From: Grant Likely <grant.likely@secretlab.ca>
To: linuxppc-dev@ozlabs.org, hans.lehmann@ritter-elektronik.de,
matt@genesi-usa.com, olaf@aepfle.de
Subject: [PATCH 3/4] powerpc/mpc5200: Disable bestcomm prefetching when ATA DMA enabled
Date: Sun, 21 Dec 2008 01:06:30 -0700 [thread overview]
Message-ID: <20081221080630.29029.51667.stgit@localhost.localdomain> (raw)
In-Reply-To: <20081221080213.29029.7646.stgit@localhost.localdomain>
From: Grant Likely <grant.likely@secretlab.ca>
When ATA DMA is enabled, bestcomm prefetching does not work. This
patch adds a function to disable bestcomm prefetch when the ATA
Bestcomm task is initialized.
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
arch/powerpc/sysdev/bestcomm/ata.c | 3 +++
arch/powerpc/sysdev/bestcomm/bestcomm.c | 7 ++-----
arch/powerpc/sysdev/bestcomm/bestcomm_priv.h | 16 ++++++++++++++++
3 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/sysdev/bestcomm/ata.c b/arch/powerpc/sysdev/bestcomm/ata.c
index 1f5258f..901c9f9 100644
--- a/arch/powerpc/sysdev/bestcomm/ata.c
+++ b/arch/powerpc/sysdev/bestcomm/ata.c
@@ -61,6 +61,9 @@ bcom_ata_init(int queue_len, int maxbufsize)
struct bcom_ata_var *var;
struct bcom_ata_inc *inc;
+ /* Prefetch breaks ATA DMA. Turn it off for ATA DMA */
+ bcom_disable_prefetch();
+
tsk = bcom_task_alloc(queue_len, sizeof(struct bcom_ata_bd), 0);
if (!tsk)
return NULL;
diff --git a/arch/powerpc/sysdev/bestcomm/bestcomm.c b/arch/powerpc/sysdev/bestcomm/bestcomm.c
index 446c9ea..378ebd9 100644
--- a/arch/powerpc/sysdev/bestcomm/bestcomm.c
+++ b/arch/powerpc/sysdev/bestcomm/bestcomm.c
@@ -279,7 +279,6 @@ bcom_engine_init(void)
int task;
phys_addr_t tdt_pa, ctx_pa, var_pa, fdt_pa;
unsigned int tdt_size, ctx_size, var_size, fdt_size;
- u16 regval;
/* Allocate & clear SRAM zones for FDT, TDTs, contexts and vars/incs */
tdt_size = BCOM_MAX_TASKS * sizeof(struct bcom_tdt);
@@ -331,10 +330,8 @@ bcom_engine_init(void)
out_8(&bcom_eng->regs->ipr[BCOM_INITIATOR_ALWAYS], BCOM_IPR_ALWAYS);
/* Disable COMM Bus Prefetch on the original 5200; it's broken */
- if ((mfspr(SPRN_SVR) & MPC5200_SVR_MASK) == MPC5200_SVR) {
- regval = in_be16(&bcom_eng->regs->PtdCntrl);
- out_be16(&bcom_eng->regs->PtdCntrl, regval | 1);
- }
+ if ((mfspr(SPRN_SVR) & MPC5200_SVR_MASK) == MPC5200_SVR)
+ bcom_disable_prefetch();
/* Init lock */
spin_lock_init(&bcom_eng->lock);
diff --git a/arch/powerpc/sysdev/bestcomm/bestcomm_priv.h b/arch/powerpc/sysdev/bestcomm/bestcomm_priv.h
index 746f155..eb0d1c8 100644
--- a/arch/powerpc/sysdev/bestcomm/bestcomm_priv.h
+++ b/arch/powerpc/sysdev/bestcomm/bestcomm_priv.h
@@ -241,6 +241,22 @@ extern void bcom_set_initiator(int task, int initiator);
#define TASK_ENABLE 0x8000
+/**
+ * bcom_disable_prefetch - Hook to disable bus prefetching
+ *
+ * ATA DMA and the original MPC5200 need this due to silicon bugs. At the
+ * moment disabling prefetch is a one-way street. There is no mechanism
+ * in place to turn prefetch back on after it has been disabled. There is
+ * no reason it couldn't be done, it would just be more complex to implement.
+ */
+static inline void bcom_disable_prefetch(void)
+{
+ u16 regval;
+
+ regval = in_be16(&bcom_eng->regs->PtdCntrl);
+ out_be16(&bcom_eng->regs->PtdCntrl, regval | 1);
+};
+
static inline void
bcom_enable_task(int task)
{
next prev parent reply other threads:[~2008-12-21 8:06 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-21 8:06 [PATCH 0/4] powerpc/mpc5200: Add support for ATA DMA Grant Likely
2008-12-21 8:06 ` [PATCH 1/4] powerpc/mpc5200: Bugfix on handling variable sized buffer descriptors Grant Likely
2008-12-21 8:06 ` [PATCH 2/4] powerpc/mpc5200: Bestcomm fixes to ATA support Grant Likely
2008-12-21 8:06 ` Grant Likely [this message]
2008-12-21 8:06 ` [PATCH 4/4] powerpc/mpc5200: Add MDMA/UDMA support to MPC5200 ATA driver Grant Likely
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=20081221080630.29029.51667.stgit@localhost.localdomain \
--to=grant.likely@secretlab.ca \
--cc=hans.lehmann@ritter-elektronik.de \
--cc=linuxppc-dev@ozlabs.org \
--cc=matt@genesi-usa.com \
--cc=olaf@aepfle.de \
/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).