Skip to contentSkip to conversation

x86/Intel: insert Ice Lake and Comet Lake model numbers

PATCH In review Jan Beulich 4 participants 10 messages Jun 5, 2020 – Jun 9, 2020

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86/Intel: insert Ice Lake and Comet Lake model numbers

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

xen/arch/x86/acpi/cpu_idle.c +6 -0
@@ -180,9 +180,15 @@ static void do_get_hw_residencies(void *
     case 0x4E:
     case 0x55:
     case 0x5E:
+    /* Ice Lake */
+    case 0x7D:
+    case 0x7E:
     /* Kaby Lake */
     case 0x8E:
     case 0x9E:
+    /* Comet Lake */
+    case 0xA5:
+    case 0xA6:
         GET_PC2_RES(hw_res->pc2);
         GET_CC7_RES(hw_res->cc7);
         /* fall through */
xen/arch/x86/hvm/vmx/vmx.c +8 -0
@@ -2438,8 +2438,12 @@ static bool __init has_if_pschange_mc(vo
     case 0x4e: /* Skylake M */
     case 0x5e: /* Skylake D */
     case 0x55: /* Skylake-X / Cascade Lake */
+    case 0x7d: /* Ice Lake */
+    case 0x7e: /* Ice Lake */
     case 0x8e: /* Kaby / Coffee / Whiskey Lake M */
     case 0x9e: /* Kaby / Coffee / Whiskey Lake D */
+    case 0xa5: /* Comet Lake H/S */
+    case 0xa6: /* Comet Lake U */
         return true;
 
         /*
@@ -2781,10 +2785,14 @@ static const struct lbr_info *last_branc
         case 0x66:
         /* Goldmont Plus */
         case 0x7a:
+        /* Ice Lake */
+        case 0x7d: case 0x7e:
         /* Tremont */
         case 0x86:
         /* Kaby Lake */
         case 0x8e: case 0x9e:
+        /* Comet Lake */
+        case 0xa5: case 0xa6:
             return sk_lbr;
         /* Atom */
         case 0x1c: case 0x26: case 0x27: case 0x35: case 0x36:
#1 lore raw

xen/arch/x86/acpi/cpu_idle.c

Quoted 16 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Thanks, Roger.

#2 lore raw
Quoted 15 lines · Roger Pau Monné
Quoted 14 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

Quoted 5 lines

xen/arch/x86/acpi/cpu_idle.c

Quoted 2 lines
xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Correct. I've looked at recent history there, and I'm uncertain they'll add further models there. They look to prefer to use ACPI _CST now again with, as it seems, not overly much of a difference to the ACPI driver (which, if we were to follow, I'd rather see us integrate there).

Jan

#3 lore raw
Quoted 22 lines · Jan Beulich
Quoted 15 lines · Roger Pau Monné
Quoted 14 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

Quoted 5 lines

xen/arch/x86/acpi/cpu_idle.c

Quoted 2 lines
xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Quoted 4 lines

Correct. I've looked at recent history there, and I'm uncertain they'll add further models there. They look to prefer to use ACPI _CST now again with, as it seems, not overly much of a difference to the ACPI driver (which, if we were to follow, I'd rather see us integrate there).

Urg, OK, that's a shame as using mwait-idle was IMO better from a Xen PoV as we didn't rely on dom0 in order to discover C states. I wonder if we could continue to update mwait-idle on our own for newer models.

FWIW, wikichip also lists 6c and 6a [0] as Ice Lake Server model versions, but I'm not sure if this has been confirmed in any way?

Roger.

[0] https://en.wikichip.org/wiki/intel/cpuid#Big_Cores_.28Server.29

#4 lore raw
Quoted 23 lines · Roger Pau Monné
Quoted 22 lines · Jan Beulich
Quoted 20 lines · Roger Pau Monné

xen/arch/x86/acpi/cpu_idle.c

Quoted 16 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Quoted 8 lines
Quoted 4 lines

Correct. I've looked at recent history there, and I'm uncertain they'll add further models there. They look to prefer to use ACPI _CST now again with, as it seems, not overly much of a difference to the ACPI driver (which, if we were to follow, I'd rather see us integrate there).

Urg, OK, that's a shame as using mwait-idle was IMO better from a Xen PoV as we didn't rely on dom0 in order to discover C states. I wonder if we could continue to update mwait-idle on our own for newer models.

This would be nice indeed, but would require Intel to provide us with the necessary data.

Quoted 2 lines

FWIW, wikichip also lists 6c and 6a [0] as Ice Lake Server model versions, but I'm not sure if this has been confirmed in any way?

SDM vol 4 confirms this, but mentions the two model numbers exclusively in the table matching signatures to model names ("Future Intel Xeon processors based on Ice Lake microarchitecture"). Without there being an actual table for these I don't think we should "speculatively" add the numbers anywhere.

