PCI: controller: Remove duplicate error message
Subsystems (6)
PATCH 1/1 PCI: controller: Remove duplicate error message
It will print an error message by itself when devm_pci_remap_cfg_resource() goes wrong. so remove the duplicate error message.
--- drivers/pci/controller/cadence/pcie-cadence-host.c | 4 +--- drivers/pci/controller/dwc/pcie-al.c | 13 +++---------- drivers/pci/controller/dwc/pcie-armada8k.c | 1 - drivers/pci/controller/dwc/pcie-spear13xx.c | 1 - 4 files changed, 4 insertions(+), 15 deletions(-)
drivers/pci/controller/cadence/pcie-cadence-host.c +1 -3
@@ -234,10 +234,8 @@ int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"cfg" ); rc->cfg_base =devm_pci_remap_cfg_resource (dev, res); -if (IS_ERR (rc->cfg_base)) { -dev_err (dev,"missing \"cfg\"\n" ); +if (IS_ERR (rc->cfg_base))return PTR_ERR (rc->cfg_base); - } rc->cfg_res = res; res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"mem" );
drivers/pci/controller/dwc/pcie-al.c +3 -10
@@ -67,13 +67,8 @@ static int al_pcie_init(struct pci_config_window *cfg)dev_dbg (dev,"Root port dbi res: %pR\n" , res); al_pcie->dbi_base =devm_pci_remap_cfg_resource (dev, res); -if (IS_ERR (al_pcie->dbi_base)) { -long err =PTR_ERR (al_pcie->dbi_base); - -dev_err (dev,"couldn't remap dbi base %pR (err:%ld)\n" , - res, err); -return err; - } +if (IS_ERR (al_pcie->dbi_base)) +return PTR_ERR (al_pcie->dbi_base); cfg->priv = al_pcie;
@@ -408,10 +403,8 @@ static int al_pcie_probe(struct platform_device *pdev) dbi_res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"dbi" ); pci->dbi_base =devm_pci_remap_cfg_resource (dev, dbi_res); -if (IS_ERR (pci->dbi_base)) { -dev_err (dev,"couldn't remap dbi base %pR\n" , dbi_res); +if (IS_ERR (pci->dbi_base))return PTR_ERR (pci->dbi_base); - } ecam_res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"config" );if (!ecam_res) {
drivers/pci/controller/dwc/pcie-armada8k.c +0 -1
@@ -317,7 +317,6 @@ static int armada8k_pcie_probe(struct platform_device *pdev) base =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"ctrl" ); pci->dbi_base =devm_pci_remap_cfg_resource (dev, base);if (IS_ERR (pci->dbi_base)) { -dev_err (dev,"couldn't remap regs base %p\n" , base); ret =PTR_ERR (pci->dbi_base); goto fail_clkreg; }
drivers/pci/controller/dwc/pcie-spear13xx.c +0 -1
@@ -273,7 +273,6 @@ static int spear13xx_pcie_probe(struct platform_device *pdev) dbi_base =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"dbi" ); pci->dbi_base =devm_pci_remap_cfg_resource (dev, dbi_base);if (IS_ERR (pci->dbi_base)) { -dev_err (dev,"couldn't remap dbi base %p\n" , dbi_base); ret =PTR_ERR (pci->dbi_base); goto fail_clk; }
Signature · 1 line
2.25.0
Chocron, Jonathan
maintainerQuoted 10 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
It will print an error message by itself when
devm_pci_remap_cfg_resource()goes wrong. so remove the duplicate error message.
It seems like that in the first error case in devm_pci_remap_cfg_resource(), the print will be less indicative. Could you please share an example print log with the duplicate print?
Thanks,
Jonathan
Quoted 7 lines · Jonathan
Quoted 6 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
Quoted 9 lines
Quoted 4 lines
It will print an error message by itself when
devm_pci_remap_cfg_resource()goes wrong. so remove the duplicate error message.It seems like that in the first error case in
devm_pci_remap_cfg_resource(), the print will be less indicative. Could you please share an example print log with the duplicate print?
Hi Jonathan:
Thank you very much for using your precious time to review my patch.
