Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
On 13/12/2022 15:31, Quentin Schulz wrote:
Hi,
On 12/13/22 15:14, Neil Armstrong wrote:
On 13/12/2022 14:45, Quentin Schulz wrote:
Hi Tom,
On 12/13/22 14:37, Tom Rini wrote:
This reverts commits
51c5c28af59c ("cmd: pxe: use strdup to copy config"),
a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
As reported by Quentin Schulz, this introduces a regression with
previously generated and working extlinux.conf files.
The situation is much complex than that:
- Since the commit d5ba6188dfbf ("cmd: pxe_utils: Check fdtcontroladdr in
label_boot"),\
all the extlinux.conf with only "KERNEL /fitImage" don't work anymore.
Those are generated by Poky/OE WIC bootimg-partition bootloader partition
generator.
- With the changeset introduced by Patrick, this doesn't permit booting
those images, but it solves the issue by explicitly specifying we want to
get the dtb from the fitImage with "FDT /fitImage", which is smart.
- If we start generating extlinux.conf with "FDT /fitImage", those won't
boot on previous u-boot versions (as I reported in the second link shared by
Quentin).
But honestly, as of today we need to patch by reverting d5ba6188dfbf to
use extlinux.conf with fitImage, so I'll vote to keep this solution and
go forward and implement the change in the Poky/OE bootimg-partition
bootloader partition generator with perhaps a check on the u-boot version.
Yocto could do this, though I'm not entirely sure they'll like it either. But
you have plenty other distributions, and I imagine it is perfectly possible it
does not have knowledge of the bootloader being used to boot the distribution.
I'm voting against because currently I see d5ba6188dfbf as a regression and
distros are free to handle them how they wish, by e.g. reverting this patch
only (which I have done for our boards for example). I'd expect Yocto to have a
local patch reverting this in their next release, where U-Boot would be bumped
to something containing this commit, for example.
By merging the patchset, we limit other possible implementations by adding one
more logic we should then maintain for backward compatibility.
Ack, you're right.
Neil
Cheers,
Quentin
Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
Hi,
On 12/13/22 15:14, Neil Armstrong wrote:
On 13/12/2022 14:45, Quentin Schulz wrote:
Hi Tom,
On 12/13/22 14:37, Tom Rini wrote:
This reverts commits
51c5c28af59c ("cmd: pxe: use strdup to copy config"),
a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
As reported by Quentin Schulz, this introduces a regression with
previously generated and working extlinux.conf files.
The situation is much complex than that:
- Since the commit d5ba6188dfbf ("cmd: pxe_utils: Check fdtcontroladdr
in label_boot"),\
all the extlinux.conf with only "KERNEL /fitImage" don't work anymore.
Those are generated by Poky/OE WIC bootimg-partition bootloader
partition generator.
- With the changeset introduced by Patrick, this doesn't permit booting
those images, but it solves the issue by explicitly specifying we want to
get the dtb from the fitImage with "FDT /fitImage", which is smart.
- If we start generating extlinux.conf with "FDT /fitImage", those won't
boot on previous u-boot versions (as I reported in the second link
shared by Quentin).
But honestly, as of today we need to patch by reverting d5ba6188dfbf to
use extlinux.conf with fitImage, so I'll vote to keep this solution and
go forward and implement the change in the Poky/OE bootimg-partition
bootloader partition generator with perhaps a check on the u-boot version.
Yocto could do this, though I'm not entirely sure they'll like it
either. But you have plenty other distributions, and I imagine it is
perfectly possible it does not have knowledge of the bootloader being
used to boot the distribution.
I'm voting against because currently I see d5ba6188dfbf as a regression
and distros are free to handle them how they wish, by e.g. reverting
this patch only (which I have done for our boards for example). I'd
expect Yocto to have a local patch reverting this in their next release,
where U-Boot would be bumped to something containing this commit, for
example.
By merging the patchset, we limit other possible implementations by
adding one more logic we should then maintain for backward compatibility.
Cheers,
Quentin
Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
On 13/12/2022 15:22, Tom Rini wrote:
On Tue, Dec 13, 2022 at 03:14:43PM +0100, Neil Armstrong wrote:
On 13/12/2022 14:45, Quentin Schulz wrote:
Hi Tom,
On 12/13/22 14:37, Tom Rini wrote:
This reverts commits
51c5c28af59c ("cmd: pxe: use strdup to copy config"),
a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
As reported by Quentin Schulz, this introduces a regression with
previously generated and working extlinux.conf files.
