Modularize ufs-bsg
v3 cover Modularize ufs-bsg
In order to improve the flexibility of ufs-bsg, modularizing it is a good choice. Introduce tri-state to ufs-bsg to allow users compile it as an external module.
Changes since v2:
- Incoperate comments from Avri Altman, Bjorn Andersson and Bart Van Assche.
- Updated the main change to use platform_device_register_data() to create a platform device ufs-bsg in ufshcd_platform_init(). Also added the ufs-bsg platform driver. In the driver probe routine, create the ufs-bsg char dev under /dev/bsg/, the ufs-bsg platform device is the parent of the created ufs-bsg char dev.
- Modified commit message.
- Removed defconfig change.
Changes since v1:
- Included one more defconfig change.
Can Guo (2):
scsi: ufs: Put SCSI host after remove it
scsi: ufs: Modularize ufs-bsg
drivers/scsi/ufs/Kconfig | 2 +- drivers/scsi/ufs/Makefile | 2 +- drivers/scsi/ufs/ufs_bsg.c | 67 +++++++++++++++++++++------------------- drivers/scsi/ufs/ufs_bsg.h | 8 ----- drivers/scsi/ufs/ufshcd-pltfrm.c | 12 +++++++ drivers/scsi/ufs/ufshcd.c | 8 ++--- drivers/scsi/ufs/ufshcd.h | 2 +- 7 files changed, 55 insertions(+), 46 deletions(-)
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project
PATCH v3 1/2 scsi: ufs: Put SCSI host after remove it
In ufshcd_remove(), after SCSI host is removed, put it once so that its resources can be released.
drivers/scsi/ufs/ufshcd.c +1 -0
@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 10 lines · Can Guo
In
ufshcd_remove(), after SCSI host is removed, put it once so that its resources can be released.Signed-off-by Can Guo <[email protected]>drivers/scsi/ufs/ufshcd.c +0
-0
drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8257
Quoted 7 lines
drivers/scsi/ufs/ufshcd.c +1
-0@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);
Hi Can,
The UFS driver may queue work asynchronously and that asynchronous work may refer to the SCSI host, e.g. ufshcd_err_handler(). Is it guaranteed that all that asynchronous work has finished before scsi_host_put() is called?
Thanks,
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 21 lines · Bart Van Assche
Quoted 10 lines · Can Guo
In
ufshcd_remove(), after SCSI host is removed, put it once so that its resources can be released.Signed-off-by Can Guo <[email protected]>drivers/scsi/ufs/ufshcd.c +0
-0drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8257
Quoted 7 lines
drivers/scsi/ufs/ufshcd.c +1
-0@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);Hi Can,
Quoted 8 lines
The UFS driver may queue work asynchronously and that asynchronous work may refer to the SCSI host, e.g.
ufshcd_err_handler(). Is it guaranteed that all that asynchronous work has finished beforescsi_host_put()is called?Thanks,
Bart.
Hi Bart,
Thanks for pointing it out. I noticed that you are changing this path too in below 2 changes.
https://marc.info/?l=linux-scsi&m=157591520015924&w=2
https://marc.info/?l=linux-scsi&m=157591519915923&w=2
Actually the async works you pointed may also affect your change, because you may tear down the hba->cmd_queue too early, as there can be devm commands sent by clock gating, eeh_work and eh_work after that point, meaning when blk_get_request is called in exec_dev_cmd(), hba->cmd_queue may have been released already.
@@ -8263,6 +8232,7 @@ void ufshcd_remove(struct ufs_hba *hba)
{
ufs_bsg_remove(hba);
ufs_sysfs_remove_nodes(hba->dev);
+ blk_cleanup_queue(hba->cmd_queue);
scsi_remove_host(hba->host);
/* disable interrupts */
ufshcd_disable_intr(hba, hba->intr_mask);How do you think if I replace my patch with below one? In this way, you can also move blk_cleanup_queue() behind cancel_work_sync(eh_work).
