From: "Jan Beulich" <JBeulich@suse.com>
To: <mingo@elte.hu>, <tglx@linutronix.de>, <hpa@zytor.com>
Cc: <linux-kernel@vger.kernel.org>
Subject: [PATCH] x86: clear HPET configuration registers on startup
Date: Mon, 02 Apr 2012 15:15:55 +0100 [thread overview]
Message-ID: <4F79D0BB020000780007C02D@nat28.tlf.novell.com> (raw)
While Linux itself has been calling hpet_disable() for quite a while,
having e.g. a secondary (kexec) kernel depend on such behavior of the
primary (crashed) environment is fragile. It particularly broke until
very recently when the primary environment was Xen based, as that
hypervisor did not clear any of the HPET settings it may have used.
Rather than blindly (and incompletely) clearing certain HPET settings
in hpet_disable(), latch the config register settings during boot and
restore then here.
(Note on the hpet_set_mode() change: Now that we're clearing the level
bit upon initialization, there's no need anymore to do so here.)
Signed-off-by: Jan Beulich <jbeulich@suse.com>
---
arch/x86/kernel/hpet.c | 59 ++++++++++++++++++++++++++++++++++++++++++-------
1 file changed, 51 insertions(+), 8 deletions(-)
--- 3.4-rc1-x86-hpet.orig/arch/x86/kernel/hpet.c
+++ 3.4-rc1-x86-hpet/arch/x86/kernel/hpet.c
@@ -319,8 +319,6 @@ static void hpet_set_mode(enum clock_eve
now = hpet_readl(HPET_COUNTER);
cmp = now + (unsigned int) delta;
cfg = hpet_readl(HPET_Tn_CFG(timer));
- /* Make sure we use edge triggered interrupts */
- cfg &= ~HPET_TN_LEVEL;
cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC |
HPET_TN_SETVAL | HPET_TN_32BIT;
hpet_writel(cfg, HPET_Tn_CFG(timer));
@@ -787,15 +785,16 @@ static int hpet_clocksource_register(voi
return 0;
}
+static u32 *hpet_boot_cfg;
+
/**
* hpet_enable - Try to setup the HPET timer. Returns 1 on success.
*/
int __init hpet_enable(void)
{
- unsigned long hpet_period;
- unsigned int id;
+ u32 hpet_period, cfg, id;
u64 freq;
- int i;
+ unsigned int i, last;
if (!is_hpet_capable())
return 0;
@@ -847,15 +846,45 @@ int __init hpet_enable(void)
id = hpet_readl(HPET_ID);
hpet_print_config();
+ last = (id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT;
+
#ifdef CONFIG_HPET_EMULATE_RTC
/*
* The legacy routing mode needs at least two channels, tick timer
* and the rtc emulation channel.
*/
- if (!(id & HPET_ID_NUMBER))
+ if (!last)
goto out_nohpet;
#endif
+ cfg = hpet_readl(HPET_CFG);
+ hpet_boot_cfg = kmalloc((last + 2) * sizeof(*hpet_boot_cfg),
+ GFP_KERNEL);
+ if (hpet_boot_cfg)
+ *hpet_boot_cfg = cfg;
+ else
+ pr_warn("HPET initial state will not be saved\n");
+ cfg &= ~(HPET_CFG_ENABLE | HPET_CFG_LEGACY);
+ hpet_writel(cfg, HPET_Tn_CFG(i));
+ if (cfg)
+ pr_warn("HPET: Unrecognized bits %#x set in global cfg\n",
+ cfg);
+
+ for (i = 0; i <= last; ++i) {
+ cfg = hpet_readl(HPET_Tn_CFG(i));
+ if (hpet_boot_cfg)
+ hpet_boot_cfg[i + 1] = cfg;
+ cfg &= ~(HPET_TN_ENABLE | HPET_TN_LEVEL | HPET_TN_FSB);
+ hpet_writel(cfg, HPET_Tn_CFG(i));
+ cfg &= ~(HPET_TN_PERIODIC | HPET_TN_PERIODIC_CAP
+ | HPET_TN_64BIT_CAP | HPET_TN_32BIT | HPET_TN_ROUTE
+ | HPET_TN_FSB | HPET_TN_FSB_CAP);
+ if (cfg)
+ pr_warn("HPET: Unrecognized bits %#x set in cfg#%u\n",
+ cfg, i);
+ }
+ hpet_print_config();
+
if (hpet_clocksource_register())
goto out_nohpet;
@@ -923,14 +952,28 @@ fs_initcall(hpet_late_init);
void hpet_disable(void)
{
if (is_hpet_capable() && hpet_virt_address) {
- unsigned int cfg = hpet_readl(HPET_CFG);
+ unsigned int cfg = hpet_readl(HPET_CFG), id, last;
- if (hpet_legacy_int_enabled) {
+ if (hpet_boot_cfg)
+ cfg = *hpet_boot_cfg;
+ else if (hpet_legacy_int_enabled) {
cfg &= ~HPET_CFG_LEGACY;
hpet_legacy_int_enabled = 0;
}
cfg &= ~HPET_CFG_ENABLE;
hpet_writel(cfg, HPET_CFG);
+
+ if (!hpet_boot_cfg)
+ return;
+
+ id = hpet_readl(HPET_ID);
+ last = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT);
+
+ for (id = 0; id <= last; ++id)
+ hpet_writel(hpet_boot_cfg[id + 1], HPET_Tn_CFG(id));
+
+ if (*hpet_boot_cfg & HPET_CFG_ENABLE)
+ hpet_writel(*hpet_boot_cfg, HPET_CFG);
}
}
next reply other threads:[~2012-04-02 14:16 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-02 14:15 Jan Beulich [this message]
2012-05-08 4:22 ` [tip:timers/core] x86: Clear " tip-bot for Jan Beulich
2012-05-24 22:06 ` [PATCH] x86: clear " Thomas Gleixner
2012-05-25 10:21 ` Jan Beulich
2012-05-25 10:33 ` Thomas Gleixner
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=4F79D0BB020000780007C02D@nat28.tlf.novell.com \
--to=jbeulich@suse.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=tglx@linutronix.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
all inboxes | Powered by JetHome®