- Fix a race condition when clearing error count bits and toggling

the error interrupt throug the same register, in synopsys_edac
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCgAdFiEEzv7L6UO9uDPlPSfHEsHwGGHeVUoFAmZAdYEACgkQEsHwGGHe
 VUrtORAAvwK9Ump8tGigXQ58aXY7a2b+iYrYiZtZkP23y0p81HJdYruanHeeLM4n
 CYKnuI+yKFP/e5jfxQj6vPj5sKk6U285+C2L/7MsHcfKf/bFzgjopzHLKYS+u6E6
 YtjJQOFbozMvsDnF0BcxQf3OAiLsEkvA4JKb8gE7YqslMlJF5nhTVRRYtX3H+RBH
 FBupB/wZqvFS41igzTexkY89L71TlhRbP2hjUKScV4N5v9Jhh0m2PmApARW2EGkW
 X4RhqIs8kHVBseuNpanV/vBLJDFekJtZD95WLFltK10pGC306gMzEGVIj3H4Dnw3
 qrgDK2hHxz9/i3ukHox5YoMKWVXBYTTo74a1kMvAoQxmCBERPQopUKu34nq9/NpQ
 ecWN6pFFwhwQAjdZkAoZgJknJtIaO8Ti4Uj9roN1PFgBDCConGXhJbFrU+z3WPfR
 aUk3VK4zBDyUSBkj9TJviWPm+8Se5HdcqgrFFqiLH5BYZIUWPYit8Q56LYzx5sKp
 +OgaguxjVwdyaPovFs6h/ae7/mJzzXn7FPRfOmeOH3KVFbo1X+Cu+o/lolXHfbvN
 KFfov33DMDuHLs6ECYrEM8OwIXXFcYBe/Pd7TePkaoEr9+7arujaBEJvMAyTQaWN
 x1OsOBo4Nxg7YYvfMVO7pKIv5Mn1LomAlOs8VZynRWAFeSqcaFk=
 =mBtC
 -----END PGP SIGNATURE-----

Merge tag 'edac_urgent_for_v6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras

Pull EDAC fix from Borislav Petkov:

 - Fix a race condition when clearing error count bits and toggling the
   error interrupt throug the same register, in synopsys_edac

* tag 'edac_urgent_for_v6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
  EDAC/synopsys: Fix ECC status and IRQ control race condition
This commit is contained in:
Linus Torvalds 2024-05-12 09:09:27 -07:00
commit ba16c1cf11
1 changed files with 37 additions and 13 deletions

