Linux CXL
 help / color / mirror / Atom feed
From: Dave Jiang <dave.jiang@intel.com>
To: Robin Murphy <robin.murphy@arm.com>,
	linux-cxl@vger.kernel.org, linux-perf-users@vger.kernel.org
Cc: jic23@kernel.org, will@kernel.org, mark.rutland@arm.com,
	dave@stgolabs.net
Subject: Re: [PATCH v3 4/9] perf/cxl: Split the MSI vector out of info->irq
Date: Mon, 3 Aug 2026 09:01:04 -0700	[thread overview]
Message-ID: <ab9b7f3a-e2a5-4627-b2ac-d53ce3a82477@intel.com> (raw)
In-Reply-To: <a4a807e6-b477-40c3-ac42-2bef8dd82d7e@arm.com>



On 8/3/26 5:04 AM, Robin Murphy wrote:
> On 01/08/2026 12:28 am, Dave Jiang wrote:
>> info->irq means two different things depending on how far probe has got.
>> cxl_pmu_parse_caps() sets it to the MSI/MSI-X message number from the CPMU
>> Capability register. cxl_pmu_probe() passes that to pci_irq_vector() and
>> then overwrites it with the Linux virq, which is what the hotplug callbacks
>> hand to irq_set_affinity().
>>
>> Give the message number its own field. No functional change.
> 
> Thanks, this makes it much less confusing;
> 
> Reviewed-by: Robin Murphy <robin.murphy@arm.com>
> 
> If the fix itself was just the comparison operator then I'd suggest folding it in so as to avoid extra backporting dependencies/conflicts, but it sounds like we might have a corner case where 0 may still mean "no IRQ", so maybe sort that first then put this refactoring on top?

Will do. I'll flip the order.

DJ

> 
> Cheers,
> Robin.
> 
>> Assisted-by: Claude:claude-opus-4-8
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
>> ---
>> v3:
>> - New patch
>> ---
>>   drivers/perf/cxl_pmu.c | 9 +++++----
>>   1 file changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/perf/cxl_pmu.c b/drivers/perf/cxl_pmu.c
>> index 3511d049aea4..56e6ae7bebf7 100644
>> --- a/drivers/perf/cxl_pmu.c
>> +++ b/drivers/perf/cxl_pmu.c
>> @@ -106,6 +106,7 @@ struct cxl_pmu_info {
>>       int on_cpu;
>>       struct hlist_node node;
>>       bool filter_hdm;
>> +    int msi_vec;
>>       int irq;
>>   };
>>   @@ -143,9 +144,9 @@ static int cxl_pmu_parse_caps(struct device *dev, struct cxl_pmu_info *info)
>>         info->filter_hdm = FIELD_GET(CXL_PMU_CAP_FILTERS_SUP_MSK, val) & CXL_PMU_FILTER_HDM;
>>       if (FIELD_GET(CXL_PMU_CAP_INT, val))
>> -        info->irq = FIELD_GET(CXL_PMU_CAP_MSI_N_MSK, val);
>> +        info->msi_vec = FIELD_GET(CXL_PMU_CAP_MSI_N_MSK, val);
>>       else
>> -        info->irq = -1;
>> +        info->msi_vec = -1;
>>         /* First handle fixed function counters; note if configurable counters found */
>>       for (i = 0; i < info->num_counters; i++) {
>> @@ -874,10 +875,10 @@ static int cxl_pmu_probe(struct device *dev)
>>           .capabilities = PERF_PMU_CAP_NO_EXCLUDE,
>>       };
>>   -    if (info->irq <= 0)
>> +    if (info->msi_vec <= 0)
>>           return -EINVAL;
>>   -    rc = pci_irq_vector(pdev, info->irq);
>> +    rc = pci_irq_vector(pdev, info->msi_vec);
>>       if (rc < 0)
>>           return rc;
>>       irq = rc;
> 


  reply	other threads:[~2026-08-03 16:01 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-31 23:28 [PATCH v3 0/9] perf/cxlpmu: Misc sashiko raised issues fixes Dave Jiang
2026-07-31 23:28 ` [PATCH v3 1/9] perf/cxl: Program the requested event group on configurable counters Dave Jiang
2026-07-31 23:38   ` sashiko-bot
2026-07-31 23:28 ` [PATCH v3 2/9] perf/cxl: Clear stale event fields before reprogramming a counter Dave Jiang
2026-07-31 23:40   ` sashiko-bot
2026-07-31 23:28 ` [PATCH v3 3/9] perf/cxl: Fix the counter overflow delta fixup Dave Jiang
2026-07-31 23:37   ` sashiko-bot
2026-07-31 23:28 ` [PATCH v3 4/9] perf/cxl: Split the MSI vector out of info->irq Dave Jiang
2026-07-31 23:45   ` sashiko-bot
2026-08-03 12:04   ` Robin Murphy
2026-08-03 16:01     ` Dave Jiang [this message]
2026-07-31 23:28 ` [PATCH v3 5/9] perf/cxl: Accept an overflow interrupt on MSI message number 0 Dave Jiang
2026-08-03  5:57   ` Richard Cheng
2026-08-03 14:30     ` Dave Jiang
2026-08-03 14:40       ` Richard Cheng
2026-07-31 23:28 ` [PATCH v3 6/9] perf/cxl: Don't share the overflow interrupt, and keep it pinned Dave Jiang
2026-07-31 23:50   ` sashiko-bot
2026-07-31 23:28 ` [PATCH v3 7/9] perf/cxl: Unfreeze counters after handling an overflow interrupt Dave Jiang
2026-07-31 23:40   ` sashiko-bot
2026-08-03  4:00   ` Richard Cheng
2026-08-03 11:26     ` Robin Murphy
2026-07-31 23:28 ` [PATCH v3 8/9] perf/cxl: Validate the hardware-reported counter width Dave Jiang
2026-07-31 23:28 ` [PATCH v3 9/9] perf/cxl: Don't log through pmu.dev in the overflow interrupt handler Dave Jiang
2026-07-31 23:46   ` sashiko-bot
2026-08-03  5:29   ` Richard Cheng
2026-08-03 16:39     ` Dave Jiang

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=ab9b7f3a-e2a5-4627-b2ac-d53ce3a82477@intel.com \
    --to=dave.jiang@intel.com \
    --cc=dave@stgolabs.net \
    --cc=jic23@kernel.org \
    --cc=linux-cxl@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=robin.murphy@arm.com \
    --cc=will@kernel.org \
    /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