[U-Boot] [PATCH] bootm: fix conditional controlling call to fixup_silent_linux
This function is only defined if CONFIG_SILENT_CONSOLE is set and CONFIG_SILENT_U_BOOT_ONLY is not set, the call to it should be based on the same conditions.
Signed-off-by: Paul B. Henson henson@acm.org --- common/cmd_bootm.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/common/cmd_bootm.c b/common/cmd_bootm.c index 046e22f..691d16b 100644 --- a/common/cmd_bootm.c +++ b/common/cmd_bootm.c @@ -636,7 +636,7 @@ static int do_bootm_states(cmd_tbl_t *cmdtp, int flag, int argc, goto err; else if (ret == BOOTM_ERR_OVERLAP) ret = 0; -#ifdef CONFIG_SILENT_CONSOLE +#if defined(CONFIG_SILENT_CONSOLE) && !defined(CONFIG_SILENT_U_BOOT_ONLY) if (images->os.os == IH_OS_LINUX) fixup_silent_linux(); #endif
Looks like this was broken in commit 35fc84fa1ff51e15ecd3e464dac87eb105ffed30, the refactor changed the conditional test when the code was moved.
On Sat, Aug 03, 2013 at 09:29:09PM -0700, Paul B. Henson wrote:
This function is only defined if CONFIG_SILENT_CONSOLE is set and CONFIG_SILENT_U_BOOT_ONLY is not set, the call to it should be based on the same conditions.
Signed-off-by: Paul B. Henson henson@acm.org
common/cmd_bootm.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/common/cmd_bootm.c b/common/cmd_bootm.c index 046e22f..691d16b 100644 --- a/common/cmd_bootm.c +++ b/common/cmd_bootm.c @@ -636,7 +636,7 @@ static int do_bootm_states(cmd_tbl_t *cmdtp, int flag, int argc, goto err; else if (ret == BOOTM_ERR_OVERLAP) ret = 0; -#ifdef CONFIG_SILENT_CONSOLE +#if defined(CONFIG_SILENT_CONSOLE) && !defined(CONFIG_SILENT_U_BOOT_ONLY) if (images->os.os == IH_OS_LINUX) fixup_silent_linux();
#endif
1.7.8.6
Hi Paul,
On Sun, Aug 4, 2013 at 1:59 PM, Paul B. Henson henson@acm.org wrote:
Looks like this was broken in commit 35fc84fa1ff51e15ecd3e464dac87eb105ffed30, the refactor changed the conditional test when the code was moved.
This looks right.
Acked-by: Simon Glass sjg@chromium.org
But which board does it actually break, please?
On Sat, Aug 03, 2013 at 09:29:09PM -0700, Paul B. Henson wrote:
This function is only defined if CONFIG_SILENT_CONSOLE is set and CONFIG_SILENT_U_BOOT_ONLY is not set, the call to it should be based on the same conditions.
Signed-off-by: Paul B. Henson henson@acm.org
common/cmd_bootm.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/common/cmd_bootm.c b/common/cmd_bootm.c index 046e22f..691d16b 100644 --- a/common/cmd_bootm.c +++ b/common/cmd_bootm.c @@ -636,7 +636,7 @@ static int do_bootm_states(cmd_tbl_t *cmdtp, int
flag, int argc,
goto err; else if (ret == BOOTM_ERR_OVERLAP) ret = 0;-#ifdef CONFIG_SILENT_CONSOLE +#if defined(CONFIG_SILENT_CONSOLE) &&
!defined(CONFIG_SILENT_U_BOOT_ONLY)
if (images->os.os == IH_OS_LINUX) fixup_silent_linux();#endif
1.7.8.6
Regards, Simon
On Sun, Aug 04, 2013 at 02:10:14PM -0600, Simon Glass wrote:
But which board does it actually break, please?
I don't think any of the default configs included in u-boot would have broken, while some of them define CONFIG_SILENT_CONSOLE, none of them define CONFIG_SILENT_U_BOOT_ONLY. I was working with a freescale mx28evk prototyping something that will eventually be a custom board, and I modified the config to drop u-boot output but not drop the linux kernel output. It was working fine with a git checkout from late May, and I noticed when I tried out the latest release it wouldn't compile anymore as it tried to call a symbol that was never defined.
So it was really only broken for a custom config, but as the behavior is documented in doc/README.silent, I assume it's still desirable for it to work :).
Thanks...
Hi Paul,
On Sun, Aug 4, 2013 at 8:43 PM, Paul B. Henson henson@acm.org wrote:
On Sun, Aug 04, 2013 at 02:10:14PM -0600, Simon Glass wrote:
But which board does it actually break, please?
I don't think any of the default configs included in u-boot would have broken, while some of them define CONFIG_SILENT_CONSOLE, none of them define CONFIG_SILENT_U_BOOT_ONLY. I was working with a freescale mx28evk prototyping something that will eventually be a custom board, and I modified the config to drop u-boot output but not drop the linux kernel output. It was working fine with a git checkout from late May, and I noticed when I tried out the latest release it wouldn't compile anymore as it tried to call a symbol that was never defined.
So it was really only broken for a custom config, but as the behavior is documented in doc/README.silent, I assume it's still desirable for it to work :).
OK, well at least that explains why I didn't see the problem. Thanks.
Thanks...
Regards, Simon
On Sat, Aug 03, 2013 at 09:29:09PM -0700, Paul B. Henson wrote:
This function is only defined if CONFIG_SILENT_CONSOLE is set and CONFIG_SILENT_U_BOOT_ONLY is not set, the call to it should be based on the same conditions.
Signed-off-by: Paul B. Henson henson@acm.org Acked-by: Simon Glass sjg@chromium.org
Applied to u-boot/master, thanks!
participants (3)
-
Paul B. Henson -
Simon Glass -
Tom Rini