PCI: iproc: Fix multi-MSI base vector number allocation

Commit fc54bae288 ("PCI: iproc: Allow allocation of multiple MSIs")
introduced multi-MSI support with a broken allocation mechanism (it failed
to reserve the proper number of bits from the inner domain).  Natural
alignment of the base vector number was also not guaranteed.

Link: https://lore.kernel.org/r/20210622152630.40842-1-sbodomerle@gmail.com
Fixes: fc54bae288 ("PCI: iproc: Allow allocation of multiple MSIs")
Reported-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Sandor Bodo-Merle <sbodomerle@gmail.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Marc Zyngier <maz@kernel.org>
Acked-by: Pali Rohár <pali@kernel.org>
Acked-by: Ray Jui <ray.jui@broadcom.com>
This commit is contained in:
Sandor Bodo-Merle 2021-06-22 17:26:29 +02:00 committed by Lorenzo Pieralisi
parent 6efb943b86
commit e673d697b9
1 changed files with 11 additions and 10 deletions

View File

@ -252,18 +252,18 @@ static int iproc_msi_irq_domain_alloc(struct irq_domain *domain,
mutex_lock(&msi->bitmap_lock);
/* Allocate 'nr_cpus' number of MSI vectors each time */
hwirq = bitmap_find_next_zero_area(msi->bitmap, msi->nr_msi_vecs, 0,
msi->nr_cpus, 0);
if (hwirq < msi->nr_msi_vecs) {
bitmap_set(msi->bitmap, hwirq, msi->nr_cpus);
} else {
mutex_unlock(&msi->bitmap_lock);
return -ENOSPC;
}
/*
* Allocate 'nr_irqs' multiplied by 'nr_cpus' number of MSI vectors
* each time
*/
hwirq = bitmap_find_free_region(msi->bitmap, msi->nr_msi_vecs,
order_base_2(msi->nr_cpus * nr_irqs));
mutex_unlock(&msi->bitmap_lock);
if (hwirq < 0)
return -ENOSPC;
for (i = 0; i < nr_irqs; i++) {
irq_domain_set_info(domain, virq + i, hwirq + i,
&iproc_msi_bottom_irq_chip,
@ -284,7 +284,8 @@ static void iproc_msi_irq_domain_free(struct irq_domain *domain,
mutex_lock(&msi->bitmap_lock);
hwirq = hwirq_to_canonical_hwirq(msi, data->hwirq);
bitmap_clear(msi->bitmap, hwirq, msi->nr_cpus);
bitmap_release_region(msi->bitmap, hwirq,
order_base_2(msi->nr_cpus * nr_irqs));
mutex_unlock(&msi->bitmap_lock);