From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
To: bzolnier@gmail.com, rah@bash.sh
Cc: linux-ide@vger.kernel.org
Subject: [PATCH 1/2] hpt366: fix PCI clock detection for HPT374
Date: Mon, 6 Aug 2007 00:06:35 +0400 [thread overview]
Message-ID: <200708060006.35511.sshtylyov@ru.mvista.com> (raw)
HPT374 BIOS seems to only save f_CNT register value for the function #0 before
re-tuning DPLL causing the driver to report obviously distorted f_CNT for the
function #1 -- fix this by always reading the saved f_CNT register value from
in the init_chipset() method the function #0 of HPT374 chip.
While at it, introduce 'chip_type' for the copy of the 'struct hpt_info' member
and replace the structure assignment by memcpy()...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
---
This is against the current Linus tree and unfortunately I was able to only
compile test it since that tree gives MODPOST warning and dies early.
Bob, please test it if/when you'll be able to and report the results...
drivers/ide/pci/hpt366.c | 27 +++++++++++++++++++--------
1 files changed, 19 insertions(+), 8 deletions(-)
Index: linux-2.6/drivers/ide/pci/hpt366.c
===================================================================
--- linux-2.6.orig/drivers/ide/pci/hpt366.c
+++ linux-2.6/drivers/ide/pci/hpt366.c
@@ -1,5 +1,5 @@
/*
- * linux/drivers/ide/pci/hpt366.c Version 1.10 Jun 29, 2007
+ * linux/drivers/ide/pci/hpt366.c Version 1.11 Aug 4, 2007
*
* Copyright (C) 1999-2003 Andre Hedrick <andre@linux-ide.org>
* Portions Copyright (C) 2001 Sun Microsystems, Inc.
@@ -981,6 +981,7 @@ static unsigned int __devinit init_chips
struct hpt_info *info = kmalloc(sizeof(struct hpt_info), GFP_KERNEL);
unsigned long io_base = pci_resource_start(dev, 4);
u8 pci_clk, dpll_clk = 0; /* PCI and DPLL clock in MHz */
+ u8 chip_type;
enum ata_clock clock;
if (info == NULL) {
@@ -992,7 +993,8 @@ static unsigned int __devinit init_chips
* Copy everything from a static "template" structure
* to just allocated per-chip hpt_info structure.
*/
- *info = *(struct hpt_info *)pci_get_drvdata(dev);
+ memcpy(info, pci_get_drvdata(dev), sizeof(struct hpt_info));
+ chip_type = info->chip_type;
pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
@@ -1002,7 +1004,7 @@ static unsigned int __devinit init_chips
/*
* First, try to estimate the PCI clock frequency...
*/
- if (info->chip_type >= HPT370) {
+ if (chip_type >= HPT370) {
u8 scr1 = 0;
u16 f_cnt = 0;
u32 temp = 0;
@@ -1016,7 +1018,7 @@ static unsigned int __devinit init_chips
* HighPoint does this for HPT372A.
* NOTE: This register is only writeable via I/O space.
*/
- if (info->chip_type == HPT372A)
+ if (chip_type == HPT372A)
outb(0x0e, io_base + 0x9c);
/*
@@ -1040,7 +1042,16 @@ static unsigned int __devinit init_chips
* reading the f_CNT register itself in hopes that nobody has
* touched the DPLL yet...
*/
- temp = inl(io_base + 0x90);
+ if (chip_type == HPT374 && (PCI_FUNC(dev->devfn) & 1)) {
+ struct pci_dev *dev1 = pci_get_slot(dev->bus,
+ dev->devfn - 1);
+ unsigned long io_base = pci_resource_start(dev1, 4);
+
+ temp = inl(io_base + 0x90);
+ pci_dev_put(dev1);
+ } else
+ temp = inl(io_base + 0x90);
+
if ((temp & 0xFFFFF000) != 0xABCDE000) {
int i;
@@ -1120,7 +1131,7 @@ static unsigned int __devinit init_chips
* We also don't like using the DPLL because this causes glitches
* on PRST-/SRST- when the state engine gets reset...
*/
- if (info->chip_type >= HPT374 || info->settings[clock] == NULL) {
+ if (chip_type >= HPT374 || info->settings[clock] == NULL) {
u16 f_low, delta = pci_clk < 50 ? 2 : 4;
int adjust;
@@ -1190,7 +1201,7 @@ static unsigned int __devinit init_chips
/* Point to this chip's own instance of the hpt_info structure. */
pci_set_drvdata(dev, info);
- if (info->chip_type >= HPT370) {
+ if (chip_type >= HPT370) {
u8 mcr1, mcr4;
/*
@@ -1209,7 +1220,7 @@ static unsigned int __devinit init_chips
* the MISC. register to stretch the UltraDMA Tss timing.
* NOTE: This register is only writeable via I/O space.
*/
- if (info->chip_type == HPT371N && clock == ATA_CLOCK_66MHZ)
+ if (chip_type == HPT371N && clock == ATA_CLOCK_66MHZ)
outb(inb(io_base + 0x9c) | 0x04, io_base + 0x9c);
next reply other threads:[~2007-08-05 20:04 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-05 20:06 Sergei Shtylyov [this message]
2007-08-06 17:10 ` [PATCH 1/2] hpt366: fix PCI clock detection for HPT374 Bob Ham
2007-08-06 17:18 ` Sergei Shtylyov
2007-08-07 9:01 ` Bob Ham
2007-08-06 19:29 ` Alan Cox
2007-08-10 15:41 ` Sergei Shtylyov
2007-08-10 17:19 ` Alan Cox
2007-08-10 17:29 ` Sergei Shtylyov
2007-08-10 21:54 ` Bartlomiej Zolnierkiewicz
2007-08-11 16:03 ` Sergei Shtylyov
2007-08-11 16:41 ` Bartlomiej Zolnierkiewicz
2007-08-11 16:54 ` Sergei Shtylyov
2007-08-11 17:07 ` Bartlomiej Zolnierkiewicz
2007-08-11 17:23 ` Sergei Shtylyov
2007-08-11 21:31 ` Bartlomiej Zolnierkiewicz
2007-08-17 17:43 ` Sergei Shtylyov
2007-08-08 21:10 ` Bartlomiej Zolnierkiewicz
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=200708060006.35511.sshtylyov@ru.mvista.com \
--to=sshtylyov@ru.mvista.com \
--cc=bzolnier@gmail.com \
--cc=linux-ide@vger.kernel.org \
--cc=rah@bash.sh \
/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).