linux-kernel - Re: [PATCH v5 16/17] qcom: llcc/edac: Support polling mode for ECC handling

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Y8Up1kjaIRLlxemH@zn.tnic>
Date:   Mon, 16 Jan 2023 11:41:26 +0100
From:   Borislav Petkov <bp@...en8.de>
To:     Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
Cc:     andersson@...nel.org, robh+dt@...nel.org,
        krzysztof.kozlowski+dt@...aro.org, tony.luck@...el.com,
        quic_saipraka@...cinc.com, konrad.dybcio@...aro.org,
        linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org,
        james.morse@....com, mchehab@...nel.org, rric@...nel.org,
        linux-edac@...r.kernel.org, quic_ppareek@...cinc.com,
        luca.weiss@...rphone.com, ahalaney@...hat.com, steev@...i.org
Subject: Re: [PATCH v5 16/17] qcom: llcc/edac: Support polling mode for ECC
 handling

On Sun, Jan 15, 2023 at 09:38:25AM +0530, Manivannan Sadhasivam wrote:
> > You need to request the IRQ first and then set edac_op_state above. I.e., this
> > devm_request_irq() needs to move in the if (ecc_irq > 0) branch above.
> 
> May I know why? I also checked other drivers, most of them are doing the same.

If the others do it, that doesn't mean it is clean.

What happens to edac_op_state if devm_request_irq() fails?

I know I know, the probe function will fail and the driver won't load but still,
this is sloppy. And it could come down to bite us later, when someone
reorganizes that function.

So, do all the error checking method determination - polling or interrupt - in
one place.  Something like this (totally untested ofc, pasting here the whole
thing to show what I mean):

static int qcom_llcc_edac_probe(struct platform_device *pdev)
{
        struct llcc_drv_data *llcc_driv_data = pdev->dev.platform_data;
        struct edac_device_ctl_info *edev_ctl;
        struct device *dev = &pdev->dev;
        int ecc_irq;
        int rc;

        rc = qcom_llcc_core_setup(llcc_driv_data->bcast_regmap);
        if (rc)
                return rc;

        /* Allocate edac control info */
        edev_ctl = edac_device_alloc_ctl_info(0, "qcom-llcc", 1, "bank",
                                              llcc_driv_data->num_banks, 1,
                                              NULL, 0,
                                              edac_device_alloc_index());

        if (!edev_ctl)
                return -ENOMEM;

        edev_ctl->dev = dev;
        edev_ctl->mod_name = dev_name(dev);
        edev_ctl->dev_name = dev_name(dev);
        edev_ctl->ctl_name = "llcc";
        edev_ctl->panic_on_ue = LLCC_ERP_PANIC_ON_UE;

        /* Check if LLCC driver has passed ECC IRQ */
        ecc_irq = llcc_driv_data->ecc_irq;
        if (ecc_irq > 0) {
                rc = devm_request_irq(dev, ecc_irq, llcc_ecc_irq_handler,
                                      IRQF_TRIGGER_HIGH, "llcc_ecc", edev_ctl);
                if (!rc) {
                        edac_op_state = EDAC_OPSTATE_INT;
                        goto irq_done;
                }
        }

        /* Fall back to polling mode otherwise */
        edev_ctl->poll_msec = ECC_POLL_MSEC;
        edev_ctl->edac_check = llcc_ecc_check;
        edac_op_state = EDAC_OPSTATE_POLL;

irq_done:
        rc = edac_device_add_device(edev_ctl);
        if (rc) {
                edac_device_free_ctl_info(edev_ctl);
                return rc;
        }

        platform_set_drvdata(pdev, edev_ctl);

        return rc;
}

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette