dm: core: Move "/chosen" and "/firmware" node scan
PATCH v2 1/1 dm: core: Move "/chosen" and "/firmware" node scan
Use the new function dm_scan_fdt_ofnode_path() to scan all the nodes which aren't devices themselves but may contain some:
- "/chosen"
- "/clocks"
- "/firmware"
The patch removes the strcmp call in recursive function dm_scan_fdt_live() and also corrects a conflict with the 2 applied patches in
the commit 1712ca21924b ("dm: core: Scan /firmware node by default") and in the commit 747558d01457 ("dm: fdt: scan for devices under /firmware too"): the subnodes of "/firmware" (optee for example) are bound 2 times.
For example the dm tree command result on STM32MP1 is:
STM32MP> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------
root 0 [ + ] root_driver root_driver firmware 0 [ ] psci |-- psci sysreset 0 [ ] psci-sysreset | `-- psci-sysreset simple_bus 0 [ + ] generic_simple_bus |-- soc
...
tee 0 [ + ] optee |-- optee
...
tee 1 [ ] optee `-- optee
---
Changes in v2:
- update commit message (Serie-cc => Series-cc)
drivers/core/root.c | 52 +++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 35 deletions(-)
drivers/core/root.c +17 -35
@@ -203,15 +203,6 @@ static int dm_scan_fdt_live(struct udevice *parent,int ret =0 , err;for (np = node_parent->child; np; np = np->sibling) { -/* "chosen" node isn't a device itself but may contain some: */ -if (!strcmp (np->name,"chosen" )) { -pr_debug ("parsing subnodes of \"chosen\"\n" ); - - err =dm_scan_fdt_live (parent, np, pre_reloc_only); -if (err && !ret) - ret = err; -continue ; - }if (!of_device_is_available (np)) {pr_debug (" - ignoring disabled device\n" );
@@ -256,21 +247,6 @@ static int dm_scan_fdt_node(struct udevice *parent, const void *blob, offset = fdt_next_subnode(blob, offset)) {const char *node_name =fdt_get_name (blob, offset,NULL ); -/* -* The "chosen" and "firmware" nodes aren't devices -* themselves but may contain some: -*/ -if (!strcmp (node_name,"chosen" ) || - !strcmp (node_name,"firmware" )) { -pr_debug ("parsing subnodes of \"%s\"\n" , node_name); - - err =dm_scan_fdt_node (parent, blob, offset, - pre_reloc_only); -if (err && !ret) - ret = err; -continue ; - } -if (!fdtdec_get_is_enabled (blob, offset)) {pr_debug (" - ignoring disabled device\n" );continue ;
@@ -315,7 +291,8 @@ int dm_scan_fdt(const void *blob, bool pre_reloc_only)return dm_scan_fdt_node (gd->dm_root, blob,0 , pre_reloc_only); } -static int dm_scan_fdt_ofnode_path (const char *path,bool pre_reloc_only) +static int dm_scan_fdt_ofnode_path (const void *blob,const char *path, +bool pre_reloc_only) {ofnode node;
@@ -327,13 +304,18 @@ static int dm_scan_fdt_ofnode_path(const char *path, bool pre_reloc_only)if (of_live_active ())return dm_scan_fdt_live (gd->dm_root, node.np, pre_reloc_only);#endif -return dm_scan_fdt_node (gd->dm_root, gd->fdt_blob, node.of_offset, +return dm_scan_fdt_node (gd->dm_root, blob, node.of_offset, pre_reloc_only); }int dm_extended_scan_fdt (const void *blob,bool pre_reloc_only) { -int ret; +int ret, i; +const char *const nodes[] = { +"/chosen" , +"/clocks" , +"/firmware" + }; ret =dm_scan_fdt (blob, pre_reloc_only);if (ret) {
@@ -341,16 +323,16 @@ int dm_extended_scan_fdt(const void *blob, bool pre_reloc_only) return ret; } - ret =dm_scan_fdt_ofnode_path ("/clocks" , pre_reloc_only); -if (ret) { -debug ("scan for /clocks failed: %d\n" , ret); -return ret; +/* Some nodes aren't devices themselves but may contain some */ +for (i =0 ; i <ARRAY_SIZE (nodes); i++) { + ret =dm_scan_fdt_ofnode_path (blob, nodes[i], pre_reloc_only); +if (ret) { +debug ("dm_scan_fdt() scan for %s failed: %d\n" , + nodes[i], ret); +return ret; + } } - ret =dm_scan_fdt_ofnode_path ("/firmware" , pre_reloc_only); -if (ret) -debug ("scan for /firmware failed: %d\n" , ret); -return ret; }#endif
Signature · 1 line
2.17.1
PATCH v2 1/1 dm: core: Move "/chosen" and "/firmware" node scan
?t 13. 2. 2020 v 19:48 odes?latel Patrick Delaunay <[email protected]> napsal:
Quoted 136 lines
Use the new function
dm_scan_fdt_ofnode_path()to scan all the nodes which aren't devices themselves but may contain some:
- "/chosen"
- "/clocks"
- "/firmware"The patch removes the strcmp call in recursive function
dm_scan_fdt_live()and also corrects a conflict with the 2 applied patches in
the commit1712ca21924b("dm: core: Scan /firmware node by default") and in the commit747558d01457("dm: fdt: scan for devices under /firmware too"): the subnodes of "/firmware" (optee for example) are bound 2 times.For example the dm tree command result on STM32MP1 is:
STM32MP> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------root 0 [ + ] root_driver root_driver firmware 0 [ ] psci |-- psci sysreset 0 [ ] psci-sysreset | `-- psci-sysreset simple_bus 0 [ + ] generic_simple_bus |-- soc...
tee 0 [ + ] optee |-- optee
...
tee 1 [ ] optee `-- opteeSigned-off-by Patrick Delaunay <[email protected]>---
Changes in v2:
- update commit message (Serie-cc => Series-cc)drivers/core/root.c | 52 +++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 35 deletions(-)drivers/core/root.c +17
-36@@ -203,15 +203,6 @@ static int dm_scan_fdt_live(struct udevice *parent,int ret =0 , err;for (np = node_parent->child; np; np = np->sibling) { -/* "chosen" node isn't a device itself but may contain some: */ -if (!strcmp (np->name,"chosen" )) { -pr_debug ("parsing subnodes of \"chosen\"\n" ); - - err =dm_scan_fdt_live (parent, np, pre_reloc_only); -if (err && !ret) - ret = err; -continue ; - }if (!of_device_is_available (np)) {pr_debug (" - ignoring disabled device\n" );@@ -256,21 +247,6 @@ static int dm_scan_fdt_node(struct udevice *parent, const void *blob, offset = fdt_next_subnode(blob, offset)) {const char *node_name =fdt_get_name (blob, offset,NULL ); -/* -* The "chosen" and "firmware" nodes aren't devices -* themselves but may contain some: -*/ -if (!strcmp (node_name,"chosen" ) || - !strcmp (node_name,"firmware" )) { -pr_debug ("parsing subnodes of \"%s\"\n" , node_name); - - err =dm_scan_fdt_node (parent, blob, offset, - pre_reloc_only); -if (err && !ret) - ret = err; -continue ; - } -if (!fdtdec_get_is_enabled (blob, offset)) {pr_debug (" - ignoring disabled device\n" );continue ;@@ -315,7 +291,8 @@ int dm_scan_fdt(const void *blob, bool pre_reloc_only)return dm_scan_fdt_node (gd->dm_root, blob,0 , pre_reloc_only); } -static int dm_scan_fdt_ofnode_path (const char *path,bool pre_reloc_only) +static int dm_scan_fdt_ofnode_path (const void *blob,const char *path, +bool pre_reloc_only) {ofnode node;@@ -327,13 +304,18 @@ static int dm_scan_fdt_ofnode_path(const char *path, bool pre_reloc_only)if (of_live_active ())return dm_scan_fdt_live (gd->dm_root, node.np, pre_reloc_only);#endif -return dm_scan_fdt_node (gd->dm_root, gd->fdt_blob, node.of_offset, +return dm_scan_fdt_node (gd->dm_root, blob, node.of_offset, pre_reloc_only); }int dm_extended_scan_fdt (const void *blob,bool pre_reloc_only) { -int ret; +int ret, i; +const char *const nodes[] = { +"/chosen" , +"/clocks" , +"/firmware" + }; ret =dm_scan_fdt (blob, pre_reloc_only);if (ret) {@@ -341,16 +323,16 @@ int dm_extended_scan_fdt(const void *blob, bool pre_reloc_only) return ret; } - ret =dm_scan_fdt_ofnode_path ("/clocks" , pre_reloc_only); -if (ret) { -debug ("scan for /clocks failed: %d\n" , ret); -return ret; +/* Some nodes aren't devices themselves but may contain some */ +for (i =0 ; i <ARRAY_SIZE (nodes); i++) { + ret =dm_scan_fdt_ofnode_path (blob, nodes[i], pre_reloc_only); +if (ret) { +debug ("dm_scan_fdt() scan for %s failed: %d\n" , + nodes[i], ret); +return ret; + } } - ret =dm_scan_fdt_ofnode_path ("/firmware" , pre_reloc_only); -if (ret) -debug ("scan for /firmware failed: %d\n" , ret); -return ret; }#endif --
Quoted 2 lines
2.17.1
It looks good to me. I have reported here.
