[PATCH v1 0/2] imx: spl_imx_romapi: fixes
From: Marcel Ziswiler marcel.ziswiler@toradex.com
The first patch is a trivial spelling fix in comments. The second fixes a recent regression that only manifests in the eMMC fast boot mode case which does actually also use streaming rather than seekable.
Thanks go to Rasmus who helped me debug this and suggested the fix.
@Stefano: This one is also still missing [1]. Thanks!
[1] https://lore.kernel.org/all/20230828213803.454303-1-marcel@ziswiler.com
Marcel Ziswiler (2): imx: spl_imx_romapi: fix comment about stream(usb) download failure imx: spl_imx_romapi: fix emmc fast boot mode case
arch/arm/mach-imx/spl_imx_romapi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
From: Marcel Ziswiler marcel.ziswiler@toradex.com
Fix comment about Stream(USB) download failure.
Fixes: 1cbebc786276 ("imx: add rom api support") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com ---
arch/arm/mach-imx/spl_imx_romapi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/mach-imx/spl_imx_romapi.c b/arch/arm/mach-imx/spl_imx_romapi.c index 93d48e56aca..cb220869b50 100644 --- a/arch/arm/mach-imx/spl_imx_romapi.c +++ b/arch/arm/mach-imx/spl_imx_romapi.c @@ -285,7 +285,7 @@ static int spl_romapi_load_image_stream(struct spl_image_info *spl_image, ret = rom_api_download_image(p, 0, pg);
if (ret != ROM_API_OKAY) { - puts("Steam(USB) download failure\n"); + puts("Stream(USB) download failure\n"); return -1; }
@@ -305,7 +305,7 @@ static int spl_romapi_load_image_stream(struct spl_image_info *spl_image, ret = rom_api_download_image(p, 0, pg);
if (ret != ROM_API_OKAY) { - puts("Steam(USB) download failure\n"); + puts("Stream(USB) download failure\n"); return -1; }
On Thu, Oct 26, 2023 at 4:32 AM Marcel Ziswiler marcel@ziswiler.com wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
Fix comment about Stream(USB) download failure.
Fixes: 1cbebc786276 ("imx: add rom api support") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Reviewed-by: Fabio Estevam festevam@gmail.com
On Thu, Oct 26, 2023 at 09:32:19AM +0200, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
Fix comment about Stream(USB) download failure.
Fixes: 1cbebc786276 ("imx: add rom api support") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com Reviewed-by: Fabio Estevam festevam@gmail.com
Applied to u-boot/master, thanks!
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
---
arch/arm/mach-imx/spl_imx_romapi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm/mach-imx/spl_imx_romapi.c b/arch/arm/mach-imx/spl_imx_romapi.c index cb220869b50..5eb5a3d3c4a 100644 --- a/arch/arm/mach-imx/spl_imx_romapi.c +++ b/arch/arm/mach-imx/spl_imx_romapi.c @@ -162,7 +162,7 @@ static ulong spl_romapi_read_stream(struct spl_load_info *load, ulong sector, return 0; }
- ss->end = end; + ss->end += bytes; }
memcpy(buf, (void *)(sector), count);
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Regards, Stefano
Hi Stefano
On Thu, 2023-10-26 at 10:36 +0200, Stefano Babic wrote:
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Any news on applying this fix? Note that master still stays broken as of today.
Regards, Stefano
Cheers
Marcel
Hi Tom
On Thu, 2023-11-02 at 08:09 +0100, Marcel Ziswiler wrote:
Hi Stefano
On Thu, 2023-10-26 at 10:36 +0200, Stefano Babic wrote:
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Any news on applying this fix? Note that master still stays broken as of today.
And two weeks later still broken!
Anyway, I feel i.MX maintainership seems rather stalled at times most of the time. I don't want to criticise anybody knowing how busy our trade can get. However, if it helps the cause Toradex would be ready to step up. Let me know what you think. Thanks!
Regards, Stefano
Cheers
Marcel
On Thu, Nov 16, 2023 at 10:57:27AM +0000, Marcel Ziswiler wrote:
Hi Tom
On Thu, 2023-11-02 at 08:09 +0100, Marcel Ziswiler wrote:
Hi Stefano
On Thu, 2023-10-26 at 10:36 +0200, Stefano Babic wrote:
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Any news on applying this fix? Note that master still stays broken as of today.
And two weeks later still broken!
I've put this in my not to forget queue for the next -rc.
Anyway, I feel i.MX maintainership seems rather stalled at times most of the time. I don't want to criticise anybody knowing how busy our trade can get. However, if it helps the cause Toradex would be ready to step up. Let me know what you think. Thanks!
I would be happy to see an additional custodian for the imx tree, if that's agreeable with Stefano.
Hi Tom, Marcel,
On 16.11.23 19:42, Tom Rini wrote:
On Thu, Nov 16, 2023 at 10:57:27AM +0000, Marcel Ziswiler wrote:
Hi Tom
On Thu, 2023-11-02 at 08:09 +0100, Marcel Ziswiler wrote:
Hi Stefano
On Thu, 2023-10-26 at 10:36 +0200, Stefano Babic wrote:
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Any news on applying this fix? Note that master still stays broken as of today.
And two weeks later still broken!
I've put this in my not to forget queue for the next -rc.
