Commit 7eadcfd6 authored by Giovanni Cabiddu's avatar Giovanni Cabiddu Committed by Herbert Xu

crypto: qat - prevent spurious MSI interrupt in VF

QAT GEN2 devices suffer from a defect where the MSI interrupt can be
sent multiple times.

If the second (spurious) interrupt is handled before the bottom half
handler runs, then the extra interrupt is effectively ignored because
the bottom half is only scheduled once.
However, if the top half runs again after the bottom half runs, this
will appear as a spurious PF to VF interrupt.

This can be avoided by checking the interrupt mask register in addition
to the interrupt source register in the interrupt handler.

This patch is based on earlier work done by Conor McLoughlin.
Signed-off-by: default avatarGiovanni Cabiddu <giovanni.cabiddu@intel.com>
Co-developed-by: default avatarMarco Chiappero <marco.chiappero@intel.com>
Signed-off-by: default avatarMarco Chiappero <marco.chiappero@intel.com>
Reviewed-by: default avatarFiona Trahe <fiona.trahe@intel.com>
Signed-off-by: default avatarHerbert Xu <herbert@gondor.apana.org.au>
parent 0a73c762
......@@ -18,6 +18,7 @@
#include "adf_pf2vf_msg.h"
#define ADF_VINTSOU_OFFSET 0x204
#define ADF_VINTMSK_OFFSET 0x208
#define ADF_VINTSOU_BUN BIT(0)
#define ADF_VINTSOU_PF2VF BIT(1)
......@@ -161,11 +162,20 @@ static irqreturn_t adf_isr(int irq, void *privdata)
&GET_BARS(accel_dev)[hw_data->get_misc_bar_id(hw_data)];
void __iomem *pmisc_bar_addr = pmisc->virt_addr;
bool handled = false;
u32 v_int;
u32 v_int, v_mask;
/* Read VF INT source CSR to determine the source of VF interrupt */
v_int = ADF_CSR_RD(pmisc_bar_addr, ADF_VINTSOU_OFFSET);
/* Read VF INT mask CSR to determine which sources are masked */
v_mask = ADF_CSR_RD(pmisc_bar_addr, ADF_VINTMSK_OFFSET);
/*
* Recompute v_int ignoring sources that are masked. This is to
* avoid rescheduling the tasklet for interrupts already handled
*/
v_int &= ~v_mask;
/* Check for PF2VF interrupt */
if (v_int & ADF_VINTSOU_PF2VF) {
/* Disable PF to VF interrupt */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment