aboutsummaryrefslogtreecommitdiff
path: root/drivers/dma/ioat
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/dma/ioat')
-rw-r--r--drivers/dma/ioat/dca.c11
-rw-r--r--drivers/dma/ioat/dma.c130
-rw-r--r--drivers/dma/ioat/dma.h65
-rw-r--r--drivers/dma/ioat/dma_v2.c126
-rw-r--r--drivers/dma/ioat/dma_v2.h2
-rw-r--r--drivers/dma/ioat/dma_v3.c1128
-rw-r--r--drivers/dma/ioat/hw.h116
-rw-r--r--drivers/dma/ioat/pci.c45
-rw-r--r--drivers/dma/ioat/registers.h4
9 files changed, 1077 insertions, 550 deletions
diff --git a/drivers/dma/ioat/dca.c b/drivers/dma/ioat/dca.c
index 9b041858d10..9e84d5bc930 100644
--- a/drivers/dma/ioat/dca.c
+++ b/drivers/dma/ioat/dca.c
@@ -470,8 +470,10 @@ struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase)
}
if (!dca2_tag_map_valid(ioatdca->tag_map)) {
- dev_err(&pdev->dev, "APICID_TAG_MAP set incorrectly by BIOS, "
- "disabling DCA\n");
+ WARN_TAINT_ONCE(1, TAINT_FIRMWARE_WORKAROUND,
+ "%s %s: APICID_TAG_MAP set incorrectly by BIOS, disabling DCA\n",
+ dev_driver_string(&pdev->dev),
+ dev_name(&pdev->dev));
free_dca_provider(dca);
return NULL;
}
@@ -689,7 +691,10 @@ struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase)
}
if (dca3_tag_map_invalid(ioatdca->tag_map)) {
- dev_err(&pdev->dev, "APICID_TAG_MAP set incorrectly by BIOS, disabling DCA\n");
+ WARN_TAINT_ONCE(1, TAINT_FIRMWARE_WORKAROUND,
+ "%s %s: APICID_TAG_MAP set incorrectly by BIOS, disabling DCA\n",
+ dev_driver_string(&pdev->dev),
+ dev_name(&pdev->dev));
free_dca_provider(dca);
return NULL;
}
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 1a68a8ba87e..4e3549a1613 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -77,7 +77,8 @@ static irqreturn_t ioat_dma_do_interrupt(int irq, void *data)
attnstatus = readl(instance->reg_base + IOAT_ATTNSTATUS_OFFSET);
for_each_set_bit(bit, &attnstatus, BITS_PER_LONG) {
chan = ioat_chan_by_index(instance, bit);
- tasklet_schedule(&chan->cleanup_task);
+ if (test_bit(IOAT_RUN, &chan->state))
+ tasklet_schedule(&chan->cleanup_task);
}
writeb(intrctrl, instance->reg_base + IOAT_INTRCTRL_OFFSET);
@@ -93,7 +94,8 @@ static irqreturn_t ioat_dma_do_interrupt_msix(int irq, void *data)
{
struct ioat_chan_common *chan = data;
- tasklet_schedule(&chan->cleanup_task);
+ if (test_bit(IOAT_RUN, &chan->state))
+ tasklet_schedule(&chan->cleanup_task);
return IRQ_HANDLED;
}
@@ -116,7 +118,6 @@ void ioat_init_channel(struct ioatdma_device *device, struct ioat_chan_common *c
chan->timer.function = device->timer_fn;
chan->timer.data = data;
tasklet_init(&chan->cleanup_task, device->cleanup_fn, data);
- tasklet_disable(&chan->cleanup_task);
}
/**
@@ -354,13 +355,49 @@ static int ioat1_dma_alloc_chan_resources(struct dma_chan *c)
writel(((u64) chan->completion_dma) >> 32,
chan->reg_base + IOAT_CHANCMP_OFFSET_HIGH);
- tasklet_enable(&chan->cleanup_task);
+ set_bit(IOAT_RUN, &chan->state);
ioat1_dma_start_null_desc(ioat); /* give chain to dma device */
dev_dbg(to_dev(chan), "%s: allocated %d descriptors\n",
__func__, ioat->desccount);
return ioat->desccount;
}
+void ioat_stop(struct ioat_chan_common *chan)
+{
+ struct ioatdma_device *device = chan->device;
+ struct pci_dev *pdev = device->pdev;
+ int chan_id = chan_num(chan);
+ struct msix_entry *msix;
+
+ /* 1/ stop irq from firing tasklets
+ * 2/ stop the tasklet from re-arming irqs
+ */
+ clear_bit(IOAT_RUN, &chan->state);
+
+ /* flush inflight interrupts */
+ switch (device->irq_mode) {
+ case IOAT_MSIX:
+ msix = &device->msix_entries[chan_id];
+ synchronize_irq(msix->vector);
+ break;
+ case IOAT_MSI:
+ case IOAT_INTX:
+ synchronize_irq(pdev->irq);
+ break;
+ default:
+ break;
+ }
+
+ /* flush inflight timers */
+ del_timer_sync(&chan->timer);
+
+ /* flush inflight tasklet runs */
+ tasklet_kill(&chan->cleanup_task);
+
+ /* final cleanup now that everything is quiesced and can't re-arm */
+ device->cleanup_fn((unsigned long) &chan->common);
+}
+
/**
* ioat1_dma_free_chan_resources - release all the descriptors
* @chan: the channel to be cleaned
@@ -379,9 +416,7 @@ static void ioat1_dma_free_chan_resources(struct dma_chan *c)
if (ioat->desccount == 0)
return;
- tasklet_disable(&chan->cleanup_task);
- del_timer_sync(&chan->timer);
- ioat1_cleanup(ioat);
+ ioat_stop(chan);
/* Delay 100ms after reset to allow internal DMA logic to quiesce
* before removing DMA descriptor resources.
@@ -526,26 +561,14 @@ ioat1_dma_prep_memcpy(struct dma_chan *c, dma_addr_t dma_dest,
static void ioat1_cleanup_event(unsigned long data)
{
struct ioat_dma_chan *ioat = to_ioat_chan((void *) data);
+ struct ioat_chan_common *chan = &ioat->base;
ioat1_cleanup(ioat);
+ if (!test_bit(IOAT_RUN, &chan->state))
+ return;
writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
}
-void ioat_dma_unmap(struct ioat_chan_common *chan, enum dma_ctrl_flags flags,
- size_t len, struct ioat_dma_descriptor *hw)
-{
- struct pci_dev *pdev = chan->device->pdev;
- size_t offset = len - hw->size;
-
- if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP))
- ioat_unmap(pdev, hw->dst_addr - offset, len,
- PCI_DMA_FROMDEVICE, flags, 1);
-
- if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP))
- ioat_unmap(pdev, hw->src_addr - offset, len,
- PCI_DMA_TODEVICE, flags, 0);
-}
-
dma_addr_t ioat_get_current_completion(struct ioat_chan_common *chan)
{
dma_addr_t phys_complete;
@@ -602,7 +625,7 @@ static void __cleanup(struct ioat_dma_chan *ioat, dma_addr_t phys_complete)
dump_desc_dbg(ioat, desc);
if (tx->cookie) {
dma_cookie_complete(tx);
- ioat_dma_unmap(chan, tx->flags, desc->len, desc->hw);
+ dma_descriptor_unmap(tx);
ioat->active -= desc->hw->tx_cnt;
if (tx->callback) {
tx->callback(tx->callback_param);
@@ -733,7 +756,7 @@ ioat_dma_tx_status(struct dma_chan *c, dma_cookie_t cookie,
enum dma_status ret;
ret = dma_cookie_status(c, cookie, txstate);
- if (ret == DMA_SUCCESS)
+ if (ret == DMA_COMPLETE)
return ret;
device->cleanup_fn((unsigned long) c);
@@ -832,15 +855,22 @@ int ioat_dma_self_test(struct ioatdma_device *device)
}
dma_src = dma_map_single(dev, src, IOAT_TEST_SIZE, DMA_TO_DEVICE);
+ if (dma_mapping_error(dev, dma_src)) {
+ dev_err(dev, "mapping src buffer failed\n");
+ goto free_resources;
+ }
dma_dest = dma_map_single(dev, dest, IOAT_TEST_SIZE, DMA_FROM_DEVICE);
- flags = DMA_COMPL_SRC_UNMAP_SINGLE | DMA_COMPL_DEST_UNMAP_SINGLE |
- DMA_PREP_INTERRUPT;
+ if (dma_mapping_error(dev, dma_dest)) {
+ dev_err(dev, "mapping dest buffer failed\n");
+ goto unmap_src;
+ }
+ flags = DMA_PREP_INTERRUPT;
tx = device->common.device_prep_dma_memcpy(dma_chan, dma_dest, dma_src,
IOAT_TEST_SIZE, flags);
if (!tx) {
dev_err(dev, "Self-test prep failed, disabling\n");
err = -ENODEV;
- goto free_resources;
+ goto unmap_dma;
}
async_tx_ack(tx);
@@ -851,7 +881,7 @@ int ioat_dma_self_test(struct ioatdma_device *device)
if (cookie < 0) {
dev_err(dev, "Self-test setup failed, disabling\n");
err = -ENODEV;
- goto free_resources;
+ goto unmap_dma;
}
dma->device_issue_pending(dma_chan);
@@ -859,10 +889,10 @@ int ioat_dma_self_test(struct ioatdma_device *device)
if (tmo == 0 ||
dma->device_tx_status(dma_chan, cookie, NULL)
- != DMA_SUCCESS) {
+ != DMA_COMPLETE) {
dev_err(dev, "Self-test copy timed out, disabling\n");
err = -ENODEV;
- goto free_resources;
+ goto unmap_dma;
}
if (memcmp(src, dest, IOAT_TEST_SIZE)) {
dev_err(dev, "Self-test copy failed compare, disabling\n");
@@ -870,6 +900,10 @@ int ioat_dma_self_test(struct ioatdma_device *device)
goto free_resources;
}
+unmap_dma:
+ dma_unmap_single(dev, dma_dest, IOAT_TEST_SIZE, DMA_FROM_DEVICE);
+unmap_src:
+ dma_unmap_single(dev, dma_src, IOAT_TEST_SIZE, DMA_TO_DEVICE);
free_resources:
dma->device_free_chan_resources(dma_chan);
out:
@@ -882,14 +916,13 @@ static char ioat_interrupt_style[32] = "msix";
module_param_string(ioat_interrupt_style, ioat_interrupt_style,
sizeof(ioat_interrupt_style), 0644);
MODULE_PARM_DESC(ioat_interrupt_style,
- "set ioat interrupt style: msix (default), "
- "msix-single-vector, msi, intx)");
+ "set ioat interrupt style: msix (default), msi, intx");
/**
* ioat_dma_setup_interrupts - setup interrupt handler
* @device: ioat device
*/
-static int ioat_dma_setup_interrupts(struct ioatdma_device *device)
+int ioat_dma_setup_interrupts(struct ioatdma_device *device)
{
struct ioat_chan_common *chan;
struct pci_dev *pdev = device->pdev;
@@ -901,8 +934,6 @@ static int ioat_dma_setup_interrupts(struct ioatdma_device *device)
if (!strcmp(ioat_interrupt_style, "msix"))
goto msix;
- if (!strcmp(ioat_interrupt_style, "msix-single-vector"))
- goto msix_single_vector;
if (!strcmp(ioat_interrupt_style, "msi"))
goto msi;
if (!strcmp(ioat_interrupt_style, "intx"))
