PCI: tegra: A couple of cleanups
Subsystems (4)
cover PCI: tegra: A couple of cleanups
Hi All,
This small series provides two cleanup patches for the Tegra PCIe driver. The overall goal is to replace custom, open-coded logic with standard kernel helper functions.
These changes improve the driver's readability and maintainability by everaging modern, well-tested APIs for clock management and register polling.
Thanks
-Anand
Anand Moon (2):
PCI: tegra: Simplify clock handling by using clk_bulk*() functions
PCI: tegra: Use readl_poll_timeout() for link status polling
drivers/pci/controller/pci-tegra.c | 109 +++++++---------------------- 1 file changed, 27 insertions(+), 82 deletions(-)
base-commit: 5c3b3264e5858813632031ba58bcd6e1eeb3b214
Signature · 1 line
2.50.1
RFC 1/2 PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
- devm_clk_bulk_get_all() API to acquire all the clocks
- clk_bulk_prepare_enable() to prepare/enable clocks
- clk_bulk_disable_unprepare() APIs to disable/unprepare them in bulk
As part of this cleanup, the legacy has_cml_clk flag and explicit handling of individual clocks (pex, afi, pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.
--- drivers/pci/controller/pci-tegra.c | 71 +++++------------------------- 1 file changed, 12 insertions(+), 59 deletions(-)
drivers/pci/controller/pci-tegra.c +12 -59
@@ -297,7 +297,6 @@ struct tegra_pcie_soc {bool has_pex_clkreq_en;bool has_pex_bias_ctrl;bool has_intr_prsnt_sense; -bool has_cml_clk;bool has_gen2;bool force_pca_enable;bool program_uphy;
@@ -330,10 +329,8 @@ struct tegra_pcie {struct resource cs; -struct clk *pex_clk; -struct clk *afi_clk; -struct clk *pll_e; -struct clk *cml_clk; +struct clk_bulk_data *clks; +int num_clks;struct reset_control *pex_rst;struct reset_control *afi_rst;
@@ -1153,15 +1150,11 @@ static void tegra_pcie_enable_controller(struct tegra_pcie *pcie)static void tegra_pcie_power_off (struct tegra_pcie *pcie) {struct device *dev = pcie->dev; -const struct tegra_pcie_soc *soc = pcie->soc;int err;reset_control_assert (pcie->afi_rst); -clk_disable_unprepare (pcie->pll_e); -if (soc->has_cml_clk) -clk_disable_unprepare (pcie->cml_clk); -clk_disable_unprepare (pcie->afi_clk); +clk_bulk_disable_unprepare (pcie->num_clks, pcie->clks);if (!dev->pm_domain)tegra_powergate_power_off (TEGRA_POWERGATE_PCIE );
@@ -1174,7 +1167,6 @@ static void tegra_pcie_power_off(struct tegra_pcie *pcie)static int tegra_pcie_power_on (struct tegra_pcie *pcie) {struct device *dev = pcie->dev; -const struct tegra_pcie_soc *soc = pcie->soc;int err;reset_control_assert (pcie->pcie_xrst);
@@ -1202,35 +1194,16 @@ static int tegra_pcie_power_on(struct tegra_pcie *pcie) } } - err =clk_prepare_enable (pcie->afi_clk); + err =clk_bulk_prepare_enable (pcie->num_clks, pcie->clks);if (err <0 ) { -dev_err (dev,"failed to enable AFI clock: %d\n" , err); +dev_err (dev,"filed to enable clocks: %d\n" , err); goto powergate; } -if (soc->has_cml_clk) { - err =clk_prepare_enable (pcie->cml_clk); -if (err <0 ) { -dev_err (dev,"failed to enable CML clock: %d\n" , err); - goto disable_afi_clk; - } - } - - err =clk_prepare_enable (pcie->pll_e); -if (err <0 ) { -dev_err (dev,"failed to enable PLLE clock: %d\n" , err); - goto disable_cml_clk; - } -reset_control_deassert (pcie->afi_rst);return 0 ; -disable_cml_clk: -if (soc->has_cml_clk) -clk_disable_unprepare (pcie->cml_clk); -disable_afi_clk: -clk_disable_unprepare (pcie->afi_clk); powergate:if (!dev->pm_domain)tegra_powergate_power_off (TEGRA_POWERGATE_PCIE );
@@ -1254,25 +1227,11 @@ static void tegra_pcie_apply_pad_settings(struct tegra_pcie *pcie)static int tegra_pcie_clocks_get (struct tegra_pcie *pcie) {struct device *dev = pcie->dev; -const struct tegra_pcie_soc *soc = pcie->soc; - - pcie->pex_clk =devm_clk_get (dev,"pex" ); -if (IS_ERR (pcie->pex_clk)) -return PTR_ERR (pcie->pex_clk); - pcie->afi_clk =devm_clk_get (dev,"afi" ); -if (IS_ERR (pcie->afi_clk)) -return PTR_ERR (pcie->afi_clk); - - pcie->pll_e =devm_clk_get (dev,"pll_e" ); -if (IS_ERR (pcie->pll_e)) -return PTR_ERR (pcie->pll_e); - -if (soc->has_cml_clk) { - pcie->cml_clk =devm_clk_get (dev,"cml" ); -if (IS_ERR (pcie->cml_clk)) -return PTR_ERR (pcie->cml_clk); - } + pcie->num_clks =devm_clk_bulk_get_all (dev, &pcie->clks); +if (pcie->num_clks <0 ) +return dev_err_probe (dev, pcie->num_clks, +"failed to get clocks\n" );return 0 ; }
@@ -2345,7 +2304,6 @@ static const struct tegra_pcie_soc tegra20_pcie = { .has_pex_clkreq_en = false, .has_pex_bias_ctrl = false, .has_intr_prsnt_sense = false, - .has_cml_clk = false, .has_gen2 = false, .force_pca_enable = false, .program_uphy = true,
@@ -2374,7 +2332,6 @@ static const struct tegra_pcie_soc tegra30_pcie = { .has_pex_clkreq_en = true, .has_pex_bias_ctrl = true, .has_intr_prsnt_sense = true, - .has_cml_clk = true, .has_gen2 = false, .force_pca_enable = false, .program_uphy = true,
@@ -2395,7 +2352,6 @@ static const struct tegra_pcie_soc tegra124_pcie = { .has_pex_clkreq_en = true, .has_pex_bias_ctrl = true, .has_intr_prsnt_sense = true, - .has_cml_clk = true, .has_gen2 = true, .force_pca_enable = false, .program_uphy = true,
@@ -2418,7 +2374,6 @@ static const struct tegra_pcie_soc tegra210_pcie = { .has_pex_clkreq_en = true, .has_pex_bias_ctrl = true, .has_intr_prsnt_sense = true, - .has_cml_clk = true, .has_gen2 = true, .force_pca_enable = true, .program_uphy = true,
@@ -2459,7 +2414,6 @@ static const struct tegra_pcie_soc tegra186_pcie = { .has_pex_clkreq_en = true, .has_pex_bias_ctrl = true, .has_intr_prsnt_sense = true, - .has_cml_clk = false, .has_gen2 = true, .force_pca_enable = false, .program_uphy = false,
@@ -2672,7 +2626,7 @@ static int tegra_pcie_pm_suspend(struct device *dev) }reset_control_assert (pcie->pex_rst); -clk_disable_unprepare (pcie->pex_clk); +clk_bulk_disable_unprepare (pcie->num_clks, pcie->clks);if (IS_ENABLED (CONFIG_PCI_MSI ))tegra_pcie_disable_msi (pcie);
@@ -2706,9 +2660,9 @@ static int tegra_pcie_pm_resume(struct device *dev)if (IS_ENABLED (CONFIG_PCI_MSI ))tegra_pcie_enable_msi (pcie); - err =clk_prepare_enable (pcie->pex_clk); + err =clk_bulk_prepare_enable (pcie->num_clks, pcie->clks);if (err) { -dev_err (dev,"failed to enable PEX clock: %d\n" , err); +dev_err (dev,"failed to enable clock: %d\n" , err); goto pex_dpd_enable; }
@@ -2729,7 +2683,6 @@ static int tegra_pcie_pm_resume(struct device *dev) disable_pex_clk:reset_control_assert (pcie->pex_rst); -clk_disable_unprepare (pcie->pex_clk); pex_dpd_enable:pinctrl_pm_select_idle_state (dev); poweroff:
Signature · 1 line
2.50.1
Jon Hunter
maintainerRe: [RFC v1 1/2] PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Quoted 10 lines · Anand Moon
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
-devm_clk_bulk_get_all()API to acquire all the clocks
-clk_bulk_prepare_enable()to prepare/enable clocks
-clk_bulk_disable_unprepare()APIs to disable/unprepare them in bulk
Quoted 4 lines
As part of this cleanup, the legacy
has_cml_clkflag and explicit handling of individual clocks (pex, afi,pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.