drivers/scsi/ufs/ufshcd.c +0 -4
@@ -8251,15 +8251,17 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); -/* disable interrupts */ -ufshcd_disable_intr (hba, hba->intr_mask); -ufshcd_hba_stop (hba, true); -ufshcd_exit_clk_scaling (hba);ufshcd_exit_clk_gating (hba);if (ufshcd_is_clkscaling_supported (hba)) device_remove_file(hba->dev,
&hba->clk_scaling.enable_attr);
+ cancel_work_sync(&hba->eeh_work);
+ cancel_work_sync(&hba->eh_work);
+ /* disable interrupts */
+ ufshcd_disable_intr(hba, hba->intr_mask);
+ ufshcd_hba_stop(hba, true);
ufshcd_hba_exit(hba);+ ufshcd_dealloc_host(hba);
}
EXPORT_SYMBOL_GPL(ufshcd_remove);
Signature · 3 lines
Thanks, Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 20 lines · On 2019-12-14 14:24, [email protected] wrote:
How do you think if I replace my patch with below one? In this way, you can also move
blk_cleanup_queue()behindcancel_work_sync(eh_work).drivers/scsi/ufs/ufshcd.c +0
-4@@ -8251,15 +8251,17 @@ void ufshcd_remove(struct ufs_hba *hba) ufs_bsg_remove(hba); ufs_sysfs_remove_nodes(hba->dev); scsi_remove_host(hba->host); - /* disable interrupts */ - ufshcd_disable_intr(hba, hba->intr_mask); - ufshcd_hba_stop(hba, true); - ufshcd_exit_clk_scaling(hba); ufshcd_exit_clk_gating(hba); if (ufshcd_is_clkscaling_supported(hba)) device_remove_file(hba->dev,
drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8267
Quoted 10 lines
drivers/scsi/ufs/ufshcd.c +6
-0&hba->clk_scaling.enable_attr); + cancel_work_sync(&hba->eeh_work); + cancel_work_sync(&hba->eh_work); + /* disable interrupts */ + ufshcd_disable_intr(hba, hba->intr_mask); + ufshcd_hba_stop(hba, true); ufshcd_hba_exit(hba); + ufshcd_dealloc_host(hba); } EXPORT_SYMBOL_GPL(ufshcd_remove);
Hi Can,
To which kernel tree does the above patch apply? I'm asking this because I don't see the recently added blk_cleanup_queue() calls in the above patch. Please start from Martin's latest scsi-queue branch when preparing SCSI patches.
Additionally, is it on purpose that there is no scsi_host_put() call in the above code? I'd like to keep that call because without that call a memory leak will occur when unloading the ufshcd-core kernel driver.
Thanks,
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 39 lines · Bart Van Assche
Quoted 20 lines · On 2019-12-14 14:24, [email protected] wrote:
How do you think if I replace my patch with below one? In this way, you can also move
blk_cleanup_queue()behindcancel_work_sync(eh_work).drivers/scsi/ufs/ufshcd.c +0
-4@@ -8251,15 +8251,17 @@ void ufshcd_remove(struct ufs_hba *hba) ufs_bsg_remove(hba); ufs_sysfs_remove_nodes(hba->dev); scsi_remove_host(hba->host); - /* disable interrupts */ - ufshcd_disable_intr(hba, hba->intr_mask); - ufshcd_hba_stop(hba, true); - ufshcd_exit_clk_scaling(hba); ufshcd_exit_clk_gating(hba); if (ufshcd_is_clkscaling_supported(hba)) device_remove_file(hba->dev,drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8267
Quoted 10 lines
drivers/scsi/ufs/ufshcd.c +6
-0&hba->clk_scaling.enable_attr); + cancel_work_sync(&hba->eeh_work); + cancel_work_sync(&hba->eh_work); + /* disable interrupts */ + ufshcd_disable_intr(hba, hba->intr_mask); + ufshcd_hba_stop(hba, true); ufshcd_hba_exit(hba); + ufshcd_dealloc_host(hba); } EXPORT_SYMBOL_GPL(ufshcd_remove);Hi Can,
To which kernel tree does the above patch apply? I'm asking this because I don't see the recently added
blk_cleanup_queue()calls in the above patch. Please start from Martin's latest scsi-queue branch when preparing SCSI patches.
Quoted 7 lines
Additionally, is it on purpose that there is no
scsi_host_put()call in the above code? I'd like to keep that call because without that call a memory leak will occur when unloading the ufshcd-core kernel driver.Thanks,
Bart.
