| Message ID | 20220111012441.3232527-1-yangyingliang@huawei.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 61263b3a11a2594b4e898f166c31162236182b5c |
| Headers | show |
| Series | [-next,v2] scsi: efct: don't use GFP_KERNEL under spin lock | expand |
On 1/10/2022 5:24 PM, Yang Yingliang wrote: > GFP_KERNEL/GFP_DMA can't be used under a spin lock, according the > comment of els_ios_lock, it's used to protect els ios list, so we > can move down the spin lock to avoid using this flag under the lock. > > Reported-by: Hulk Robot <hulkci@huawei.com> > Fixes: 8f406ef72859 ("scsi: elx: libefc: Extended link Service I/O handling") > Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> > --- > v2: > move up unlock to just protect list_add_tail() > --- > drivers/scsi/elx/libefc/efc_els.c | 8 ++------ > 1 file changed, 2 insertions(+), 6 deletions(-) > Looks fine. Reviewed-by: James Smart <jsmart2021@gmail.com> -- james
Yang, > GFP_KERNEL/GFP_DMA can't be used under a spin lock, according the > comment of els_ios_lock, it's used to protect els ios list, so we > can move down the spin lock to avoid using this flag under the lock. Applied to 5.17/scsi-staging, thanks!
On Tue, 11 Jan 2022 09:24:41 +0800, Yang Yingliang wrote: > GFP_KERNEL/GFP_DMA can't be used under a spin lock, according the > comment of els_ios_lock, it's used to protect els ios list, so we > can move down the spin lock to avoid using this flag under the lock. > > Applied to 5.17/scsi-fixes, thanks! [1/1] scsi: efct: don't use GFP_KERNEL under spin lock https://git.kernel.org/mkp/scsi/c/61263b3a11a2
diff --git a/drivers/scsi/elx/libefc/efc_els.c b/drivers/scsi/elx/libefc/efc_els.c index 7bb4f9aad2c8..84bc81d7ce76 100644 --- a/drivers/scsi/elx/libefc/efc_els.c +++ b/drivers/scsi/elx/libefc/efc_els.c @@ -46,18 +46,14 @@ efc_els_io_alloc_size(struct efc_node *node, u32 reqlen, u32 rsplen) efc = node->efc; - spin_lock_irqsave(&node->els_ios_lock, flags); - if (!node->els_io_enabled) { efc_log_err(efc, "els io alloc disabled\n"); - spin_unlock_irqrestore(&node->els_ios_lock, flags); return NULL; } els = mempool_alloc(efc->els_io_pool, GFP_ATOMIC); if (!els) { atomic_add_return(1, &efc->els_io_alloc_failed_count); - spin_unlock_irqrestore(&node->els_ios_lock, flags); return NULL; } @@ -74,7 +70,6 @@ efc_els_io_alloc_size(struct efc_node *node, u32 reqlen, u32 rsplen) &els->io.req.phys, GFP_KERNEL); if (!els->io.req.virt) { mempool_free(els, efc->els_io_pool); - spin_unlock_irqrestore(&node->els_ios_lock, flags); return NULL; } @@ -94,10 +89,11 @@ efc_els_io_alloc_size(struct efc_node *node, u32 reqlen, u32 rsplen) /* add els structure to ELS IO list */ INIT_LIST_HEAD(&els->list_entry); + spin_lock_irqsave(&node->els_ios_lock, flags); list_add_tail(&els->list_entry, &node->els_ios_list); + spin_unlock_irqrestore(&node->els_ios_lock, flags); } - spin_unlock_irqrestore(&node->els_ios_lock, flags); return els; }
GFP_KERNEL/GFP_DMA can't be used under a spin lock, according the comment of els_ios_lock, it's used to protect els ios list, so we can move down the spin lock to avoid using this flag under the lock. Reported-by: Hulk Robot <hulkci@huawei.com> Fixes: 8f406ef72859 ("scsi: elx: libefc: Extended link Service I/O handling") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> --- v2: move up unlock to just protect list_add_tail() --- drivers/scsi/elx/libefc/efc_els.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-)