@@ -917,10 +948,8 @@ msix:
device->msix_entries[i].entry = i;
err = pci_enable_msix(pdev, device->msix_entries, msixcnt);
- if (err < 0)
+ if (err)
goto msi;
- if (err > 0)
- goto msix_single_vector;
for (i = 0; i < msixcnt; i++) {
msix = &device->msix_entries[i];
@@ -934,25 +963,11 @@ msix:
chan = ioat_chan_by_index(device, j);
devm_free_irq(dev, msix->vector, chan);
}
- goto msix_single_vector;
+ goto msi;
}
}
intrctrl |= IOAT_INTRCTRL_MSIX_VECTOR_CONTROL;
- goto done;
-
-msix_single_vector:
- msix = &device->msix_entries[0];
- msix->entry = 0;
- err = pci_enable_msix(pdev, device->msix_entries, 1);
- if (err)
- goto msi;
-
- err = devm_request_irq(dev, msix->vector, ioat_dma_do_interrupt, 0,
- "ioat-msix", device);
- if (err) {
- pci_disable_msix(pdev);
- goto msi;
- }
+ device->irq_mode = IOAT_MSIX;
goto done;
msi:
@@ -966,6 +981,7 @@ msi:
pci_disable_msi(pdev);
goto intx;
}
+ device->irq_mode = IOAT_MSI;
goto done;
intx:
@@ -974,6 +990,7 @@ intx:
if (err)
goto err_no_irq;
+ device->irq_mode = IOAT_INTX;
done:
if (device->intr_quirk)
device->intr_quirk(device);
@@ -984,9 +1001,11 @@ done:
err_no_irq:
/* Disable all interrupt generation */
writeb(0, device->reg_base + IOAT_INTRCTRL_OFFSET);
+ device->irq_mode = IOAT_NOIRQ;
dev_err(dev, "no usable interrupts\n");
return err;
}
+EXPORT_SYMBOL(ioat_dma_setup_interrupts);
static void ioat_disable_interrupts(struct ioatdma_device *device)
{
@@ -1096,12 +1115,11 @@ static ssize_t cap_show(struct dma_chan *c, char *page)
{
struct dma_device *dma = c->device;
- return sprintf(page, "copy%s%s%s%s%s%s\n",
+ return sprintf(page, "copy%s%s%s%s%s\n",
dma_has_cap(DMA_PQ, dma->cap_mask) ? " pq" : "",
dma_has_cap(DMA_PQ_VAL, dma->cap_mask) ? " pq_val" : "",
dma_has_cap(DMA_XOR, dma->cap_mask) ? " xor" : "",
dma_has_cap(DMA_XOR_VAL, dma->cap_mask) ? " xor_val" : "",
- dma_has_cap(DMA_MEMSET, dma->cap_mask) ? " fill" : "",
dma_has_cap(DMA_INTERRUPT, dma->cap_mask) ? " intr" : "");
}
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h
index 087935f1565..e982f00a984 100644
--- a/drivers/dma/ioat/dma.h
+++ b/drivers/dma/ioat/dma.h
@@ -39,6 +39,7 @@
#define to_ioat_desc(lh) container_of(lh, struct ioat_desc_sw, node)
#define tx_to_ioat_desc(tx) container_of(tx, struct ioat_desc_sw, txd)
#define to_dev(ioat_chan) (&(ioat_chan)->device->pdev->dev)
+#define to_pdev(ioat_chan) ((ioat_chan)->device->pdev)
#define chan_num(ch) ((int)((ch)->reg_base - (ch)->device->reg_base) / 0x80)
@@ -48,6 +49,13 @@
*/
#define NULL_DESC_BUFFER_SIZE 1
+enum ioat_irq_mode {
+ IOAT_NOIRQ = 0,
+ IOAT_MSIX,
+ IOAT_MSI,
+ IOAT_INTX
+};
+
/**
* struct ioatdma_device - internal representation of a IOAT device
* @pdev: PCI-Express device
@@ -72,11 +80,15 @@ struct ioatdma_device {
void __iomem *reg_base;
struct pci_pool *dma_pool;
struct pci_pool *completion_pool;
+#define MAX_SED_POOLS 5
+ struct dma_pool *sed_hw_pool[MAX_SED_POOLS];
struct dma_device common;
u8 version;
struct msix_entry msix_entries[4];
struct ioat_chan_common *idx[4];
struct dca_provider *dca;
+ enum ioat_irq_mode irq_mode;
+ u32 cap;
void (*intr_quirk)(struct ioatdma_device *device);
int (*enumerate_channels)(struct ioatdma_device *device);
int (*reset_hw)(struct ioat_chan_common *chan);
@@ -97,6 +109,7 @@ struct ioat_chan_common {
#define IOAT_KOBJ_INIT_FAIL 3
#define IOAT_RESHAPE_PENDING 4
#define IOAT_RUN 5
+ #define IOAT_CHAN_ACTIVE 6
struct timer_list timer;
#define COMPLETION_TIMEOUT msecs_to_jiffies(100)
#define IDLE_TIMEOUT msecs_to_jiffies(2000)
@@ -130,6 +143,20 @@ struct ioat_dma_chan {
u16 active;
};
+/**
+ * struct ioat_sed_ent - wrapper around super extended hardware descriptor
+ * @hw: hardware SED
+ * @sed_dma: dma address for the SED
+ * @list: list member
+ * @parent: point to the dma descriptor that's the parent
+ */
+struct ioat_sed_ent {
+ struct ioat_sed_raw_descriptor *hw;
+ dma_addr_t dma;
+ struct ioat_ring_ent *parent;
+ unsigned int hw_pool;
+};
+
static inline struct ioat_chan_common *to_chan_common(struct dma_chan *c)
{
return container_of(c, struct ioat_chan_common, common);
@@ -178,7 +205,7 @@ __dump_desc_dbg(struct ioat_chan_common *chan, struct ioat_dma_descriptor *hw,
struct device *dev = to_dev(chan);
dev_dbg(dev, "desc[%d]: (%#llx->%#llx) cookie: %d flags: %#x"
- " ctl: %#x (op: %d int_en: %d compl: %d)\n", id,
+ " ctl: %#10.8x (op: %#x int_en: %d compl: %d)\n", id,
(unsigned long long) tx->phys,
(unsigned long long) hw->next, tx->cookie, tx->flags,
hw->ctl, hw->ctl_f.op, hw->ctl_f.int_en, hw->ctl_f.compl_write);
@@ -200,7 +227,7 @@ ioat_chan_by_index(struct ioatdma_device *device, int index)
return device->idx[index];
}
-static inline u64 ioat_chansts(struct ioat_chan_common *chan)
+static inline u64 ioat_chansts_32(struct ioat_chan_common *chan)
{
u8 ver = chan->device->version;
u64 status;
@@ -217,6 +244,26 @@ static inline u64 ioat_chansts(struct ioat_chan_common *chan)
return status;
}
+#if BITS_PER_LONG == 64
+
+static inline u64 ioat_chansts(struct ioat_chan_common *chan)
+{
+ u8 ver = chan->device->version;
+ u64 status;
+
+ /* With IOAT v3.3 the status register is 64bit. */
+ if (ver >= IOAT_VER_3_3)
+ status = readq(chan->reg_base + IOAT_CHANSTS_OFFSET(ver));
+ else
+ status = ioat_chansts_32(chan);
+
+ return status;
+}
+
+#else
+#define ioat_chansts ioat_chansts_32
+#endif
+
static inline void ioat_start(struct ioat_chan_common *chan)
{
u8 ver = chan->device->version;
@@ -293,16 +340,6 @@ static inline bool is_ioat_bug(unsigned long err)
return !!err;
}
-static inline void ioat_unmap(struct pci_dev *pdev, dma_addr_t addr, size_t len,
- int direction, enum dma_ctrl_flags flags, bool dst)
-{
- if ((dst && (flags & DMA_COMPL_DEST_UNMAP_SINGLE)) ||
- (!dst && (flags & DMA_COMPL_SRC_UNMAP_SINGLE)))
- pci_unmap_single(pdev, addr, len, direction);
- else
- pci_unmap_page(pdev, addr, len, direction);
-}
-
int ioat_probe(struct ioatdma_device *device);
int ioat_register(struct ioatdma_device *device);
int ioat1_dma_probe(struct ioatdma_device *dev, int dca);
@@ -314,12 +351,12 @@ void ioat_init_channel(struct ioatdma_device *device,
struct ioat_chan_common *chan, int idx);
enum dma_status ioat_dma_tx_status(struct dma_chan *c, dma_cookie_t cookie,
struct dma_tx_state *txstate);
-void ioat_dma_unmap(struct ioat_chan_common *chan, enum dma_ctrl_flags flags,
- size_t len, struct ioat_dma_descriptor *hw);
bool ioat_cleanup_preamble(struct ioat_chan_common *chan,
dma_addr_t *phys_complete);
void ioat_kobject_add(struct ioatdma_device *device, struct kobj_type *type);
void ioat_kobject_del(struct ioatdma_device *device);
+int ioat_dma_setup_interrupts(struct ioatdma_device *device);
+void ioat_stop(struct ioat_chan_common *chan);
extern const struct sysfs_ops ioat_sysfs_ops;
extern struct ioat_sysfs_entry ioat_version_attr;
extern struct ioat_sysfs_entry ioat_cap_attr;
diff --git a/drivers/dma/ioat/dma_v2.c b/drivers/dma/ioat/dma_v2.c
index 82d4e306c32..8d1058085ee 100644
--- a/drivers/dma/ioat/dma_v2.c
+++ b/drivers/dma/ioat/dma_v2.c
@@ -148,7 +148,7 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
tx = &desc->txd;
dump_desc_dbg(ioat, desc);
if (tx->cookie) {
- ioat_dma_unmap(chan, tx->flags, desc->len, desc->hw);
+ dma_descriptor_unmap(tx);
dma_cookie_complete(tx);
if (tx->callback) {
tx->callback(tx->callback_param);
@@ -190,8 +190,11 @@ static void ioat2_cleanup(struct ioat2_dma_chan *ioat)
void ioat2_cleanup_event(unsigned long data)
{
struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+ struct ioat_chan_common *chan = &ioat->base;
ioat2_cleanup(ioat);
+ if (!test_bit(IOAT_RUN, &chan->state))
+ return;
writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
}
@@ -269,61 +272,22 @@ static void ioat2_restart_channel(struct ioat2_dma_chan *ioat)
__ioat2_restart_chan(ioat);
}
-void ioat2_timer_event(unsigned long data)
+static void check_active(struct ioat2_dma_chan *ioat)
{
- struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
struct ioat_chan_common *chan = &ioat->base;
- if (test_bit(IOAT_COMPLETION_PENDING, &chan->state)) {
- dma_addr_t phys_complete;
- u64 status;
-
- status = ioat_chansts(chan);
-
- /* when halted due to errors check for channel
- * programming errors before advancing the completion state
- */
- if (is_ioat_halted(status)) {
- u32 chanerr;
-
- chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
- dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
- __func__, chanerr);
- if (test_bit(IOAT_RUN, &chan->state))
- BUG_ON(is_ioat_bug(chanerr));
- else /* we never got off the ground */
- return;
- }
-
- /* if we haven't made progress and we have already
