serial: Make full device search optional
PATCH 1/1 serial: Make full device search optional
Commit 608b0c4ad4e5ec0c ("serial: Use next serial device if probing fails") added code to search for more serial devices if the default one was not probed correctly.
Unfortunately, that breaks omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.
---
Derald, could you please test this patch and verify it does indeed unbreak omap3_evm?
Thanks!
--- arch/arm/Kconfig | 1 + drivers/serial/Kconfig | 12 ++++++++++++ drivers/serial/serial-uclass.c | 13 +++++++++++++ 3 files changed, 26 insertions(+)
arch/arm/Kconfig +1 -0
@@ -550,6 +550,7 @@ config ARCH_BCM283X select DM_GPIO select OF_CONTROL select PL01X_SERIAL + select SERIAL_SEARCH_ALL imply FAT_WRITE config TARGET_VEXPRESS_CA15_TC2
drivers/serial/Kconfig +12 -0
@@ -79,6 +79,18 @@ config SERIAL_RX_BUFFER_SIZE help The size of the RX buffer (needs to be power of 2) +config SERIAL_SEARCH_ALL + bool "Search for serial devices after default one failed" + depends on DM_SERIAL + help + The serial subsystem only searches for a single serial device + that was instantiated, but does not check whether it was probed + correctly. With this option set, we make successful probing + mandatory and search for fallback serial devices if the default + device does not work. + + If unsure, say N. + config SPL_DM_SERIAL bool "Enable Driver Model for serial drivers in SPL" depends on DM_SERIAL
drivers/serial/serial-uclass.c +13 -0
@@ -74,7 +74,9 @@ static void serial_find_console_or_panic(void) {const void *blob = gd->fdt_blob;struct udevice *dev; +#ifdef CONFIG_SERIAL_SEARCH_ALL int ret; +#endif if (CONFIG_IS_ENABLED (OF_PLATDATA )) {uclass_first_device (UCLASS_SERIAL , &dev);
@@ -113,6 +115,8 @@ static void serial_find_console_or_panic(void)#else #define INDEX 0#endif + +#ifdef CONFIG_SERIAL_SEARCH_ALL if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev)) {if (dev->flags &DM_FLAG_ACTIVATED ) {
@@ -131,6 +135,15 @@ static void serial_find_console_or_panic(void) return; } } +#else +if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || + !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev) || + (!uclass_first_device (UCLASS_SERIAL , &dev) && dev)) { + gd->cur_serial_dev = dev; +return ; + } +#endif +#undef INDEX }
Signature · 1 line
2.12.3
PATCH 1/1 serial: Make full device search optional
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected]> wrote:
Commit 608b0c4ad4e5ec0c ("serial: Use next serial device if probing fails") added code to search for more serial devices if the default one was not probed correctly.
Unfortunately, that breaks omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.
---
Derald, could you please test this patch and verify it does indeed unbreak omap3_evm?
I will check later today. The board is not with me at the moment.
Derald
Thanks!
--- arch/arm/Kconfig | 1 + drivers/serial/Kconfig | 12 ++++++++++++ drivers/serial/serial-uclass.c | 13 +++++++++++++ 3 files changed, 26 insertions(+)
arch/arm/Kconfig +1 -0
@@ -550,6 +550,7 @@ config ARCH_BCM283X select DM_GPIO select OF_CONTROL select PL01X_SERIAL + select SERIAL_SEARCH_ALL imply FAT_WRITE config TARGET_VEXPRESS_CA15_TC2
drivers/serial/Kconfig +12 -0
@@ -79,6 +79,18 @@ config SERIAL_RX_BUFFER_SIZE help The size of the RX buffer (needs to be power of 2) +config SERIAL_SEARCH_ALL + bool "Search for serial devices after default one failed" + depends on DM_SERIAL + help + The serial subsystem only searches for a single serial device + that was instantiated, but does not check whether it was probed + correctly. With this option set, we make successful probing + mandatory and search for fallback serial devices if the default + device does not work. + + If unsure, say N. + config SPL_DM_SERIAL bool "Enable Driver Model for serial drivers in SPL" depends on DM_SERIAL
drivers/serial/serial-uclass.c +13 -0
@@ -74,7 +74,9 @@ static void serial_find_console_or_panic(void) {const void *blob = gd->fdt_blob;struct udevice *dev; +#ifdef CONFIG_SERIAL_SEARCH_ALL int ret; +#endif if (CONFIG_IS_ENABLED (OF_PLATDATA )) {uclass_first_device (UCLASS_SERIAL , &dev);
@@ -113,6 +115,8 @@ static void serial_find_console_or_panic(void)#else #define INDEX 0#endif + +#ifdef CONFIG_SERIAL_SEARCH_ALL if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev)) {if (dev->flags &DM_FLAG_ACTIVATED ) {
@@ -131,6 +135,15 @@ static void serial_find_console_or_panic(void) return; } } +#else +if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || + !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev) || + (!uclass_first_device (UCLASS_SERIAL , &dev) && dev)) { + gd->cur_serial_dev = dev; +return ; + } +#endif +#undef INDEX }
--
2.12.3
PATCH 1/1 serial: Make full device search optional
Quoted 10 lines · Derald Woods
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected]> wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails") added code to search for more serial devices if the default one was not probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.
