_ALIGN)); info->queue = alloc_pages_exact(size, GFP_KERNEL|__GFP_ZERO); @@ -247,7 +404,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index, /* create the vring */ vq = vring_new_virtqueue(info->num, VIRTIO_PCI_VRING_ALIGN, - vdev, info->queue, vp_notify, callback); + vdev, info->queue, vp_notify, callback, name); if (!vq) { err = -ENOMEM; goto out_activate_queue; @@ -256,12 +413,43 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index, vq->priv = info; info->vq = vq; + /* allocate per-vq vector if available and necessary */ + if (callback && vp_dev->msix_used_vectors < vp_dev->msix_vectors) { + vector = vp_dev->msix_used_vectors; + snprintf(vp_dev->msix_names[vector], sizeof *vp_dev->msix_names, + "%s-%s", dev_name(&vp_dev->vdev.dev), name); + err = request_irq(vp_dev->msix_entries[vector].vector, + vring_interrupt, 0, + vp_dev->msix_names[vector], vq); + if (err) + goto out_request_irq; + info->vector = vector; + ++vp_dev->msix_used_vectors; + } else + vector = VP_MSIX_VQ_VECTOR; + + if (callback && vp_dev->msix_enabled) { + iowrite16(vector, vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR); + vector = ioread16(vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR); + if (vector == VIRTIO_MSI_NO_VECTOR) { + err = -EBUSY; + goto out_assign; + } + } + spin_lock_irqsave(&vp_dev->lock, flags); list_add(&info->node, &vp_dev->virtqueues); spin_unlock_irqrestore(&vp_dev->lock, flags); return vq; +out_assign: + if (info->vector != VIRTIO_MSI_NO_VECTOR) { + free_irq(vp_dev->msix_entries[info->vector].vector, vq); + --vp_dev->msix_used_vectors; + } +out_request_irq: + vring_del_virtqueue(vq); out_activate_queue: iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN); free_pages_exact(info->queue, size); @@ -270,21 +458,27 @@ out_info: return ERR_PTR(err); } -/* the config->del_vq() implementation */ static void vp_del_vq(struct virtqueue *vq) { struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev); struct virtio_pci_vq_info *info = vq->priv; - unsigned long flags, size; + unsigned long size; - spin_lock_irqsave(&vp_dev->lock, flags); - list_del(&info->node); - spin_unlock_irqrestore(&vp_dev->lock, flags); + iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_SEL); + + if (info->vector != VIRTIO_MSI_NO_VECTOR) + free_irq(vp_dev->msix_entries[info->vector].vector, vq); + + if (vp_dev->msix_enabled) { + iowrite16(VIRTIO_MSI_NO_VECTOR, + vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR); + /* Flush the write out to device */ + ioread8(vp_dev->ioaddr + VIRTIO_PCI_ISR); + } vring_del_virtqueue(vq); /* Select and deactivate the queue */ - iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_SEL); iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN); size = PAGE_ALIGN(vring_size(info->num, VIRTIO_PCI_VRING_ALIGN)); @@ -292,14 +486,57 @@ static void vp_del_vq(struct virtqueue *vq) kfree(info); } +/* the config->del_vqs() implementation */ +static void vp_del_vqs(struct virtio_device *vdev) +{ + struct virtqueue *vq, *n; + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + vp_del_vq(vq); + + vp_free_vectors(vdev); +} + +/* the config->find_vqs() implementation */ +static int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + int vectors = 0; + int i, err; + + /* How many vectors would we like? */ + for (i = 0; i < nvqs; ++i) + if (callbacks[i]) + ++vectors; + + err = vp_request_vectors(vdev, vectors); + if (err) + goto error_request; + + for (i = 0; i < nvqs; ++i) { + vqs[i] = vp_find_vq(vdev, i, callbacks[i], names[i]); + if (IS_ERR(vqs[i])) + goto error_find; + } + return 0; + +error_find: + vp_del_vqs(vdev); + +error_request: + return PTR_ERR(vqs[i]); +} + static struct virtio_config_ops virtio_pci_config_ops = { .get = vp_get, .set = vp_set, .get_status = vp_get_status, .set_status = vp_set_status, .reset = vp_reset, - .find_vq = vp_find_vq, - .del_vq = vp_del_vq, + .find_vqs = vp_find_vqs, + .del_vqs = vp_del_vqs, .get_features = vp_get_features, .finalize_features = vp_finalize_features, }; @@ -310,7 +547,7 @@ static void virtio_pci_release_dev(struct device *_d) struct virtio_pci_device *vp_dev = to_vp_device(dev); struct pci_dev *pci_dev = vp_dev->pci_dev; - free_irq(pci_dev->irq, vp_dev); + vp_del_vqs(dev); pci_set_drvdata(pci_dev, NULL); pci_iounmap(pci_dev, vp_dev->ioaddr); pci_release_regions(pci_dev); @@ -369,21 +606,13 @@ static int __devinit virtio_pci_probe(struct pci_dev *pci_dev, vp_dev->vdev.id.vendor = pci_dev->subsystem_vendor; vp_dev->vdev.id.device = pci_dev->subsystem_device; - /* register a handler for the queue with the PCI device's interrupt */ - err = request_irq(vp_dev->pci_dev->irq, vp_interrupt, IRQF_SHARED, - dev_name(&vp_dev->vdev.dev), vp_dev); - if (err) - goto out_set_drvdata; - /* finally register the virtio device */ err = register_virtio_device(&vp_dev->vdev); if (err) - goto out_req_irq; + goto out_set_drvdata; return 0; -out_req_irq: - free_irq(pci_dev->irq, vp_dev); out_set_drvdata: pci_set_drvdata(pci_dev, NULL); pci_iounmap(pci_dev, vp_dev->ioaddr); diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 5c52369ab9b..a882f260651 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -23,21 +23,30 @@ #ifdef DEBUG /* For development, we want to crash whenever the ring is screwed. */ -#define BAD_RING(_vq, fmt...) \ - do { dev_err(&(_vq)->vq.vdev->dev, fmt); BUG(); } while(0) +#define BAD_RING(_vq, fmt, args...) \ + do { \ + dev_err(&(_vq)->vq.vdev->dev, \ + "%s:"fmt, (_vq)->vq.name, ##args); \ + BUG(); \ + } while (0) /* Caller is supposed to guarantee no reentry. */ #define START_USE(_vq) \ do { \ if ((_vq)->in_use) \ - panic("in_use = %i\n", (_vq)->in_use); \ + panic("%s:in_use = %i\n", \ + (_vq)->vq.name, (_vq)->in_use); \ (_vq)->in_use = __LINE__; \ mb(); \ - } while(0) + } while (0) #define END_USE(_vq) \ do { BUG_ON(!(_vq)->in_use); (_vq)->in_use = 0; mb(); } while(0) #else -#define BAD_RING(_vq, fmt...) \ - do { dev_err(&_vq->vq.vdev->dev, fmt); (_vq)->broken = true; } while(0) +#define BAD_RING(_vq, fmt, args...) \ + do { \ + dev_err(&_vq->vq.vdev->dev, \ + "%s:"fmt, (_vq)->vq.name, ##args); \ + (_vq)->broken = true; \ + } while (0) #define START_USE(vq) #define END_USE(vq) #endif @@ -52,6 +61,9 @@ struct vring_virtqueue /* Other side has made a mess, don't try any more. */ bool broken; + /* Host supports indirect buffers */ + bool indirect; + /* Number of free buffers */ unsigned int num_free; /* Head of free buffer list. */ @@ -76,6 +88,55 @@ struct vring_virtqueue #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq) +/* Set up an indirect table of descriptors and add it to the queue. */ +static int vring_add_indirect(struct vring_virtqueue *vq, + struct scatterlist sg[], + unsigned int out, + unsigned int in) +{ + struct vring_desc *desc; + unsigned head; + int i; + + desc = kmalloc((out + in) * sizeof(struct vring_desc), GFP_ATOMIC); + if (!desc) + return vq->vring.num; + + /* Transfer entries from the sg list into