- * acknowledged a pending completion once, then be more
- * forceful with a restart
- */
- spin_lock_bh(&chan->cleanup_lock);
- if (ioat_cleanup_preamble(chan, &phys_complete)) {
- __cleanup(ioat, phys_complete);
- } else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
- spin_lock_bh(&ioat->prep_lock);
- ioat2_restart_channel(ioat);
- spin_unlock_bh(&ioat->prep_lock);
- } else {
- set_bit(IOAT_COMPLETION_ACK, &chan->state);
- mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
- }
- spin_unlock_bh(&chan->cleanup_lock);
- } else {
- u16 active;
+ if (ioat2_ring_active(ioat)) {
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ return;
+ }
+ if (test_and_clear_bit(IOAT_CHAN_ACTIVE, &chan->state))
+ mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+ else if (ioat->alloc_order > ioat_get_alloc_order()) {
/* if the ring is idle, empty, and oversized try to step
* down the size
*/
- spin_lock_bh(&chan->cleanup_lock);
- spin_lock_bh(&ioat->prep_lock);
- active = ioat2_ring_active(ioat);
- if (active == 0 && ioat->alloc_order > ioat_get_alloc_order())
- reshape_ring(ioat, ioat->alloc_order-1);
- spin_unlock_bh(&ioat->prep_lock);
- spin_unlock_bh(&chan->cleanup_lock);
+ reshape_ring(ioat, ioat->alloc_order - 1);
/* keep shrinking until we get back to our minimum
* default size
@@ -331,6 +295,60 @@ void ioat2_timer_event(unsigned long data)
if (ioat->alloc_order > ioat_get_alloc_order())
mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
}
+
+}
+
+void ioat2_timer_event(unsigned long data)
+{
+ struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+ struct ioat_chan_common *chan = &ioat->base;
+ dma_addr_t phys_complete;
+ u64 status;
+
+ status = ioat_chansts(chan);
+
+ /* when halted due to errors check for channel
+ * programming errors before advancing the completion state
+ */
+ if (is_ioat_halted(status)) {
+ u32 chanerr;
+
+ chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+ dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
+ __func__, chanerr);
+ if (test_bit(IOAT_RUN, &chan->state))
+ BUG_ON(is_ioat_bug(chanerr));
+ else /* we never got off the ground */
+ return;
+ }
+
+ /* if we haven't made progress and we have already
+ * acknowledged a pending completion once, then be more
+ * forceful with a restart
+ */
+ spin_lock_bh(&chan->cleanup_lock);
+ if (ioat_cleanup_preamble(chan, &phys_complete))
+ __cleanup(ioat, phys_complete);
+ else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
+ spin_lock_bh(&ioat->prep_lock);
+ ioat2_restart_channel(ioat);
+ spin_unlock_bh(&ioat->prep_lock);
+ spin_unlock_bh(&chan->cleanup_lock);
+ return;
+ } else {
+ set_bit(IOAT_COMPLETION_ACK, &chan->state);
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ }
+
+
+ if (ioat2_ring_active(ioat))
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ else {
+ spin_lock_bh(&ioat->prep_lock);
+ check_active(ioat);
+ spin_unlock_bh(&ioat->prep_lock);
+ }
+ spin_unlock_bh(&chan->cleanup_lock);
}
static int ioat2_reset_hw(struct ioat_chan_common *chan)
@@ -404,7 +422,7 @@ static dma_cookie_t ioat2_tx_submit_unlock(struct dma_async_tx_descriptor *tx)
cookie = dma_cookie_assign(tx);
dev_dbg(to_dev(&ioat->base), "%s: cookie: %d\n", __func__, cookie);
- if (!test_and_set_bit(IOAT_COMPLETION_PENDING, &chan->state))
+ if (!test_and_set_bit(IOAT_CHAN_ACTIVE, &chan->state))
mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
/* make descriptor updates visible before advancing ioat->head,
@@ -538,10 +556,10 @@ int ioat2_alloc_chan_resources(struct dma_chan *c)
ioat->issued = 0;
ioat->tail = 0;
ioat->alloc_order = order;
+ set_bit(IOAT_RUN, &chan->state);
spin_unlock_bh(&ioat->prep_lock);
spin_unlock_bh(&chan->cleanup_lock);
- tasklet_enable(&chan->cleanup_task);
ioat2_start_null_desc(ioat);
/* check that we got off the ground */
@@ -551,7 +569,6 @@ int ioat2_alloc_chan_resources(struct dma_chan *c)
} while (i++ < 20 && !is_ioat_active(status) && !is_ioat_idle(status));
if (is_ioat_active(status) || is_ioat_idle(status)) {
- set_bit(IOAT_RUN, &chan->state);
return 1 << ioat->alloc_order;
} else {
u32 chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
@@ -794,11 +811,8 @@ void ioat2_free_chan_resources(struct dma_chan *c)
if (!ioat->ring)
return;
- tasklet_disable(&chan->cleanup_task);
- del_timer_sync(&chan->timer);
- device->cleanup_fn((unsigned long) c);
+ ioat_stop(chan);
device->reset_hw(chan);
- clear_bit(IOAT_RUN, &chan->state);
spin_lock_bh(&chan->cleanup_lock);
spin_lock_bh(&ioat->prep_lock);
diff --git a/drivers/dma/ioat/dma_v2.h b/drivers/dma/ioat/dma_v2.h
index e100f644e34..470292767e6 100644
--- a/drivers/dma/ioat/dma_v2.h
+++ b/drivers/dma/ioat/dma_v2.h
@@ -123,7 +123,6 @@ static inline u16 ioat2_xferlen_to_descs(struct ioat2_dma_chan *ioat, size_t len
struct ioat_ring_ent {
union {
struct ioat_dma_descriptor *hw;
- struct ioat_fill_descriptor *fill;
struct ioat_xor_descriptor *xor;
struct ioat_xor_ext_descriptor *xor_ex;
struct ioat_pq_descriptor *pq;
@@ -137,6 +136,7 @@ struct ioat_ring_ent {
#ifdef DEBUG
int id;
#endif
+ struct ioat_sed_ent *sed;
};
static inline struct ioat_ring_ent *
diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c
index 3e9d66920eb..b9b38a1cf92 100644
--- a/drivers/dma/ioat/dma_v3.c
+++ b/drivers/dma/ioat/dma_v3.c
@@ -55,7 +55,7 @@
/*
* Support routines for v3+ hardware
*/
-
+#include <linux/module.h>
#include <linux/pci.h>
#include <linux/gfp.h>
#include <linux/dmaengine.h>
@@ -67,9 +67,15 @@
#include "dma.h"
#include "dma_v2.h"
+extern struct kmem_cache *ioat3_sed_cache;
+
/* ioat hardware assumes at least two sources for raid operations */
#define src_cnt_to_sw(x) ((x) + 2)
#define src_cnt_to_hw(x) ((x) - 2)
+#define ndest_to_sw(x) ((x) + 1)
+#define ndest_to_hw(x) ((x) - 1)
+#define src16_cnt_to_sw(x) ((x) + 9)
+#define src16_cnt_to_hw(x) ((x) - 9)
/* provide a lookup table for setting the source address in the base or
* extended descriptor of an xor or pq descriptor
@@ -77,14 +83,13 @@
static const u8 xor_idx_to_desc = 0xe0;
static const u8 xor_idx_to_field[] = { 1, 4, 5, 6, 7, 0, 1, 2 };
static const u8 pq_idx_to_desc = 0xf8;
+static const u8 pq16_idx_to_desc[] = { 0, 0, 1, 1, 1, 1, 1, 1, 1,
+ 2, 2, 2, 2, 2, 2, 2 };
static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 };
+static const u8 pq16_idx_to_field[] = { 1, 4, 1, 2, 3, 4, 5, 6, 7,
+ 0, 1, 2, 3, 4, 5, 6 };
-static dma_addr_t xor_get_src(struct ioat_raw_descriptor *descs[2], int idx)
-{
- struct ioat_raw_descriptor *raw = descs[xor_idx_to_desc >> idx & 1];
-
- return raw->field[xor_idx_to_field[idx]];
-}
+static void ioat3_eh(struct ioat2_dma_chan *ioat);
static void xor_set_src(struct ioat_raw_descriptor *descs[2],
dma_addr_t addr, u32 offset, int idx)
@@ -101,6 +106,13 @@ static dma_addr_t pq_get_src(struct ioat_raw_descriptor *descs[2], int idx)
return raw->field[pq_idx_to_field[idx]];
}
+static dma_addr_t pq16_get_src(struct ioat_raw_descriptor *desc[3], int idx)
+{
+ struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];
+
+ return raw->field[pq16_idx_to_field[idx]];
+}
+
static void pq_set_src(struct ioat_raw_descriptor *descs[2],
dma_addr_t addr, u32 offset, u8 coef, int idx)
{
@@ -111,122 +123,159 @@ static void pq_set_src(struct ioat_raw_descriptor *descs[2],
pq->coef[idx] = coef;
}
-static void ioat3_dma_unmap(struct ioat2_dma_chan *ioat,
- struct ioat_ring_ent *desc, int idx)
+static bool is_jf_ioat(struct pci_dev *pdev)
{
- struct ioat_chan_common *chan = &ioat->base;
- struct pci_dev *pdev = chan->device->pdev;
- size_t len = desc->len;
- size_t offset = len - desc->hw->size;
- struct dma_async_tx_descriptor *tx = &desc->txd;
- enum dma_ctrl_flags flags = tx->flags;
-
- switch (desc->hw->ctl_f.op) {
- case IOAT_OP_COPY:
- if (!desc->hw->ctl_f.null) /* skip 'interrupt' ops */
- ioat_dma_unmap(chan, flags, len, desc->hw);
- break;
- case IOAT_OP_FILL: {
- struct ioat_fill_descriptor *hw = desc->fill;
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF0:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF1:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF2:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF3:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF4:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF5:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF6:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF7:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF8:
+ case PCI_DEVICE_ID_INTEL_IOAT_JSF9:
+ return true;
+ default:
+ return false;
+ }
+}
- if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP))
- ioat_unmap(pdev, hw->dst_addr - offset, len,
- PCI_DMA_FROMDEVICE, flags, 1);