Quoted 9 lines
Fixes 608b0c4ad4e5ec0c ("serial: Use next serial device if probing fails")Reported-by Derald D. Woods <[email protected]>Signed-off-by Alexander Graf <[email protected]>---
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?
The omap3_evm boots now with this patch applied on master. If I enable SERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the default config only. On failure, there is no console input/output and the board unresponsive.
Derald
Quoted 91 lines
I will check later today. The board is not with me at the moment.
Derald
Thanks!
--- arch/arm/Kconfig | 1 + drivers/serial/Kconfig | 12 ++++++++++++ drivers/serial/serial-uclass.c | 13 +++++++++++++ 3 files changed, 26 insertions(+)arch/arm/Kconfig +1
-0@@ -550,6 +550,7 @@ config ARCH_BCM283X select DM_GPIO select OF_CONTROL select PL01X_SERIAL + select SERIAL_SEARCH_ALL imply FAT_WRITE config TARGET_VEXPRESS_CA15_TC2drivers/serial/Kconfig +12
-0@@ -79,6 +79,18 @@ config SERIAL_RX_BUFFER_SIZE help The size of the RX buffer (needs to be power of 2) +config SERIAL_SEARCH_ALL + bool "Search for serial devices after default one failed" + depends on DM_SERIAL + help + The serial subsystem only searches for a single serial device + that was instantiated, but does not check whether it was probed + correctly. With this option set, we make successful probing + mandatory and search for fallback serial devices if the default + device does not work. + + If unsure, say N. + config SPL_DM_SERIAL bool "Enable Driver Model for serial drivers in SPL" depends on DM_SERIALdrivers/serial/serial-uclass.c +13
-1@@ -74,7 +74,9 @@ static void serial_find_console_or_panic(void) {const void *blob = gd->fdt_blob;struct udevice *dev; +#ifdef CONFIG_SERIAL_SEARCH_ALL int ret; +#endif if (CONFIG_IS_ENABLED (OF_PLATDATA )) {uclass_first_device (UCLASS_SERIAL , &dev);@@ -113,6 +115,8 @@ static void serial_find_console_or_panic(void)#else #define INDEX 0#endif + +#ifdef CONFIG_SERIAL_SEARCH_ALL if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev)) {if (dev->flags &DM_FLAG_ACTIVATED ) {@@ -131,6 +135,15 @@ static void serial_find_console_or_panic(void) return; } } +#else +if (!uclass_get_device_by_seq (UCLASS_SERIAL ,INDEX , &dev) || + !uclass_get_device (UCLASS_SERIAL ,INDEX , &dev) || + (!uclass_first_device (UCLASS_SERIAL , &dev) && dev)) { + gd->cur_serial_dev = dev; +return ; + } +#endif +#undef INDEX } --2.12.3
PATCH 1/1 serial: Make full device search optional
Quoted 17 lines · Derald D. Woods
Quoted 16 lines · Derald Woods
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected]> wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails") added code to search for more serial devices if the default one was not probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial device if probing fails")Reported-by Derald D. Woods <[email protected]>Signed-off-by Alexander Graf <[email protected]>---
Quoted 7 lines
Quoted 3 lines
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on master. If I enableSERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the default config only. On failure, there is no console input/output and the board unresponsive.
So SPL is already broken? Can you try a known working SPL with SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
Alex
PATCH 1/1 serial: Make full device search optional
On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected]> wrote:
Quoted 20 lines · Derald D. Woods
Quoted 18 lines · Derald Woods
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected]> wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails")
added code to search for more serial devices if the default one was not probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial device if probingfails")
Reported-by Derald D. Woods <[email protected]>Signed-off-by Alexander Graf <[email protected]>---
Quoted 8 lines
Quoted 4 lines
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on master. If I enable
SERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the default config only. On failure, there is no console input/output and the board unresponsive.
So SPL is already broken? Can you try a known working SPL with SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the omap3_evm is ahead of some other OMAP34XX boards currently, a good working example would be helpful. If omap3_evm becomes the example, let's make it a good one.
Derald
PATCH 1/1 serial: Make full device search optional
Quoted 48 lines · Derald Woods
On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected] <mailto:[email protected]>> wrote:
On 01/30/2018 12:41 AM, Derald D. Woods wrote:
On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods wrote:
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected] <mailto:[email protected]>> wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails")
added code to search for more serial devices if the default one was not
probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is
the case, let's disable the full search for everyone but bcm283x where it
is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial deviceif probing fails")
Reported-by Derald D. Woods <[email protected]<mailto:[email protected]>>
Signed-off-by Alexander Graf <[email protected]<mailto:[email protected]>>
---
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on master. If I enable
SERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the
default config only. On failure, there is no console input/output and
the board unresponsive.So SPL is already broken? Can you try a known working SPL with SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
Quoted 6 lines
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the
omap3_evmis ahead of some other OMAP34XX boards currently, a good working example would be helpful. Ifomap3_evmbecomes the example, let's make it a good one.
If you want to make it a good example, don't disable CONFIG_EFI_LOADER :).
But really, the only major difference I saw between beagle and evm was the fact that evm used DM in SPL. I patched that up locally (had to remove ext support as the binary became too big otherwise), but that didn't show the issue either. So we'll have to wait on your tests.
Alex
PATCH 1/1 serial: Make full device search optional
Quoted 57 lines · Alexander Graf
Quoted 48 lines · Derald Woods
On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected] <mailto:
agraf at suse.de>> wrote:On 01/30/2018 12:41 AM, Derald D. Woods wrote:
On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods wrote:
On Jan 29, 2018 6:57 AM, "Alexander Graf" <[email protected] <mailto:[email protected]>> wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails")
added code to search for more serial devices if the default one was not
probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is
the case, let's disable the full search for everyone but bcm283x where it
is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial deviceif probing fails")
Reported-by Derald D. Woods <[email protected]<mailto:[email protected]>>
Signed-off-by Alexander Graf <[email protected]<mailto:[email protected]>>
---
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on master. If I enable
SERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the
default config only. On failure, there is no console input/output and
the board unresponsive.So SPL is already broken? Can you try a known working SPL with SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
Quoted 6 lines
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the
omap3_evmis ahead of some other OMAP34XX boards currently, a good working example would be helpful. Ifomap3_evmbecomes the example, let's make it a good one.
Quoted 8 lines
If you want to make it a good example, don't disable
CONFIG_EFI_LOADER:).But really, the only major difference I saw between beagle and evm was the fact that evm used DM in SPL. I patched that up locally (had to remove ext support as the binary became too big otherwise), but that didn't show the issue either. So we'll have to wait on your test
s.
It looks like some compiler issue is causing the problem. I was using GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with SERIAL_SEARCH_ALL=y. I also verified this by enabling SPL_DM_SERIAL on 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC versions compiled without error. GCC 6.4.0 is the compiler version that is working for me now. The actual offending generated code will take more time, for me, to sort through.
Derald
PATCH 1/1 serial: Make full device search optional
Quoted 78 lines · Derald Woods
On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <[email protected] <mailto:[email protected]>> wrote:
On 01/30/2018 02:09 PM, Derald Woods wrote:
On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected] <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> wrote:
On 01/30/2018 12:41 AM, Derald D. Woods wrote:
On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods wrote:
On Jan 29, 2018 6:57 AM, "Alexander Graf" <agraf at suse.de <mailto:[email protected]>
<mailto:agraf at suse.de <mailto:[email protected]>>> wrote:Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails")
added code to search for more serial devices if the default one was not
probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is
the case, let's disable the full search for everyone but bcm283x where it
is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial deviceif probing fails")
Reported-by: Derald D. Woods <woods.technical at gmail.com <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>Signed-off-by Alexander Graf <[email protected]<mailto:[email protected]>
<mailto:agraf at suse.de <mailto:[email protected]>>>---
Derald, could you please test this patch and verify it does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on master. If
I enable
SERIAL_SEARCH_ALL, it does not boot. I always build cleanly using the
default config only. On failure, there is no console input/output and
the board unresponsive.So SPL is already broken? Can you try a known working SPL with SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the
omap3_evmis ahead of some other OMAP34XX boards currently, a good working example would be helpful. Ifomap3_evmbecomes the example, let's make it a good one.If you want to make it a good example, don't disable
CONFIG_EFI_LOADER:).But really, the only major difference I saw between beagle and evm was the fact that evm used DM in SPL. I patched that up locally (had to remove ext support as the binary became too big otherwise), but that didn't show the issue either. So we'll have to wait on your test s.
Quoted 8 lines
It looks like some compiler issue is causing the problem. I was using GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with SERIAL_SEARCH_ALL=y. I also verified this by enablingSPL_DM_SERIALon 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC versions compiled without error. GCC 6.4.0 is the compiler version that is working for me now. The actual offending generated code will take more time, for me, to sort through.
Can you somehow make it break with omap3_beagle? I have one of those and could then help debug.