View File

@ -9,6 +9,7 @@
#include <linux/edac.h> #include <linux/edac.h>
#include <linux/module.h> #include <linux/module.h>
#include <linux/platform_device.h> #include <linux/platform_device.h>
#include <linux/spinlock.h>
#include <linux/interrupt.h> #include <linux/interrupt.h>
#include <linux/of.h> #include <linux/of.h>
@ -299,6 +300,7 @@ struct synps_ecc_status {
/** /**
* struct synps_edac_priv - DDR memory controller private instance data. * struct synps_edac_priv - DDR memory controller private instance data.
* @baseaddr: Base address of the DDR controller. * @baseaddr: Base address of the DDR controller.
* @reglock: Concurrent CSRs access lock.
* @message: Buffer for framing the event specific info. * @message: Buffer for framing the event specific info.
* @stat: ECC status information. * @stat: ECC status information.
* @p_data: Platform data. * @p_data: Platform data.
@ -313,6 +315,7 @@ struct synps_ecc_status {
*/ */
struct synps_edac_priv { struct synps_edac_priv {
void __iomem *baseaddr; void __iomem *baseaddr;
spinlock_t reglock;
char message[SYNPS_EDAC_MSG_SIZE]; char message[SYNPS_EDAC_MSG_SIZE];
struct synps_ecc_status stat; struct synps_ecc_status stat;
const struct synps_platform_data *p_data; const struct synps_platform_data *p_data;
@ -408,7 +411,8 @@ out:
static int zynqmp_get_error_info(struct synps_edac_priv *priv) static int zynqmp_get_error_info(struct synps_edac_priv *priv)
{ {
struct synps_ecc_status *p; struct synps_ecc_status *p;
u32 regval, clearval = 0; u32 regval, clearval;
unsigned long flags;
void __iomem *base; void __iomem *base;
base = priv->baseaddr; base = priv->baseaddr;
@ -452,10 +456,14 @@ ue_err:
p->ueinfo.blknr = (regval & ECC_CEADDR1_BLKNR_MASK); p->ueinfo.blknr = (regval & ECC_CEADDR1_BLKNR_MASK);
p->ueinfo.data = readl(base + ECC_UESYND0_OFST); p->ueinfo.data = readl(base + ECC_UESYND0_OFST);
out: out:
clearval = ECC_CTRL_CLR_CE_ERR | ECC_CTRL_CLR_CE_ERRCNT; spin_lock_irqsave(&priv->reglock, flags);
clearval |= ECC_CTRL_CLR_UE_ERR | ECC_CTRL_CLR_UE_ERRCNT;
clearval = readl(base + ECC_CLR_OFST) |
ECC_CTRL_CLR_CE_ERR | ECC_CTRL_CLR_CE_ERRCNT |
ECC_CTRL_CLR_UE_ERR | ECC_CTRL_CLR_UE_ERRCNT;
writel(clearval, base + ECC_CLR_OFST); writel(clearval, base + ECC_CLR_OFST);
writel(0x0, base + ECC_CLR_OFST);
spin_unlock_irqrestore(&priv->reglock, flags);
return 0; return 0;
} }
@ -515,24 +523,41 @@ static void handle_error(struct mem_ctl_info *mci, struct synps_ecc_status *p)
static void enable_intr(struct synps_edac_priv *priv) static void enable_intr(struct synps_edac_priv *priv)
{ {
unsigned long flags;
/* Enable UE/CE Interrupts */ /* Enable UE/CE Interrupts */
if (priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR) if (!(priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR)) {
writel(DDR_UE_MASK | DDR_CE_MASK,
priv->baseaddr + ECC_CLR_OFST);
else
writel(DDR_QOSUE_MASK | DDR_QOSCE_MASK, writel(DDR_QOSUE_MASK | DDR_QOSCE_MASK,
priv->baseaddr + DDR_QOS_IRQ_EN_OFST); priv->baseaddr + DDR_QOS_IRQ_EN_OFST);
return;
}
spin_lock_irqsave(&priv->reglock, flags);
writel(DDR_UE_MASK | DDR_CE_MASK,
priv->baseaddr + ECC_CLR_OFST);
spin_unlock_irqrestore(&priv->reglock, flags);
} }
static void disable_intr(struct synps_edac_priv *priv) static void disable_intr(struct synps_edac_priv *priv)
{ {
unsigned long flags;
/* Disable UE/CE Interrupts */ /* Disable UE/CE Interrupts */
if (priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR) if (!(priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR)) {
writel(0x0, priv->baseaddr + ECC_CLR_OFST);
else
writel(DDR_QOSUE_MASK | DDR_QOSCE_MASK, writel(DDR_QOSUE_MASK | DDR_QOSCE_MASK,
priv->baseaddr + DDR_QOS_IRQ_DB_OFST); priv->baseaddr + DDR_QOS_IRQ_DB_OFST);
return;
}
spin_lock_irqsave(&priv->reglock, flags);
writel(0, priv->baseaddr + ECC_CLR_OFST);
spin_unlock_irqrestore(&priv->reglock, flags);
} }
/** /**
@ -576,8 +601,6 @@ static irqreturn_t intr_handler(int irq, void *dev_id)
/* v3.0 of the controller does not have this register */ /* v3.0 of the controller does not have this register */
if (!(priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR)) if (!(priv->p_data->quirks & DDR_ECC_INTR_SELF_CLEAR))
writel(regval, priv->baseaddr + DDR_QOS_IRQ_STAT_OFST); writel(regval, priv->baseaddr + DDR_QOS_IRQ_STAT_OFST);
else
enable_intr(priv);
return IRQ_HANDLED; return IRQ_HANDLED;
} }
@ -1357,6 +1380,7 @@ static int mc_probe(struct platform_device *pdev)
priv = mci->pvt_info; priv = mci->pvt_info;
priv->baseaddr = baseaddr; priv->baseaddr = baseaddr;
priv->p_data = p_data; priv->p_data = p_data;
spin_lock_init(&priv->reglock);
mc_init(mci, pdev); mc_init(mci, pdev);