Hi Bart,
This is applied to 5.5/scsi-queue. The two changes I patsed from you are not merged yet, I am still doing code review to them, so there is no blk_cleanup_queue() calls in my code base. I am just saying you may move your blk_cleanup_queue() calls below cancel_work_sync(&hba->eh_work) if my change applies. How do you think?
scsi_host_put() was there before but explicitly removed by afa3dfd42d205b106787476647735aa1de1a5d02. I agree with you, without this change, there is memory leak.
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 9 lines · On 2019-12-15 17:34, [email protected] wrote:
This is applied to 5.5/scsi-queue. The two changes I patsed from you are not merged yet, I am still doing code review to them, so there is no
blk_cleanup_queue()calls in my code base. I am just saying you may move yourblk_cleanup_queue()calls belowcancel_work_sync(&hba->eh_work)if my change applies. How do you think?
scsi_host_put()was there before but explicitly removed byafa3dfd42d205b106787476647735aa1de1a5d02. I agree with you, without this change, there is memory leak.
Hi Can,
Since your patch restores a call that was removed earlier, please consider adding a Fixes: tag to your patch.
Please also have a look at
https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git/log/?h=5.6/scsi-queue.
As one can see my patches that introduce blk_cleanup_queue() and blk_mq_free_tag_set() calls have already been queued on Martin's 5.6/scsi-queue branch.
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 16 lines · Bart Van Assche
Quoted 9 lines · On 2019-12-15 17:34, [email protected] wrote:
This is applied to 5.5/scsi-queue. The two changes I patsed from you are not merged yet, I am still doing code review to them, so there is no
blk_cleanup_queue()calls in my code base. I am just saying you may move yourblk_cleanup_queue()calls belowcancel_work_sync(&hba->eh_work)if my change applies. How do you think?
scsi_host_put()was there before but explicitly removed byafa3dfd42d205b106787476647735aa1de1a5d02. I agree with you, without this change, there is memory leak.Hi Can,
Since your patch restores a call that was removed earlier, please consider adding a Fixes: tag to your patch.
Quoted 7 lines
Please also have a look at
https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git/log/?h=5.6/scsi-queue.
As one can see my patches that introduceblk_cleanup_queue()andblk_mq_free_tag_set()calls have already been queued on Martin's 5.6/scsi-queue branch.Bart.
Hi Bart,
Sure, I will add the Fixes tag and rebase my changes. How about the logic part of this change? Does it look good to you?
Sorry I was not aware of that your changes have been applied to 5.6/scsi-queue. I am still trying to get it tested on my setups...
Anyways, aside of hba->cmd_queue, tearing down hba->tmf_queue before scsi_remove_host() may be problem too. Requests can still be
sent before and during scsi_remove_host(). If a request timed out, task abort will be invoked to abort the request, during which hba->tmf_queue is expected to be present. Please correct me if I am wrong.
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 38 lines · On 2019-12-16 11:12, [email protected] wrote:
Quoted 16 lines · Bart Van Assche
Quoted 9 lines · On 2019-12-15 17:34, [email protected] wrote:
This is applied to 5.5/scsi-queue. The two changes I patsed from you are not merged yet, I am still doing code review to them, so there is no
blk_cleanup_queue()calls in my code base. I am just saying you may move yourblk_cleanup_queue()calls belowcancel_work_sync(&hba->eh_work)if my change applies. How do you think?
scsi_host_put()was there before but explicitly removed byafa3dfd42d205b106787476647735aa1de1a5d02. I agree with you, without this change, there is memory leak.Hi Can,
Since your patch restores a call that was removed earlier, please consider adding a Fixes: tag to your patch.
Quoted 7 lines
Please also have a look at
https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git/log/?h=5.6/scsi-queue.
As one can see my patches that introduceblk_cleanup_queue()andblk_mq_free_tag_set()calls have already been queued on Martin's 5.6/scsi-queue branch.Bart.
Hi Bart,
Sure, I will add the Fixes tag and rebase my changes. How about the logic part of this change? Does it look good to you?
Sorry I was not aware of that your changes have been applied to 5.6/scsi-queue. I am still trying to get it tested on my setups...
