From: Andi Kleen <ak@suse.de>
To: patches@x86-64.org, linux-kernel@vger.kernel.org
Subject: [PATCH] [7/45] x86_64: Replace nvidia timer override quirk with pci id list and unify quirks
Date: Fri, 21 Sep 2007 22:44:48 +0200 (CEST) [thread overview]
Message-ID: <20070921204448.5E0EB14EFF@wotan.suse.de> (raw)
In-Reply-To: <200709211044.901175000@suse.de>
This replaces the old NF3/NF4 reference BIOS timer override quirk with a device
ID list. We need to ignore the timer override on these systems, but not
ignore it on NF5 based systems. Previously this was distingushed by checking
for HPET, but a lot of BIOS vendors didn't enable HPET in their pre Vista BIOSes.
Replace the old "for all of nvidia" quirk with a quirk containing pci device
ID. I goobled this list together from pci.ids and googling and it may be incomplete.
I'm still not 100% sure the list is correct, but the only way
to find out is to do testing in mainline. So let's do that.
Signed-off-by: Andi Kleen <ak@suse.de>
---
arch/x86_64/kernel/early-quirks.c | 50 ++++++++++++++++++--------------------
1 file changed, 24 insertions(+), 26 deletions(-)
Index: linux/arch/x86_64/kernel/early-quirks.c
===================================================================
--- linux.orig/arch/x86_64/kernel/early-quirks.c
+++ linux/arch/x86_64/kernel/early-quirks.c
@@ -33,36 +33,20 @@ static void __init via_bugs(void)
#endif
}
-#ifdef CONFIG_ACPI
-
-static int __init nvidia_hpet_check(struct acpi_table_header *header)
-{
- return 0;
-}
-#endif
-
static void __init nvidia_bugs(void)
{
#ifdef CONFIG_ACPI
#ifdef CONFIG_X86_IO_APIC
/*
- * All timer overrides on Nvidia are
- * wrong unless HPET is enabled.
- * Unfortunately that's not true on many Asus boards.
- * We don't know yet how to detect this automatically, but
- * at least allow a command line override.
+ * All timer overrides on Nvidia NF3/NF4 are
+ * wrong.
*/
if (acpi_use_timer_override)
return;
- if (acpi_table_parse(ACPI_SIG_HPET, nvidia_hpet_check)) {
- acpi_skip_timer_override = 1;
- printk(KERN_INFO "Nvidia board "
- "detected. Ignoring ACPI "
- "timer override.\n");
- printk(KERN_INFO "If you got timer trouble "
- "try acpi_use_timer_override\n");
- }
+ acpi_skip_timer_override = 1;
+ printk(KERN_INFO "Nvidia board detected. Ignoring ACPI timer override.\n");
+ printk(KERN_INFO "If you got timer trouble try acpi_use_timer_override\n");
#endif
#endif
/* RED-PEN skip them on mptables too? */
@@ -83,10 +67,19 @@ static void __init ati_bugs(void)
struct chipset {
u16 vendor;
void (*f)(void);
+ int id;
};
static struct chipset early_qrk[] __initdata = {
- { PCI_VENDOR_ID_NVIDIA, nvidia_bugs },
+ /* This list should cover at least one PCI ID from each NF3 or NF4
+ mainboard to handle a bug in their reference BIOS. May be incomplete. */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x00dd }, /* nforce 3 */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x00e1 }, /* nforce 3 */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x00ed }, /* nforce 3 */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x003d }, /* mcp 04 ?? */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x005c }, /* ck 804 */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x026f }, /* mcp 51 / nf4 ? */
+ { PCI_VENDOR_ID_NVIDIA, nvidia_bugs, 0x02f0 }, /* mcp 51 / nf4 ? */
{ PCI_VENDOR_ID_VIA, via_bugs },
{ PCI_VENDOR_ID_ATI, ati_bugs },
{}
@@ -99,12 +92,13 @@ void __init early_quirks(void)
if (!early_pci_allowed())
return;
- /* Poor man's PCI discovery */
+ /* Poor man's PCI discovery.