I did not have this log and just found it by review codes. the function of devm_pci_remap_cfg_resource() is designed to handle error messages by itself. and Its recommended usage is as follows in the function description
base = devm_pci_remap_cfg_resource(&pdev->dev, res); if (IS_ERR(base)) return PTR_ERR(base);
In fact, I think its error handling is clear enough, It just goes wrong in three places, as follows:
void __iomem *devm_pci_remap_cfg_resource(struct device *dev,
struct resource *res)
{
resource_size_t size;
const char *name;
void __iomem *dest_ptr;
BUG_ON(!dev);
if (!res || resource_type(res) != IORESOURCE_MEM) {
dev_err(dev, "invalid resource\n");
return IOMEM_ERR_PTR(-EINVAL);
}
size = resource_size(res);
name = res->name ?: dev_name(dev);
if (!devm_request_mem_region(dev, res->start, size, name)) {
dev_err(dev, "can't request region for resource %pR\n", res);
return IOMEM_ERR_PTR(-EBUSY);
}
dest_ptr = devm_pci_remap_cfgspace(dev, res->start, size);
if (!dest_ptr) {
dev_err(dev, "ioremap failed for resource %pR\n", res);
devm_release_mem_region(dev, res->start, size);
dest_ptr = IOMEM_ERR_PTR(-ENOMEM);
}
return dest_ptr;
}BR,
Dejin
Quoted 2 lines
Thanks,
Jonathan
Chocron, Jonathan
maintainerQuoted 27 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
Quoted 14 lines · Jonathan
Quoted 8 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do
not click links or open attachments unless you can confirm the sender
and know the content is safe.Quoted 4 lines
It will print an error message by itself when
devm_pci_remap_cfg_resource()goes wrong. so remove the duplicate error message.Quoted 5 lines
It seems like that in the first error case in
devm_pci_remap_cfg_resource(), the print will be less indicative. Could
you please share an example print log with the duplicate print?
Quoted 4 lines
Hi Jonathan:
Thank you very much for using your precious time to review my patch.
Sure, no problem.
Quoted 11 lines
I did not have this log and just found it by review codes. the function
ofdevm_pci_remap_cfg_resource()is designed to handle error messages by
itself. and Its recommended usage is as follows in the function descriptionbase = devm_pci_remap_cfg_resource(&pdev->dev, res); if (IS_ERR(base)) return PTR_ERR(base);
I assume that the recommended usage's main intent was to point out that IS_ERR() should be used, but this is mainly speculation.
Quoted 6 lines
In fact, I think its error handling is clear enough, It just goes wrong
in three places, as follows:void
__iomem*devm_pci_remap_cfg_resource(struct device *dev,
struct resource *res)
Quoted 12 lines
{
resource_size_t size;
const char *name;
void __iomem *dest_ptr;
BUG_ON(!dev);
if (!res || resource_type(res) != IORESOURCE_MEM) {
dev_err(dev, "invalid resource\n");
return IOMEM_ERR_PTR(-EINVAL);
}In the above error case there is no indication of which resource failed (mainly relevant if the resource name is missing in the devicetree, since in the drivers you are changing platform_get_resource_byname() is mostly used). In the existing drivers' code, on return from this function in this case, the name would be printed by the caller.
Quoted 9 lines
size = resource_size(res);
name = res->name ?: dev_name(dev);
if (!devm_request_mem_region(dev, res->start, size, name)) {
dev_err(dev, "can't request region for resource
%pR\n", res);
return IOMEM_ERR_PTR(-EBUSY);
}Quoted 8 lines
dest_ptr = devm_pci_remap_cfgspace(dev, res->start, size);
if (!dest_ptr) {
dev_err(dev, "ioremap failed for resource %pR\n",
res);
devm_release_mem_region(dev, res->start, size);
dest_ptr = IOMEM_ERR_PTR(-ENOMEM);
}The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Quoted 8 lines
return dest_ptr; }BR,
DejinQuoted 2 lines
Thanks,
Jonathan
Rob Herring
reviewerQuoted 42 lines · Jonathan
Quoted 27 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
Quoted 14 lines · Jonathan
Quoted 8 lines · Dejin Zheng
CAUTION: This email originated from outside of the organization. Do
not click links or open attachments unless you can confirm the sender
and know the content is safe.Quoted 4 lines
It will print an error message by itself when
devm_pci_remap_cfg_resource()goes wrong. so remove the duplicate error message.Quoted 5 lines
It seems like that in the first error case in
devm_pci_remap_cfg_resource(), the print will be less indicative. Could
you please share an example print log with the duplicate print?Quoted 4 lines
Hi Jonathan:
Thank you very much for using your precious time to review my patch.