Anyways, aside ofhba->cmd_queue, tearing downhba->tmf_queuebeforescsi_remove_host()may be problem too. Requests can still be
sent before and duringscsi_remove_host(). If a request timed out, task abort will be invoked to abort the request, during whichhba->tmf_queueis expected to be present. Please correct me if I am wrong.
Quoted 3 lines
Thanks,
Can Guo.
Hi Bart
Just found that I should also remove the ufshcd_dealloc_host() called in ufshcd_pci_remove() to make sure the deallocation is only handled by ufshcd_remove().
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
In reply to [email protected]Quoted 2 lines · On 12/15/19 7:12 PM, [email protected] wrote:
Sure, I will add the Fixes tag and rebase my changes. How about the logic part of this change? Does it look good to you?
Hi Can,
You may want to ask someone who is more familiar with the UFS driver than I to have a look. I'm not a UFS expert ...
Quoted 7 lines
Sorry I was not aware of that your changes have been applied to 5.6/scsi-queue. I am still trying to get it tested on my setups...
Anyways, aside ofhba->cmd_queue, tearing downhba->tmf_queuebeforescsi_remove_host()may be problem too. Requests can still be
sent before and duringscsi_remove_host(). If a request timed out, task abort will be invoked to abort the request, during whichhba->tmf_queueis expected to be present. Please correct me if I am wrong.
I agree that the code I added in ufshcd_remove() probably needs to be moved somewhere below the scsi_remove_host() call.
Thanks,
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
In reply to Bart Van AsscheQuoted 21 lines · Bart Van Assche
Quoted 10 lines · Can Guo
In
ufshcd_remove(), after SCSI host is removed, put it once so that its resources can be released.Signed-off-by Can Guo <[email protected]>drivers/scsi/ufs/ufshcd.c +0
-0drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8257
Quoted 7 lines
drivers/scsi/ufs/ufshcd.c +1
-0@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);Hi Can,
Quoted 8 lines
The UFS driver may queue work asynchronously and that asynchronous work may refer to the SCSI host, e.g.
ufshcd_err_handler(). Is it guaranteed that all that asynchronous work has finished beforescsi_host_put()is called?Thanks,
Bart.
Hi Bart,
As SCSI host is allocated in ufshcd_platform_init() during platform drive probe, it is much more appropriate if platform driver calls ufshcd_dealloc_host() in their own drv->remove() path. How do you think if I change it as below? If it is OK to you, please ignore my previous mails.
drivers/scsi/ufs/ufs-qcom.c +1 -0
@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev)pm_runtime_get_sync (&(pdev)->dev);ufshcd_remove (hba); +ufshcd_dealloc_host (hba);return 0 ; }
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 11 lines · On 12/16/19 6:31 AM, [email protected] wrote:
As SCSI host is allocated in
ufshcd_platform_init()during platform drive probe, it is much more appropriate if platform driver callsufshcd_dealloc_host()in their owndrv->remove()path. How do you think if I change it as below? If it is OK to you, please ignore my previous mails.drivers/scsi/ufs/ufs-qcom.c +0
-0
drivers/scsi/ufs/ufs-qcom.c static int ufs_qcom_remove(struct platform_device · line 3245
Quoted 7 lines
drivers/scsi/ufs/ufs-qcom.c +1
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev) pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0; }
Hi Can,
Apparently some UFS drivers call ufshcd_remove() only and others (PCIe) call both ufshcd_remove() and ufshcd_dealloc_host(). I think that the above change will cause trouble for the PCIe driver unless the ufshcd_dealloc_host() call is removed from ufshcd_pci_remove().
Thanks,
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 22 lines · Bart Van Assche
Quoted 11 lines · On 12/16/19 6:31 AM, [email protected] wrote:
As SCSI host is allocated in
ufshcd_platform_init()during platform drive probe, it is much more appropriate if platform driver callsufshcd_dealloc_host()in their owndrv->remove()path. How do you think if I change it as below? If it is OK to you, please ignore my previous mails.drivers/scsi/ufs/ufs-qcom.c +0
-0drivers/scsi/ufs/ufs-qcom.c static int ufs_qcom_remove(struct · line 3245
Quoted 7 lines
drivers/scsi/ufs/ufs-qcom.c +1
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev) pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0; }Hi Can,
Quoted 8 lines
Apparently some UFS drivers call
ufshcd_remove()only and others (PCIe) call bothufshcd_remove()andufshcd_dealloc_host(). I think that the above change will cause trouble for the PCIe driver unless theufshcd_dealloc_host()call is removed fromufshcd_pci_remove().Thanks,
Bart.