- break;
+static bool is_snb_ioat(struct pci_dev *pdev)
+{
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB0:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB1:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB2:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB3:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB4:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB5:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB6:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB7:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB8:
+ case PCI_DEVICE_ID_INTEL_IOAT_SNB9:
+ return true;
+ default:
+ return false;
}
- case IOAT_OP_XOR_VAL:
- case IOAT_OP_XOR: {
- struct ioat_xor_descriptor *xor = desc->xor;
- struct ioat_ring_ent *ext;
- struct ioat_xor_ext_descriptor *xor_ex = NULL;
- int src_cnt = src_cnt_to_sw(xor->ctl_f.src_cnt);
- struct ioat_raw_descriptor *descs[2];
- int i;
+}
- if (src_cnt > 5) {
- ext = ioat2_get_ring_ent(ioat, idx + 1);
- xor_ex = ext->xor_ex;
- }
+static bool is_ivb_ioat(struct pci_dev *pdev)
+{
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB0:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB1:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB2:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB3:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB4:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB5:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB6:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB7:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB8:
+ case PCI_DEVICE_ID_INTEL_IOAT_IVB9:
+ return true;
+ default:
+ return false;
+ }
- if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
- descs[0] = (struct ioat_raw_descriptor *) xor;
- descs[1] = (struct ioat_raw_descriptor *) xor_ex;
- for (i = 0; i < src_cnt; i++) {
- dma_addr_t src = xor_get_src(descs, i);
+}
- ioat_unmap(pdev, src - offset, len,
- PCI_DMA_TODEVICE, flags, 0);
- }
+static bool is_hsw_ioat(struct pci_dev *pdev)
+{
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW0:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW1:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW2:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW3:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW4:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW5:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW6:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW7:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW8:
+ case PCI_DEVICE_ID_INTEL_IOAT_HSW9:
+ return true;
+ default:
+ return false;
+ }
- /* dest is a source in xor validate operations */
- if (xor->ctl_f.op == IOAT_OP_XOR_VAL) {
- ioat_unmap(pdev, xor->dst_addr - offset, len,
- PCI_DMA_TODEVICE, flags, 1);
- break;
- }
- }
+}
- if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP))
- ioat_unmap(pdev, xor->dst_addr - offset, len,
- PCI_DMA_FROMDEVICE, flags, 1);
- break;
+static bool is_xeon_cb32(struct pci_dev *pdev)
+{
+ return is_jf_ioat(pdev) || is_snb_ioat(pdev) || is_ivb_ioat(pdev) ||
+ is_hsw_ioat(pdev);
+}
+
+static bool is_bwd_ioat(struct pci_dev *pdev)
+{
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD0:
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD1:
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
+ return true;
+ default:
+ return false;
}
- case IOAT_OP_PQ_VAL:
- case IOAT_OP_PQ: {
- struct ioat_pq_descriptor *pq = desc->pq;
- struct ioat_ring_ent *ext;
- struct ioat_pq_ext_descriptor *pq_ex = NULL;
- int src_cnt = src_cnt_to_sw(pq->ctl_f.src_cnt);
- struct ioat_raw_descriptor *descs[2];
- int i;
+}
- if (src_cnt > 3) {
- ext = ioat2_get_ring_ent(ioat, idx + 1);
- pq_ex = ext->pq_ex;
- }
+static bool is_bwd_noraid(struct pci_dev *pdev)
+{
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
+ case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
+ return true;
+ default:
+ return false;
+ }
- /* in the 'continue' case don't unmap the dests as sources */
- if (dmaf_p_disabled_continue(flags))
- src_cnt--;
- else if (dmaf_continue(flags))
- src_cnt -= 3;
-
- if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
- descs[0] = (struct ioat_raw_descriptor *) pq;
- descs[1] = (struct ioat_raw_descriptor *) pq_ex;
- for (i = 0; i < src_cnt; i++) {
- dma_addr_t src = pq_get_src(descs, i);
-
- ioat_unmap(pdev, src - offset, len,
- PCI_DMA_TODEVICE, flags, 0);
- }
+}
- /* the dests are sources in pq validate operations */
- if (pq->ctl_f.op == IOAT_OP_XOR_VAL) {
- if (!(flags & DMA_PREP_PQ_DISABLE_P))
- ioat_unmap(pdev, pq->p_addr - offset,
- len, PCI_DMA_TODEVICE, flags, 0);
- if (!(flags & DMA_PREP_PQ_DISABLE_Q))
- ioat_unmap(pdev, pq->q_addr - offset,
- len, PCI_DMA_TODEVICE, flags, 0);
- break;
- }
- }
+static void pq16_set_src(struct ioat_raw_descriptor *desc[3],
+ dma_addr_t addr, u32 offset, u8 coef, unsigned idx)
+{
+ struct ioat_pq_descriptor *pq = (struct ioat_pq_descriptor *)desc[0];
+ struct ioat_pq16a_descriptor *pq16 =
+ (struct ioat_pq16a_descriptor *)desc[1];
+ struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];
- if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
- if (!(flags & DMA_PREP_PQ_DISABLE_P))
- ioat_unmap(pdev, pq->p_addr - offset, len,
- PCI_DMA_BIDIRECTIONAL, flags, 1);
- if (!(flags & DMA_PREP_PQ_DISABLE_Q))
- ioat_unmap(pdev, pq->q_addr - offset, len,
- PCI_DMA_BIDIRECTIONAL, flags, 1);
- }
- break;
- }
- default:
- dev_err(&pdev->dev, "%s: unknown op type: %#x\n",
- __func__, desc->hw->ctl_f.op);
+ raw->field[pq16_idx_to_field[idx]] = addr + offset;
+
+ if (idx < 8)
+ pq->coef[idx] = coef;
+ else
+ pq16->coef[idx - 8] = coef;
+}
+
+static struct ioat_sed_ent *
+ioat3_alloc_sed(struct ioatdma_device *device, unsigned int hw_pool)
+{
+ struct ioat_sed_ent *sed;
+ gfp_t flags = __GFP_ZERO | GFP_ATOMIC;
+
+ sed = kmem_cache_alloc(ioat3_sed_cache, flags);
+ if (!sed)
+ return NULL;
+
+ sed->hw_pool = hw_pool;
+ sed->hw = dma_pool_alloc(device->sed_hw_pool[hw_pool],
+ flags, &sed->dma);
+ if (!sed->hw) {
+ kmem_cache_free(ioat3_sed_cache, sed);
+ return NULL;
}
+
+ return sed;
+}
+
+static void ioat3_free_sed(struct ioatdma_device *device, struct ioat_sed_ent *sed)
+{
+ if (!sed)
+ return;
+
+ dma_pool_free(device->sed_hw_pool[sed->hw_pool], sed->hw, sed->dma);
+ kmem_cache_free(ioat3_sed_cache, sed);
}
static bool desc_has_ext(struct ioat_ring_ent *desc)
@@ -250,6 +299,63 @@ static bool desc_has_ext(struct ioat_ring_ent *desc)
return false;
}
+static u64 ioat3_get_current_completion(struct ioat_chan_common *chan)
+{
+ u64 phys_complete;
+ u64 completion;
+
+ completion = *chan->completion;
+ phys_complete = ioat_chansts_to_addr(completion);
+
+ dev_dbg(to_dev(chan), "%s: phys_complete: %#llx\n", __func__,
+ (unsigned long long) phys_complete);
+
+ return phys_complete;
+}
+
+static bool ioat3_cleanup_preamble(struct ioat_chan_common *chan,
+ u64 *phys_complete)
+{
+ *phys_complete = ioat3_get_current_completion(chan);
+ if (*phys_complete == chan->last_completion)
+ return false;
+
+ clear_bit(IOAT_COMPLETION_ACK, &chan->state);
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+
+ return true;
+}
+
+static void
+desc_get_errstat(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc)
+{
+ struct ioat_dma_descriptor *hw = desc->hw;
+
+ switch (hw->ctl_f.op) {
+ case IOAT_OP_PQ_VAL:
+ case IOAT_OP_PQ_VAL_16S:
+ {
+ struct ioat_pq_descriptor *pq = desc->pq;
+
+ /* check if there's error written */
+ if (!pq->dwbes_f.wbes)
+ return;
+
+ /* need to set a chanerr var for checking to clear later */
+
+ if (pq->dwbes_f.p_val_err)
+ *desc->result |= SUM_CHECK_P_RESULT;
+
+ if (pq->dwbes_f.q_val_err)
+ *desc->result |= SUM_CHECK_Q_RESULT;
+
+ return;
+ }
+ default:
+ return;
+ }
+}
+
/**
* __cleanup - reclaim used descriptors
* @ioat: channel (ring) to clean
@@ -260,6 +366,7 @@ static bool desc_has_ext(struct ioat_ring_ent *desc)
static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
{
struct ioat_chan_common *chan = &ioat->base;
+ struct ioatdma_device *device = chan->device;
struct ioat_ring_ent *desc;
bool seen_current = false;
int idx = ioat->tail, i;
@@ -268,6 +375,16 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
dev_dbg(to_dev(chan), "%s: head: %#x tail: %#x issued: %#x\n",
__func__, ioat->head, ioat->tail, ioat->issued);
+ /*
+ * At restart of the channel, the completion address and the
+ * channel status will be 0 due to starting a new chain. Since
+ * it's new chain and the first descriptor "fails", there is
+ * nothing to clean up. We do not want to reap the entire submitted
+ * chain due to this 0 address value and then BUG.