https://lists.denx.de/pipermail/u-boot/2020-January/395427.html
Thanks,
Michal
Signature · 5 lines
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91 w: www.monstr.eu p: +42-0-721842854 Maintainer of Linux kernel - Xilinx Microblaze Maintainer of Linux kernel - Xilinx Zynq ARM and ZynqMP ARM64 SoCs U-Boot custodian - Xilinx Microblaze/Zynq/ZynqMP/Versal SoCs
PATCH v2 1/1 dm: core: Move "/chosen" and "/firmware" node scan
In reply to Patrick DelaunayHi Patrick,
Quoted 28 lines · Patrick Delaunay
Use the new function
dm_scan_fdt_ofnode_path()to scan all the nodes which aren't devices themselves but may contain some:
- "/chosen"
- "/clocks"
- "/firmware"The patch removes the strcmp call in recursive function
dm_scan_fdt_live()and also corrects a conflict with the 2 applied patches in
the commit1712ca21924b("dm: core: Scan /firmware node by default") and in the commit747558d01457("dm: fdt: scan for devices under /firmware too"): the subnodes of "/firmware" (optee for example) are bound 2 times.For example the dm tree command result on STM32MP1 is:
STM32MP> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------root 0 [ + ] root_driver root_driver firmware 0 [ ] psci |-- psci sysreset 0 [ ] psci-sysreset | `-- psci-sysreset simple_bus 0 [ + ] generic_simple_bus |-- soc...
tee 0 [ + ] optee |-- optee
...
tee 1 [ ] optee `-- optee
Quoted 8 lines
Signed-off-by Patrick Delaunay <[email protected]>---
Changes in v2:
- update commit message (Serie-cc => Series-cc)drivers/core/root.c | 52 +++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 35 deletions(-)
This looks good to me, but please can you address the test failure (make qcheck)?
Regards,
Simon
PATCH v2 1/1 dm: core: Move "/chosen" and "/firmware" node scan
Hi,
Quoted 41 lines
From: Simon Glass <[email protected]>
Sent: lundi 17 f?vrier 2020 04:56Hi Patrick,
Quoted 34 lines · Patrick Delaunay
Use the new function
dm_scan_fdt_ofnode_path()to scan all the nodes which aren't devices themselves but may contain some:
- "/chosen"
- "/clocks"
- "/firmware"The patch removes the strcmp call in recursive function
dm_scan_fdt_live()and also corrects a conflict with the 2 applied patches in the commit1712ca21924b("dm: core: Scan /firmware node by default") and in the commit747558d01457("dm: fdt: scan for devices under /firmware too"): the subnodes of "/firmware" (optee for example) are bound 2 times.For example the dm tree command result on STM32MP1 is:
STM32MP> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------root 0 [ + ] root_driver root_driver firmware 0 [ ] psci |-- psci sysreset 0 [ ] psci-sysreset | `-- psci-sysreset simple_bus 0 [ + ] generic_simple_bus |-- soc...
tee 0 [ + ] optee |-- optee
...
tee 1 [ ] optee `-- opteeSigned-off-by Patrick Delaunay <[email protected]>---
Changes in v2:
- update commit message (Serie-cc => Series-cc)
Quoted 7 lines
Quoted 3 lines
drivers/core/root.c | 52 +++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 35 deletions(-)This looks good to me, but please can you address the test failure (make qcheck)?
I forget to execute it, sorry.
Work in progress....
Quoted 2 lines
Regards,
Simon
Patrick