Hi Bart,
You may get me wrong. I mean we should do like what ufshcd-pci.c does. As driver probe routine allocates SCSI host, then driver remove() should de-allocate it. Meaning ufs_qcom_remove() should call both ufshcd_remove() and ufshcd_dealloc_host().
drivers/scsi/ufs/ufs-qcom.c +0 -0
@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev) pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0; }
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 16 lines · On 12/16/19 4:46 PM, [email protected] wrote:
Quoted 4 lines · Bart Van Assche
Apparently some UFS drivers call
ufshcd_remove()only and others (PCIe) call bothufshcd_remove()andufshcd_dealloc_host(). I think that the above change will cause trouble for the PCIe driver unless theufshcd_dealloc_host()call is removed fromufshcd_pci_remove().You may get me wrong. I mean we should do like what
ufshcd-pci.cdoes. As driver probe routine allocates SCSI host, then driverremove()should de-allocate it. Meaningufs_qcom_remove()should call bothufshcd_remove()andufshcd_dealloc_host().drivers/scsi/ufs/ufs-qcom.c +0
-0
drivers/scsi/ufs/ufs-qcom.c static int ufs_qcom_remove(struct platform_device · line 3244
Quoted 7 lines
drivers/scsi/ufs/ufs-qcom.c +0
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev) pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0; }
Hi Can,
If it is possible to move the ufshcd_dealloc_host() into ufshcd_remove() then I would prefer to do that. If all UFS transport drivers need that call then I think that call should happen from the UFS core instead of from the transport drivers.
Thanks,
Bart.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 27 lines · Bart Van Assche
Quoted 16 lines · On 12/16/19 4:46 PM, [email protected] wrote:
Quoted 4 lines · Bart Van Assche
Apparently some UFS drivers call
ufshcd_remove()only and others (PCIe) call bothufshcd_remove()andufshcd_dealloc_host(). I think that the above change will cause trouble for the PCIe driver unless theufshcd_dealloc_host()call is removed fromufshcd_pci_remove().You may get me wrong. I mean we should do like what
ufshcd-pci.cdoes. As driver probe routine allocates SCSI host, then driverremove()should de-allocate it. Meaningufs_qcom_remove()should call bothufshcd_remove()andufshcd_dealloc_host().drivers/scsi/ufs/ufs-qcom.c +0
-0drivers/scsi/ufs/ufs-qcom.c static int ufs_qcom_remove(struct · line 3244
Quoted 7 lines
drivers/scsi/ufs/ufs-qcom.c +0
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev) pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0; }Hi Can,
Quoted 8 lines
If it is possible to move the
ufshcd_dealloc_host()intoufshcd_remove()then I would prefer to do that. If all UFS transport drivers need that call then I think that call should happen from the UFS core instead of from the transport drivers.Thanks,
Bart.
Yeah, that is an once for all solution, but I not sure if PCI folks are OK if I remove the ufshcd_dealloc_host() call from their driver. In next version, I will try to make such change and see.
Thanks,
Can Guo.
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
In reply to [email protected]Quoted 45 lines · On Mon, Dec 16, 2019 at 10:31:29PM +0800, [email protected] wrote:
Quoted 22 lines · Bart Van Assche
Quoted 11 lines · Can Guo
In
ufshcd_remove(), after SCSI host is removed, put it once so that its
resources can be released.Signed-off-by Can Guo <[email protected]>drivers/scsi/ufs/ufshcd.c +0
-0drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8257
Quoted 7 lines
drivers/scsi/ufs/ufshcd.c +1
-0@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);Hi Can,
Quoted 8 lines
The UFS driver may queue work asynchronously and that asynchronous work may refer to the SCSI host, e.g.
ufshcd_err_handler(). Is it guaranteed that all that asynchronous work has finished beforescsi_host_put()is called?Thanks,
Bart.