+ We just look for a chipset unique PCI bridge; not scan all devices */
for (num = 0; num < 32; num++) {
for (slot = 0; slot < 32; slot++) {
for (func = 0; func < 8; func++) {
u32 class;
- u32 vendor;
+ u32 vendor, device;
u8 type;
int i;
class = read_pci_config(num,slot,func,
@@ -117,13 +111,17 @@ void __init early_quirks(void)
vendor = read_pci_config(num, slot, func,
PCI_VENDOR_ID);
+ device = vendor >> 16;
+
vendor &= 0xffff;
- for (i = 0; early_qrk[i].f; i++)
- if (early_qrk[i].vendor == vendor) {
+ for (i = 0; early_qrk[i].f; i++) {
+ struct chipset *c = &early_qrk[i];
+ if (c->vendor == vendor && (!c->id || (c->id && c->id==device))) {
early_qrk[i].f();
return;
}
+ }
type = read_pci_config_byte(num, slot, func,
PCI_HEADER_TYPE);
next prev parent reply other threads:[~2007-09-21 20:47 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-21 20:44 [PATCH] [0/45] x86 2.6.24 patches review I Andi Kleen
2007-09-21 20:44 ` [PATCH] [1/45] x86: misc. constifications Andi Kleen
2007-09-21 20:44 ` [PATCH] [2/45] x86: constify stacktrace_ops Andi Kleen
2007-09-21 20:44 ` [PATCH] [3/45] x86: Separate checking of unsynchronized and unstable TSC Andi Kleen
2007-09-21 20:44 ` [PATCH] [4/45] i386: Rewrite sched_clock Andi Kleen
2007-09-21 20:44 ` [PATCH] [5/45] x86_64: Use new shared sched_clock in x86-64 too Andi Kleen
2007-09-21 20:44 ` [PATCH] [6/45] x86: Unify i386 and x86-64 early quirks Andi Kleen
2007-09-21 20:44 ` Andi Kleen [this message]
2007-09-21 20:44 ` [PATCH] [8/45] x86_64: Use string instruction memcpy on AMD Fam11h Andi Kleen
2007-09-22 9:29 ` Jan Engelhardt
2007-09-21 20:44 ` [PATCH] [9/45] i386: Clean up duplicate includes in arch/i386/kernel/ Andi Kleen
2007-09-21 20:44 ` [PATCH] [10/45] x86_64: x86_64 Sanitize user specified e820 memmap values Andi Kleen
2007-09-21 20:44 ` [PATCH] [11/45] x86_64: Remove rogue default m in drivers/video/Kconfig Andi Kleen
2007-09-22 3:15 ` Len Brown
2007-09-21 20:44 ` [PATCH] [12/45] i386: Fix arch/i386/kernel/nmi.c - 'unknown_nmi_panic_callback' declared 'static' but never defined warning Andi Kleen
2007-09-21 21:08 ` Dave Jones
2007-09-21 20:44 ` [PATCH] [13/45] x86_64: Increase VDSO_TEXT_OFFSET for ancient binutils Andi Kleen
2007-11-14 0:21 ` CaT
2007-11-14 0:25 ` Andi Kleen
2007-09-21 20:44 ` [PATCH] [14/45] x86: Create clflush() inline, remove hardcoded wbinvd Andi Kleen
2007-09-21 20:44 ` [PATCH] [15/45] i386: i386 add AMD64 Barcelona PMU MSR definitions to msr.h Andi Kleen
2007-09-21 20:44 ` [PATCH] [16/45] i386: do not BUG_ON() when MSR is unknown Andi Kleen
2007-09-21 20:44 ` [PATCH] [17/45] i386: make Oprofile call shutdown() only once per session Andi Kleen
2007-09-21 20:44 ` [PATCH] [18/45] x86_64: 0 -> NULL, for arch/x86_64 Andi Kleen
2007-09-21 20:45 ` [PATCH] [19/45] x86_64: Always accumulate compiler options in CFLAGS Andi Kleen
