From: Tejun Heo <htejun@gmail.com>
To: jgarzik@pobox.com, linux-ide@vger.kernel.org
Cc: Tejun Heo <htejun@gmail.com>
Subject: [PATCH 4/4] sata_sil: implement Mod15Write workaround by split request processing
Date: Thu, 19 Oct 2006 14:51:39 +0900 [thread overview]
Message-ID: <11612370993859-git-send-email-htejun@gmail.com> (raw)
In-Reply-To: <1161237098656-git-send-email-htejun@gmail.com>
Instead of limiting max_sectors to 15 unconditionally, this patch
leaves max_sectors at the default value and process write requests
larger than 15 sectors by processing them in <= 15 sectors chunks.
This results in unhamplered read performance and better write
performance.
Split request processing is performed only for DMA write requests > 15
sectors. dev->max_sectors is limited to 15 sectors when the device is
put into PIO mode.
Signed-off-by: Tejun Heo <htejun@gmail.com>
---
drivers/ata/sata_sil.c | 307 ++++++++++++++++++++++++++++++++++++++++++++++--
1 files changed, 294 insertions(+), 13 deletions(-)
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index f844a1f..37e186f 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -43,6 +43,7 @@ #include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/device.h>
#include <scsi/scsi_host.h>
+#include <scsi/scsi_device.h>
#include <linux/libata.h>
#define DRV_NAME "sata_sil"
@@ -106,6 +107,9 @@ enum {
*/
SIL_QUIRK_MOD15WRITE = (1 << 0),
SIL_QUIRK_UDMA5MAX = (1 << 1),
+
+ SIL_M15W_ENABLED = (1 << 0),
+ SIL_M15W_ACTIVATED = (1 << 1),
};
static int sil_init_one (struct pci_dev *pdev, const struct pci_device_id *ent);
@@ -113,12 +117,16 @@ #ifdef CONFIG_PM
static int sil_pci_device_resume(struct pci_dev *pdev);
#endif
static void sil_dev_config(struct ata_port *ap, struct ata_device *dev);
+static void sil_qc_prep(struct ata_queued_cmd *qc);
static u32 sil_scr_read (struct ata_port *ap, unsigned int sc_reg);
static void sil_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val);
static void sil_post_set_mode (struct ata_port *ap);
static irqreturn_t sil_interrupt(int irq, void *dev_instance);
static void sil_freeze(struct ata_port *ap);
static void sil_thaw(struct ata_port *ap);
+static void sil_error_handler(struct ata_port *ap);
+static int sil_port_start(struct ata_port *ap);
+static void sil_port_stop(struct ata_port *ap);
static const struct pci_device_id sil_pci_tbl[] = {
@@ -198,19 +206,19 @@ static const struct ata_port_operations
.bmdma_start = ata_bmdma_start,
.bmdma_stop = ata_bmdma_stop,
.bmdma_status = ata_bmdma_status,
- .qc_prep = ata_qc_prep,
+ .qc_prep = sil_qc_prep,
.qc_issue = ata_qc_issue_prot,
.data_xfer = ata_mmio_data_xfer,
.freeze = sil_freeze,
.thaw = sil_thaw,
- .error_handler = ata_bmdma_error_handler,
+ .error_handler = sil_error_handler,
.post_internal_cmd = ata_bmdma_post_internal_cmd,
.irq_handler = sil_interrupt,
.irq_clear = ata_bmdma_irq_clear,
.scr_read = sil_scr_read,
.scr_write = sil_scr_write,
- .port_start = ata_port_start,
- .port_stop = ata_port_stop,
+ .port_start = sil_port_start,
+ .port_stop = sil_port_stop,
.host_stop = ata_pci_host_stop,
};
@@ -275,6 +283,48 @@ static const struct {
/* ... port 3 */
};
+/*
+ * Context to loop over write requests > 15 sectors for Mod15Write bug.
+ *
+ * The following libata layer fields are saved at the beginning and
+ * mangled as necessary.
+ *
+ * qc->sg : To fool ata_fill_sg().
+ * qc->n_elem : ditto.
+ * qc->flags : Except for the last iteration, ATA_QCFLAG_DMAMAP
+ * should be off on entering ata_interrupt() such
+ * that ata_qc_complete() doesn't call ata_sg_clean()
+ * before sil_m15w_chunk_complete(), but the flags
+ * should be set for ata_qc_prep() to work.
+ * qc->complete_fn : Overrided to sil_m15w_chunk_complete().
+ *
+ * The following cxt fields are used to iterate over write requests.
+ *
+ * next_block : The starting block of the next chunk.
+ * next_sg : The first sg entry of the next chunk.
+ * left : Total bytes left.
+ * cur_sg_ofs : Number of processed bytes in the first sg entry
+ * of this chunk.