Alex
PATCH 1/1 serial: Make full device search optional
Quoted 97 lines · Alexander Graf
Quoted 17 lines · Derald Woods
On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <[email protected] <mailto:[email protected]>> wrote:
On 01/30/2018 02:09 PM, Derald Woods wrote: On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected] <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> wrote: On 01/30/2018 12:41 AM, Derald D. Woods wrote: On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods wrote: On Jan 29, 2018 6:57 AM, "Alexander Graf" <agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>Quoted 2 lines · wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serialdevice
Quoted 8 lines
if probing fails")
added code to search for more serial devices if the default one was not
probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is
the case, let's disable the full search for everyonebut
Quoted 4 lines
bcm283x where it
is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serialdevice
Quoted 5 lines
if probing fails")
Reported-by: Derald D. Woods
<woods.technical at gmail.com <mailto:[email protected]>
<mailto:[email protected]
<mailto:[email protected]>>>Quoted 7 lines
Signed-off-by Alexander Graf <[email protected]<mailto:[email protected]>
<mailto:agraf at suse.de <mailto:[email protected]>>>---
Derald, could you please test this patch and verify
it
Quoted 7 lines
does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on
master. If
I enable
SERIAL_SEARCH_ALL, it does not boot. I always buildcleanly
Quoted 7 lines
using the
default config only. On failure, there is no console input/output and
the board unresponsive.So SPL is already broken? Can you try a known working SPL
with
Quoted 22 lines
SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the
omap3_evmis ahead of some other OMAP34XX boards currently, a good working example would be helpful. Ifomap3_evmbecomes the example, let's make it a good one.If you want to make it a good example, don't disable
CONFIG_EFI_LOADER:).But really, the only major difference I saw between beagle and evm was the fact that evm used DM in SPL. I patched that up locally (had to remove ext support as the binary became too big otherwise), but that didn't show the issue either. So we'll have to wait on your test s.
Quoted 7 lines
It looks like some compiler issue is causing the problem. I was using GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with SERIAL_SEARCH_ALL=y. I also verified this by enabling
SPL_DM_SERIALon 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC versions compiled without error. GCC 6.4.0 is the compiler version that is working for me now. The actual offending generated code will take more time, for me, to sort through.
Quoted 3 lines
Can you somehow make it break with
omap3_beagle? I have one of those and could then help debug.
No. Not with the current default configurations. I have both the beagleboard(3530) and beagleboard-xM(3730) at home. Their configuration currently does not enable DM_SERIAL/SPL_DM_SERIAL. I just recently added OF_CONTROL for them on U-Boot master. The code path is different for without DM_SERIAL. Enabling DM_SERIAL will aid in comparison, but will require dropping some config options to make it fit into SRAM. The '-xM' does not have NAND. On the omap3_evm, I enable UBI in the default config. So their are at least a couple of options that would not apply to both beagle variants. There should probably be a separate default config file for each variant. The 3530 beagle variant would/should be identical to the omap3_evm. They are somewhat related from a historical perspective. I will put together an update to the default configurations this week. Once that is done, I will be able to make a true comparison. (omap3_evm has a 3730 variant also)
Derald
PATCH 1/1 serial: Make full device search optional
Quoted 120 lines · Derald Woods
Quoted 98 lines · Alexander Graf
Quoted 17 lines · Derald Woods
On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <[email protected] <mailto:[email protected]>> wrote:
On 01/30/2018 02:09 PM, Derald Woods wrote: On Jan 30, 2018 3:17 AM, "Alexander Graf" <[email protected] <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> wrote: On 01/30/2018 12:41 AM, Derald D. Woods wrote: On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods wrote: On Jan 29, 2018 6:57 AM, "Alexander Graf" <agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>Quoted 2 lines · wrote:
Commit
608b0c4ad4e5ec0c("serial: Use next serialdevice
Quoted 8 lines
if probing fails")
added code to search for more serial devices if the default one was not
probed correctly.Unfortunately, that breaks
omap3_evm. So while investigating why that is
the case, let's disable the full search foreveryone but
Quoted 4 lines
bcm283x where it
is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serialdevice
Quoted 5 lines
if probing fails")
Reported-by: Derald D. Woods
<woods.technical at gmail.com <mailto:[email protected]>
<mailto:[email protected]
<mailto:[email protected]>>>Quoted 7 lines
Signed-off-by Alexander Graf <[email protected]<mailto:[email protected]>
<mailto:agraf at suse.de <mailto:[email protected]>>>---
Derald, could you please test this patch and verify
it
Quoted 7 lines
does indeed unbreak
omap3_evm?The
omap3_evmboots now with this patch applied on
master. If
I enable
SERIAL_SEARCH_ALL, it does not boot. I always buildcleanly
Quoted 7 lines
using the
default config only. On failure, there is no console input/output and
the board unresponsive.So SPL is already broken? Can you try a known working SPL
with
Quoted 11 lines
SERIAL_SEARCH_ALL=y U-Boot payload on top? Does that work?
I will give that path a try and see what I can discover. Again, it will be later today or tomorrow before I can get to this. This is why I asked what should the board code actually look like. As the
omap3_evmis ahead of some other OMAP34XX boards currently, a good working example would be helpful. Ifomap3_evmbecomes the example, let's make it a good one.Quoted 7 lines
If you want to make it a good example, don't disable
CONFIG_EFI_LOADER:).But really, the only major difference I saw between beagle and evm was the fact that evm used DM in SPL. I patched that up locally (had to remove ext support as the binary became too big otherwise), but that didn't show the issue either. So we'll have to wait on your
test
Quoted 11 lines
s.