2007-09-21 20:45 ` [PATCH] [20/45] x86_64: Use 8 byte stack alignment when possible Andi Kleen
2007-09-21 21:13 ` Dave Jones
2007-09-21 22:34 ` Andi Kleen
2007-09-21 22:52 ` Dave Jones
2007-09-21 21:19 ` Jakub Jelinek
2007-09-30 22:07 ` Andi Kleen
2007-09-21 20:45 ` [PATCH] [21/45] x86_64: Some cleanups for pci gart code Andi Kleen
2007-09-21 20:45 ` [PATCH] [22/45] x86_64: Enable iommu_merge by default Andi Kleen
2007-09-21 20:45 ` [PATCH] [23/45] i386: Make callgraph use dump_trace() on i386/x86_64 Andi Kleen
2007-09-21 20:45 ` [PATCH] [24/45] x86: Introduce frame_pointer() and stack_pointer() Andi Kleen
2007-09-22 9:26 ` Christoph Hellwig
2007-10-01 16:48 ` Andi Kleen
2007-09-21 20:45 ` [PATCH] [25/45] x86_64: remove sync_Arb_IDs Andi Kleen
2007-09-21 20:45 ` [PATCH] [26/45] x86_64: clear IO_APIC before enabing apic error vector Andi Kleen
2007-09-21 20:45 ` [PATCH] [27/45] x86_64: convert mm_context_t semaphore to a mutex Andi Kleen
2007-09-21 20:45 ` [PATCH] [28/45] x86_64: clean up apicid_to_node declaration Andi Kleen
2007-09-21 20:45 ` [PATCH] [29/45] i386: Consolidate show_regs and show_registers for i386 Andi Kleen
2007-09-21 20:45 ` [PATCH] [30/45] i386: Remove local CPU logic in MTRR call to smp_call_function_single Andi Kleen
2007-09-21 20:45 ` [PATCH] [31/45] i386: make struct apic_probe static Andi Kleen
2007-09-21 20:45 ` [PATCH] [32/45] x86_64: hide cond_syscall behind __KERNEL__ Andi Kleen
2007-09-22 9:30 ` Christoph Hellwig
2007-09-21 20:45 ` [PATCH] [33/45] i386: es7000 minor cleanups Andi Kleen
2007-09-21 20:45 ` [PATCH] [34/45] i386: no need to make enable_cpu_hotplug a variable Andi Kleen
2007-09-21 20:45 ` [PATCH] [35/45] i386: make some variables static Andi Kleen
2007-09-21 20:45 ` [PATCH] [36/45] x86_64: kmalloc + memset conversion to kzalloc Andi Kleen
2007-09-21 20:45 ` [PATCH] [37/45] i386: remove -maccumulate-outgoing-args Andi Kleen
2007-09-21 20:45 ` [PATCH] [38/45] i386: setup_trampoline() must be __cpuinit Andi Kleen
2007-09-21 20:45 ` [PATCH] [39/45] x86_64: block irq balancing for timer Andi Kleen
2007-09-21 20:45 ` [PATCH] [40/45] i386: deactivate the test for the dead CONFIG_DEBUG_PAGE_TYPE Andi Kleen
2007-09-21 20:45 ` [PATCH] [41/45] i386: remove unnecessary code Andi Kleen
2007-09-21 20:45 ` [PATCH] [42/45] x86_64: use descriptor's functions instead of inline assembly Andi Kleen
2007-09-21 20:45 ` [PATCH] [43/45] i386: Clean up duplicate includes in arch/i386/xen/ Andi Kleen
2007-09-21 20:45 ` [PATCH] [44/45] i386: simplify smp_call_function_single() call sequence in cpuid Andi Kleen
2007-09-21 20:45 ` [PATCH] [45/45] i386: simplify smp_call_function_single() call sequence Andi Kleen
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=20070921204448.5E0EB14EFF@wotan.suse.de \
--to=ak@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=patches@x86-64.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
all inboxes | Powered by JetHome®