+ */
+ if (!phys_complete)
+ return;
+
active = ioat2_ring_active(ioat);
for (i = 0; i < active && !seen_current; i++) {
struct dma_async_tx_descriptor *tx;
@@ -276,10 +393,15 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
prefetch(ioat2_get_ring_ent(ioat, idx + i + 1));
desc = ioat2_get_ring_ent(ioat, idx + i);
dump_desc_dbg(ioat, desc);
+
+ /* set err stat if we are using dwbes */
+ if (device->cap & IOAT_CAP_DWBES)
+ desc_get_errstat(ioat, desc);
+
tx = &desc->txd;
if (tx->cookie) {
dma_cookie_complete(tx);
- ioat3_dma_unmap(ioat, desc, idx + i);
+ dma_descriptor_unmap(tx);
if (tx->callback) {
tx->callback(tx->callback_param);
tx->callback = NULL;
@@ -294,6 +416,12 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
BUG_ON(i + 1 >= active);
i++;
}
+
+ /* cleanup super extended descriptors */
+ if (desc->sed) {
+ ioat3_free_sed(device, desc->sed);
+ desc->sed = NULL;
+ }
}
smp_mb(); /* finish all descriptor reads before incrementing tail */
ioat->tail = idx + i;
@@ -314,89 +442,127 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
static void ioat3_cleanup(struct ioat2_dma_chan *ioat)
{
struct ioat_chan_common *chan = &ioat->base;
- dma_addr_t phys_complete;
+ u64 phys_complete;
spin_lock_bh(&chan->cleanup_lock);
- if (ioat_cleanup_preamble(chan, &phys_complete))
+
+ if (ioat3_cleanup_preamble(chan, &phys_complete))
__cleanup(ioat, phys_complete);
+
+ if (is_ioat_halted(*chan->completion)) {
+ u32 chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+
+ if (chanerr & IOAT_CHANERR_HANDLE_MASK) {
+ mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+ ioat3_eh(ioat);
+ }
+ }
+
spin_unlock_bh(&chan->cleanup_lock);
}
static void ioat3_cleanup_event(unsigned long data)
{
struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+ struct ioat_chan_common *chan = &ioat->base;
ioat3_cleanup(ioat);
+ if (!test_bit(IOAT_RUN, &chan->state))
+ return;
writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
}
static void ioat3_restart_channel(struct ioat2_dma_chan *ioat)
{
struct ioat_chan_common *chan = &ioat->base;
- dma_addr_t phys_complete;
+ u64 phys_complete;
ioat2_quiesce(chan, 0);
- if (ioat_cleanup_preamble(chan, &phys_complete))
+ if (ioat3_cleanup_preamble(chan, &phys_complete))
__cleanup(ioat, phys_complete);
__ioat2_restart_chan(ioat);
}
-static void ioat3_timer_event(unsigned long data)
+static void ioat3_eh(struct ioat2_dma_chan *ioat)
{
- struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
struct ioat_chan_common *chan = &ioat->base;
+ struct pci_dev *pdev = to_pdev(chan);
+ struct ioat_dma_descriptor *hw;
+ u64 phys_complete;
+ struct ioat_ring_ent *desc;
+ u32 err_handled = 0;
+ u32 chanerr_int;
+ u32 chanerr;
- if (test_bit(IOAT_COMPLETION_PENDING, &chan->state)) {
- dma_addr_t phys_complete;
- u64 status;
+ /* cleanup so tail points to descriptor that caused the error */
+ if (ioat3_cleanup_preamble(chan, &phys_complete))
+ __cleanup(ioat, phys_complete);
- status = ioat_chansts(chan);
+ chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+ pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr_int);
- /* when halted due to errors check for channel
- * programming errors before advancing the completion state
- */
- if (is_ioat_halted(status)) {
- u32 chanerr;
-
- chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
- dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
- __func__, chanerr);
- if (test_bit(IOAT_RUN, &chan->state))
- BUG_ON(is_ioat_bug(chanerr));
- else /* we never got off the ground */
- return;
- }
+ dev_dbg(to_dev(chan), "%s: error = %x:%x\n",
+ __func__, chanerr, chanerr_int);
- /* if we haven't made progress and we have already
- * acknowledged a pending completion once, then be more
- * forceful with a restart
- */
- spin_lock_bh(&chan->cleanup_lock);
- if (ioat_cleanup_preamble(chan, &phys_complete))
- __cleanup(ioat, phys_complete);
- else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
- spin_lock_bh(&ioat->prep_lock);
- ioat3_restart_channel(ioat);
- spin_unlock_bh(&ioat->prep_lock);
- } else {
- set_bit(IOAT_COMPLETION_ACK, &chan->state);
- mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ desc = ioat2_get_ring_ent(ioat, ioat->tail);
+ hw = desc->hw;
+ dump_desc_dbg(ioat, desc);
+
+ switch (hw->ctl_f.op) {
+ case IOAT_OP_XOR_VAL:
+ if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
+ *desc->result |= SUM_CHECK_P_RESULT;
+ err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
}
- spin_unlock_bh(&chan->cleanup_lock);
- } else {
- u16 active;
+ break;
+ case IOAT_OP_PQ_VAL:
+ case IOAT_OP_PQ_VAL_16S:
+ if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
+ *desc->result |= SUM_CHECK_P_RESULT;
+ err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
+ }
+ if (chanerr & IOAT_CHANERR_XOR_Q_ERR) {
+ *desc->result |= SUM_CHECK_Q_RESULT;
+ err_handled |= IOAT_CHANERR_XOR_Q_ERR;
+ }
+ break;
+ }
+
+ /* fault on unhandled error or spurious halt */
+ if (chanerr ^ err_handled || chanerr == 0) {
+ dev_err(to_dev(chan), "%s: fatal error (%x:%x)\n",
+ __func__, chanerr, err_handled);
+ BUG();
+ }
+
+ writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
+ pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr_int);
+
+ /* mark faulting descriptor as complete */
+ *chan->completion = desc->txd.phys;
+ spin_lock_bh(&ioat->prep_lock);
+ ioat3_restart_channel(ioat);
+ spin_unlock_bh(&ioat->prep_lock);
+}
+
+static void check_active(struct ioat2_dma_chan *ioat)
+{
+ struct ioat_chan_common *chan = &ioat->base;
+
+ if (ioat2_ring_active(ioat)) {
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ return;
+ }
+
+ if (test_and_clear_bit(IOAT_CHAN_ACTIVE, &chan->state))
+ mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+ else if (ioat->alloc_order > ioat_get_alloc_order()) {
/* if the ring is idle, empty, and oversized try to step
* down the size
*/
- spin_lock_bh(&chan->cleanup_lock);
- spin_lock_bh(&ioat->prep_lock);
- active = ioat2_ring_active(ioat);
- if (active == 0 && ioat->alloc_order > ioat_get_alloc_order())
- reshape_ring(ioat, ioat->alloc_order-1);
- spin_unlock_bh(&ioat->prep_lock);
- spin_unlock_bh(&chan->cleanup_lock);
+ reshape_ring(ioat, ioat->alloc_order - 1);
/* keep shrinking until we get back to our minimum
* default size
@@ -404,6 +570,60 @@ static void ioat3_timer_event(unsigned long data)
if (ioat->alloc_order > ioat_get_alloc_order())
mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
}
+
+}
+
+static void ioat3_timer_event(unsigned long data)
+{
+ struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+ struct ioat_chan_common *chan = &ioat->base;
+ dma_addr_t phys_complete;
+ u64 status;
+
+ status = ioat_chansts(chan);
+
+ /* when halted due to errors check for channel
+ * programming errors before advancing the completion state
+ */
+ if (is_ioat_halted(status)) {
+ u32 chanerr;
+
+ chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+ dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
+ __func__, chanerr);
+ if (test_bit(IOAT_RUN, &chan->state))
+ BUG_ON(is_ioat_bug(chanerr));
+ else /* we never got off the ground */
+ return;
+ }
+
+ /* if we haven't made progress and we have already
+ * acknowledged a pending completion once, then be more
+ * forceful with a restart
+ */
+ spin_lock_bh(&chan->cleanup_lock);
+ if (ioat_cleanup_preamble(chan, &phys_complete))
+ __cleanup(ioat, phys_complete);
+ else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
+ spin_lock_bh(&ioat->prep_lock);
+ ioat3_restart_channel(ioat);
+ spin_unlock_bh(&ioat->prep_lock);
+ spin_unlock_bh(&chan->cleanup_lock);
+ return;
+ } else {
+ set_bit(IOAT_COMPLETION_ACK, &chan->state);
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ }
+
+
+ if (ioat2_ring_active(ioat))
+ mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+ else {
+ spin_lock_bh(&ioat->prep_lock);
+ check_active(ioat);
+ spin_unlock_bh(&ioat->prep_lock);
+ }
+ spin_unlock_bh(&chan->cleanup_lock);
}
static enum dma_status
@@ -414,7 +634,7 @@ ioat3_tx_status(struct dma_chan *c, dma_cookie_t cookie,
enum dma_status ret;
ret = dma_cookie_status(c, cookie, txstate);
- if (ret == DMA_SUCCESS)
+ if (ret == DMA_COMPLETE)
return ret;
ioat3_cleanup(ioat);
@@ -423,51 +643,6 @@ ioat3_tx_status(struct dma_chan *c, dma_cookie_t cookie,
}
static struct dma_async_tx_descriptor *
-ioat3_prep_memset_lock(struct dma_chan *c, dma_addr_t dest, int value,
- size_t len, unsigned long flags)
-{
- struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
- struct ioat_ring_ent *desc;
- size_t total_len = len;
- struct ioat_fill_descriptor *fill;
- u64 src_data = (0x0101010101010101ULL) * (value & 0xff);
- int num_descs, idx, i;
-
- num_descs = ioat2_xferlen_to_descs(ioat, len);
- if (likely(num_descs) && ioat2_check_space_lock(ioat, num_descs) == 0)
- idx = ioat->head;
- else
- return NULL;
- i = 0;
- do {
- size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);
-
- desc = ioat2_get_ring_ent(ioat, idx + i);
- fill = desc->fill;
-
- fill->size = xfer_size;
- fill->src_data = src_data;
- fill->dst_addr = dest;
- fill->ctl = 0;
- fill->ctl_f.op = IOAT_OP_FILL;
-
- len -= xfer_size;
- dest += xfer_size;
- dump_desc_dbg(ioat, desc);
- } while (++i < num_descs);
-
- desc->txd.flags = flags;
- desc->len = total_len;
- fill->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
- fill->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
- fill->ctl_f.compl_write = 1;
- dump_desc_dbg(ioat, desc);
-
- /* we leave the channel locked to ensure in order submission */
- return &desc->txd;
-}
-
-static struct dma_async_tx_descriptor *
__ioat3_prep_xor_lock(struct dma_chan *c, enum sum_check_flags *result,
dma_addr_t dest, dma_addr_t *src, unsigned int src_cnt,
size_t len, unsigned long flags)
@@ -590,7 +765,8 @@ dump_pq_desc_dbg(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc, struct
int i;
dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
- " sz: %#x ctl: %#x (op: %d int: %d compl: %d pq: '%s%s' src_cnt: %d)\n",
+ " sz: %#10.8x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
+ " src_cnt: %d)\n",
desc_id(desc), (unsigned long long) desc->txd.phys,
(unsigned long long) (pq_ex ? pq_ex->next : pq->next),
desc->txd.flags, pq->size, pq->ctl, pq->ctl_f.op, pq->ctl_f.int_en,
@@ -602,6 +778,42 @@ dump_pq_desc_dbg(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc, struct
(unsigned long long) pq_get_src(descs, i), pq->coef[i]);
dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
+ dev_dbg(dev, "\tNEXT: %#llx\n", pq->next);
+}
+
+static void dump_pq16_desc_dbg(struct ioat2_dma_chan *ioat,
+ struct ioat_ring_ent *desc)
+{
+ struct device *dev = to_dev(&ioat->base);
+ struct ioat_pq_descriptor *pq = desc->pq;
+ struct ioat_raw_descriptor *descs[] = { (void *)pq,
+ (void *)pq,
+ (void *)pq };
+ int src_cnt = src16_cnt_to_sw(pq->ctl_f.src_cnt);
+ int i;
+
+ if (desc->sed) {
+ descs[1] = (void *)desc->sed->hw;
+ descs[2] = (void *)desc->sed->hw + 64;
+ }
+
+ dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
+ " sz: %#x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
+ " src_cnt: %d)\n",
+ desc_id(desc), (unsigned long long) desc->txd.phys,
+ (unsigned long long) pq->next,
+ desc->txd.flags, pq->size, pq->ctl,
+ pq->ctl_f.op, pq->ctl_f.int_en,
+ pq->ctl_f.compl_write,
+ pq->ctl_f.p_disable ? "" : "p", pq->ctl_f.q_disable ? "" : "q",
+ pq->ctl_f.src_cnt);
+ for (i = 0; i < src_cnt; i++) {
+ dev_dbg(dev, "\tsrc[%d]: %#llx coef: %#x\n", i,
+ (unsigned long long) pq16_get_src(descs, i),
+ pq->coef[i]);
+ }
+ dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
+ dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
}
static struct dma_async_tx_descriptor *
@@ -612,6 +824,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
{
struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
struct ioat_chan_common *chan = &ioat->base;
+ struct ioatdma_device *device = chan->device;
struct ioat_ring_ent *compl_desc;
struct ioat_ring_ent *desc;
struct ioat_ring_ent *ext;
@@ -622,6 +835,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
u32 offset = 0;
u8 op = result ? IOAT_OP_PQ_VAL : IOAT_OP_PQ;
int i, s, idx, with_ext, num_descs;
+ int cb32 = (device->version < IOAT_VER_3_3) ? 1 : 0;
dev_dbg(to_dev(chan), "%s\n", __func__);
/* the engine requires at least two sources (we provide
@@ -647,7 +861,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
* order.