+ * next_sg_ofs : Number of bytes to be processed in the last sg
+ * entry of this chunk.
+ * next_sg_len : Number of bytes to be processed in the first sg
+ * entry of the next chunk.
+ */
+struct sil_m15w_cxt {
+ unsigned int flags;
+ u64 next_block;
+ struct scatterlist * next_sg;
+ unsigned int left;
+ unsigned int cur_sg_ofs;
+ unsigned int next_sg_ofs;
+ unsigned int next_sg_len;
+
+ struct scatterlist * orig_sg;
+ unsigned int orig_nelem;
+ unsigned long orig_flags;
+ ata_qc_cb_t orig_complete_fn;
+};
+
MODULE_AUTHOR("Jeff Garzik");
MODULE_DESCRIPTION("low-level driver for Silicon Image SATA controller");
MODULE_LICENSE("GPL");
@@ -295,6 +345,7 @@ static unsigned char sil_get_device_cach
static void sil_post_set_mode (struct ata_port *ap)
{
+ struct sil_m15w_cxt *m15w_cxt = ap->private_data;
struct ata_host *host = ap->host;
struct ata_device *dev;
void __iomem *addr = host->mmio_base + sil_port[ap->port_no].xfer_mode;
@@ -318,6 +369,193 @@ static void sil_post_set_mode (struct at
tmp |= (dev_mode[1] << 4);
writel(tmp, addr);
readl(addr); /* flush */
+
+ if (m15w_cxt->flags & SIL_M15W_ENABLED) {
+ struct ata_device *dev = ap->device;
+ struct scsi_device *sdev = dev->sdev;
+
+ if (!(dev->flags & ATA_DFLAG_PIO))
+ m15w_cxt->flags |= SIL_M15W_ACTIVATED;
+ else
+ dev->max_sectors = 15;
+
+ if (sdev)
+ blk_queue_max_sectors(sdev->request_queue,
+ dev->max_sectors);
+ }
+}
+
+static inline u64 sil_m15w_read_tf_block(struct ata_taskfile *tf)
+{
+ u64 block = 0;
+
+ block |= (u64)tf->lbal;
+ block |= (u64)tf->lbam << 8;
+ block |= (u64)tf->lbah << 16;
+
+ if (tf->flags & ATA_TFLAG_LBA48) {
+ block |= (u64)tf->hob_lbal << 24;
+ block |= (u64)tf->hob_lbam << 32;
+ block |= (u64)tf->hob_lbah << 40;
+ } else
+ block |= (u64)(tf->device & 0xf) << 24;
+
+ return block;
+}
+
+static inline void sil_m15w_rewrite_tf(struct ata_taskfile *tf,
+ u64 block, u16 nsect)
+{
+ tf->nsect = nsect & 0xff;
+ tf->lbal = block & 0xff;
+ tf->lbam = (block >> 8) & 0xff;
+ tf->lbah = (block >> 16) & 0xff;
+
+ if (tf->flags & ATA_TFLAG_LBA48) {
+ tf->hob_nsect = (nsect >> 8) & 0xff;
+ tf->hob_lbal = (block >> 24) & 0xff;
+ tf->hob_lbam = (block >> 32) & 0xff;
+ tf->hob_lbah = (block >> 40) & 0xff;
+ } else {
+ tf->device &= ~0xf;
+ tf->device |= (block >> 24) & 0xf;
+ }
+}
+
+static void sil_m15w_next(struct ata_queued_cmd *qc)
+{
+ struct sil_m15w_cxt *cxt = qc->ap->private_data;
+ struct scatterlist *sg;
+ unsigned int todo, res, nelem;
+
+ if (qc->__sg != cxt->next_sg) {
+ sg_dma_address(qc->__sg) -= cxt->cur_sg_ofs;
+ sg_dma_len(qc->__sg) += cxt->cur_sg_ofs;
+ cxt->cur_sg_ofs = 0;
+ }
+ cxt->cur_sg_ofs += cxt->next_sg_ofs;
+
+ qc->__sg = sg = cxt->next_sg;
+ sg_dma_address(sg) += cxt->next_sg_ofs;
+ sg_dma_len(sg) = cxt->next_sg_len;
+
+ res = todo = min_t(unsigned int, cxt->left, 15 << 9);
+
+ nelem = 0;
+ while (sg_dma_len(sg) <= res) {
+ res -= sg_dma_len(sg);
+ sg++;
+ nelem++;
+ }
+
+ if (todo < cxt->left) {
+ cxt->next_sg = sg;
+ cxt->next_sg_ofs = res;
+ cxt->next_sg_len = sg_dma_len(sg) - res;
+ if (res) {
+ nelem++;
+ sg_dma_len(sg) = res;
+ }
+ } else {
+ cxt->next_sg = NULL;
+ cxt->next_sg_ofs = 0;
+ cxt->next_sg_len = 0;
+ }
+
+ DPRINTK("block=%llu nelem=%u todo=%u left=%u\n",
+ cxt->next_block, nelem, todo, cxt->left);
+
+ qc->n_elem = nelem;
+ sil_m15w_rewrite_tf(&qc->tf, cxt->next_block, todo >> 9);
