The patch titled intel-iommu: make intel iommu_wait_op() work when jiffies aren't running has been added to the -mm tree. Its filename is intel-iommu-make-intel-iommu_wait_op-work-when-jiffies-arent-running.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: intel-iommu: make intel iommu_wait_op() work when jiffies aren't running From: mark gross <mgross@xxxxxxxxxxxxxxx> I've noticed that a few times when the bios is buggy the code for detecting failing DMAR hardware can get stuck in an infinite loop in early boot waiting for jiffies to time out. Which are not running at the time we get stuck in this macro. This patch replaces the use of the jiffies to detect the failing hardware with the tsc counter. Not a big change, instead of getting a black screen forever, you get a panic... I also changed the time out from an arbitrary 60 seconds to an arbitrary 10 seconds. Signed-off-by: mark gross <mgross@xxxxxxxxxxxxxxx> Cc: Greg KH <greg@xxxxxxxxx> Cc: Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- drivers/pci/intel-iommu.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff -puN drivers/pci/intel-iommu.c~intel-iommu-make-intel-iommu_wait_op-work-when-jiffies-arent-running drivers/pci/intel-iommu.c --- a/drivers/pci/intel-iommu.c~intel-iommu-make-intel-iommu_wait_op-work-when-jiffies-arent-running +++ a/drivers/pci/intel-iommu.c @@ -49,7 +49,7 @@ #define DEFAULT_DOMAIN_ADDRESS_WIDTH 48 -#define DMAR_OPERATION_TIMEOUT (HZ*60) /* 1m */ +#define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) /* 10sec */ #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1) @@ -490,13 +490,13 @@ static int iommu_alloc_root_entry(struct #define IOMMU_WAIT_OP(iommu, offset, op, cond, sts) \ {\ - unsigned long start_time = jiffies;\ + cycles_t start_time = get_cycles();\ while (1) {\ sts = op (iommu->reg + offset);\ if (cond)\ break;\ - if (time_after(jiffies, start_time + DMAR_OPERATION_TIMEOUT))\ - panic("DMAR hardware is malfunctioning\n");\ + if (DMAR_OPERATION_TIMEOUT < (get_cycles() - start_time))\ + panic(KERN_ERR "DMAR hardware is malfunctioning\n");\ cpu_relax();\ }\ } _ Patches currently in -mm which might be from mgross@xxxxxxxxxxxxxxx are intel-iommu-make-intel-iommu_wait_op-work-when-jiffies-arent-running.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html