Skip to contentSkip to conversation

linux-next acpi changes make vmware server unusable

DISCUSSION Eric Paris 3 participants 4 messages Mar 4, 2009 – Mar 16, 2009

Subsystems (2)
lore
Share & tools
lore

Eric Paris

author
#0 lore raw

I built a linux-next kernel yesterday and discovered that the free vmware server 2 would start to boot the kernel and very quickly just power off. vmware gave some crap message about the MBR being wrong, which obviously wasn't the case since grub started and the kernel started booting. Looking at the last line on the serial console for the new kernel and the next line in a working kernel I knew the next output was supposed to be in the ACPI code. I bisected drivers/acpi and found that d6c349993fc7c9dabf873796c4e82bb94544b3ce is first bad commit.

I have no idea what vmware is doing, or what we are doing, but before that patch I was able to boot and after it, vmware just shuts itself off.

-Eric

#1 lore raw
Quoted 8 lines · Eric Paris

I built a linux-next kernel yesterday and discovered that the free vmware server 2 would start to boot the kernel and very quickly just power off. vmware gave some crap message about the MBR being wrong, which obviously wasn't the case since grub started and the kernel started booting. Looking at the last line on the serial console for the new kernel and the next line in a working kernel I knew the next output was supposed to be in the ACPI code. I bisected drivers/acpi and found that d6c349993fc7c9dabf873796c4e82bb94544b3ce is first bad commit.

Would you please try below patch?

ACPICA: Check for non-zero address before being converted to GAS

---
 drivers/acpi/acpica/tbfadt.c |   16 +++++++++-------
 1 files changed, 9 insertions(+), 7 deletions(-)
drivers/acpi/acpica/tbfadt.c +9 -7
index 042d239..af8fbe1 100644
@@ -625,12 +625,14 @@ static void acpi_tb_setup_fadt_registers(void)
 		    ACPI_ADD_PTR(struct acpi_generic_address, &acpi_gbl_FADT,
 				 fadt_pm_info_table[i].source);
 
-		acpi_tb_init_generic_address(fadt_pm_info_table[i].target,
-					     source64->space_id,
-					     pm1_register_byte_width,
-					     source64->address +
-					     (fadt_pm_info_table[i].
-					      register_num *
-					      pm1_register_byte_width));
+		if (source64->address) {
+			acpi_tb_init_generic_address(fadt_pm_info_table[i].
+						     target, source64->space_id,
+						     pm1_register_byte_width,
+						     source64->address +
+						     (fadt_pm_info_table[i].
+						      register_num *
+						      pm1_register_byte_width));
+		}
 	}
 }
Quoted 8 lines

I have no idea what vmware is doing, or what we are doing, but before that patch I was able to boot and after it, vmware just shuts itself off.

-Eric

author
#2 lore raw
Quoted 11 lines · Lin Ming
Quoted 8 lines · Eric Paris

I built a linux-next kernel yesterday and discovered that the free vmware server 2 would start to boot the kernel and very quickly just power off. vmware gave some crap message about the MBR being wrong, which obviously wasn't the case since grub started and the kernel started booting. Looking at the last line on the serial console for the new kernel and the next line in a working kernel I knew the next output was supposed to be in the ACPI code. I bisected drivers/acpi and found that d6c349993fc7c9dabf873796c4e82bb94544b3ce is first bad commit.

Would you please try below patch?

linux-next + this patch boots just fine.

Thanks!!

-Eric

Quoted 44 lines

ACPICA: Check for non-zero address before being converted to GAS

---
 drivers/acpi/acpica/tbfadt.c |   16 +++++++++-------
 1 files changed, 9 insertions(+), 7 deletions(-)
drivers/acpi/acpica/tbfadt.c +9 -7
index 042d239..af8fbe1 100644
@@ -625,12 +625,14 @@ static void acpi_tb_setup_fadt_registers(void)
 		    ACPI_ADD_PTR(struct acpi_generic_address, &acpi_gbl_FADT,
 				 fadt_pm_info_table[i].source);
 
-		acpi_tb_init_generic_address(fadt_pm_info_table[i].target,
-					     source64->space_id,
-					     pm1_register_byte_width,
-					     source64->address +
-					     (fadt_pm_info_table[i].
-					      register_num *
-					      pm1_register_byte_width));
+		if (source64->address) {
+			acpi_tb_init_generic_address(fadt_pm_info_table[i].
+						     target, source64->space_id,
+						     pm1_register_byte_width,
+						     source64->address +
+						     (fadt_pm_info_table[i].
+						      register_num *
+						      pm1_register_byte_width));
+		}
 	}
 }
Quoted 10 lines

I have no idea what vmware is doing, or what we are doing, but before that patch I was able to boot and after it, vmware just shuts itself off.

-Eric

reviewer
#3 lore raw
In reply to Lin Ming

this one is inserted into the acpica branch to fix the -next regression.

thanks,
Len Brown, Intel Open Source Technology Center

Quoted 62 lines · Lin Ming
Quoted 8 lines · Eric Paris

I built a linux-next kernel yesterday and discovered that the free vmware server 2 would start to boot the kernel and very quickly just power off. vmware gave some crap message about the MBR being wrong, which obviously wasn't the case since grub started and the kernel started booting. Looking at the last line on the serial console for the new kernel and the next line in a working kernel I knew the next output was supposed to be in the ACPI code. I bisected drivers/acpi and found that d6c349993fc7c9dabf873796c4e82bb94544b3ce is first bad commit.

Would you please try below patch?

ACPICA: Check for non-zero address before being converted to GAS

---
 drivers/acpi/acpica/tbfadt.c |   16 +++++++++-------
 1 files changed, 9 insertions(+), 7 deletions(-)
drivers/acpi/acpica/tbfadt.c +9 -7
index 042d239..af8fbe1 100644
@@ -625,12 +625,14 @@ static void acpi_tb_setup_fadt_registers(void)
 		    ACPI_ADD_PTR(struct acpi_generic_address, &acpi_gbl_FADT,
 				 fadt_pm_info_table[i].source);
 
-		acpi_tb_init_generic_address(fadt_pm_info_table[i].target,
-					     source64->space_id,
-					     pm1_register_byte_width,
-					     source64->address +
-					     (fadt_pm_info_table[i].
-					      register_num *
-					      pm1_register_byte_width));
+		if (source64->address) {
+			acpi_tb_init_generic_address(fadt_pm_info_table[i].
+						     target, source64->space_id,
+						     pm1_register_byte_width,
+						     source64->address +
+						     (fadt_pm_info_table[i].
+						      register_num *
+						      pm1_register_byte_width));
+		}
 	}
 }
Quoted 10 lines

I have no idea what vmware is doing, or what we are doing, but before that patch I was able to boot and after it, vmware just shuts itself off.

-Eric

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html