Hi Bart,
As SCSI host is allocated in
ufshcd_platform_init()during platform drive probe, it is much more appropriate if platform driver callsufshcd_dealloc_host()in their owndrv->remove()path. How do you think if I change it as below? If it is OK to you, please ignore my previous mails.drivers/scsi/ufs/ufs-qcom.c +0
-0
drivers/scsi/ufs/ufs-qcom.c static int ufs_qcom_remove(struct platform_device · line 3245
Quoted 7 lines
drivers/scsi/ufs/ufs-qcom.c +1
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct platform_device *pdev)pm_runtime_get_sync (&(pdev)->dev);ufshcd_remove (hba); +ufshcd_dealloc_host (hba);return 0 ; }
Wait, why is this a platform device? Don't you hang off of a pci device? Or am I missing something earlier in this patchset?
thanks,
greg k-h
Re: [PATCH 1/2] scsi: ufs: Put SCSI host after remove it
Quoted 58 lines · Greg KH
Quoted 48 lines · On Mon, Dec 16, 2019 at 10:31:29PM +0800, [email protected] wrote:
Quoted 22 lines · Bart Van Assche
Quoted 11 lines · Can Guo
In
ufshcd_remove(), after SCSI host is removed, put it once so that its
resources can be released.Signed-off-by Can Guo <[email protected]>drivers/scsi/ufs/ufshcd.c +0
-0drivers/scsi/ufs/ufshcd.c void ufshcd_remove(struct ufs_hba *hba) · line 8257
Quoted 7 lines
drivers/scsi/ufs/ufshcd.c +1
-0@@ -8251,6 +8251,7 @@ void ufshcd_remove(struct ufs_hba *hba)ufs_bsg_remove (hba);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host); +scsi_host_put (hba->host);/* disable interrupts */ ufshcd_disable_intr (hba, hba->intr_mask);ufshcd_hba_stop (hba, true);Hi Can,
Quoted 8 lines
The UFS driver may queue work asynchronously and that asynchronous work may refer to the SCSI host, e.g.
ufshcd_err_handler(). Is it guaranteed that all that asynchronous work has finished beforescsi_host_put()is called?Thanks,
Bart.
Hi Bart,
As SCSI host is allocated in
ufshcd_platform_init()during platform drive probe, it is much more appropriate if platform driver callsufshcd_dealloc_host()in their owndrv->remove()path. How do you think if I change it as below? If it is OK to you, please ignore my previous mails.drivers/scsi/ufs/ufs-qcom.c +0
-0@@ -3239,6 +3239,7 @@ static int ufs_qcom_remove(struct
platform_device
*pdev)Quoted 5 lines
pm_runtime_get_sync(&(pdev)->dev); ufshcd_remove(hba); + ufshcd_dealloc_host(hba); return 0;}
Wait, why is this a platform device? Don't you hang off of a pci device? Or am I missing something earlier in this patchset?
Quoted 3 lines
thanks,
greg k-h
Hi Greg,
I am not saying someone is a platform device here. My point is whoever allocates the SCSI host in its drv->probe(), should de-allocate it in its own drv->remove(), just like what ufshcd-pci.c does.
Thanks,
Can Guo.
PATCH v3 2/2 scsi: ufs: Modularize ufs-bsg
In reply to Can GuoIn order to improve the flexibility of ufs-bsg, modularizing it is a good choice. This change introduces tristate to ufs-bsg to allow users compile it as an external module.
