public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Renninger <trenn@suse.de>
To: megaraidlinux.pdl@broadcom.com
Cc: shivasharan.srikanteshwara@broadcom.com,
	linux-scsi@vger.kernel.org, nhorman@tuxdriver.com
Subject: Blacklist megaraid_sas driver driven irqs in irqbalance - Longterm kernel solution thought about?
Date: Fri, 14 Mar 2025 15:36:05 +0100	[thread overview]
Message-ID: <2724461.lGaqSPkdTl@laptop.fritzbox> (raw)

[-- Attachment #1: Type: text/plain, Size: 1431 bytes --]

Hi,

I have a request to blacklist megaraid_sas driver in irqbalance.

According to some developers it's known that megaraid (sas?) does have issues 
with migrating IRQs at runtime which could lead to Null pointer dereferences 
when handling IRQs (an example backtrace pasted at the end).

I understand the problem like:
- These cards can initialize on any (or at least on affinity hint pointing) irqs 
without problems. The issue/freeze rises when those IRQs are migrated while 
the card is operating.
- If smp_affinity_enable=0 is passed the PCI/platform irq affinity hints will/
might be ignore which could lead to a dramatically performance regression
- A longterm mainline solution might be to split smp_affinity (hint) 
initialization feature and IRQ migration feature for PCI drivers and disable 
the latter by default for affected cards.

Does that make sense?

A confirmation from developers working on these drivers/HW would be very much 
appreciated. I'd like to submit the blacklisting to mainline irqbalance 
project as well if it makes sense if it cannot be handled in the kernel yet.

Thanks,

             Thomas


Call Trace:
 <IRQ>
 megasas_isr_fusion+0x84/0x90 [megaraid_sas 
d7ec7902299b887bdbcf4651784a30e2e26d5c22]
 __handle_irq_event_percpu+0x36/0x1a0
 handle_irq_event_percpu+0x30/0x70
 handle_irq_event+0x34/0x60
 handle_edge_irq+0x7e/0x1a0
 __common_interrupt+0x3b/0xb0
 common_interrupt+0x58/0xa0
 </IRQ>

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

                 reply	other threads:[~2025-03-14 14:36 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=2724461.lGaqSPkdTl@laptop.fritzbox \
    --to=trenn@suse.de \
    --cc=linux-scsi@vger.kernel.org \
    --cc=megaraidlinux.pdl@broadcom.com \
    --cc=nhorman@tuxdriver.com \
    --cc=shivasharan.srikanteshwara@broadcom.com \
    /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