What platforms have you tested this change on?
Thanks
Jon
Signature · 1 line
nvpublic
Re: [RFC v1 1/2] PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Hi Jon,
Quoted 13 lines · Jon Hunter
Quoted 10 lines · Anand Moon
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
-devm_clk_bulk_get_all()API to acquire all the clocks
-clk_bulk_prepare_enable()to prepare/enable clocks
-clk_bulk_disable_unprepare()APIs to disable/unprepare them in bulk
Quoted 7 lines
Quoted 4 lines
As part of this cleanup, the legacy
has_cml_clkflag and explicit handling of individual clocks (pex, afi,pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.What platforms have you tested this change on?
I'm using a Jetson Nano 4GB model as my test platform.
Quoted 2 lines
Thanks
Jon
Thanks
-Anand
Jon Hunter
maintainerRe: [RFC v1 1/2] PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Quoted 16 lines · Anand Moon
Hi Jon,
Quoted 13 lines · Jon Hunter
Quoted 10 lines · Anand Moon
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
-devm_clk_bulk_get_all()API to acquire all the clocks
-clk_bulk_prepare_enable()to prepare/enable clocks
-clk_bulk_disable_unprepare()APIs to disable/unprepare them in bulk
Quoted 8 lines
Quoted 7 lines
Quoted 4 lines
As part of this cleanup, the legacy
has_cml_clkflag and explicit handling of individual clocks (pex, afi,pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.What platforms have you tested this change on?
I'm using a Jetson Nano 4GB model as my test platform.
Thanks. One concern I have about this is that right now the DT binding doc for this device is still in the legacy text format and not converted to yaml. Therefore, there is no way to validate the device-tree bindings for this driver. So by making this change we are susceptible to people getting the device-tree incorrect and there is no way to check. Right now the driver will fail is a given clock is missing but after this change we are completely reliant that the device-tree is correct but no way to validate.
It would be great to convert the text binding doc to yaml as part of this series.
Also if you look at the dwmac-tegra.c driver this one still populates the clock names when using the bulk APIs so that we know that the clocks that we require are present.
Jon
[0] drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c
Signature · 1 line
nvpublic
Re: [RFC v1 1/2] PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Hi Jon,
Quoted 37 lines · Jon Hunter
Quoted 16 lines · Anand Moon
Hi Jon,
Quoted 13 lines · Jon Hunter
Quoted 10 lines · Anand Moon
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
-devm_clk_bulk_get_all()API to acquire all the clocks
-clk_bulk_prepare_enable()to prepare/enable clocks
-clk_bulk_disable_unprepare()APIs to disable/unprepare them in bulkQuoted 8 lines
Quoted 7 lines
Quoted 4 lines
As part of this cleanup, the legacy
has_cml_clkflag and explicit handling of individual clocks (pex, afi,pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.What platforms have you tested this change on?
I'm using a Jetson Nano 4GB model as my test platform.
Thanks. One concern I have about this is that right now the DT binding doc for this device is still in the legacy text format and not converted to yaml. Therefore, there is no way to validate the device-tree bindings for this driver. So by making this change we are susceptible to people getting the device-tree incorrect and there is no way to check. Right now the driver will fail is a given clock is missing but after this change we are completely reliant that the device-tree is correct but no way to validate.
Quoted 3 lines
It would be great to convert the text binding doc to yaml as part of this series.
I will convert the legacy text binding to a YAML file as part of this series.
[0] Documentation/devicetree/bindings/pci/nvidia,tegra20-pcie.txt
Quoted 4 lines
Also if you look at the
dwmac-tegra.cdriver this one still populates the clock names when using the bulk APIs so that we know that the clocks that we require are present.
