[PATCH 02/14] scsi: hisi_sas: Remove preemptible().

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: linux-scsi@vger.kernel.org
Cc: Finn Thain <fthain@telegraphics.com.au>,
	GR-QLogic-Storage-Upstream@marvell.com,
	Hannes Reinecke <hare@kernel.org>,
	Jack Wang <jinpu.wang@cloud.ionos.com>,
	John Garry <john.garry@huawei.com>,
	linux-m68k@lists.linux-m68k.org,
	Manish Rangankar <mrangankar@marvell.com>,
	Michael Schmitz <schmitzmic@gmail.com>,
	MPT-FusionLinux.pdl@broadcom.com,
	Nilesh Javali <njavali@marvell.com>,
	Sathya Prakash <sathya.prakash@broadcom.com>,
	Sreekanth Reddy <sreekanth.reddy@broadcom.com>,
	Suganath Prabu Subramani  <suganath-prabu.subramani@broadcom.com>,
	Vikram Auradkar <auradkar@google.com>,
	Xiang Chen <chenxiang66@hisilicon.com>,
	Xiaofei Tan <tanxiaofei@huawei.com>,
	"James E . J . Bottomley" <jejb@linux.ibm.com>,
	"Martin K . Petersen" <martin.petersen@oracle.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	"Ahmed S . Darwish" <a.darwish@linutronix.de>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: [PATCH 02/14] scsi: hisi_sas: Remove preemptible().
Date: Thu, 26 Nov 2020 14:29:40 +0100	[thread overview]
Message-ID: <20201126132952.2287996-3-bigeasy@linutronix.de> (raw)
In-Reply-To: <20201126132952.2287996-1-bigeasy@linutronix.de>

From: "Ahmed S. Darwish" <a.darwish@linutronix.de>

hisi_sas_task_exec() uses preemptible() to see if it's safe to block.
This does not work for CONFIG_PREEMPT_COUNT=n kernels in which
preemptible() always returns 0.

The problem is masked when enabling some of the common Kconfig.debug
options (like CONFIG_DEBUG_ATOMIC_SLEEP), as they implicitly enable the
preemption counter.

In general, driver leaf functions should not make logic decisions based
on the context they're called from. The caller should be the entity
responsible for explicitly indicating context.

Since hisi_sas_task_exec() already has a gfp_t flags parameter, use it
as the explicit context marker.

Fixes: 214e702d4b70 ("scsi: hisi_sas: Adjust task reject period during host reset")
Fixes: 550c0d89d52d ("scsi: hisi_sas: Replace in_softirq() check in hisi_sas_task_exec()")
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Xiaofei Tan <tanxiaofei@huawei.com>
Cc: Xiang Chen <chenxiang66@hisilicon.com>
Cc: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index c8dd8588f800e..06e65c461f027 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -585,13 +585,7 @@ static int hisi_sas_task_exec(struct sas_task *task, gfp_t gfp_flags,
 	dev = hisi_hba->dev;
 
 	if (unlikely(test_bit(HISI_SAS_REJECT_CMD_BIT, &hisi_hba->flags))) {
-		/*
-		 * For IOs from upper layer, it may already disable preempt
-		 * in the IO path, if disable preempt again in down(),
-		 * function schedule() will report schedule_bug(), so check
-		 * preemptible() before goto down().
-		 */
-		if (!preemptible())
+		if (!gfpflags_allow_blocking(gfp_flags))
 			return -EINVAL;
 
 		down(&hisi_hba->sem);
-- 
2.29.2