It looks like some compiler issue is causing the problem. I was using GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with SERIAL_SEARCH_ALL=y. I also verified this by enabling
SPL_DM_SERIALon 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC versions compiled without error. GCC 6.4.0 is the compiler version that is working for me now. The actual offending generated code will take more time, for me, to sort through.Quoted 3 lines
Can you somehow make it break with
omap3_beagle? I have one of those and could then help debug.No. Not with the current default configurations. I have both the beagleboard(3530) and beagleboard-xM(3730) at home. Their configuration currently does not enable DM_SERIAL/SPL_DM_SERIAL. I just recently added
OF_CONTROLfor them on U-Boot master. The code path is different for withoutDM_SERIAL. EnablingDM_SERIALwill aid in comparison, but will require dropping some config options to make it fit into SRAM. The '-xM' does not have NAND. On theomap3_evm, I enable UBI in the default config. So their are at least a couple of options that would not apply to both beagle variants. There should probably be a separate default config file for each variant. The 3530 beagle variant would/should be identical to theomap3_evm. They are somewhat related from a historical perspective. I will put together an update to the default configurations this week. Once that is done, I will be able to make a true comparison. (omap3_evmhas a 3730 variant also)
Quoted 2 lines
Derald
The issue appears to have been related to the use of CONFIG_SPL_SYS_MALLOC_SIMPLE and CONFIG_SYS_MALLOC_F_LEN in the default config. The RFC patch series is given below:
https://lists.denx.de/pipermail/u-boot/2018-February/320152.html
The BeagleBoard config never attempted to use CONFIG_SPL_SYS_MALLOC_SIMPLE.
Derald
PATCH 1/1 serial: Make full device search optional
Quoted 148 lines · Derald Woods
On Mon, Feb 5, 2018 at 7:13 AM, Derald Woods <[email protected] <mailto:[email protected]>> wrote:
On Mon, Feb 5, 2018 at 3:42 AM, Alexander Graf <[email protected] <mailto:[email protected]>>wrote:
On 05.02.18 01:39, Derald Woods wrote:
> On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <agraf at suse.de <mailto:[email protected]> > <mailto:agraf at suse.de <mailto:[email protected]>>> wrote:
>
> On 01/30/2018 02:09 PM, Derald Woods wrote:
>
> On Jan 30, 2018 3:17 AM, "Alexander Graf" <agraf at suse.de <mailto:[email protected]> > <mailto:agraf at suse.de <mailto:[email protected]>>
<mailto:agraf at suse.de <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>>>> wrote:
>
> On 01/30/2018 12:41 AM, Derald D. Woods wrote:
>
> On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods
> wrote:
>
> On Jan 29, 2018 6:57 AM, "Alexander Graf"
> <agraf at suse.de <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
> <mailto:agraf at suse.de <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>> wrote:
>
> Commit608b0c4ad4e5ec0c("serial: Use next serial device > if probing fails")
> added code to search for more serial devices if the
> default one was not
> probed correctly.
>
> Unfortunately, that breaksomap3_evm. So while
> investigating why that is
> the case, let's disable the full search for everyone but > bcm283x where it
> is needed.
>
> Fixes:608b0c4ad4e5ec0c("serial: Use next serial device > if probing fails")
> Reported-by: Derald D. Woods
> <woods.technical at gmail.com <mailto:[email protected]> <mailto:[email protected]
<mailto:[email protected]>>
> <mailto:[email protected] <mailto:[email protected]>
> <mailto:woods.technical at gmail.com <mailto:[email protected]>>>> > Signed-off-by: Alexander Graf <agraf at suse.de <mailto:[email protected]> > <mailto:agraf at suse.de <mailto:[email protected]>>
> <mailto:[email protected]
<mailto:[email protected]> <mailto:[email protected]
<mailto:[email protected]>>>>
>
> ---
>
> Derald, could you please test this patch
and verify it
> does indeed unbreak
>omap3_evm?
>
> Theomap3_evmboots now with this patch applied on
> master. If
> I enable
>SERIAL_SEARCH_ALL, it does not boot. I always
build cleanly
> using the
> default config only. On failure, there is no
console
> input/output and
> the board unresponsive.
>
>
> So SPL is already broken? Can you try a known
working SPL with
> SERIAL_SEARCH_ALL=y U-Boot payload on top? Does
that work?
>
>
> I will give that path a try and see what I can
discover. Again,
> it will be later today or tomorrow before I can get to
this.
> This is why I asked what should the board code
actually look
> like. As theomap3_evmis ahead of some other OMAP34XX
boards
> currently, a good working example would be helpful. If
omap3_evm
> becomes the example, let's make it a good one.
>
>
> If you want to make it a good example, don't disable
> CONFIG_EFI_LOADER :).
>
> But really, the only major difference I saw between beagle
and evm
> was the fact that evm used DM in SPL. I patched that up
locally (had
> to remove ext support as the binary became too big
otherwise), but
> that didn't show the issue either. So we'll have to wait
on your test
> s.