*/
if (likely(num_descs) &&
- ioat2_check_space_lock(ioat, num_descs+1) == 0)
+ ioat2_check_space_lock(ioat, num_descs + cb32) == 0)
idx = ioat->head;
else
return NULL;
@@ -685,6 +899,9 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
pq->q_addr = dst[1] + offset;
pq->ctl = 0;
pq->ctl_f.op = op;
+ /* we turn on descriptor write back error status */
+ if (device->cap & IOAT_CAP_DWBES)
+ pq->ctl_f.wb_en = result ? 1 : 0;
pq->ctl_f.src_cnt = src_cnt_to_hw(s);
pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);
@@ -701,22 +918,140 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
dump_pq_desc_dbg(ioat, desc, ext);
- /* completion descriptor carries interrupt bit */
- compl_desc = ioat2_get_ring_ent(ioat, idx + i);
- compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
- hw = compl_desc->hw;
- hw->ctl = 0;
- hw->ctl_f.null = 1;
- hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
- hw->ctl_f.compl_write = 1;
- hw->size = NULL_DESC_BUFFER_SIZE;
- dump_desc_dbg(ioat, compl_desc);
+ if (!cb32) {
+ pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+ pq->ctl_f.compl_write = 1;
+ compl_desc = desc;
+ } else {
+ /* completion descriptor carries interrupt bit */
+ compl_desc = ioat2_get_ring_ent(ioat, idx + i);
+ compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
+ hw = compl_desc->hw;
+ hw->ctl = 0;
+ hw->ctl_f.null = 1;
+ hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+ hw->ctl_f.compl_write = 1;
+ hw->size = NULL_DESC_BUFFER_SIZE;
+ dump_desc_dbg(ioat, compl_desc);
+ }
+
/* we leave the channel locked to ensure in order submission */
return &compl_desc->txd;
}
static struct dma_async_tx_descriptor *
+__ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
+ const dma_addr_t *dst, const dma_addr_t *src,
+ unsigned int src_cnt, const unsigned char *scf,
+ size_t len, unsigned long flags)
+{
+ struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
+ struct ioat_chan_common *chan = &ioat->base;
+ struct ioatdma_device *device = chan->device;
+ struct ioat_ring_ent *desc;
+ size_t total_len = len;
+ struct ioat_pq_descriptor *pq;
+ u32 offset = 0;
+ u8 op;
+ int i, s, idx, num_descs;
+
+ /* this function is only called with 9-16 sources */
+ op = result ? IOAT_OP_PQ_VAL_16S : IOAT_OP_PQ_16S;
+
+ dev_dbg(to_dev(chan), "%s\n", __func__);
+
+ num_descs = ioat2_xferlen_to_descs(ioat, len);
+
+ /*
+ * 16 source pq is only available on cb3.3 and has no completion
+ * write hw bug.
+ */
+ if (num_descs && ioat2_check_space_lock(ioat, num_descs) == 0)
+ idx = ioat->head;
+ else
+ return NULL;
+
+ i = 0;
+
+ do {
+ struct ioat_raw_descriptor *descs[4];
+ size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);
+
+ desc = ioat2_get_ring_ent(ioat, idx + i);
+ pq = desc->pq;
+
+ descs[0] = (struct ioat_raw_descriptor *) pq;
+
+ desc->sed = ioat3_alloc_sed(device, (src_cnt-2) >> 3);
+ if (!desc->sed) {
+ dev_err(to_dev(chan),
+ "%s: no free sed entries\n", __func__);
+ return NULL;
+ }
+
+ pq->sed_addr = desc->sed->dma;
+ desc->sed->parent = desc;
+
+ descs[1] = (struct ioat_raw_descriptor *)desc->sed->hw;
+ descs[2] = (void *)descs[1] + 64;
+
+ for (s = 0; s < src_cnt; s++)
+ pq16_set_src(descs, src[s], offset, scf[s], s);
+
+ /* see the comment for dma_maxpq in include/linux/dmaengine.h */
+ if (dmaf_p_disabled_continue(flags))
+ pq16_set_src(descs, dst[1], offset, 1, s++);
+ else if (dmaf_continue(flags)) {
+ pq16_set_src(descs, dst[0], offset, 0, s++);
+ pq16_set_src(descs, dst[1], offset, 1, s++);
+ pq16_set_src(descs, dst[1], offset, 0, s++);
+ }
+
+ pq->size = xfer_size;
+ pq->p_addr = dst[0] + offset;
+ pq->q_addr = dst[1] + offset;
+ pq->ctl = 0;
+ pq->ctl_f.op = op;
+ pq->ctl_f.src_cnt = src16_cnt_to_hw(s);
+ /* we turn on descriptor write back error status */
+ if (device->cap & IOAT_CAP_DWBES)
+ pq->ctl_f.wb_en = result ? 1 : 0;
+ pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
+ pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);
+
+ len -= xfer_size;
+ offset += xfer_size;
+ } while (++i < num_descs);
+
+ /* last pq descriptor carries the unmap parameters and fence bit */
+ desc->txd.flags = flags;
+ desc->len = total_len;
+ if (result)
+ desc->result = result;
+ pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
+
+ /* with cb3.3 we should be able to do completion w/o a null desc */
+ pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+ pq->ctl_f.compl_write = 1;
+
+ dump_pq16_desc_dbg(ioat, desc);
+
+ /* we leave the channel locked to ensure in order submission */
+ return &desc->txd;
+}
+
+static int src_cnt_flags(unsigned int src_cnt, unsigned long flags)
+{
+ if (dmaf_p_disabled_continue(flags))
+ return src_cnt + 1;
+ else if (dmaf_continue(flags))
+ return src_cnt + 3;
+ else
+ return src_cnt;
+}
+
+static struct dma_async_tx_descriptor *
ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
unsigned int src_cnt, const unsigned char *scf, size_t len,
unsigned long flags)
@@ -740,11 +1075,20 @@ ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
single_source_coef[0] = scf[0];
single_source_coef[1] = 0;
- return __ioat3_prep_pq_lock(chan, NULL, dst, single_source, 2,
- single_source_coef, len, flags);
- } else
- return __ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt, scf,
- len, flags);
+ return src_cnt_flags(src_cnt, flags) > 8 ?
+ __ioat3_prep_pq16_lock(chan, NULL, dst, single_source,
+ 2, single_source_coef, len,
+ flags) :
+ __ioat3_prep_pq_lock(chan, NULL, dst, single_source, 2,
+ single_source_coef, len, flags);
+
+ } else {
+ return src_cnt_flags(src_cnt, flags) > 8 ?
+ __ioat3_prep_pq16_lock(chan, NULL, dst, src, src_cnt,
+ scf, len, flags) :
+ __ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt,
+ scf, len, flags);
+ }
}
struct dma_async_tx_descriptor *
@@ -763,8 +1107,11 @@ ioat3_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
*/
*pqres = 0;
- return __ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
- flags);
+ return src_cnt_flags(src_cnt, flags) > 8 ?
+ __ioat3_prep_pq16_lock(chan, pqres, pq, src, src_cnt, scf, len,
+ flags) :
+ __ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
+ flags);
}
static struct dma_async_tx_descriptor *
@@ -779,8 +1126,11 @@ ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
flags |= DMA_PREP_PQ_DISABLE_Q;
pq[1] = dst; /* specify valid address for disabled result */
- return __ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
- flags);
+ return src_cnt_flags(src_cnt, flags) > 8 ?
+ __ioat3_prep_pq16_lock(chan, NULL, pq, src, src_cnt, scf, len,
+ flags) :
+ __ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
+ flags);
}
struct dma_async_tx_descriptor *
@@ -801,8 +1151,11 @@ ioat3_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
flags |= DMA_PREP_PQ_DISABLE_Q;
pq[1] = pq[0]; /* specify valid address for disabled result */
- return __ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1, scf,
- len, flags);
+ return src_cnt_flags(src_cnt, flags) > 8 ?
