Revert "hw/block/pflash_cfi0{1, 2}: Error out if device length isn't a power of two"
PATCH 1/1 Revert "hw/block/pflash_cfi0{1, 2}: Error out if device length isn't a power of two"
This commit caused a regression [1] that prevents machines that uses Open Virtual Machine Firmware (OVMF) to boot.
This is a long standing behavior with how pflash handles images. More information about why this happens can be found in [2] and commit 06f1521795 ("pflash: Require backend size to match device, improve errors").
This reverts commit 334c388f25707a234c4a0dea05b9df08d7746638.
[1] https://gitlab.com/qemu-project/qemu/-/issues/1294
[2] https://lore.kernel.org/qemu-devel/[email protected]/
Cc: Bernhard Beschow <[email protected]>
Cc: Philippe Mathieu-Daudé <[email protected]>
Cc: Stefan Hajnoczi <[email protected]>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1294
--- hw/block/pflash_cfi01.c | 8 ++------ hw/block/pflash_cfi02.c | 5 ----- 2 files changed, 2 insertions(+), 11 deletions(-)
hw/block/pflash_cfi01.c +2 -6
@@ -690,7 +690,7 @@ static const MemoryRegionOps pflash_cfi01_ops = { .endianness = DEVICE_NATIVE_ENDIAN, }; -static void pflash_cfi01_fill_cfi_table (PFlashCFI01 *pfl,Error **errp) +static void pflash_cfi01_fill_cfi_table (PFlashCFI01 *pfl) {uint64_t blocks_per_device, sector_len_per_device, device_len;int num_devices;
@@ -708,10 +708,6 @@ static void pflash_cfi01_fill_cfi_table(PFlashCFI01 *pfl, Error **errp) sector_len_per_device = pfl->sector_len / num_devices; } device_len = sector_len_per_device * blocks_per_device; -if (!is_power_of_2 (device_len)) { -error_setg (errp,"Device size must be a power of two." ); -return ; - }/* Hardcoded CFI table */ /* Standard "QRY" string */
@@ -869,7 +865,7 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)*/ pfl->cmd =0x00 ; pfl->status =0x80 ;/* WSM ready */ -pflash_cfi01_fill_cfi_table (pfl, errp); +pflash_cfi01_fill_cfi_table (pfl); }static void pflash_cfi01_system_reset (DeviceState *dev)
hw/block/pflash_cfi02.c +0 -5
@@ -880,11 +880,6 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)return ; } -if (!is_power_of_2 (pfl->chip_len)) { -error_setg (errp,"Device size must be a power of two." ); -return ; - } -memory_region_init_rom_device (&pfl->orig_mem,OBJECT (pfl), &pflash_cfi02_ops, pfl, pfl->name, pfl->chip_len, errp);
Signature · 1 line
2.37.3
Quoted 17 lines · Daniel Henrique Barboza
This commit caused a regression [1] that prevents machines that uses Open Virtual Machine Firmware (OVMF) to boot.
This is a long standing behavior with how pflash handles images. More information about why this happens can be found in [2] and commit
06f1521795("pflash: Require backend size to match device, improve errors").This reverts commit
334c388f25707a234c4a0dea05b9df08d7746638.[1] https://gitlab.com/qemu-project/qemu/-/issues/1294
[2] https://lore.kernel.org/qemu-devel/[email protected]/Cc: Bernhard Beschow <[email protected]>
Cc: Philippe Mathieu-Daudé <[email protected]>
Cc: Stefan Hajnoczi <[email protected]>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1294
Quoted 5 lines
Signed-off-by Daniel Henrique Barboza <[email protected]>--- hw/block/pflash_cfi01.c | 8 ++------ hw/block/pflash_cfi02.c | 5 ----- 2 files changed, 2 insertions(+), 11 deletions(-)
Thanks, our patches crossed :)
https://lore.kernel.org/qemu-devel/[email protected]/
I'm queuing yours which is first and will amend the description (if you don't disagree).
Regards,
Phil.
Quoted 25 lines · Philippe Mathieu-Daudé
Quoted 17 lines · Daniel Henrique Barboza
This commit caused a regression [1] that prevents machines that uses Open Virtual Machine Firmware (OVMF) to boot.
This is a long standing behavior with how pflash handles images. More information about why this happens can be found in [2] and commit
06f1521795("pflash: Require backend size to match device, improve errors").This reverts commit
334c388f25707a234c4a0dea05b9df08d7746638.[1] https://gitlab.com/qemu-project/qemu/-/issues/1294
[2] https://lore.kernel.org/qemu-devel/[email protected]/Cc: Bernhard Beschow <[email protected]>
Cc: Philippe Mathieu-Daudé <[email protected]>
Cc: Stefan Hajnoczi <[email protected]>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1294Quoted 5 lines
Signed-off-by Daniel Henrique Barboza <[email protected]>--- hw/block/pflash_cfi01.c | 8 ++------ hw/block/pflash_cfi02.c | 5 ----- 2 files changed, 2 insertions(+), 11 deletions(-)
Quoted 7 lines
Reviewed-by Philippe Mathieu-Daudé <[email protected]>Thanks, our patches crossed :)
https://lore.kernel.org/qemu-devel/[email protected]/I'm queuing yours which is first and will amend the description (if you don't disagree).
I've already applied yours, Philippe, because the description is more comprehensive.
Daniel, thank you for sending your version of the patch!
Stefan
Quoted 26 lines · Stefan Hajnoczi
Quoted 25 lines · Philippe Mathieu-Daudé
Quoted 22 lines · Daniel Henrique Barboza
This commit caused a regression [1] that prevents machines that uses Open Virtual Machine Firmware (OVMF) to boot.
This is a long standing behavior with how pflash handles images. More information about why this happens can be found in [2] and commit
06f1521795("pflash: Require backend size to match device, improve errors").This reverts commit
334c388f25707a234c4a0dea05b9df08d7746638.[1] https://gitlab.com/qemu-project/qemu/-/issues/1294
[2] https://lore.kernel.org/qemu-devel/[email protected]/Cc: Bernhard Beschow <[email protected]>
Cc: Philippe Mathieu-Daudé <[email protected]>
Cc: Stefan Hajnoczi <[email protected]>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1294Signed-off-by Daniel Henrique Barboza <[email protected]>--- hw/block/pflash_cfi01.c | 8 ++------ hw/block/pflash_cfi02.c | 5 ----- 2 files changed, 2 insertions(+), 11 deletions(-)
Quoted 4 lines
Quoted 4 lines
Reviewed-by Philippe Mathieu-Daudé <[email protected]>Thanks, our patches crossed :)
https://lore.kernel.org/qemu-devel/[email protected]/
Well of course that we both decided to send the PR almost at the same time :D
Quoted 8 lines
Quoted 3 lines
I'm queuing yours which is first and will amend the description (if you don't disagree).
I've already applied yours, Philippe, because the description is more comprehensive.
Daniel, thank you for sending your version of the patch!
Thank you Phil for sending the revert and Stefan for quickly queueing it. The sooner we get rid of the regression the better.
Daniel
Quoted 2 lines
Stefan