Only the Tegra20 SoC has three clocks.
compatible = "nvidia,tegra20-pcie";
clocks = <&tegra_car TEGRA20_CLK_PEX>,
<&tegra_car TEGRA20_CLK_AFI>,
<&tegra_car TEGRA20_CLK_PLL_E>;
clock-names = "pex", "afi", "pll_e";Whereas all the rest of the SoCs have 4 clocks.
compatible = "nvidia,tegra30-pcie"; compatible = "nvidia,tegra124-pcie"; compatible = "nvidia,tegra210-pcie"; compatible = "nvidia,tegra186-pcie";
clocks = <&tegra_car TEGRA30_CLK_PCIE>,
<&tegra_car TEGRA30_CLK_AFI>,
<&tegra_car TEGRA30_CLK_PLL_E>,
<&tegra_car TEGRA30_CLK_CML0>;
clock-names = "pex", "afi", "pll_e", "cml";As suggested, I need to create two clock arrays for the clocks of the SoC.
But the code will introduce more overhead:
bulk clks -> devm_kcalloc (for clocks) -> assign id to clocks -> devm_clk_bulk_get -> clk_bulk_prepare_enable.
I believe the use of devm_clk_bulk_get_all() is a cleaner and more modern approach for the following reasons:
It simplifies the code by removing the need for manual memory allocation (devm_kcalloc) and populating an array of clock specifications. It is more efficient, as all clocks are fetched in a single API call, reducing overhead.
Please let me know if this plan addresses your concerns.
Quoted 7 lines
Jon
[0]
drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c--
nvpublic
Thanks
-Anand
Jon Hunter
maintainerRe: [RFC v1 1/2] PCI: tegra: Simplify clock handling by using clk_bulk*() functions
Quoted 40 lines · Anand Moon
Hi Jon,
Quoted 37 lines · Jon Hunter
Quoted 16 lines · Anand Moon
Hi Jon,
Quoted 13 lines · Jon Hunter
Quoted 10 lines · Anand Moon
Currently, the driver acquires clocks and prepare/enable/disable/unprepare the clocks individually thereby making the driver complex to read.
The driver can be simplified by using the clk_bulk*() APIs.
Use:
-devm_clk_bulk_get_all()API to acquire all the clocks
-clk_bulk_prepare_enable()to prepare/enable clocks
-clk_bulk_disable_unprepare()APIs to disable/unprepare them in bulkQuoted 8 lines
Quoted 7 lines
Quoted 4 lines
As part of this cleanup, the legacy
has_cml_clkflag and explicit handling of individual clocks (pex, afi,pll_e, cml) are removed. Clock sequencing is now implicitly determined by the order defined in the device tree, eliminating hardcoded logic and improving maintainability.What platforms have you tested this change on?
I'm using a Jetson Nano 4GB model as my test platform.
Thanks. One concern I have about this is that right now the DT binding doc for this device is still in the legacy text format and not converted to yaml. Therefore, there is no way to validate the device-tree bindings for this driver. So by making this change we are susceptible to people getting the device-tree incorrect and there is no way to check. Right now the driver will fail is a given clock is missing but after this change we are completely reliant that the device-tree is correct but no way to validate.
Quoted 6 lines
Quoted 3 lines
It would be great to convert the text binding doc to yaml as part of this series.
I will convert the legacy text binding to a YAML file as part of this series.
[0]
Documentation/devicetree/bindings/pci/nvidia,tegra20-pcie.txt
Thanks!
Quoted 31 lines
Quoted 4 lines
Also if you look at the
dwmac-tegra.cdriver this one still populates the clock names when using the bulk APIs so that we know that the clocks that we require are present.Only the Tegra20 SoC has three clocks.
compatible = "nvidia,tegra20-pcie"; clocks = <&tegra_car TEGRA20_CLK_PEX>, <&tegra_car TEGRA20_CLK_AFI>, <&tegra_car TEGRA20_CLK_PLL_E>; clock-names = "pex", "afi", "pll_e";Whereas all the rest of the SoCs have 4 clocks.
compatible = "nvidia,tegra30-pcie"; compatible = "nvidia,tegra124-pcie"; compatible = "nvidia,tegra210-pcie"; compatible = "nvidia,tegra186-pcie";clocks = <&tegra_car TEGRA30_CLK_PCIE>, <&tegra_car TEGRA30_CLK_AFI>, <&tegra_car TEGRA30_CLK_PLL_E>, <&tegra_car TEGRA30_CLK_CML0>; clock-names = "pex", "afi", "pll_e", "cml";As suggested, I need to create two clock arrays for the clocks of the SoC.
