Re: [PATCH] uio: disable lazy irq disable to avoid double fire

2020-05-26 Thread Dan Carpenter
Hi Thommy, url: https://github.com/0day-ci/linux/commits/Thommy-Jakobsson/uio-disable-lazy-irq-disable-to-avoid-double-fire/20200521-225755 base: https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git c9d7e3da1f3c4cf5dddfc5d7ce4d76d013aba1cc config: i386-allyesconfig (attache

Re: [PATCH] uio: disable lazy irq disable to avoid double fire

2020-05-23 Thread Thommy Jakobsson
On 2020-05-22 11:14, Greg KH wrote: > On Thu, May 21, 2020 at 04:42:09PM +0200, Thommy Jakobsson wrote: >> +if (uioinfo->irq) { > > How is this not true at this point in time based on the code above this? > ->irq should always be set here, right? It seems to me like there is a path to continue

Re: [PATCH] uio: disable lazy irq disable to avoid double fire

2020-05-22 Thread Greg KH
On Thu, May 21, 2020 at 04:42:09PM +0200, Thommy Jakobsson wrote: > uio_pdrv_genirq and uio_dmem_genirq interrupts are handled in > userspace. So the condition for the interrupt hasn't normally not been > cleared when top half returns. disable_irq_nosync is called in top half, > but since that norm

[PATCH] uio: disable lazy irq disable to avoid double fire

2020-05-21 Thread Thommy Jakobsson
uio_pdrv_genirq and uio_dmem_genirq interrupts are handled in userspace. So the condition for the interrupt hasn't normally not been cleared when top half returns. disable_irq_nosync is called in top half, but since that normally is lazy the irq isn't actually disabled. For level triggered interru