Sure, no problem.
Quoted 7 lines
I did not have this log and just found it by review codes. the function
ofdevm_pci_remap_cfg_resource()is designed to handle error messages by
itself. and Its recommended usage is as follows in the function description
Quoted 6 lines
Quoted 4 lines
base = devm_pci_remap_cfg_resource(&pdev->dev, res); if (IS_ERR(base)) return PTR_ERR(base);I assume that the recommended usage's main intent was to point out that
IS_ERR()should be used, but this is mainly speculation.
It's generally preferred to print error messages in a called function rather than in return error handling.
Quoted 18 lines
Quoted 6 lines
In fact, I think its error handling is clear enough, It just goes wrong
in three places, as follows:void
__iomem*devm_pci_remap_cfg_resource(struct device *dev,
struct resource *res)Quoted 12 lines
{ resource_size_t size; const char *name; void __iomem *dest_ptr; BUG_ON(!dev); if (!res || resource_type(res) != IORESOURCE_MEM) { dev_err(dev, "invalid resource\n"); return IOMEM_ERR_PTR(-EINVAL); }
Quoted 5 lines
In the above error case there is no indication of which resource failed (mainly relevant if the resource name is missing in the devicetree, since in the drivers you are changing
platform_get_resource_byname()is mostly used). In the existing drivers' code, on return from this function in this case, the name would be printed by the caller.
A driver should only have one call to devm_pci_remap_cfg_resource() as there's only 1 config space. However, it looks like this function is frequently used on what is not config space which is a bigger issue.
If this error happens, it's almost always going to be a NULL ptr as platform_get_resource_byname() would have set IORESOURCE_MEM. Perhaps a WARN here so you get a backtrace to the caller location.
Quoted 17 lines
Quoted 9 lines
size = resource_size(res); name = res->name ?: dev_name(dev); if (!devm_request_mem_region(dev, res->start, size, name)) { dev_err(dev, "can't request region for resource %pR\n", res); return IOMEM_ERR_PTR(-EBUSY); }Quoted 8 lines
dest_ptr = devm_pci_remap_cfgspace(dev, res->start, size); if (!dest_ptr) { dev_err(dev, "ioremap failed for resource %pR\n", res); devm_release_mem_region(dev, res->start, size); dest_ptr = IOMEM_ERR_PTR(-ENOMEM); }
Quoted 2 lines
The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Start/end are what are important for why either of these functions failed.
But sure, we could add 'name' here. That's a separate patch IMO.
Rob
On Tue, Jun 02, 2020 at 09:01:13AM -0600, Rob Herring wrote:
[...]
Quoted 24 lines
Quoted 18 lines
Quoted 6 lines
In fact, I think its error handling is clear enough, It just goes wrong
in three places, as follows:void
__iomem*devm_pci_remap_cfg_resource(struct device *dev,
struct resource *res)Quoted 12 lines
{ resource_size_t size; const char *name; void __iomem *dest_ptr; BUG_ON(!dev); if (!res || resource_type(res) != IORESOURCE_MEM) { dev_err(dev, "invalid resource\n"); return IOMEM_ERR_PTR(-EINVAL); }Quoted 5 lines
In the above error case there is no indication of which resource failed (mainly relevant if the resource name is missing in the devicetree, since in the drivers you are changing
platform_get_resource_byname()is mostly used). In the existing drivers' code, on return from this function in this case, the name would be printed by the caller.
Quoted 3 lines
A driver should only have one call to
devm_pci_remap_cfg_resource()as there's only 1 config space. However, it looks like this function is frequently used on what is not config space which is a bigger issue.
That certainly is and should be fixed.