drivers/scsi/ufs/Kconfig +1 -1
@@ -143,7 +143,7 @@ config SCSI_UFS_TI_J721E If unsure, say N. config SCSI_UFS_BSG - bool "Universal Flash Storage BSG device node" + tristate "Universal Flash Storage BSG device node" depends on SCSI_UFSHCD select BLK_DEV_BSGLIB help
drivers/scsi/ufs/Makefile +1 -1
@@ -6,7 +6,7 @@ obj-$(CONFIG_SCSI_UFS_CDNS_PLATFORM) += cdns-pltfrm.o obj-$(CONFIG_SCSI_UFS_QCOM) += ufs-qcom.o obj-$(CONFIG_SCSI_UFSHCD) += ufshcd-core.o ufshcd-core-y += ufshcd.o ufs-sysfs.o -ufshcd-core-$(CONFIG_SCSI_UFS_BSG) += ufs_bsg.o +obj-$(CONFIG_SCSI_UFS_BSG) += ufs_bsg.o obj-$(CONFIG_SCSI_UFSHCD_PCI) += ufshcd-pci.o obj-$(CONFIG_SCSI_UFSHCD_PLATFORM) += ufshcd-pltfrm.o obj-$(CONFIG_SCSI_UFS_HISI) += ufs-hisi.o
drivers/scsi/ufs/ufs_bsg.c +36 -31
102 lines not on this page · raw message
drivers/scsi/ufs/ufs_bsg.h +0 -8
@@ -12,12 +12,4 @@#include "ufshcd.h" #include "ufs.h" -#ifdef CONFIG_SCSI_UFS_BSG -void ufs_bsg_remove (struct ufs_hba *hba); -int ufs_bsg_probe (struct ufs_hba *hba); -#else -static inline void ufs_bsg_remove (struct ufs_hba *hba) {} -static inline int ufs_bsg_probe (struct ufs_hba *hba) {return 0 ; } -#endif -#endif /* UFS_BSG_H */
drivers/scsi/ufs/ufshcd-pltfrm.c +12 -0
@@ -393,6 +393,7 @@ int ufshcd_pltfrm_init(struct platform_device *pdev,void __iomem *mmio_base;int irq, err;struct device *dev = &pdev->dev; +struct platform_device *bsg_pdev; mmio_base =devm_platform_ioremap_resource (pdev,0 );if (IS_ERR (mmio_base)) {
@@ -440,6 +441,17 @@ int ufshcd_pltfrm_init(struct platform_device *pdev,pm_runtime_set_active (&pdev->dev);pm_runtime_enable (&pdev->dev); + bsg_pdev =platform_device_register_data (dev,"ufs-bsg" , + hba->host->host_no, hba, +sizeof (struct ufs_hba )); +/* Failure here is non-fatal */ +if (IS_ERR (bsg_pdev)) { + err =PTR_ERR (bsg_pdev); +dev_warn (dev,"Register bsg platform device failed %d\n" , err); + }else { + hba->bsg_pdev = bsg_pdev; + } +return 0 ; dealloc_host:
drivers/scsi/ufs/ufshcd.c +4 -4
@@ -42,6 +42,7 @@#include <linux/nls.h> #include <linux/of.h> #include <linux/bitfield.h> +#include <linux/platform_device.h> #include "ufshcd.h" #include "ufs_quirks.h" #include "unipro.h"
@@ -2093,6 +2094,7 @@ int ufshcd_send_uic_cmd(struct ufs_hba *hba, struct uic_command *uic_cmd)ufshcd_release (hba);return ret; } +EXPORT_SYMBOL_GPL (ufshcd_send_uic_cmd);/** * ufshcd_map_sg - Map scatter-gather list to prdt
@@ -6024,6 +6026,7 @@ int ufshcd_exec_raw_upiu_cmd(struct ufs_hba *hba,return err; } +EXPORT_SYMBOL_GPL (ufshcd_exec_raw_upiu_cmd);/** * ufshcd_eh_device_reset_handler - device reset handler registered to
@@ -7043,9 +7046,6 @@ static int ufshcd_probe_hba(struct ufs_hba *hba) } hba->clk_scaling.is_allowed = true; } - -ufs_bsg_probe (hba); -scsi_scan_host (hba->host);pm_runtime_put_sync (hba->dev); }
@@ -8248,7 +8248,7 @@ int ufshcd_shutdown(struct ufs_hba *hba)*/ void ufshcd_remove (struct ufs_hba *hba) { -ufs_bsg_remove (hba); +platform_device_unregister (hba->bsg_pdev);ufs_sysfs_remove_nodes (hba->dev);scsi_remove_host (hba->host);scsi_host_put (hba->host);
drivers/scsi/ufs/ufshcd.h +1 -1
@@ -734,7 +734,7 @@ struct ufs_hba {struct ufs_desc_size desc_size;atomic_t scsi_block_reqs_cnt; -struct device bsg_dev; +struct platform_device *bsg_pdev;struct request_queue *bsg_queue; };
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project