But the code will introduce more overhead:
bulk clks ->
devm_kcalloc(for clocks) -> assign id to clocks ->devm_clk_bulk_get->clk_bulk_prepare_enable.
Quoted 6 lines
I believe the use of
devm_clk_bulk_get_all()is a cleaner and more modern approach for the following reasons:
It simplifies the code by removing the need for manual memory allocation (devm_kcalloc) and populating an array of clock specifications. It is more efficient, as all clocks are fetched in a single API call, reducing overhead.
Yes that's correct and I don't think it is that much overhead. The clk names array can just be part of the 'tegra_pcie_soc' structure.
Jon
Signature · 1 line
nvpublic
RFC 2/2 PCI: tegra: Use readl_poll_timeout() for link status polling
In reply to Anand MoonReplace the manual do-while polling loops with the readl_poll_timeout() helper when checking the link DL_UP and DL_LINK_ACTIVE status bits during link bring-up. This simplifies the code by removing the open-coded timeout logic in favor of the standard, more robust iopoll framework. The change improves readability and reduces code duplication.
--- drivers/pci/controller/pci-tegra.c | 38 ++++++++++++------------------ 1 file changed, 15 insertions(+), 23 deletions(-)
drivers/pci/controller/pci-tegra.c +15 -23
@@ -24,6 +24,7 @@#include <linux/irqchip/chained_irq.h> #include <linux/irqchip/irq-msi-lib.h> #include <linux/irqdomain.h> +#include <linux/iopoll.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h>
@@ -2157,37 +2158,28 @@ static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) value |=RP_PRIV_MISC_PRSNT_MAP_EP_PRSNT ;writel (value, port->base +RP_PRIV_MISC ); -do { -unsigned int timeout =TEGRA_PCIE_LINKUP_TIMEOUT ; - -do { - value =readl (port->base +RP_VEND_XP ); - -if (value &RP_VEND_XP_DL_UP ) -break ; - -usleep_range (1000 ,2000 ); - }while (--timeout); +while (retries--) { +int err; -if (!timeout) { + err =readl_poll_timeout (port->base +RP_VEND_XP , value, + value &RP_VEND_XP_DL_UP , +1000 , +TEGRA_PCIE_LINKUP_TIMEOUT *1000 ); +if (err) {dev_dbg (dev,"link %u down, retrying\n" , port->index); goto retry; } - timeout =TEGRA_PCIE_LINKUP_TIMEOUT ; - -do { - value =readl (port->base +RP_LINK_CONTROL_STATUS ); - -if (value &RP_LINK_CONTROL_STATUS_DL_LINK_ACTIVE ) -return true; - -usleep_range (1000 ,2000 ); - }while (--timeout); + err =readl_poll_timeout (port->base +RP_LINK_CONTROL_STATUS , + value, + value &RP_LINK_CONTROL_STATUS_DL_LINK_ACTIVE , +1000 ,TEGRA_PCIE_LINKUP_TIMEOUT *1000 ); +if (!err) +return true; retry:tegra_pcie_port_reset (port); - }while (--retries); + }return false; }
Signature · 1 line
2.50.1
Re: [RFC v1 2/2] PCI: tegra: Use readl_poll_timeout() for link status polling
Quoted 17 lines · Anand Moon
Replace the manual
do-whilepolling loops with thereadl_poll_timeout()helper when checking the linkDL_UPandDL_LINK_ACTIVEstatus bits during link bring-up. This simplifies the code by removing the open-coded timeout logic in favor of the standard, more robust iopoll framework. The change improves readability and reduces code duplication.Cc Thierry Reding <[email protected]>Signed-off-by Anand Moon <[email protected]>--- drivers/pci/controller/pci-tegra.c | 38 ++++++++++++------------------ 1 file changed, 15 insertions(+), 23 deletions(-)drivers/pci/controller/pci-tegra.c +0
-0
drivers/pci/controller/pci-tegra.c line 27
Quoted 4 lines
drivers/pci/controller/pci-tegra.c +1
-0@@ -24,6 +24,7 @@#include <linux/irqchip/chained_irq.h> #include <linux/irqchip/irq-msi-lib.h> #include <linux/irqdomain.h> +#include <linux/iopoll.h>
There is already an iopoll.h include in this file, so this adds a duplicate.