The situation is much complex than that:
- Since the commit d5ba6188dfbf ("cmd: pxe_utils: Check fdtcontroladdr in
label_boot"),\
all the extlinux.conf with only "KERNEL /fitImage" don't work anymore.
Those are generated by Poky/OE WIC bootimg-partition bootloader partition
generator.
- With the changeset introduced by Patrick, this doesn't permit booting
those images, but it solves the issue by explicitly specifying we want to
get the dtb from the fitImage with "FDT /fitImage", which is smart.
- If we start generating extlinux.conf with "FDT /fitImage", those won't
boot on previous u-boot versions (as I reported in the second link shared by
Quentin).
But honestly, as of today we need to patch by reverting d5ba6188dfbf to
use extlinux.conf with fitImage, so I'll vote to keep this solution and
go forward and implement the change in the Poky/OE bootimg-partition
bootloader partition generator with perhaps a check on the u-boot version.
So, I'll post a revert for d5ba6188dfbf in a moment then, which of these
should I use for Link's
For reference:
https://lore.kernel.org/u-boot/b7e891d1-d134-b489-eb2d-6125d4c7b...@theobroma-systems.com/
for my worry
https://lore.kernel.org/u-boot/f0dd213c-4a34-926d-3f3b-f2ed49bb9...@linaro.org/
for confirmation from Neil it is breaking existing extlinux.conf
Or something else? And then to be clear, you're saying the 3 changes
here don't break anything really?
Yea no they will only break old u-boot with future extlinux.conf files with "FDT
/fitImage",
but without booting with fitImage as KERNEL is already broken since
d5ba6188dfbf anyway.
Neil
Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
On Tue, Dec 13, 2022 at 03:14:43PM +0100, Neil Armstrong wrote:
> On 13/12/2022 14:45, Quentin Schulz wrote:
> > Hi Tom,
> >
> > On 12/13/22 14:37, Tom Rini wrote:
> > > This reverts commits
> > > 51c5c28af59c ("cmd: pxe: use strdup to copy config"),
> > > a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
> > > f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
> > >
> > > As reported by Quentin Schulz, this introduces a regression with
> > > previously generated and working extlinux.conf files.
> > >
>
> The situation is much complex than that:
> - Since the commit d5ba6188dfbf ("cmd: pxe_utils: Check fdtcontroladdr in
> label_boot"),\
> all the extlinux.conf with only "KERNEL /fitImage" don't work anymore.
> Those are generated by Poky/OE WIC bootimg-partition bootloader partition
> generator.
> - With the changeset introduced by Patrick, this doesn't permit booting
> those images, but it solves the issue by explicitly specifying we want to
> get the dtb from the fitImage with "FDT /fitImage", which is smart.
> - If we start generating extlinux.conf with "FDT /fitImage", those won't
> boot on previous u-boot versions (as I reported in the second link shared by
> Quentin).
>
> But honestly, as of today we need to patch by reverting d5ba6188dfbf to
> use extlinux.conf with fitImage, so I'll vote to keep this solution and
> go forward and implement the change in the Poky/OE bootimg-partition
> bootloader partition generator with perhaps a check on the u-boot version.
So, I'll post a revert for d5ba6188dfbf in a moment then, which of these
should I use for Link's
> > For reference:
> > https://lore.kernel.org/u-boot/b7e891d1-d134-b489-eb2d-6125d4c7b...@theobroma-systems.com/
> > for my worry
> > https://lore.kernel.org/u-boot/f0dd213c-4a34-926d-3f3b-f2ed49bb9...@linaro.org/
> > for confirmation from Neil it is breaking existing extlinux.conf
Or something else? And then to be clear, you're saying the 3 changes
here don't break anything really?
--
Tom
signature.asc
Description: PGP signature
Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
On 13/12/2022 14:45, Quentin Schulz wrote:
Hi Tom,
On 12/13/22 14:37, Tom Rini wrote:
This reverts commits
51c5c28af59c ("cmd: pxe: use strdup to copy config"),
a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
As reported by Quentin Schulz, this introduces a regression with
previously generated and working extlinux.conf files.
The situation is much complex than that:
- Since the commit d5ba6188dfbf ("cmd: pxe_utils: Check fdtcontroladdr in
label_boot"),\
all the extlinux.conf with only "KERNEL /fitImage" don't work anymore.
Those are generated by Poky/OE WIC bootimg-partition bootloader partition
generator.