Quoted 3 lines
If this error happens, it's almost always going to be a NULL ptr as
platform_get_resource_byname()would have setIORESOURCE_MEM. Perhaps a WARN here so you get a backtrace to the caller location.
+1
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
size = resource_size(res); name = res->name ?: dev_name(dev); if (!devm_request_mem_region(dev, res->start, size, name)) { dev_err(dev, "can't request region for resource %pR\n", res); return IOMEM_ERR_PTR(-EBUSY); } dest_ptr = devm_pci_remap_cfgspace(dev, res->start, size); if (!dest_ptr) { dev_err(dev, "ioremap failed for resource %pR\n", res); devm_release_mem_region(dev, res->start, size); dest_ptr = IOMEM_ERR_PTR(-ENOMEM); }
Quoted 7 lines
Quoted 2 lines
The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Start/end are what are important for why either of these functions failed.
But sure, we could add 'name' here. That's a separate patch IMO.
I agree. In sum, I think it is OK to proceed with this patch, provided we send follow-ups as discussed here, are we in agreement ?
Lorenzo
Quoted 11 lines · Lorenzo Pieralisi
On Tue, Jun 02, 2020 at 09:01:13AM -0600, Rob Herring wrote:
[...]
Quoted 7 lines
Quoted 2 lines
The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Start/end are what are important for why either of these functions failed.
But sure, we could add 'name' here. That's a separate patch IMO.
Hi Lorenzo, Bob and Jonathan:
Thank you very much for helping me review this patch, I sent a new patch for print the resource name when the request memory region or remapping of configuration space fails. and it is here:
https://patchwork.kernel.org/patch/11657801/
BR,
Dejin
Quoted 5 lines
I agree. In sum, I think it is OK to proceed with this patch, provided we send follow-ups as discussed here, are we in agreement ?
Lorenzo
Quoted 8 lines · Dejin Zheng
Quoted 7 lines · Lorenzo Pieralisi
On Tue, Jun 02, 2020 at 09:01:13AM -0600, Rob Herring wrote:
[...]
Quoted 3 lines
Quoted 2 lines
The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Start/end are what are important for why either of these functions failed.
But sure, we could add 'name' here. That's a separate patch IMO.
Hi Lorenzo, Bob and Jonathan:
Thank you very much for helping me review this patch,
I merge this patch in pci/misc, thanks.
Quoted 3 lines
I sent a new patch for print the resource name when the request memory region or remapping of configuration space fails. and it is here:
https://patchwork.kernel.org/patch/11657801/
We will get to it soon.
Lorenzo
Quoted 20 lines · Lorenzo Pieralisi
Quoted 8 lines · Dejin Zheng
Quoted 7 lines · Lorenzo Pieralisi
On Tue, Jun 02, 2020 at 09:01:13AM -0600, Rob Herring wrote:
[...]
Quoted 3 lines
Quoted 2 lines
The other 2 error cases as well don't print the resource name as far as I recall (they will at least print the resource start/end).
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Start/end are what are important for why either of these functions failed.
But sure, we could add 'name' here. That's a separate patch IMO.
Hi Lorenzo, Bob and Jonathan:
Thank you very much for helping me review this patch,
I merge this patch in pci/misc, thanks.
Quoted 7 lines
Quoted 3 lines
I sent a new patch for print the resource name when the request memory region or remapping of configuration space fails. and it is here:
https://patchwork.kernel.org/patch/11657801/We will get to it soon.
Lorenzo
Lorenzo, Thanks very much!
Dejin
Quoted 4 lines · Dejin Zheng
It will print an error message by itself when
devm_pci_remap_cfg_resource()goes wrong. so remove the duplicate error message.
Quoted 8 lines
Signed-off-by Dejin Zheng <[email protected]>--- drivers/pci/controller/cadence/pcie-cadence-host.c | 4 +--- drivers/pci/controller/dwc/pcie-al.c | 13 +++---------- drivers/pci/controller/dwc/pcie-armada8k.c | 1 - drivers/pci/controller/dwc/pcie-spear13xx.c | 1 - 4 files changed, 4 insertions(+), 15 deletions(-)