On Tue, 9 Feb 2021, tanxiaofei wrote:
Hi Finn, Thanks for reviewing the patch set.
On 2021/2/8 15:57, Finn Thain wrote:
On Sun, 7 Feb 2021, Xiaofei Tan wrote:
Replace spin_lock_irqsave with spin_lock in hard IRQ of SCSI drivers. There are no function changes, but may speed up if interrupt happen too often.
This change doesn't necessarily work on platforms that support nested interrupts.
Linux doesn't support nested interrupts anymore after the following patch, so please don't worry this.
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
Clearly that patch did not disable interrupts. It removed a statement that enabled them.
Were you able to measure any benefit from this change on some other platform?
It's hard to measure the benefit of this change.
It's hard to see any benefit. But it's easy to see risk, when there's no indication that you've confirmed that the affected drivers do not rely on the irq lock, nor tested them for regressions, nor checked whether the affected platforms meet your assumuptions.
Hmm, you could take this patch set as cleanup. thanks.
A "cleanup" does not change program behaviour. Can you demonstrate that program behaviour is unchanged?
Please see also, https://lore.kernel.org/linux-scsi/89c5cb05cb844939ae684db0077f675f@h3c.com/
.