Anyway, I feel i.MX maintainership seems rather stalled at times most of the time. I don't want to criticise anybody knowing how busy our trade can get. However, if it helps the cause Toradex would be ready to step up. Let me know what you think. Thanks!
I would be happy to see an additional custodian for the imx tree, if that's agreeable with Stefano.
I have rather less time for U-Boot and it won't be better in the near future. Fabio is co-maintainer and he has accepted to take over the repository and go on maintaining u-boot-imx. Just let him some time to be confident with patchwork and CI.
Thanks, Stefano
On Fri, Nov 17, 2023 at 09:17:27AM +0100, Stefano Babic wrote:
Hi Tom, Marcel,
On 16.11.23 19:42, Tom Rini wrote:
On Thu, Nov 16, 2023 at 10:57:27AM +0000, Marcel Ziswiler wrote:
Hi Tom
On Thu, 2023-11-02 at 08:09 +0100, Marcel Ziswiler wrote:
Hi Stefano
On Thu, 2023-10-26 at 10:36 +0200, Stefano Babic wrote:
On 26.10.23 10:29, Rasmus Villemoes wrote:
On 26/10/2023 09.32, Marcel Ziswiler wrote: > From: Marcel Ziswiler marcel.ziswiler@toradex.com > > This fixes a regression in the eMMC fast boot mode case where the buffer > was missing 464 bytes. > > The code figures out how many bytes must at least be fetched to honor > the current read, rounds that up to the ss->pagesize [which is a no-op > in the USB download case because that has ->pagesize==1], fetches that > many bytes, but then recorded the original upper bound as the new end of > the valid data. However, this did not take into account the rounding up > to the ss->pagesize. Fix this by recording the actual bytes downloaded. > > Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT > size") > Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com >
Thanks for reporting and fixing this, and sorry for the trouble.
Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk
And thanks for all your work - I will apply this fix soon.
Any news on applying this fix? Note that master still stays broken as of today.
And two weeks later still broken!
I've put this in my not to forget queue for the next -rc.
Anyway, I feel i.MX maintainership seems rather stalled at times most of the time. I don't want to criticise anybody knowing how busy our trade can get. However, if it helps the cause Toradex would be ready to step up. Let me know what you think. Thanks!
I would be happy to see an additional custodian for the imx tree, if that's agreeable with Stefano.
I have rather less time for U-Boot and it won't be better in the near future. Fabio is co-maintainer and he has accepted to take over the repository and go on maintaining u-boot-imx. Just let him some time to be confident with patchwork and CI.
OK. Fabio, please send me your patchwork username off-list and I'll get you added and you can start managing patches there. I've started using some of the b4 + patchwork integration just recently and it's very nice.
Hi Tom,
On Fri, Nov 17, 2023 at 11:25 AM Tom Rini trini@konsulko.com wrote:
OK. Fabio, please send me your patchwork username off-list and I'll get you added and you can start managing patches there. I've started using some of the b4 + patchwork integration just recently and it's very nice.
In the meantime, could you pleas directly apply this series from Marcel?
Thanks
On Fri, Nov 17, 2023 at 01:13:07PM -0300, Fabio Estevam wrote:
Hi Tom,
On Fri, Nov 17, 2023 at 11:25 AM Tom Rini trini@konsulko.com wrote:
OK. Fabio, please send me your patchwork username off-list and I'll get you added and you can start managing patches there. I've started using some of the b4 + patchwork integration just recently and it's very nice.
In the meantime, could you pleas directly apply this series from Marcel?
I've put it in my list for before the next -rc.
On Thu, Oct 26, 2023 at 4:32 AM Marcel Ziswiler marcel@ziswiler.com wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com
Thanks for testing mainline U-Boot and fixing it:
Reviewed-by: Fabio Estevam festevam@gmail.com
On Thu, Oct 26, 2023 at 09:32:20AM +0200, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk Reviewed-by: Fabio Estevam festevam@gmail.com
Applied to u-boot/master, thanks!
On Fri, 2023-11-17 at 14:42 -0500, Tom Rini wrote:
On Thu, Oct 26, 2023 at 09:32:20AM +0200, Marcel Ziswiler wrote:
From: Marcel Ziswiler marcel.ziswiler@toradex.com
This fixes a regression in the eMMC fast boot mode case where the buffer was missing 464 bytes.
The code figures out how many bytes must at least be fetched to honor the current read, rounds that up to the ss->pagesize [which is a no-op in the USB download case because that has ->pagesize==1], fetches that many bytes, but then recorded the original upper bound as the new end of the valid data. However, this did not take into account the rounding up to the ss->pagesize. Fix this by recording the actual bytes downloaded.
Fixes: 4b4472438f5a ("imx: spl_imx_romapi: avoid tricky use of spl_load_simple_fit() to get full FIT size") Signed-off-by: Marcel Ziswiler marcel.ziswiler@toradex.com Acked-by: Rasmus Villemoes rasmus.villemoes@prevas.dk Reviewed-by: Fabio Estevam festevam@gmail.com
Applied to u-boot/master, thanks!
Finally our weekly master/upstream CI is passing on all i.MX 8M Plus SKUs again!
Thanks!
Cheers
Marcel
participants (6)
-
Fabio Estevam -
Marcel Ziswiler -
Marcel Ziswiler -
Rasmus Villemoes -
Stefano Babic -
Tom Rini