+ __ioat3_prep_pq16_lock(chan, result, pq, &src[1], src_cnt - 1,
+ scf, len, flags) :
+ __ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1,
+ scf, len, flags);
}
static struct dma_async_tx_descriptor *
@@ -851,7 +1204,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
struct page *xor_srcs[IOAT_NUM_SRC_TEST];
struct page *xor_val_srcs[IOAT_NUM_SRC_TEST + 1];
dma_addr_t dma_srcs[IOAT_NUM_SRC_TEST + 1];
- dma_addr_t dma_addr, dest_dma;
+ dma_addr_t dest_dma;
struct dma_async_tx_descriptor *tx;
struct dma_chan *dma_chan;
dma_cookie_t cookie;
@@ -863,6 +1216,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
unsigned long tmo;
struct device *dev = &device->pdev->dev;
struct dma_device *dma = &device->common;
+ u8 op = 0;
dev_dbg(dev, "%s\n", __func__);
@@ -908,6 +1262,8 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
}
/* test xor */
+ op = IOAT_OP_XOR;
+
dest_dma = dma_map_page(dev, dest, 0, PAGE_SIZE, DMA_FROM_DEVICE);
for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
dma_srcs[i] = dma_map_page(dev, xor_srcs[i], 0, PAGE_SIZE,
@@ -919,7 +1275,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (!tx) {
dev_err(dev, "Self-test xor prep failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
async_tx_ack(tx);
@@ -930,18 +1286,22 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (cookie < 0) {
dev_err(dev, "Self-test xor setup failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
dma->device_issue_pending(dma_chan);
tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
- if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+ if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
dev_err(dev, "Self-test xor timed out\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
+ dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
+ for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
+ dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
dma_sync_single_for_cpu(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
for (i = 0; i < (PAGE_SIZE / sizeof(u32)); i++) {
u32 *ptr = page_address(dest);
@@ -957,6 +1317,8 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (!dma_has_cap(DMA_XOR_VAL, dma_chan->device->cap_mask))
goto free_resources;
+ op = IOAT_OP_XOR_VAL;
+
/* validate the sources with the destintation page */
for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
xor_val_srcs[i] = xor_srcs[i];
@@ -973,7 +1335,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (!tx) {
dev_err(dev, "Self-test zero prep failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
async_tx_ack(tx);
@@ -984,69 +1346,32 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (cookie < 0) {
dev_err(dev, "Self-test zero setup failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
dma->device_issue_pending(dma_chan);
tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
- if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+ if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
dev_err(dev, "Self-test validate timed out\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
+ for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+ dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
if (xor_val_result != 0) {
dev_err(dev, "Self-test validate failed compare\n");
err = -ENODEV;
goto free_resources;
}
- /* skip memset if the capability is not present */
- if (!dma_has_cap(DMA_MEMSET, dma_chan->device->cap_mask))
- goto free_resources;
-
- /* test memset */
- dma_addr = dma_map_page(dev, dest, 0,
- PAGE_SIZE, DMA_FROM_DEVICE);
- tx = dma->device_prep_dma_memset(dma_chan, dma_addr, 0, PAGE_SIZE,
- DMA_PREP_INTERRUPT);
- if (!tx) {
- dev_err(dev, "Self-test memset prep failed\n");
- err = -ENODEV;
- goto free_resources;
- }
-
- async_tx_ack(tx);
- init_completion(&cmp);
- tx->callback = ioat3_dma_test_callback;
- tx->callback_param = &cmp;
- cookie = tx->tx_submit(tx);
- if (cookie < 0) {
- dev_err(dev, "Self-test memset setup failed\n");
- err = -ENODEV;
- goto free_resources;
- }
- dma->device_issue_pending(dma_chan);
-
- tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
-
- if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
- dev_err(dev, "Self-test memset timed out\n");
- err = -ENODEV;
- goto free_resources;
- }
-
- for (i = 0; i < PAGE_SIZE/sizeof(u32); i++) {
- u32 *ptr = page_address(dest);
- if (ptr[i]) {
- dev_err(dev, "Self-test memset failed compare\n");
- err = -ENODEV;
- goto free_resources;
- }
- }
+ memset(page_address(dest), 0, PAGE_SIZE);
/* test for non-zero parity sum */
+ op = IOAT_OP_XOR_VAL;
+
xor_val_result = 0;
for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
dma_srcs[i] = dma_map_page(dev, xor_val_srcs[i], 0, PAGE_SIZE,
@@ -1057,7 +1382,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (!tx) {
dev_err(dev, "Self-test 2nd zero prep failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
async_tx_ack(tx);
@@ -1068,24 +1393,39 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
if (cookie < 0) {
dev_err(dev, "Self-test 2nd zero setup failed\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
dma->device_issue_pending(dma_chan);
tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
- if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+ if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
dev_err(dev, "Self-test 2nd validate timed out\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
if (xor_val_result != SUM_CHECK_P_RESULT) {
dev_err(dev, "Self-test validate failed compare\n");
err = -ENODEV;
- goto free_resources;
+ goto dma_unmap;
}
+ for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+ dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
+ goto free_resources;
+dma_unmap:
+ if (op == IOAT_OP_XOR) {
+ dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
+ for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
+ dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
+ DMA_TO_DEVICE);
+ } else if (op == IOAT_OP_XOR_VAL) {
+ for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+ dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
+ DMA_TO_DEVICE);
+ }
free_resources:
dma->device_free_chan_resources(dma_chan);
out:
@@ -1110,6 +1450,40 @@ static int ioat3_dma_self_test(struct ioatdma_device *device)
return 0;
}
+static int ioat3_irq_reinit(struct ioatdma_device *device)
+{
+ struct pci_dev *pdev = device->pdev;
+ int irq = pdev->irq, i;
+
+ if (!is_bwd_ioat(pdev))
+ return 0;
+
+ switch (device->irq_mode) {
+ case IOAT_MSIX:
+ for (i = 0; i < device->common.chancnt; i++) {
+ struct msix_entry *msix = &device->msix_entries[i];
+ struct ioat_chan_common *chan;
+
+ chan = ioat_chan_by_index(device, i);
+ devm_free_irq(&pdev->dev, msix->vector, chan);
+ }
+
+ pci_disable_msix(pdev);
+ break;
+ case IOAT_MSI:
+ pci_disable_msi(pdev);
+ /* fall through */
+ case IOAT_INTX:
+ devm_free_irq(&pdev->dev, irq, device);
+ break;
+ default:
+ return 0;
+ }
+ device->irq_mode = IOAT_NOIRQ;
+
+ return ioat_dma_setup_interrupts(device);
+}
+
static int ioat3_reset_hw(struct ioat_chan_common *chan)
{
/* throw away whatever the channel was doing and get it
@@ -1126,64 +1500,62 @@ static int ioat3_reset_hw(struct ioat_chan_common *chan)
chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
- /* -= IOAT ver.3 workarounds =- */
- /* Write CHANERRMSK_INT with 3E07h to mask out the errors
- * that can cause stability issues for IOAT ver.3, and clear any
- * pending errors
- */
- pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0x3e07);
- err = pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
- if (err) {
- dev_err(&pdev->dev, "channel error register unreachable\n");
- return err;
+ if (device->version < IOAT_VER_3_3) {
+ /* clear any pending errors */
+ err = pci_read_config_dword(pdev,
+ IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
+ if (err) {
+ dev_err(&pdev->dev,
+ "channel error register unreachable\n");
+ return err;
+ }
+ pci_write_config_dword(pdev,
+ IOAT_PCI_CHANERR_INT_OFFSET, chanerr);
+
+ /* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
+ * (workaround for spurious config parity error after restart)
+ */
+ pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
+ if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0) {
+ pci_write_config_dword(pdev,
+ IOAT_PCI_DMAUNCERRSTS_OFFSET,
+ 0x10);
+ }
}
- pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr);
- /* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
- * (workaround for spurious config parity error after restart)
- */
- pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
- if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0)
- pci_write_config_dword(pdev, IOAT_PCI_DMAUNCERRSTS_OFFSET, 0x10);
+ err = ioat2_reset_sync(chan, msecs_to_jiffies(200));
+ if (!err)
+ err = ioat3_irq_reinit(device);
- return ioat2_reset_sync(chan, msecs_to_jiffies(200));
-}
+ if (err)
+ dev_err(&pdev->dev, "Failed to reset: %d\n", err);
-static bool is_jf_ioat(struct pci_dev *pdev)
-{
- switch (pdev->device) {
- case PCI_DEVICE_ID_INTEL_IOAT_JSF0:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF1:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF2:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF3:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF4:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF5:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF6:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF7:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF8:
- case PCI_DEVICE_ID_INTEL_IOAT_JSF9:
- return true;
- default:
- return false;
- }
+ return err;
}
-static bool is_snb_ioat(struct pci_dev *pdev)
+static void ioat3_intr_quirk(struct ioatdma_device *device)
{
- switch (pdev->device) {
- case PCI_DEVICE_ID_INTEL_IOAT_SNB0:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB1:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB2:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB3:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB4:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB5:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB6:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB7:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB8:
- case PCI_DEVICE_ID_INTEL_IOAT_SNB9:
- return true;
- default:
- return false;
+ struct dma_device *dma;
+ struct dma_chan *c;
+ struct ioat_chan_common *chan;
+ u32 errmask;
+
+ dma = &device->common;
+
+ /*
+ * if we have descriptor write back error status, we mask the
+ * error interrupts
+ */
+ if (device->cap & IOAT_CAP_DWBES) {
+ list_for_each_entry(c, &dma->channels, device_node) {
+ chan = to_chan_common(c);
+ errmask = readl(chan->reg_base +
+ IOAT_CHANERR_MASK_OFFSET);
+ errmask |= IOAT_CHANERR_XOR_P_OR_CRC_ERR |
+ IOAT_CHANERR_XOR_Q_ERR;
+ writel(errmask, chan->reg_base +
+ IOAT_CHANERR_MASK_OFFSET);
+ }
}
}
@@ -1196,33 +1568,32 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
struct ioat_chan_common *chan;
bool is_raid_device = false;
int err;
- u32 cap;
device->enumerate_channels = ioat2_enumerate_channels;
device->reset_hw = ioat3_reset_hw;
device->self_test = ioat3_dma_self_test;
+ device->intr_quirk = ioat3_intr_quirk;
dma = &device->common;
dma->device_prep_dma_memcpy = ioat2_dma_prep_memcpy_lock;
dma->device_issue_pending = ioat2_issue_pending;
dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
dma->device_free_chan_resources = ioat2_free_chan_resources;
- if (is_jf_ioat(pdev) || is_snb_ioat(pdev))
- dma->copy_align = 6;
-
dma_cap_set(DMA_INTERRUPT, dma->cap_mask);
