From: Bjorn Helgaas <email@example.com> To: Oliver O'Halloran <firstname.lastname@example.org> Cc: "Pali Rohár" <email@example.com>, "Aaron Ma" <firstname.lastname@example.org>, email@example.com, firstname.lastname@example.org, "David S. Miller" <email@example.com>, "Jakub Kicinski" <firstname.lastname@example.org>, email@example.com, firstname.lastname@example.org, "Linux Kernel Mailing List" <email@example.com>, "Krzysztof Wilczyński" <firstname.lastname@example.org>, linux-pci <email@example.com> Subject: Re: [PATCH 1/2] igc: don't rd/wr iomem when PCI is removed Date: Thu, 8 Jul 2021 10:45:50 -0500 [thread overview] Message-ID: <20210708154550.GA1019947@bjorn-Precision-5520> (raw) In-Reply-To: <CAOSf1CGVpogQGAatuY_N0db6OL2BFegGtj6VTLA9KFz0TqYBQg@mail.gmail.com> On Thu, Jul 08, 2021 at 12:04:02PM +1000, Oliver O'Halloran wrote: > On Thu, Jul 8, 2021 at 8:40 AM Bjorn Helgaas <firstname.lastname@example.org> wrote: > > > > If we add the check as proposed in this patch, I think people will > > read it and think this is the correct way to avoid MMIO errors. It > > does happen to avoid some MMIO errors, but it cannot avoid them all, > > so it's not a complete solution and it gives a false sense of > > security. > > I think it's helpful to classify MMIO errors as either benign or > poisonous with the poison MMIOs causing some kind of crash. Most of > the discussions about pci_dev_is_disconnected(), including this one, > seem to stem from people trying to use it to avoid the poison case. I > agree that using pci_dev_is_disconnected() that way is hacky and > doesn't really fix the problem, but considering poison MMIOs usually > stem from broken hardware or firmware maybe we should allow it > anyway. We can't do anything better and it's an improvement compared > to crashing. Apologies for rehashing what's probably obvious to everybody but me. I'm trying to get a better handle on benign vs poisonous errors. MMIO means CPU reads or writes to the device. In PCI, writes are posted and don't receive a response, so a driver will never see writel() return an error (although an error may be reported asynchronously via AER or similar). So I think we're mostly talking about CPU reads here. We expect a PCI response containing the data. Sometimes there's no response or an error response. The behavior of the host bridge in these error cases is not defined by PCI, so what the CPU sees is not consistent across platforms. In some cases, the bridge handles this as a catastrophic error that forces a system restart. But in most cases, at least on x86, the bridge logs an error and fabricates ~0 data so the CPU read can complete. Then it's up to software to recognize that an error occurred and decide what to do about it. Is this a benign or a poisonous error? I'd say this is a benign error. It certainly can't be ignored, but as long as the driver recognizes the error, it should be able to deal with it without crashing the whole system and forcing a restart. Bjorn
next prev parent reply other threads:[~2021-07-08 15:45 UTC|newest] Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-07-02 4:51 Aaron Ma 2021-07-02 4:51 ` [PATCH 2/2] igc: wait for the MAC copy when enabled MAC passthrough Aaron Ma 2021-07-04 5:36 ` [Intel-wired-lan] " Neftin, Sasha 2021-07-05 7:38 ` Aaron Ma 2021-07-05 11:54 ` Neftin, Sasha 2021-07-06 6:46 ` Aaron Ma 2021-07-08 4:24 ` Neftin, Sasha 2021-07-13 13:45 ` Aaron Ma 2021-07-14 9:13 ` Ruinskiy, Dima 2021-07-04 14:28 ` [PATCH 1/2] igc: don't rd/wr iomem when PCI is removed Pali Rohár 2021-07-05 7:23 ` Aaron Ma 2021-07-05 23:02 ` Krzysztof Wilczyński 2021-07-06 14:23 ` Pali Rohár 2021-07-05 7:47 ` Dave Airlie 2021-07-06 6:42 ` Aaron Ma 2021-07-06 20:12 ` Bjorn Helgaas 2021-07-07 21:53 ` Pali Rohár 2021-07-07 22:10 ` Bjorn Helgaas 2021-07-08 2:04 ` Oliver O'Halloran 2021-07-08 15:45 ` Bjorn Helgaas [this message] 2021-07-18 16:31 ` Oliver O'Halloran 2021-07-18 22:50 ` Pali Rohár 2021-07-19 2:49 ` Oliver O'Halloran 2021-07-19 8:13 ` Pali Rohár 2021-07-20 0:17 ` Bjorn Helgaas 2021-07-13 13:00 ` [PATCH v2] igc: fix page fault when thunderbolt is unplugged Aaron Ma 2021-08-04 12:06 ` [Intel-wired-lan] " Fuxbrumer, Dvora
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=20210708154550.GA1019947@bjorn-Precision-5520 \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --subject='Re: [PATCH 1/2] igc: don'\''t rd/wr iomem when PCI is removed' \ /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
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).