- With the changeset introduced by Patrick, this doesn't permit booting
those images, but it solves the issue by explicitly specifying we want to
get the dtb from the fitImage with "FDT /fitImage", which is smart.
- If we start generating extlinux.conf with "FDT /fitImage", those won't
boot on previous u-boot versions (as I reported in the second link shared by
Quentin).
But honestly, as of today we need to patch by reverting d5ba6188dfbf to
use extlinux.conf with fitImage, so I'll vote to keep this solution and
go forward and implement the change in the Poky/OE bootimg-partition
bootloader partition generator with perhaps a check on the u-boot version.
Neil
For reference:
https://lore.kernel.org/u-boot/b7e891d1-d134-b489-eb2d-6125d4c7b...@theobroma-systems.com/
for my worry
https://lore.kernel.org/u-boot/f0dd213c-4a34-926d-3f3b-f2ed49bb9...@linaro.org/
for confirmation from Neil it is breaking existing extlinux.conf
Thanks,
Quentin
Cc: Neil Armstrong
Cc: Patrick Delaunay
Reported-by: Quentin Schulz
Signed-off-by: Tom Rini
---
boot/pxe_utils.c | 69 -
doc/README.pxe | 8 --
include/pxe_utils.h | 2 --
3 files changed, 30 insertions(+), 49 deletions(-)
diff --git a/boot/pxe_utils.c b/boot/pxe_utils.c
index 272431381763..075a0f28309e 100644
--- a/boot/pxe_utils.c
+++ b/boot/pxe_utils.c
@@ -258,7 +258,6 @@ static struct pxe_label *label_create(void)
static void label_destroy(struct pxe_label *label)
{
free(label->name);
- free(label->kernel_label);
free(label->kernel);
free(label->config);
free(label->append);
@@ -522,44 +521,28 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
return 1;
}
- if (get_relfile_envaddr(ctx, label->kernel, "kernel_addr_r",
- NULL) < 0) {
- printf("Skipping %s for failure retrieving kernel\n",
- label->name);
- return 1;
- }
-
- kernel_addr = env_get("kernel_addr_r");
- /* for FIT, append the configuration identifier */
- if (label->config) {
- int len = strlen(kernel_addr) + strlen(label->config) + 1;
-
- fit_addr = malloc(len);
- if (!fit_addr) {
- printf("malloc fail (FIT address)\n");
- return 1;
- }
- snprintf(fit_addr, len, "%s%s", kernel_addr, label->config);
- kernel_addr = fit_addr;
- }
-
- /* For FIT, the label can be identical to kernel one */
- if (label->initrd && !strcmp(label->kernel_label, label->initrd)) {
- initrd_addr_str = kernel_addr;
- } else if (label->initrd) {
+ if (label->initrd) {
ulong size;
+
if (get_relfile_envaddr(ctx, label->initrd, "ramdisk_addr_r",
&size) < 0) {
printf("Skipping %s for failure retrieving initrd\n",
label->name);
- goto cleanup;
+ return 1;
}
initrd_addr_str = env_get("ramdisk_addr_r");
size = snprintf(initrd_str, sizeof(initrd_str), "%s:%lx",
initrd_addr_str, size);
if (size >= sizeof(initrd_str))
- goto cleanup;
+ return 1;
+ }
+
+ if (get_relfile_envaddr(ctx, label->kernel, "kernel_addr_r",
+ NULL) < 0) {
+ printf("Skipping %s for failure retrieving kernel\n",
+ label->name);
+ return 1;
}
if (label->ipappend & 0x1) {
@@ -589,7 +572,7 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
strlen(label->append ?: ""),
strlen(ip_str), strlen(mac_str),
sizeof(bootargs));
- goto cleanup;
+ return 1;
}
if (label->append)
@@ -604,6 +587,21 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
printf("append: %s\n", finalbootargs);
}
+ kernel_addr = env_get("kernel_addr_r");
+
+ /* for FIT, append the configuration identifier */
+ if (label->config) {
+ int len = strlen(kernel_addr) + strlen(label->config) + 1;
+
+ fit_addr = malloc(len);
+ if (!fit_addr) {
+ printf("malloc fail (FIT address)\n");
+
Re: [PATCH] Revert "cmd: pxe: use strdup to copy config" et al
Hi Tom,
On 12/13/22 14:37, Tom Rini wrote:
This reverts commits
51c5c28af59c ("cmd: pxe: use strdup to copy config"),
a5dacef7380e ("cmd: pxe: support INITRD and FDT selection with FIT") and
f723c2778cf8 ("cmd: pxe: reorder kernel treatment in label_boot").