+ cxt->left -= todo;
+ cxt->next_block += todo >> 9;
+}
+
+static inline void sil_m15w_restore_qc(struct ata_queued_cmd *qc)
+{
+ struct sil_m15w_cxt *cxt = qc->ap->private_data;
+
+ DPRINTK("ENTER\n");
+
+ sg_dma_address(qc->__sg) -= cxt->cur_sg_ofs;
+ sg_dma_len(qc->__sg) += cxt->cur_sg_ofs;
+ if (cxt->next_sg_ofs)
+ sg_dma_len(cxt->next_sg) += cxt->next_sg_len;
+ qc->__sg = cxt->orig_sg;
+ qc->n_elem = cxt->orig_nelem;
+ qc->flags |= cxt->orig_flags;
+ qc->complete_fn = cxt->orig_complete_fn;
+}
+
+static void sil_m15w_chunk_complete(struct ata_queued_cmd *qc)
+{
+ struct ata_port *ap = qc->ap;
+ struct sil_m15w_cxt *cxt = ap->private_data;
+
+ DPRINTK("ENTER\n");
+
+ /* sil_error_handler() restores qc before EH and this function
+ * must never be invoked for a failed qc.
+ */
+ BUG_ON(qc->err_mask);
+
+ /* this command is still active, turn ACTIVE back on */
+ qc->flags |= ATA_QCFLAG_ACTIVE;
+ ap->active_tag = qc->tag;
+ ap->qc_active |= 1 << qc->tag;
+
+ /* iterate to the next chunk and fill sg */
+ sil_m15w_next(qc);
+ qc->flags |= cxt->orig_flags;
+ ata_qc_prep(qc);
+ qc->flags &= ~ATA_QCFLAG_DMAMAP;
+
+ /* If this is the last iteration, restore fields such that
+ * normal completion path is run on chunk completion.
+ */
+ if (!cxt->left)
+ sil_m15w_restore_qc(qc);
+
+ /* issue the next chunk */
+ sil_ops.qc_issue(qc);
+}
+
+static void sil_qc_prep(struct ata_queued_cmd *qc)
+{
+ struct sil_m15w_cxt *m15w_cxt = qc->ap->private_data;
+
+ if (unlikely((m15w_cxt->flags & SIL_M15W_ACTIVATED) &&
+ (qc->tf.flags & ATA_TFLAG_WRITE) && qc->nsect > 15)) {
+ BUG_ON(m15w_cxt->left || qc->tf.protocol != ATA_PROT_DMA);
+
+ /* okay, begin mod15write workaround */
+ m15w_cxt->next_block = sil_m15w_read_tf_block(&qc->tf);
+ m15w_cxt->next_sg = qc->__sg;
+ m15w_cxt->left = qc->nsect << 9;
+ m15w_cxt->cur_sg_ofs = 0;
+ m15w_cxt->next_sg_ofs = 0;
+ m15w_cxt->next_sg_len = sg_dma_len(qc->__sg);
+
+ /* Save fields we're gonna mess with. Read comments
+ * above struct sil_m15w_cxt for more info.
+ */
+ m15w_cxt->orig_sg = qc->__sg;
+ m15w_cxt->orig_nelem = qc->n_elem;
+ m15w_cxt->orig_flags = qc->flags & ATA_QCFLAG_DMAMAP;
+ m15w_cxt->orig_complete_fn = qc->complete_fn;
+ qc->complete_fn = sil_m15w_chunk_complete;
+
+ DPRINTK("MOD15WRITE, block=%llu nsect=%u\n",
+ m15w_cxt->next_block, qc->nsect);
+
+ sil_m15w_next(qc);
+
+ ata_qc_prep(qc);
+ qc->flags &= ~ATA_QCFLAG_DMAMAP;
+ return;
+ }
+
+ ata_qc_prep(qc);
}
static inline unsigned long sil_scr_addr(struct ata_port *ap, unsigned int sc_reg)
@@ -503,6 +741,23 @@ static void sil_thaw(struct ata_port *ap
writel(tmp, mmio_base + SIL_SYSCFG);
}
+static void sil_error_handler(struct ata_port *ap)
+{
+ struct sil_m15w_cxt *m15w_cxt = ap->private_data;
+ struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, ap->active_tag);
+
+ if (m15w_cxt->left) {
+ BUG_ON(!qc);
+ ata_port_printk(ap, KERN_INFO, "exception while processing "
+ "m15w chunks, left=%u, restoring qc\n",
+ m15w_cxt->left);
+ sil_m15w_restore_qc(qc);
+ m15w_cxt->left = 0;
+ }
+
+ ata_bmdma_error_handler(ap);
+}
+
/**
* sil_dev_config - Apply device/host-specific errata fixups
* @ap: Port containing device to be examined
@@ -513,17 +768,12 @@ static void sil_thaw(struct ata_port *ap
* We apply two errata fixups which are specific to Silicon Image,
* a Seagate and a Maxtor fixup.