>
>
>
> It looks like some compiler issue is causing the problem. I
was using
> GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with
> SERIAL_SEARCH_ALL=y. I also verified this by enabling
SPL_DM_SERIALon
> 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC
versions
> compiled without error. GCC 6.4.0 is the compiler version that is
> working for me now. The actual offending generated code will
take more
> time, for me, to sort through.Can you somehow make it break with
omap3_beagle? I have one of those and
could then help debug.No. Not with the current default configurations. I have both the beagleboard(3530) and beagleboard-xM(3730) at home. Their configuration currently does not enable DM_SERIAL/SPL_DM_SERIAL. I just recently added
OF_CONTROLfor them on U-Boot master. The code path is different for withoutDM_SERIAL. EnablingDM_SERIALwill aid in comparison, but will require dropping some config options to make it fit into SRAM. The '-xM' does not have NAND. On theomap3_evm, I enable UBI in the default config. So their are at least a couple of options that would not apply to both beagle variants. There should probably be a separate default config file for each variant. The 3530 beagle variant would/should be identical to theomap3_evm. They are somewhat related from a historical perspective. I will put together an update to the default configurations this week. Once that is done, I will be able to make a true comparison. (omap3_evmhas a 3730 variant also)Derald
Quoted 7 lines
The issue appears to have been related to the use of
CONFIG_SPL_SYS_MALLOC_SIMPLEandCONFIG_SYS_MALLOC_F_LENin the default config. The RFC patch series is given below:https://lists.denx.de/pipermail/u-boot/2018-February/320152.html
The BeagleBoard config never attempted to use
CONFIG_SPL_SYS_MALLOC_SIMPLE.
So you're saying it's probably just running out of memory?
Alex
A message that is not in the archive (sent off-list or not yet received): CA+CtpRjLxgM9TshXht0QeXQ1=ncm0nR2BYzGH9uXqJfVatLGCw@mail.gmail.com
A message that is not in the archive (sent off-list or not yet received): CA+CtpRhx6NsXAaSUuqi--7y6R0Xkmf09GFHFmneMWsEtyt4rUA@mail.gmail.com
A message that is not in the archive (sent off-list or not yet received): CA+CtpRjJMXrY8o-01wCSATuVkTjJZkDm_WkhT32mp1N=42UcvQ@mail.gmail.com
A message that is not in the archive (sent off-list or not yet received): CA+CtpRit+iLikbOkJe356u320j_sdOixnoPq1vDvdDzqknBcGQ@mail.gmail.com
PATCH 1/1 serial: Make full device search optional
On Feb 15, 2018 9:00 AM, "Alexander Graf" <[email protected]> wrote:
Quoted 10 lines · Derald Woods
On Mon, Feb 5, 2018 at 7:13 AM, Derald Woods <[email protected] <mailto:[email protected]>> wrote:
On Mon, Feb 5, 2018 at 3:42 AM, Alexander Graf <[email protected] <mailto:[email protected]>>wrote:
On 05.02.18 01:39, Derald Woods wrote:
> On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <agraf at suse.de
<mailto:[email protected]>
Quoted 5 lines
> <mailto:agraf at suse.de <mailto:[email protected]>>> wrote: > > On 01/30/2018 02:09 PM, Derald Woods wrote: > > On Jan 30, 2018 3:17 AM, "Alexander Graf" <agraf at suse.de
<mailto:[email protected]>
Quoted 7 lines
> <mailto:agraf at suse.de <mailto:[email protected]>> <mailto:agraf at suse.de <mailto:[email protected]> > <mailto:agraf at suse.de <mailto:[email protected]>>>> wrote: > > On 01/30/2018 12:41 AM, Derald D. Woods wrote: > > On Mon, Jan 29, 2018 at 07:46:09AM -0600,
Derald Woods
Quoted 4 lines
> wrote:
>
> On Jan 29, 2018 6:57 AM, "Alexander Graf"
> <agraf at suse.de <mailto:[email protected]> <mailto:
Quoted 2 lines
<mailto:[email protected]>>
> <mailto:agraf at suse.de <mailto:[email protected]>
<mailto:[email protected]
Quoted 3 lines
<mailto:[email protected]>>>> wrote: > > Commit 608b0c4ad4e5ec0c ("serial: Use next
serial device
Quoted 2 lines
> if probing fails")
> added code to search for more serial
devices if the
Quoted 4 lines
> default one was not
> probed correctly.
>
> Unfortunately, that breaksomap3_evm. So
while
Quoted 2 lines
> investigating why that is
> the case, let's disable the full search for
everyone but
Quoted 4 lines
> bcm283x where it
> is needed.
>
> Fixes:608b0c4ad4e5ec0c("serial: Use next
serial device
Quoted 3 lines
> if probing fails")
> Reported-by: Derald D. Woods
> <woods.technical at gmail.com <mailto:
Quoted 5 lines
<mailto:[email protected]
<mailto:[email protected]>>
> <mailto:[email protected]
<mailto:[email protected]>
> <mailto:woods.technical at gmail.com <mailto:
Quoted 1 line
> Signed-off-by: Alexander Graf <agraf at suse.de
<mailto:[email protected]>
Quoted 5 lines
> <mailto:agraf at suse.de <mailto:[email protected]>>
> <mailto:[email protected]
<mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>>
>
Quoted 8 lines
> ---
>
> Derald, could you please test this patch
and verify it
> does indeed unbreak
>omap3_evm?