dma->device_prep_dma_interrupt = ioat3_prep_interrupt_lock;
- cap = readl(device->reg_base + IOAT_DMA_CAP_OFFSET);
+ device->cap = readl(device->reg_base + IOAT_DMA_CAP_OFFSET);
+
+ if (is_xeon_cb32(pdev) || is_bwd_noraid(pdev))
+ device->cap &= ~(IOAT_CAP_XOR | IOAT_CAP_PQ | IOAT_CAP_RAID16SS);
/* dca is incompatible with raid operations */
- if (dca_en && (cap & (IOAT_CAP_XOR|IOAT_CAP_PQ)))
- cap &= ~(IOAT_CAP_XOR|IOAT_CAP_PQ);
+ if (dca_en && (device->cap & (IOAT_CAP_XOR|IOAT_CAP_PQ)))
+ device->cap &= ~(IOAT_CAP_XOR|IOAT_CAP_PQ);
- if (cap & IOAT_CAP_XOR) {
+ if (device->cap & IOAT_CAP_XOR) {
is_raid_device = true;
dma->max_xor = 8;
- dma->xor_align = 6;
dma_cap_set(DMA_XOR, dma->cap_mask);
dma->device_prep_dma_xor = ioat3_prep_xor;
@@ -1230,53 +1601,56 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
dma->device_prep_dma_xor_val = ioat3_prep_xor_val;
}
- if (cap & IOAT_CAP_PQ) {
+
+ if (device->cap & IOAT_CAP_PQ) {
is_raid_device = true;
- dma_set_maxpq(dma, 8, 0);
- dma->pq_align = 6;
- dma_cap_set(DMA_PQ, dma->cap_mask);
dma->device_prep_dma_pq = ioat3_prep_pq;
-
- dma_cap_set(DMA_PQ_VAL, dma->cap_mask);
dma->device_prep_dma_pq_val = ioat3_prep_pq_val;
+ dma_cap_set(DMA_PQ, dma->cap_mask);
+ dma_cap_set(DMA_PQ_VAL, dma->cap_mask);
- if (!(cap & IOAT_CAP_XOR)) {
- dma->max_xor = 8;
- dma->xor_align = 6;
+ if (device->cap & IOAT_CAP_RAID16SS) {
+ dma_set_maxpq(dma, 16, 0);
+ } else {
+ dma_set_maxpq(dma, 8, 0);
+ }
- dma_cap_set(DMA_XOR, dma->cap_mask);
+ if (!(device->cap & IOAT_CAP_XOR)) {
dma->device_prep_dma_xor = ioat3_prep_pqxor;
-
- dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
dma->device_prep_dma_xor_val = ioat3_prep_pqxor_val;
+ dma_cap_set(DMA_XOR, dma->cap_mask);
+ dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
+
+ if (device->cap & IOAT_CAP_RAID16SS) {
+ dma->max_xor = 16;
+ } else {
+ dma->max_xor = 8;
+ }
}
}
- if (is_raid_device && (cap & IOAT_CAP_FILL_BLOCK)) {
- dma_cap_set(DMA_MEMSET, dma->cap_mask);
- dma->device_prep_dma_memset = ioat3_prep_memset_lock;
- }
+ dma->device_tx_status = ioat3_tx_status;
+ device->cleanup_fn = ioat3_cleanup_event;
+ device->timer_fn = ioat3_timer_event;
- if (is_raid_device) {
- dma->device_tx_status = ioat3_tx_status;
- device->cleanup_fn = ioat3_cleanup_event;
- device->timer_fn = ioat3_timer_event;
- } else {
- dma->device_tx_status = ioat_dma_tx_status;
- device->cleanup_fn = ioat2_cleanup_event;
- device->timer_fn = ioat2_timer_event;
- }
+ /* starting with CB3.3 super extended descriptors are supported */
+ if (device->cap & IOAT_CAP_RAID16SS) {
+ char pool_name[14];
+ int i;
- #ifdef CONFIG_ASYNC_TX_DISABLE_PQ_VAL_DMA
- dma_cap_clear(DMA_PQ_VAL, dma->cap_mask);
- dma->device_prep_dma_pq_val = NULL;
- #endif
+ for (i = 0; i < MAX_SED_POOLS; i++) {
+ snprintf(pool_name, 14, "ioat_hw%d_sed", i);
- #ifdef CONFIG_ASYNC_TX_DISABLE_XOR_VAL_DMA
- dma_cap_clear(DMA_XOR_VAL, dma->cap_mask);
- dma->device_prep_dma_xor_val = NULL;
- #endif
+ /* allocate SED DMA pool */
+ device->sed_hw_pool[i] = dmam_pool_create(pool_name,
+ &pdev->dev,
+ SED_SIZE * (i + 1), 64, 0);
+ if (!device->sed_hw_pool[i])
+ return -ENOMEM;
+
+ }
+ }
err = ioat_probe(device);
if (err)
diff --git a/drivers/dma/ioat/hw.h b/drivers/dma/ioat/hw.h
index d2ff3fda0b1..62f83e983d8 100644
--- a/drivers/dma/ioat/hw.h
+++ b/drivers/dma/ioat/hw.h
@@ -30,10 +30,39 @@
#define IOAT_PCI_DID_SCNB 0x65FF
#define IOAT_PCI_DID_SNB 0x402F
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB0 0x0e20
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB1 0x0e21
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB2 0x0e22
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB3 0x0e23
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB4 0x0e24
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB5 0x0e25
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB6 0x0e26
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB7 0x0e27
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB8 0x0e2e
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB9 0x0e2f
+
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW0 0x2f20
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW1 0x2f21
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW2 0x2f22
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW3 0x2f23
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW4 0x2f24
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW5 0x2f25
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW6 0x2f26
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW7 0x2f27
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW8 0x2f2e
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW9 0x2f2f
+
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD0 0x0C50
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD1 0x0C51
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD2 0x0C52
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD3 0x0C53
+
#define IOAT_VER_1_2 0x12 /* Version 1.2 */
#define IOAT_VER_2_0 0x20 /* Version 2.0 */
#define IOAT_VER_3_0 0x30 /* Version 3.0 */
#define IOAT_VER_3_2 0x32 /* Version 3.2 */
+#define IOAT_VER_3_3 0x33 /* Version 3.3 */
+
int system_has_dca_enabled(struct pci_dev *pdev);
@@ -71,33 +100,6 @@ struct ioat_dma_descriptor {
uint64_t user2;
};
-struct ioat_fill_descriptor {
- uint32_t size;
- union {
- uint32_t ctl;
- struct {
- unsigned int int_en:1;
- unsigned int rsvd:1;
- unsigned int dest_snoop_dis:1;
- unsigned int compl_write:1;
- unsigned int fence:1;
- unsigned int rsvd2:2;
- unsigned int dest_brk:1;
- unsigned int bundle:1;
- unsigned int rsvd4:15;
- #define IOAT_OP_FILL 0x01
- unsigned int op:8;
- } ctl_f;
- };
- uint64_t src_data;
- uint64_t dst_addr;
- uint64_t next;
- uint64_t rsv1;
- uint64_t next_dst_addr;
- uint64_t user1;
- uint64_t user2;
-};
-
struct ioat_xor_descriptor {
uint32_t size;
union {
@@ -136,7 +138,17 @@ struct ioat_xor_ext_descriptor {
};
struct ioat_pq_descriptor {
- uint32_t size;
+ union {
+ uint32_t size;
+ uint32_t dwbes;
+ struct {
+ unsigned int rsvd:25;
+ unsigned int p_val_err:1;
+ unsigned int q_val_err:1;
+ unsigned int rsvd1:4;
+ unsigned int wbes:1;
+ } dwbes_f;
+ };
union {
uint32_t ctl;
struct {
@@ -151,9 +163,14 @@ struct ioat_pq_descriptor {
unsigned int hint:1;
unsigned int p_disable:1;
unsigned int q_disable:1;
- unsigned int rsvd:11;
+ unsigned int rsvd2:2;
+ unsigned int wb_en:1;
+ unsigned int prl_en:1;
+ unsigned int rsvd3:7;
#define IOAT_OP_PQ 0x89
#define IOAT_OP_PQ_VAL 0x8a
+ #define IOAT_OP_PQ_16S 0xa0
+ #define IOAT_OP_PQ_VAL_16S 0xa1
unsigned int op:8;
} ctl_f;
};
@@ -161,7 +178,10 @@ struct ioat_pq_descriptor {
uint64_t p_addr;
uint64_t next;
uint64_t src_addr2;
- uint64_t src_addr3;
+ union {
+ uint64_t src_addr3;
+ uint64_t sed_addr;
+ };
uint8_t coef[8];
uint64_t q_addr;
};
@@ -210,4 +230,40 @@ struct ioat_pq_update_descriptor {
struct ioat_raw_descriptor {
uint64_t field[8];
};
+
+struct ioat_pq16a_descriptor {
+ uint8_t coef[8];
+ uint64_t src_addr3;
+ uint64_t src_addr4;
+ uint64_t src_addr5;
+ uint64_t src_addr6;
+ uint64_t src_addr7;
+ uint64_t src_addr8;
+ uint64_t src_addr9;
+};
+
+struct ioat_pq16b_descriptor {
+ uint64_t src_addr10;
+ uint64_t src_addr11;
+ uint64_t src_addr12;
+ uint64_t src_addr13;
+ uint64_t src_addr14;
+ uint64_t src_addr15;
+ uint64_t src_addr16;
+ uint64_t rsvd;
+};
+
+union ioat_sed_pq_descriptor {
+ struct ioat_pq16a_descriptor a;
+ struct ioat_pq16b_descriptor b;
+};
+
+#define SED_SIZE 64
+
+struct ioat_sed_raw_descriptor {
+ uint64_t a[8];
+ uint64_t b[8];
+ uint64_t c[8];
+};
+
#endif
diff --git a/drivers/dma/ioat/pci.c b/drivers/dma/ioat/pci.c
index 4f686c527ab..1d051cd045d 100644
--- a/drivers/dma/ioat/pci.c
+++ b/drivers/dma/ioat/pci.c
@@ -40,17 +40,6 @@ MODULE_VERSION(IOAT_DMA_VERSION);
MODULE_LICENSE("Dual BSD/GPL");
MODULE_AUTHOR("Intel Corporation");
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB0 0x0e20
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB1 0x0e21
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB2 0x0e22
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB3 0x0e23
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB4 0x0e24
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB5 0x0e25
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB6 0x0e26
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB7 0x0e27
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB8 0x0e2e
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB9 0x0e2f
-
static struct pci_device_id ioat_pci_tbl[] = {
/* I/OAT v1 platforms */
{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT) },
@@ -105,6 +94,23 @@ static struct pci_device_id ioat_pci_tbl[] = {
{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_IVB8) },
{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_IVB9) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW0) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW1) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW2) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW3) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW4) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW5) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW6) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW7) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW8) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW9) },
+
+ /* I/OAT v3.3 platforms */
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD0) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD1) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD2) },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD3) },
+
{ 0, }
};
MODULE_DEVICE_TABLE(pci, ioat_pci_tbl);
@@ -117,6 +123,7 @@ module_param(ioat_dca_enabled, int, 0644);
MODULE_PARM_DESC(ioat_dca_enabled, "control support of dca service (default: 1)");
struct kmem_cache *ioat2_cache;
+struct kmem_cache *ioat3_sed_cache;
#define DRV_NAME "ioatdma"
@@ -212,7 +219,7 @@ static void ioat_remove(struct pci_dev *pdev)
static int __init ioat_init_module(void)
{
- int err;
+ int err = -ENOMEM;
pr_info("%s: Intel(R) QuickData Technology Driver %s\n",
DRV_NAME, IOAT_DMA_VERSION);
@@ -222,9 +229,21 @@ static int __init ioat_init_module(void)
if (!ioat2_cache)
return -ENOMEM;
+ ioat3_sed_cache = KMEM_CACHE(ioat_sed_ent, 0);
+ if (!ioat3_sed_cache)
+ goto err_ioat2_cache;
+
err = pci_register_driver(&ioat_pci_driver);
if (err)
- kmem_cache_destroy(ioat2_cache);
+ goto err_ioat3_cache;
+
+ return 0;
+
+ err_ioat3_cache:
+ kmem_cache_destroy(ioat3_sed_cache);
+
+ err_ioat2_cache:
+ kmem_cache_destroy(ioat2_cache);
return err;
}
diff --git a/drivers/dma/ioat/registers.h b/drivers/dma/ioat/registers.h
index 1391798542b..2f1cfa0f1f4 100644
--- a/drivers/dma/ioat/registers.h
+++ b/drivers/dma/ioat/registers.h
@@ -79,6 +79,8 @@
#define IOAT_CAP_APIC 0x00000080
#define IOAT_CAP_XOR 0x00000100
#define IOAT_CAP_PQ 0x00000200
+#define IOAT_CAP_DWBES 0x00002000
+#define IOAT_CAP_RAID16SS 0x00020000
#define IOAT_CHANNEL_MMIO_SIZE 0x80 /* Each Channel MMIO space is this size */
@@ -93,6 +95,8 @@
#define IOAT_CHANCTRL_ERR_COMPLETION_EN 0x0004
#define IOAT_CHANCTRL_INT_REARM 0x0001
#define IOAT_CHANCTRL_RUN (IOAT_CHANCTRL_INT_REARM |\
+ IOAT_CHANCTRL_ERR_INT_EN |\
+ IOAT_CHANCTRL_ERR_COMPLETION_EN |\
IOAT_CHANCTRL_ANY_ERR_ABORT_EN)
#define IOAT_DMA_COMP_OFFSET 0x02 /* 16-bit DMA channel compatibility */
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 22:10:47 +0000