Quoted 16 lines
drivers/pci/controller/pci-tegra.c +0
-9#include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> @@ -2157,37 +2158,28 @@ static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) value |=RP_PRIV_MISC_PRSNT_MAP_EP_PRSNT ;writel (value, port->base +RP_PRIV_MISC ); - do { - unsigned int timeout = TEGRA_PCIE_LINKUP_TIMEOUT; - - do { - value = readl(port->base + RP_VEND_XP); - - if (value & RP_VEND_XP_DL_UP) - break; -
drivers/pci/controller/pci-tegra.c static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) · line 2167
Quoted 10 lines
drivers/pci/controller/pci-tegra.c +6
-3-usleep_range (1000 ,2000 ); - }while (--timeout); +while (retries--) { +int err; -if (!timeout) { + err =readl_poll_timeout (port->base +RP_VEND_XP , value, + value &RP_VEND_XP_DL_UP , +1000 , +TEGRA_PCIE_LINKUP_TIMEOUT *1000 );
The logic change here looks OK to me. This makes the timeout 200ms (TEGRA_PCIE_LINKUP_TIMEOUT is 200). Previously, the code looped 200 times with a 1 to 2ms sleep on each iteration. So the timeout could have been longer than 200ms previously, but not in a way that could be relied on.
Quoted 30 lines
drivers/pci/controller/pci-tegra.c +8
-11+if (err) {dev_dbg (dev,"link %u down, retrying\n" , port->index); goto retry; } - timeout =TEGRA_PCIE_LINKUP_TIMEOUT ; - -do { - value =readl (port->base +RP_LINK_CONTROL_STATUS ); - -if (value &RP_LINK_CONTROL_STATUS_DL_LINK_ACTIVE ) -return true; - -usleep_range (1000 ,2000 ); - }while (--timeout); + err =readl_poll_timeout (port->base +RP_LINK_CONTROL_STATUS , + value, + value &RP_LINK_CONTROL_STATUS_DL_LINK_ACTIVE , +1000 ,TEGRA_PCIE_LINKUP_TIMEOUT *1000 ); +if (!err) +return true; retry:tegra_pcie_port_reset (port); - }while (--retries); + }return false; }
Re: [RFC v1 2/2] PCI: tegra: Use readl_poll_timeout() for link status polling
Hi Mikko,
Thanks for your review comments.
Quoted 14 lines · Mikko Perttunen
Quoted 12 lines · Anand Moon
Replace the manual
do-whilepolling loops with thereadl_poll_timeout()helper when checking the linkDL_UPandDL_LINK_ACTIVEstatus bits during link bring-up. This simplifies the code by removing the open-coded timeout logic in favor of the standard, more robust iopoll framework. The change improves readability and reduces code duplication.Cc Thierry Reding <[email protected]>Signed-off-by Anand Moon <[email protected]>--- drivers/pci/controller/pci-tegra.c | 38 ++++++++++++------------------ 1 file changed, 15 insertions(+), 23 deletions(-)
Quoted 12 lines
drivers/pci/controller/pci-tegra.c line 27
Quoted 9 lines
drivers/pci/controller/pci-tegra.c +1
-0@@ -24,6 +24,7 @@#include <linux/irqchip/chained_irq.h> #include <linux/irqchip/irq-msi-lib.h> #include <linux/irqdomain.h> +#include <linux/iopoll.h> There is already an
iopoll.hinclude in this file, so this adds a duplicate.
Opps, I missed this in rebasing my code.
Quoted 21 lines
Quoted 21 lines
drivers/pci/controller/pci-tegra.c +2
-11#include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> @@ -2157,37 +2158,28 @@ static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) value |=RP_PRIV_MISC_PRSNT_MAP_EP_PRSNT ;writel (value, port->base +RP_PRIV_MISC ); - do { - unsigned int timeout = TEGRA_PCIE_LINKUP_TIMEOUT; - - do { - value = readl(port->base + RP_VEND_XP); - - if (value & RP_VEND_XP_DL_UP) - break; - - usleep_range(1000, 2000); - } while (--timeout); +while (retries--) { +int err;
Quoted 7 lines
drivers/pci/controller/pci-tegra.c static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) · line 2167
Quoted 5 lines
drivers/pci/controller/pci-tegra.c +4
-1-if (!timeout) { + err =readl_poll_timeout (port->base +RP_VEND_XP , value, + value &RP_VEND_XP_DL_UP , +1000 , +TEGRA_PCIE_LINKUP_TIMEOUT *1000 );The logic change here looks OK to me. This makes the timeout 200ms (
TEGRA_PCIE_LINKUP_TIMEOUTis 200). Previously, the code looped 200 times with a 1 to 2ms sleep on each iteration. So the timeout could have been longer than 200ms previously, but not in a way that could be relied on.
You're right; the original usleep_range(1000, 2000) had a variable sleep time. To replicate the worst-case behavior of the old loop, the readl_poll_timeout should
use a delay_us of 1000 and a timeout_us that matches the original maximum duration.
Since the previous code looped 200 times with a maximum 2ms sleep, the correct timeout is 400ms, so update (TEGRA_PCIE_LINKUP_TIMEOUT * 2000). or increase TEGRA_PCIE_LINKUP_TIMEOUT to 400.
Are these changes ok with you?
Thank
-Anand
Re: [RFC v1 2/2] PCI: tegra: Use readl_poll_timeout() for link status polling
Quoted 71 lines · Anand Moon
Hi Mikko,
Thanks for your review comments.
Quoted 14 lines · Mikko Perttunen
Quoted 12 lines · Anand Moon
Replace the manual
do-whilepolling loops with thereadl_poll_timeout()helper when checking the linkDL_UPandDL_LINK_ACTIVEstatus bits during link bring-up. This simplifies the code by removing the open-coded timeout logic in favor of the standard, more robust iopoll framework. The change improves readability and reduces code duplication.Cc Thierry Reding <[email protected]>Signed-off-by Anand Moon <[email protected]>--- drivers/pci/controller/pci-tegra.c | 38 ++++++++++++------------------ 1 file changed, 15 insertions(+), 23 deletions(-)Quoted 12 lines
drivers/pci/controller/pci-tegra.c line 27
Quoted 9 lines
drivers/pci/controller/pci-tegra.c +1
-0@@ -24,6 +24,7 @@#include <linux/irqchip/chained_irq.h> #include <linux/irqchip/irq-msi-lib.h> #include <linux/irqdomain.h> +#include <linux/iopoll.h> There is already an
iopoll.hinclude in this file, so this adds a duplicate.Opps, I missed this in rebasing my code.
Quoted 21 lines
Quoted 21 lines
drivers/pci/controller/pci-tegra.c +2
-11#include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> @@ -2157,37 +2158,28 @@ static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) value |=RP_PRIV_MISC_PRSNT_MAP_EP_PRSNT ;writel (value, port->base +RP_PRIV_MISC ); - do { - unsigned int timeout = TEGRA_PCIE_LINKUP_TIMEOUT; - - do { - value = readl(port->base + RP_VEND_XP); - - if (value & RP_VEND_XP_DL_UP) - break; - - usleep_range(1000, 2000); - } while (--timeout); +while (retries--) { +int err;Quoted 7 lines
drivers/pci/controller/pci-tegra.c static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) · line 2167
Quoted 5 lines
drivers/pci/controller/pci-tegra.c +4
-1-if (!timeout) { + err =readl_poll_timeout (port->base +RP_VEND_XP , value, + value &RP_VEND_XP_DL_UP , +1000 , +TEGRA_PCIE_LINKUP_TIMEOUT *1000 );The logic change here looks OK to me. This makes the timeout 200ms (
TEGRA_PCIE_LINKUP_TIMEOUTis 200). Previously, the code looped 200 times with a 1 to 2ms sleep on each iteration. So the timeout could have been longer than 200ms previously, but not in a way that could be relied on.You're right; the original
usleep_range(1000, 2000)had a variable sleep time. To replicate the worst-case behavior of the old loop, thereadl_poll_timeoutshould
use adelay_usof 1000 and atimeout_usthat matches the original maximum duration.