>
> Theomap3_evmboots now with this patch applied
on
Quoted 45 lines
> master. If
> I enable
>SERIAL_SEARCH_ALL, it does not boot. I always
build cleanly
> using the
> default config only. On failure, there is no
console
> input/output and
> the board unresponsive.
>
>
> So SPL is already broken? Can you try a known
working SPL with
> SERIAL_SEARCH_ALL=y U-Boot payload on top? Does
that work?
>
>
> I will give that path a try and see what I can
discover. Again,
> it will be later today or tomorrow before I can get to
this.
> This is why I asked what should the board code
actually look
> like. As theomap3_evmis ahead of some other OMAP34XX
boards
> currently, a good working example would be helpful. If
omap3_evm
> becomes the example, let's make it a good one.
>
>
> If you want to make it a good example, don't disable
>CONFIG_EFI_LOADER:).
>
> But really, the only major difference I saw between beagle
and evm
> was the fact that evm used DM in SPL. I patched that up
locally (had
> to remove ext support as the binary became too big
otherwise), but
> that didn't show the issue either. So we'll have to wait
on your test
> s.
>
>
>
Quoted 8 lines
> It looks like some compiler issue is causing the problem. I was using
> GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with > SERIAL_SEARCH_ALL=y. I also verified this by enablingSPL_DM_SERIALon
> 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC versions
> compiled without error. GCC 6.4.0 is the compiler version that
is
Quoted 30 lines
> working for me now. The actual offending generated code will take more
> time, for me, to sort through.Can you somehow make it break with
omap3_beagle? I have one of those and
could then help debug.No. Not with the current default configurations. I have both the beagleboard(3530) and beagleboard-xM(3730) at home. Their configuration currently does not enable DM_SERIAL/SPL_DM_SERIAL. I just recently added
OF_CONTROLfor them on U-Boot master. The code path is different for withoutDM_SERIAL. EnablingDM_SERIALwill aid in comparison, but will require dropping some config options to make it fit into SRAM. The '-xM' does not have NAND. On theomap3_evm, I enable UBI in the default config. So their are at least a couple of options that would not apply to both beagle variants. There should probably be a separate default config file for each variant. The 3530 beagle variant would/should be identical to theomap3_evm. They are somewhat related from a historical perspective. I will put together an update to the default configurations this week. Once that is done, I will be able to make a true comparison. (omap3_evmhas a 3730 variant also)Derald
Quoted 7 lines
The issue appears to have been related to the use of
CONFIG_SPL_SYS_MALLOC_SIMPLEandCONFIG_SYS_MALLOC_F_LENin the default config. The RFC patch series is given below:https://lists.denx.de/pipermail/u-boot/2018-February/320152.html
The BeagleBoard config never attempted to use
CONFIG_SPL_SYS_MALLOC_SIMPLE.
So you're saying it's probably just running out of memory?
Alex
The configuration mix was the key difference. The patch series has the details. The default malloc length set by configuration is actually smaller. So there are likely a mix of factors.
Derald
PATCH 1/1 serial: Make full device search optional
Quoted 218 lines · Derald Woods
On Feb 15, 2018 9:00 AM, "Alexander Graf" <[email protected] <mailto:[email protected]>> wrote:
On 13.02.18 01:00, Derald Woods wrote:
> On Mon, Feb 5, 2018 at 7:13 AM, Derald Woods <woods.technical at gmail.com <mailto:[email protected]> > <mailto:[email protected] <mailto:[email protected]>>> wrote:
>
> On Mon, Feb 5, 2018 at 3:42 AM, Alexander Graf <[email protected] <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>>>wrote:
>
>
>
> On 05.02.18 01:39, Derald Woods wrote:
> > On Tue, Jan 30, 2018 at 7:34 AM, Alexander Graf <agraf at suse.de <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
> > <mailto:agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>> wrote:
> >
> > On 01/30/2018 02:09 PM, Derald Woods wrote:
> >
> > On Jan 30, 2018 3:17 AM, "Alexander Graf" <agraf at suse.de <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
> > <mailto:agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>
> <mailto:agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>
> > <mailto:agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>>> wrote:
> >
> > On 01/30/2018 12:41 AM, Derald D. Woods wrote:
> >
> > On Mon, Jan 29, 2018 at 07:46:09AM -0600, Derald Woods
> > wrote:
> >
> > On Jan 29, 2018 6:57 AM, "Alexander Graf"
> > <agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>> <mailto:[email protected] <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>>> > > <mailto:[email protected] <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>> <mailto:agraf at suse.de <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>>>>> wrote:
> >
> > Commit608b0c4ad4e5ec0c("serial:
Use next serial device
> > if probing fails")
> > added code to search for more serial devices if the
> > default one was not
> > probed correctly.
> >
> > Unfortunately, that breaksomap3_evm. So while
> > investigating why that is
> > the case, let's disable the full search for everyone but
> > bcm283x where it
> > is needed.