Jan

Reviewed-by
#5 lore raw
Quoted 36 lines · Jan Beulich
Quoted 23 lines · Roger Pau Monné
Quoted 22 lines · Jan Beulich
Quoted 20 lines · Roger Pau Monné

xen/arch/x86/acpi/cpu_idle.c

Quoted 16 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Quoted 8 lines
Quoted 4 lines

Correct. I've looked at recent history there, and I'm uncertain they'll add further models there. They look to prefer to use ACPI _CST now again with, as it seems, not overly much of a difference to the ACPI driver (which, if we were to follow, I'd rather see us integrate there).

Urg, OK, that's a shame as using mwait-idle was IMO better from a Xen PoV as we didn't rely on dom0 in order to discover C states. I wonder if we could continue to update mwait-idle on our own for newer models.

This would be nice indeed, but would require Intel to provide us with the necessary data.

Quoted 8 lines
Quoted 2 lines

FWIW, wikichip also lists 6c and 6a [0] as Ice Lake Server model versions, but I'm not sure if this has been confirmed in any way?

SDM vol 4 confirms this, but mentions the two model numbers exclusively in the table matching signatures to model names ("Future Intel Xeon processors based on Ice Lake microarchitecture"). Without there being an actual table for these I don't think we should "speculatively" add the numbers anywhere.

Ack.

Thanks.

#6 lore raw
In reply to Jan Beulich
Quoted 32 lines · Jan Beulich
Quoted 21 lines · Roger Pau Monné
Quoted 20 lines · Jan Beulich
Quoted 19 lines · Roger Pau Monné

xen/arch/x86/acpi/cpu_idle.c

Quoted 16 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

xen/arch/x86/acpi/cpu_idle.c +0 -0

What about mwait-idle? I guess we pick that from Linux and no patch has been added so far?

Quoted 7 lines
Quoted 4 lines

Correct. I've looked at recent history there, and I'm uncertain they'll add further models there. They look to prefer to use ACPI _CST now again with, as it seems, not overly much of a difference to the ACPI driver (which, if we were to follow, I'd rather see us integrate there).

Urg, OK, that's a shame as using mwait-idle was IMO better from a Xen PoV as we didn't rely on dom0 in order to discover C states. I wonder if we could continue to update mwait-idle on our own for newer models.

This would be nice indeed, but would require Intel to provide us with the necessary data.

Quoted 7 lines
Quoted 2 lines

FWIW, wikichip also lists 6c and 6a [0] as Ice Lake Server model versions, but I'm not sure if this has been confirmed in any way?

SDM vol 4 confirms this, but mentions the two model numbers exclusively in the table matching signatures to model names ("Future Intel Xeon processors based on Ice Lake microarchitecture"). Without there being an actual table for these I don't think we should "speculatively" add the numbers anywhere.

0x6a is server, 0x6c is microserver.

From this patch, 0x7d is regular client and 0x7e mobile, but there is also 0x9d which is separate model for inference (I believe its server with extra AVX512).

Its high time we borrowed intel-family.h from Linux and used that.

~Andrew

Acked-by on v1 1/1
#7 lore raw
In reply to Jan Beulich
Quoted 5 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

Quoted 8 lines

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

I've enquired about this.  At a guess, there was another hole found, so MDS_NO has been cleared and VERW flushing reinstated.

Either way, changes there can wait until we've got confirmation.

#8 lore raw
Quoted 15 lines · Andrew Cooper
Quoted 5 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

Quoted 8 lines

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

Quoted 6 lines

I've enquired about this.  At a guess, there was another hole found, so MDS_NO has been cleared and VERW flushing reinstated.

Either way, changes there can wait until we've got confirmation.

Paul - any thoughts about this one either way for 4.14?

Jan

#9 lore raw
Quoted 24 lines

-----Original Message-----
From: Jan Beulich <[email protected]>
Sent: 09 June 2020 11:33
To: Paul Durrant <[email protected]>
Cc: Andrew Cooper <[email protected]>; [email protected]; Roger Pau Monné <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH] x86/Intel: insert Ice Lake and Comet Lake model numbers

Quoted 15 lines · Andrew Cooper
Quoted 13 lines · Jan Beulich

Both match prior generation processors as far as LBR and C-state MSRs go (SDM rev 072) as well as applicability of the if_pschange_mc erratum (recent spec updates).

---
Such changes having been subject to backporting in the past, this change may want considering for 4.14.
---
I'm leaving alone spec_ctrl.c, albeit there's a scary looking erratum for Ice Lake indicating that MDS_NO may wrongly be set. But this is apparently addressed by ucode update, so we may not need to deal with it in software.

Quoted 9 lines
Quoted 6 lines

I've enquired about this. At a guess, there was another hole found, so MDS_NO has been cleared and VERW flushing reinstated.

Either way, changes there can wait until we've got confirmation.

Paul - any thoughts about this one either way for 4.14?

I don't see any harm in it going in at this stage.