Since the previous code looped 200 times with a maximum 2ms sleep, the correct timeout is 400ms, so update (TEGRA_PCIE_LINKUP_TIMEOUT* 2000). or increaseTEGRA_PCIE_LINKUP_TIMEOUTto 400.
Quoted 1 line
Are these changes ok with you?
I think the code is fine as is. Before, the shortest the timeout could be was 200ms, i.e. there should be no situation where we need a timeout longer than that, or otherwise that would fail randomly depending on the sleep duration. So I think the 200ms is correct here and the only change necessary is the removal of the second iopoll.h
Cheers,
Mikko
Quoted 3 lines
Thank
-Anand
Re: [RFC v1 2/2] PCI: tegra: Use readl_poll_timeout() for link status polling
Hi Mikko,
Quoted 73 lines · Mikko Perttunen
Quoted 71 lines · Anand Moon
Hi Mikko,
Thanks for your review comments.
Quoted 14 lines · Mikko Perttunen
Quoted 12 lines · Anand Moon
Replace the manual
do-whilepolling loops with thereadl_poll_timeout()helper when checking the linkDL_UPandDL_LINK_ACTIVEstatus bits during link bring-up. This simplifies the code by removing the open-coded timeout logic in favor of the standard, more robust iopoll framework. The change improves readability and reduces code duplication.Cc Thierry Reding <[email protected]>Signed-off-by Anand Moon <[email protected]>--- drivers/pci/controller/pci-tegra.c | 38 ++++++++++++------------------ 1 file changed, 15 insertions(+), 23 deletions(-)Quoted 12 lines
drivers/pci/controller/pci-tegra.c line 27
Quoted 9 lines
drivers/pci/controller/pci-tegra.c +1
-0@@ -24,6 +24,7 @@#include <linux/irqchip/chained_irq.h> #include <linux/irqchip/irq-msi-lib.h> #include <linux/irqdomain.h> +#include <linux/iopoll.h> There is already an
iopoll.hinclude in this file, so this adds a duplicate.Opps, I missed this in rebasing my code.
Quoted 21 lines
Quoted 21 lines
drivers/pci/controller/pci-tegra.c +2
-11#include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> @@ -2157,37 +2158,28 @@ static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) value |=RP_PRIV_MISC_PRSNT_MAP_EP_PRSNT ;writel (value, port->base +RP_PRIV_MISC ); - do { - unsigned int timeout = TEGRA_PCIE_LINKUP_TIMEOUT; - - do { - value = readl(port->base + RP_VEND_XP); - - if (value & RP_VEND_XP_DL_UP) - break; - - usleep_range(1000, 2000); - } while (--timeout); +while (retries--) { +int err;Quoted 7 lines
drivers/pci/controller/pci-tegra.c static bool tegra_pcie_port_check_link(struct tegra_pcie_port *port) · line 2167
Quoted 5 lines
drivers/pci/controller/pci-tegra.c +4
-1-if (!timeout) { + err =readl_poll_timeout (port->base +RP_VEND_XP , value, + value &RP_VEND_XP_DL_UP , +1000 , +TEGRA_PCIE_LINKUP_TIMEOUT *1000 );The logic change here looks OK to me. This makes the timeout 200ms (
TEGRA_PCIE_LINKUP_TIMEOUTis 200). Previously, the code looped 200 times with a 1 to 2ms sleep on each iteration. So the timeout could have been longer than 200ms previously, but not in a way that could be relied on.You're right; the original
usleep_range(1000, 2000)had a variable sleep time. To replicate the worst-case behavior of the old loop, thereadl_poll_timeoutshould
use adelay_usof 1000 and atimeout_usthat matches the original maximum duration.
Since the previous code looped 200 times with a maximum 2ms sleep, the correct timeout is 400ms, so update (TEGRA_PCIE_LINKUP_TIMEOUT* 2000). or increaseTEGRA_PCIE_LINKUP_TIMEOUTto 400.
Quoted 4 lines
Quoted 1 line
Are these changes ok with you?
I think the code is fine as is. Before, the shortest the timeout could be was 200ms, i.e. there should be no situation where we need a timeout longer than that, or otherwise that would fail randomly depending on the sleep duration. So I think the 200ms is correct here and the only change necessary is the removal of the second
iopoll.h
Thanks for your input. I'll remove the header in the next version.
Quoted 3 lines
Cheers,
Mikko
Thanks
-Anand