the indirect page */ + for (i = 0; i < out; i++) { + desc[i].flags = VRING_DESC_F_NEXT; + desc[i].addr = sg_phys(sg); + desc[i].len = sg->length; + desc[i].next = i+1; + sg++; + } + for (; i < (out + in); i++) { + desc[i].flags = VRING_DESC_F_NEXT|VRING_DESC_F_WRITE; + desc[i].addr = sg_phys(sg); + desc[i].len = sg->length; + desc[i].next = i+1; + sg++; + } + + /* Last one doesn't continue. */ + desc[i-1].flags &= ~VRING_DESC_F_NEXT; + desc[i-1].next = 0; + + /* We're about to use a buffer */ + vq->num_free--; + + /* Use a single buffer which doesn't continue */ + head = vq->free_head; + vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; + vq->vring.desc[head].addr = virt_to_phys(desc); + vq->vring.desc[head].len = i * sizeof(struct vring_desc); + + /* Update free pointer */ + vq->free_head = vq->vring.desc[head].next; + + return head; +} + static int vring_add_buf(struct virtqueue *_vq, struct scatterlist sg[], unsigned int out, @@ -85,12 +146,21 @@ static int vring_add_buf(struct virtqueue *_vq, struct vring_virtqueue *vq = to_vvq(_vq); unsigned int i, avail, head, uninitialized_var(prev); + START_USE(vq); + BUG_ON(data == NULL); + + /* If the host supports indirect descriptor tables, and we have multiple + * buffers, then go indirect. FIXME: tune this threshold */ + if (vq->indirect && (out + in) > 1 && vq->num_free) { + head = vring_add_indirect(vq, sg, out, in); + if (head != vq->vring.num) + goto add_head; + } + BUG_ON(out + in > vq->vring.num); BUG_ON(out + in == 0); - START_USE(vq); - if (vq->num_free < out + in) { pr_debug("Can't add buf len %i - avail = %i\n", out + in, vq->num_free); @@ -127,6 +197,7 @@ static int vring_add_buf(struct virtqueue *_vq, /* Update free pointer */ vq->free_head = i; +add_head: /* Set token. */ vq->data[head] = data; @@ -170,6 +241,11 @@ static void detach_buf(struct vring_virtqueue *vq, unsigned int head) /* Put back on free list: find end */ i = head; + + /* Free the indirect table */ + if (vq->vring.desc[i].flags & VRING_DESC_F_INDIRECT) + kfree(phys_to_virt(vq->vring.desc[i].addr)); + while (vq->vring.desc[i].flags & VRING_DESC_F_NEXT) { i = vq->vring.desc[i].next; vq->num_free++; @@ -284,7 +360,8 @@ struct virtqueue *vring_new_virtqueue(unsigned int num, struct virtio_device *vdev, void *pages, void (*notify)(struct virtqueue *), - void (*callback)(struct virtqueue *)) + void (*callback)(struct virtqueue *), + const char *name) { struct vring_virtqueue *vq; unsigned int i; @@ -303,14 +380,18 @@ struct virtqueue *vring_new_virtqueue(unsigned int num, vq->vq.callback = callback; vq->vq.vdev = vdev; vq->vq.vq_ops = &vring_vq_ops; + vq->vq.name = name; vq->notify = notify; vq->broken = false; vq->last_used_idx = 0; vq->num_added = 0; + list_add_tail(&vq->vq.list, &vdev->vqs); #ifdef DEBUG vq->in_use = false; #endif + vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC); + /* No callback? Tell other side not to bother us. */ if (!callback) vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT; @@ -327,6 +408,7 @@ EXPORT_SYMBOL_GPL(vring_new_virtqueue); void vring_del_virtqueue(struct virtqueue *vq) { + list_del(&vq->list); kfree(to_vvq(vq)); } EXPORT_SYMBOL_GPL(vring_del_virtqueue); @@ -338,6 +420,8 @@ void vring_transport_features(struct virtio_device *vdev) for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++) { switch (i) { + case VIRTIO_RING_F_INDIRECT_DESC: + break; default: /* We don't understand this bit. */ clear_bit(i, vdev->features); |