> >
> > Fixes:608b0c4ad4e5ec0c("serial:
Use next serial device
> > if probing fails")
> > Reported-by: Derald D. Woods > > <[email protected] <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
> <mailto:[email protected] <mailto:[email protected]>
> <mailto:[email protected] <mailto:[email protected]>>>
> > <mailto:[email protected] <mailto:[email protected]>
> <mailto:[email protected] <mailto:[email protected]>>
> > <mailto:[email protected] <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>>>
> > Signed-off-by: Alexander Graf <agraf at suse.de <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
> > <mailto:agraf at suse.de <mailto:[email protected]> <mailto:agraf at suse.de <mailto:[email protected]>>>
> > <mailto:[email protected] <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>> <mailto:agraf at suse.de <mailto:[email protected]>
> <mailto:agraf at suse.de <mailto:[email protected]>>>>> > >
> > ---
> >
> > Derald, could you please test this patch > and verify it
> > does indeed unbreak
> >omap3_evm?
> >
> > Theomap3_evmboots now with this patch applied on
> > master. If
> > I enable
> >SERIAL_SEARCH_ALL, it does not boot. I always
> build cleanly
> > using the
> > default config only. On failure, there is no > console
> > input/output and
> > the board unresponsive.
> >
> >
> > So SPL is already broken? Can you try a known > working SPL with
> > SERIAL_SEARCH_ALL=y U-Boot payload on top? Does > that work?
> >
> >
> > I will give that path a try and see what I can > discover. Again,
> > it will be later today or tomorrow before I can get to
> this.
> > This is why I asked what should the board code > actually look
> > like. As theomap3_evmis ahead of some other OMAP34XX
> boards
> > currently, a good working example would be helpful. If
> omap3_evm
> > becomes the example, let's make it a good one. > >
> >
> > If you want to make it a good example, don't disable > > CONFIG_EFI_LOADER :).
> >
> > But really, the only major difference I saw between beagle
> and evm
> > was the fact that evm used DM in SPL. I patched that up > locally (had
> > to remove ext support as the binary became too big > otherwise), but
> > that didn't show the issue either. So we'll have to wait > on your test
> > s.
> >
> >
> >
> > It looks like some compiler issue is causing the problem. I > was using
> > GCC 7.2.0. When I go back to GCC 6.4.0 the board boots with > > SERIAL_SEARCH_ALL=y. I also verified this by enabling > SPL_DM_SERIAL on
> > 'omap3_beagle' and booting with SERIAL_SEARCH_ALL=y. Both GCC
> versions
> > compiled without error. GCC 6.4.0 is the compiler version that is
> > working for me now. The actual offending generated code will > take more
> > time, for me, to sort through.
>
> Can you somehow make it break withomap3_beagle? I have one of > those and
> could then help debug.
>
>
>
> No. Not with the current default configurations. I have both the > beagleboard(3530) and beagleboard-xM(3730) at home. Their > configuration currently does not enable DM_SERIAL/SPL_DM_SERIAL. I > just recently addedOF_CONTROLfor them on U-Boot master. The code > path is different for withoutDM_SERIAL. EnablingDM_SERIALwill aid
> in comparison, but will require dropping some config options to make
> it fit into SRAM. The '-xM' does not have NAND. On theomap3_evm, I
> enable UBI in the default config. So their are at least a couple of
> options that would not apply to both beagle variants. There should > probably be a separate default config file for each variant. The > 3530 beagle variant would/should be identical to theomap3_evm. They
> are somewhat related from a historical perspective. I will put > together an update to the default configurations this week. Once > that is done, I will be able to make a true comparison. (omap3_evm> has a 3730 variant also)
>
> Derald
>
>
>
> The issue appears to have been related to the use of
>CONFIG_SPL_SYS_MALLOC_SIMPLEandCONFIG_SYS_MALLOC_F_LENin the default
> config. The RFC patch series is given below:
>
> https://lists.denx.de/pipermail/u-boot/2018-February/320152.html
<https://lists.denx.de/pipermail/u-boot/2018-February/320152.html>
>
> The BeagleBoard config never attempted to useCONFIG_SPL_SYS_MALLOC_SIMPLE.So you're saying it's probably just running out of memory?
Quoted 6 lines
Alex
The configuration mix was the key difference. The patch series has the details. The default malloc length set by configuration is actually smaller. So there are likely a mix of factors.
Ah, so you're saying with your changes you can reproduce the problem on the beagleboard?
I still fail to see the exact connection between the serial bug and the patch set :)
Alex
PATCH 1/1 serial: Make full device search optional
In reply to Alexander GrafQuoted 4 lines · Alexander Graf
Commit
608b0c4ad4e5ec0c("serial: Use next serial device if probing fails") added code to search for more serial devices if the default one was not probed correctly.
Quoted 8 lines
Unfortunately, that breaks
omap3_evm. So while investigating why that is the case, let's disable the full search for everyone but bcm283x where it is needed.Fixes 608b0c4ad4e5ec0c ("serial: Use next serial device if probing fails")Reported-by Derald D. Woods <[email protected]>Signed-off-by Alexander Graf <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: not available URL: <http://lists.denx.de/pipermail/u-boot/attachments/20180129/8e610a39/attachment.sig>