[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1437435958-15421-5-git-send-email-kys@microsoft.com>
Date: Mon, 20 Jul 2015 16:45:58 -0700
From: "K. Y. Srinivasan" <kys@...rosoft.com>
To: gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org,
devel@...uxdriverproject.org, olaf@...fle.de, apw@...onical.com,
vkuznets@...hat.com, jasowang@...hat.com
Cc: "K. Y. Srinivasan" <kys@...rosoft.com>
Subject: [PATCH V2 5/5] Drivers: hv: vmbus: Implement a clocksource based on the TSC page
The current Hyper-V clock source is based on the per-partition reference counter
and this counter is being accessed via s synthetic MSR - HV_X64_MSR_TIME_REF_COUNT.
Hyper-V has a more efficient way of computing the per-partition reference
counter value that does not involve reading a synthetic MSR. We implement
a time source based on this mechanism.
Tested-by: Vivek Yadav <vyadav@...rosoft.com>
Signed-off-by: K. Y. Srinivasan <kys@...rosoft.com>
---
V2: Cleanup and unregister the clocksource as part of cleanup Dexuan Cui <decui@...rosoft.com>
arch/x86/include/uapi/asm/hyperv.h | 2 +
drivers/hv/hv.c | 81 ++++++++++++++++++++++++++++++++++++
drivers/hv/hyperv_vmbus.h | 9 ++++
3 files changed, 92 insertions(+), 0 deletions(-)
diff --git a/arch/x86/include/uapi/asm/hyperv.h b/arch/x86/include/uapi/asm/hyperv.h
index 8fba544..89f15e3 100644
--- a/arch/x86/include/uapi/asm/hyperv.h
+++ b/arch/x86/include/uapi/asm/hyperv.h
@@ -27,6 +27,8 @@
#define HV_X64_MSR_VP_RUNTIME_AVAILABLE (1 << 0)
/* Partition Reference Counter (HV_X64_MSR_TIME_REF_COUNT) available*/
#define HV_X64_MSR_TIME_REF_COUNT_AVAILABLE (1 << 1)
+/* Partition reference TSC MSR is available */
+#define HV_X64_MSR_REFERENCE_TSC_AVAILABLE (1 << 9)
/* A partition's reference time stamp counter (TSC) page */
#define HV_X64_MSR_REFERENCE_TSC 0x40000021
diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index b9dd5e8..ab8d4b7 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -130,6 +130,54 @@ static u64 do_hypercall(u64 control, void *input, void *output)
#endif /* !x86_64 */
}
+static cycle_t read_hv_clock_tsc(struct clocksource *arg)
+{
+ cycle_t current_tick;
+ struct ms_hyperv_tsc_page *tsc_pg = hv_context.tsc_page;
+
+ if (tsc_pg->tsc_sequence != -1) {
+ /*
+ * Use the tsc page to compute the value.
+ */
+
+ while (1) {
+ cycle_t tmp;
+ u32 sequence = tsc_pg->tsc_sequence;
+ u64 cur_tsc;
+ u64 scale = tsc_pg->tsc_scale;
+ s64 offset = tsc_pg->tsc_offset;
+
+ rdtscll(cur_tsc);
+ /* current_tick = ((cur_tsc *scale) >> 64) + offset */
+ asm("mulq %3"
+ : "=d" (current_tick), "=a" (tmp)
+ : "a" (cur_tsc), "r" (scale));
+
+ current_tick += offset;
+ if (tsc_pg->tsc_sequence == sequence)
+ return current_tick;
+
+ if (tsc_pg->tsc_sequence != -1)
+ continue;
+ /*
+ * Fallback using MSR method.
+ */
+ break;
+ }
+ }
+ rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
+ return current_tick;
+}
+
+static struct clocksource hyperv_cs_tsc = {
+ .name = "hyperv_clocksource_tsc_page",
+ .rating = 425,
+ .read = read_hv_clock_tsc,
+ .mask = CLOCKSOURCE_MASK(64),
+ .flags = CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+
/*
* hv_init - Main initialization routine.
*
@@ -139,7 +187,9 @@ int hv_init(void)
{
int max_leaf;
union hv_x64_msr_hypercall_contents hypercall_msr;
+ union hv_x64_msr_hypercall_contents tsc_msr;
void *virtaddr = NULL;
+ void *va_tsc = NULL;
memset(hv_context.synic_event_page, 0, sizeof(void *) * NR_CPUS);
memset(hv_context.synic_message_page, 0,
@@ -183,6 +233,22 @@ int hv_init(void)
hv_context.hypercall_page = virtaddr;
+#ifdef CONFIG_X86_64
+ if (ms_hyperv.features & HV_X64_MSR_REFERENCE_TSC_AVAILABLE) {
+ va_tsc = __vmalloc(PAGE_SIZE, GFP_KERNEL, PAGE_KERNEL);
+ if (!va_tsc)
+ goto cleanup;
+ hv_context.tsc_page = va_tsc;
+
+ rdmsrl(HV_X64_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
+
+ tsc_msr.enable = 1;
+ tsc_msr.guest_physical_address = vmalloc_to_pfn(va_tsc);
+
+ wrmsrl(HV_X64_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
+ clocksource_register_hz(&hyperv_cs_tsc, NSEC_PER_SEC/100);
+ }
+#endif
return 0;
cleanup:
@@ -216,6 +282,21 @@ void hv_cleanup(void)
vfree(hv_context.hypercall_page);
hv_context.hypercall_page = NULL;
}
+
+#ifdef CONFIG_X86_64
+ /*
+ * Cleanup the TSC page based CS.
+ */
+ if (ms_hyperv.features & HV_X64_MSR_REFERENCE_TSC_AVAILABLE) {
+ clocksource_change_rating(&hyperv_cs_tsc, 10);
+ clocksource_unregister(&hyperv_cs_tsc);
+
+ hypercall_msr.as_uint64 = 0;
+ wrmsrl(HV_X64_MSR_REFERENCE_TSC, hypercall_msr.as_uint64);
+ vfree(hv_context.tsc_page);
+ hv_context.tsc_page = NULL;
+ }
+#endif
}
/*
diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index 7bdef66..4b1eb6d 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -517,6 +517,7 @@ struct hv_context {
u64 guestid;
void *hypercall_page;
+ void *tsc_page;
bool synic_initialized;
@@ -560,6 +561,14 @@ struct hv_context {
extern struct hv_context hv_context;
+struct ms_hyperv_tsc_page {
+ volatile u32 tsc_sequence;
+ u32 reserved1;
+ volatile u64 tsc_scale;
+ volatile s64 tsc_offset;
+ u64 reserved2[509];
+};
+
struct hv_ring_buffer_debug_info {
u32 current_interrupt_mask;
u32 current_read_index;
--
1.7.4.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists