[PATCH 3/3] x86/mce/AMD: Use saved threshold block info in interrupt handler

From: Yazen Ghannam
Date: Wed May 24 2017 - 16:42:19 EST


From: Yazen Ghannam <yazen.ghannam@xxxxxxx>

In the amd_threshold_interrupt() handler, we loop through every possible
block in each bank and rediscover the block's address and if it's valid,
e.g. valid, counter present and not locked. However, we already have the
address saved in the threshold blocks list for each CPU and bank. The list
only contains blocks that have passed all the valid checks.

Besides the redundancy, there's also an smp_call_function* in
get_block_address() and this causes a warning when servicing the interrupt.

WARNING: CPU: 0 PID: 0 at kernel/smp.c:281 smp_call_function_single+0xdd/0xf0
...
Call Trace:
<IRQ>
rdmsr_safe_on_cpu+0x5d/0x90
get_block_address.isra.2+0x97/0x100
amd_threshold_interrupt+0xae/0x220
smp_threshold_interrupt+0x1b/0x40
threshold_interrupt+0x89/0x90

Iterate over the threshold blocks list and use the saved address. Also,
drop the redundant valid checks.

Signed-off-by: Yazen Ghannam <yazen.ghannam@xxxxxxx>
---
arch/x86/kernel/cpu/mcheck/mce_amd.c | 27 ++++++++++-----------------
1 file changed, 10 insertions(+), 17 deletions(-)

diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c
index 2074b870..14e4dab 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -873,30 +873,23 @@ static void log_error_thresholding(unsigned int bank, u64 misc)
*/
static void amd_threshold_interrupt(void)
{
- u32 low = 0, high = 0, address = 0;
- unsigned int bank, block, cpu = smp_processor_id();
+ u32 low = 0, high = 0;
+ unsigned int bank, cpu = smp_processor_id();
struct thresh_restart tr;
+ struct threshold_block *block = NULL, *tmp = NULL;
+ struct list_head *head = NULL;

for (bank = 0; bank < mca_cfg.banks; ++bank) {
if (!(per_cpu(bank_map, cpu) & (1 << bank)))
continue;
- for (block = 0; block < NR_BLOCKS; ++block) {
- address = get_block_address(cpu, address, low, high, bank, block);
- if (!address)
- break;

- if (rdmsr_safe(address, &low, &high))
- break;
+ if (!per_cpu(threshold_banks, cpu)[bank]->blocks)
+ continue;

- if (!(high & MASK_VALID_HI)) {
- if (block)
- continue;
- else
- break;
- }
+ head = &per_cpu(threshold_banks, cpu)[bank]->blocks_head;

- if (!(high & MASK_CNTP_HI) ||
- (high & MASK_LOCKED_HI))
+ list_for_each_entry_safe(block, tmp, head, miscj) {
+ if (rdmsr_safe(block->address, &low, &high))
continue;

if (!(high & MASK_OVERFLOW_HI))
@@ -907,7 +900,7 @@ static void amd_threshold_interrupt(void)

/* Reset threshold block after logging error. */
memset(&tr, 0, sizeof(tr));
- tr.b = &per_cpu(threshold_banks, cpu)[bank]->blocks[block];
+ tr.b = block;
threshold_restart_bank(&tr);
}
}
--
2.7.4