*
- * For certain Seagate devices, we must limit the maximum sectors
- * to under 8K.
+ * For certain Seagate devices, we cannot issue write requests
+ * larger than 15 sectors.
*
* For certain Maxtor devices, we must not program the drive
* beyond udma5.
*
- * Both fixups are unfairly pessimistic. As soon as I get more
- * information on these errata, I will create a more exhaustive
- * list, and apply the fixups to only the specific
- * devices/hosts/firmwares that need it.
- *
* 20040111 - Seagate drives affected by the Mod15Write bug are blacklisted
* The Maxtor quirk is in the blacklist, but I'm keeping the original
* pessimistic fix for the following reasons...
@@ -531,9 +781,19 @@ static void sil_thaw(struct ata_port *ap
* Windows driver, maybe only one is affected. More info would be greatly
* appreciated.
* - But then again UDMA5 is hardly anything to complain about
+ *
+ * 20050316 Tejun Heo - Proper Mod15Write workaround implemented.
+ * sata_sil doesn't report affected Seagate drives as having max
+ * sectors of 15 anymore, but handle write requests larger than
+ * 15 sectors by looping over it inside this driver proper. This
+ * is messy but it's better to isolate this kind of peculiar bug
+ * handling inside individual drivers than tainting libata layer.
+ * This workaround results in unhampered read performance and
+ * much better write performance.
*/
static void sil_dev_config(struct ata_port *ap, struct ata_device *dev)
{
+ struct sil_m15w_cxt *m15w_cxt = ap->private_data;
int print_info = ap->eh_context.i.flags & ATA_EHI_PRINTINFO;
unsigned int n, quirks = 0;
unsigned char model_num[41];
@@ -546,14 +806,15 @@ static void sil_dev_config(struct ata_po
break;
}
- /* limit requests to 15 sectors */
+ /* activate mod15write quirk workaround */
+ m15w_cxt->flags = 0;
if (slow_down ||
((ap->flags & SIL_FLAG_MOD15WRITE) &&
(quirks & SIL_QUIRK_MOD15WRITE))) {
if (print_info)
ata_dev_printk(dev, KERN_INFO, "applying Seagate "
"errata fix (mod15write workaround)\n");
- dev->max_sectors = 15;
+ m15w_cxt->flags |= SIL_M15W_ENABLED;
return;
}
@@ -613,6 +874,26 @@ static void sil_init_controller(struct p
}
}
+static int sil_port_start(struct ata_port *ap)
+{
+ struct sil_m15w_cxt *m15w_cxt;
+
+ m15w_cxt = kzalloc(sizeof(m15w_cxt[0]), GFP_KERNEL);
+ if (!m15w_cxt)
+ return -ENOMEM;
+
+ ap->private_data = m15w_cxt;
+
+ return ata_port_start(ap);
+}
+
+static void sil_port_stop(struct ata_port *ap)
+{
+ kfree(ap->private_data);
+
+ ata_port_stop(ap);
+}
+
static int sil_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
{
static int printed_version;
--
1.4.2.3
next prev parent reply other threads:[~2006-10-19 5:51 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-10-19 5:51 [PATCHSET] sata_sil: implement Mod15Write workaround by split request processing Tejun Heo
2006-10-19 5:51 ` [PATCH 1/4] libata: implement ATA_EHI_PRINTINFO Tejun Heo
2006-11-01 1:57 ` Jeff Garzik
2006-10-19 5:51 ` Tejun Heo [this message]
2006-11-01 1:59 ` [PATCH 4/4] sata_sil: implement Mod15Write workaround by split request processing Jeff Garzik
2006-11-01 2:17 ` Tejun Heo
2006-11-01 2:31 ` Jeff Garzik
2006-11-01 9:16 ` Tejun Heo
2006-10-19 5:51 ` [PATCH 3/4] libata: move dev->max_sectors configuration into ata_dev_configure() Tejun Heo
2006-10-19 5:51 ` [PATCH 2/4] libata: implement ATA_EHI_SETMODE and ATA_EHI_POST_SETMODE Tejun Heo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=11612370993859-git-send-email-htejun@gmail.com \
--to=htejun@gmail.com \
--cc=jgarzik@pobox.com \
--cc=linux-ide@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).