As reported by Quentin Schulz, this introduces a regression with
previously generated and working extlinux.conf files.
For reference:
https://lore.kernel.org/u-boot/b7e891d1-d134-b489-eb2d-6125d4c7b...@theobroma-systems.com/
for my worry
https://lore.kernel.org/u-boot/f0dd213c-4a34-926d-3f3b-f2ed49bb9...@linaro.org/
for confirmation from Neil it is breaking existing extlinux.conf
Thanks,
Quentin
Cc: Neil Armstrong
Cc: Patrick Delaunay
Reported-by: Quentin Schulz
Signed-off-by: Tom Rini
---
boot/pxe_utils.c| 69 -
doc/README.pxe | 8 --
include/pxe_utils.h | 2 --
3 files changed, 30 insertions(+), 49 deletions(-)
diff --git a/boot/pxe_utils.c b/boot/pxe_utils.c
index 272431381763..075a0f28309e 100644
--- a/boot/pxe_utils.c
+++ b/boot/pxe_utils.c
@@ -258,7 +258,6 @@ static struct pxe_label *label_create(void)
static void label_destroy(struct pxe_label *label)
{
free(label->name);
- free(label->kernel_label);
free(label->kernel);
free(label->config);
free(label->append);
@@ -522,44 +521,28 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
return 1;
}
- if (get_relfile_envaddr(ctx, label->kernel, "kernel_addr_r",
- NULL) < 0) {
- printf("Skipping %s for failure retrieving kernel\n",
- label->name);
- return 1;
- }
-
- kernel_addr = env_get("kernel_addr_r");
- /* for FIT, append the configuration identifier */
- if (label->config) {
- int len = strlen(kernel_addr) + strlen(label->config) + 1;
-
- fit_addr = malloc(len);
- if (!fit_addr) {
- printf("malloc fail (FIT address)\n");
- return 1;
- }
- snprintf(fit_addr, len, "%s%s", kernel_addr, label->config);
- kernel_addr = fit_addr;
- }
-
- /* For FIT, the label can be identical to kernel one */
- if (label->initrd && !strcmp(label->kernel_label, label->initrd)) {
- initrd_addr_str = kernel_addr;
- } else if (label->initrd) {
+ if (label->initrd) {
ulong size;
+
if (get_relfile_envaddr(ctx, label->initrd, "ramdisk_addr_r",
&size) < 0) {
printf("Skipping %s for failure retrieving initrd\n",
label->name);
- goto cleanup;
+ return 1;
}
initrd_addr_str = env_get("ramdisk_addr_r");
size = snprintf(initrd_str, sizeof(initrd_str), "%s:%lx",
initrd_addr_str, size);
if (size >= sizeof(initrd_str))
- goto cleanup;
+ return 1;
+ }
+
+ if (get_relfile_envaddr(ctx, label->kernel, "kernel_addr_r",
+ NULL) < 0) {
+ printf("Skipping %s for failure retrieving kernel\n",
+ label->name);
+ return 1;
}
if (label->ipappend & 0x1) {
@@ -589,7 +572,7 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
strlen(label->append ?: ""),
strlen(ip_str), strlen(mac_str),
sizeof(bootargs));
- goto cleanup;
+ return 1;
}
if (label->append)
@@ -604,6 +587,21 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
printf("append: %s\n", finalbootargs);
}
+ kernel_addr = env_get("kernel_addr_r");
+
+ /* for FIT, append the configuration identifier */
+ if (label->config) {
+ int len = strlen(kernel_addr) + strlen(label->config) + 1;
+
+ fit_addr = malloc(len);
+ if (!fit_addr) {
+ printf("malloc fail (FIT address)\n");
+ return 1;
+ }
+ snprintf(fit_addr, len, "%s%s", kernel_addr, label->config);
+ kernel_addr = fit_addr;
+ }
+
/*
* fdt usage is optional:
* It handles the following scenarios.
@@ -625,11 +623,8 @@ static int label_boot(struct pxe_context *ctx, struct
pxe_label *label)
*/
bootm_argv[3] = env_get("fdt_addr_r");
- /* For FIT, the label can be identical to kernel one */
- if (label->fdt && !strcmp(label->kernel_label, label->fdt)) {
- bootm_argv[3]
