* [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
@ 2019-02-14 15:57 Alex Bennée
2019-02-14 19:54 ` Laszlo Ersek
0 siblings, 1 reply; 6+ messages in thread
From: Alex Bennée @ 2019-02-14 15:57 UTC (permalink / raw)
To: qemu-devel; +Cc: ard.biesheuvel, pkg-qemu-devel, Alex Bennée
It looks like there was going to be code to check we had some sort of
alignment so lets replace it with an actual check. This is a bit more
useful than the enigmatic "failed to read the initial flash content"
when we attempt to read the number of bytes the device should have.
This is a potential confusing stumbling block when you move from using
-bios to using -drive if=pflash,file=blob,format=raw,readonly for
loading your firmware code.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
hw/block/pflash_cfi01.c | 19 +++++++++++++------
1 file changed, 13 insertions(+), 6 deletions(-)
diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
index bffb4c40e7..f3251b236c 100644
--- a/hw/block/pflash_cfi01.c
+++ b/hw/block/pflash_cfi01.c
@@ -722,12 +722,19 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
}
device_len = sector_len_per_device * blocks_per_device;
- /* XXX: to be fixed */
-#if 0
- if (total_len != (8 * 1024 * 1024) && total_len != (16 * 1024 * 1024) &&
- total_len != (32 * 1024 * 1024) && total_len != (64 * 1024 * 1024))
- return NULL;
-#endif
+ /*
+ * Validate the backing store is the right size for pflash
+ * devices. It has to be padded to a multiple of the flash block
+ * size.
+ */
+ if (pfl->blk) {
+ uint64_t backing_len = blk_getlength(pfl->blk);
+ if (device_len != backing_len) {
+ error_setg(errp, "backing file wrong size "
+ "(%" PRId64 " != %" PRId64")", backing_len, device_len);
+ return;
+ }
+ }
memory_region_init_rom_device(
&pfl->mem, OBJECT(dev),
--
2.20.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
2019-02-14 15:57 [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned Alex Bennée
@ 2019-02-14 19:54 ` Laszlo Ersek
2019-02-14 22:38 ` Alex Bennée
0 siblings, 1 reply; 6+ messages in thread
From: Laszlo Ersek @ 2019-02-14 19:54 UTC (permalink / raw)
To: Alex Bennée, qemu-devel; +Cc: pkg-qemu-devel, ard.biesheuvel
On 02/14/19 16:57, Alex Bennée wrote:
> It looks like there was going to be code to check we had some sort of
> alignment so lets replace it with an actual check. This is a bit more
> useful than the enigmatic "failed to read the initial flash content"
> when we attempt to read the number of bytes the device should have.
>
> This is a potential confusing stumbling block when you move from using
> -bios to using -drive if=pflash,file=blob,format=raw,readonly for
> loading your firmware code.
>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
> hw/block/pflash_cfi01.c | 19 +++++++++++++------
> 1 file changed, 13 insertions(+), 6 deletions(-)
>
> diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
> index bffb4c40e7..f3251b236c 100644
> --- a/hw/block/pflash_cfi01.c
> +++ b/hw/block/pflash_cfi01.c
> @@ -722,12 +722,19 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
> }
> device_len = sector_len_per_device * blocks_per_device;
>
> - /* XXX: to be fixed */
> -#if 0
> - if (total_len != (8 * 1024 * 1024) && total_len != (16 * 1024 * 1024) &&
> - total_len != (32 * 1024 * 1024) && total_len != (64 * 1024 * 1024))
> - return NULL;
> -#endif
> + /*
> + * Validate the backing store is the right size for pflash
> + * devices. It has to be padded to a multiple of the flash block
> + * size.
> + */
> + if (pfl->blk) {
> + uint64_t backing_len = blk_getlength(pfl->blk);
> + if (device_len != backing_len) {
> + error_setg(errp, "backing file wrong size "
> + "(%" PRId64 " != %" PRId64")", backing_len, device_len);
> + return;
> + }
> + }
>
> memory_region_init_rom_device(
> &pfl->mem, OBJECT(dev),
>
I have two suggestions:
- backing_len and device_len are both uint64_t; we should print them
with PRIu64
- from a user POV, I find it more useful if the error message also shows
which quantity is which, not just two inequal numbers.
I don't feel too strongly about this, so up to you.
Thanks,
Laszlo
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
2019-02-14 19:54 ` Laszlo Ersek
@ 2019-02-14 22:38 ` Alex Bennée
2019-02-15 6:23 ` Geert Stappers
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Alex Bennée @ 2019-02-14 22:38 UTC (permalink / raw)
To: Laszlo Ersek; +Cc: qemu-devel, pkg-qemu-devel, ard.biesheuvel
Laszlo Ersek <lersek@redhat.com> writes:
> On 02/14/19 16:57, Alex Bennée wrote:
>> It looks like there was going to be code to check we had some sort of
>> alignment so lets replace it with an actual check. This is a bit more
>> useful than the enigmatic "failed to read the initial flash content"
>> when we attempt to read the number of bytes the device should have.
>>
>> This is a potential confusing stumbling block when you move from using
>> -bios to using -drive if=pflash,file=blob,format=raw,readonly for
>> loading your firmware code.
>>
>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>> ---
>> hw/block/pflash_cfi01.c | 19 +++++++++++++------
>> 1 file changed, 13 insertions(+), 6 deletions(-)
>>
>> diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
>> index bffb4c40e7..f3251b236c 100644
>> --- a/hw/block/pflash_cfi01.c
>> +++ b/hw/block/pflash_cfi01.c
>> @@ -722,12 +722,19 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
>> }
>> device_len = sector_len_per_device * blocks_per_device;
>>
>> - /* XXX: to be fixed */
>> -#if 0
>> - if (total_len != (8 * 1024 * 1024) && total_len != (16 * 1024 * 1024) &&
>> - total_len != (32 * 1024 * 1024) && total_len != (64 * 1024 * 1024))
>> - return NULL;
>> -#endif
>> + /*
>> + * Validate the backing store is the right size for pflash
>> + * devices. It has to be padded to a multiple of the flash block
>> + * size.
>> + */
>> + if (pfl->blk) {
>> + uint64_t backing_len = blk_getlength(pfl->blk);
>> + if (device_len != backing_len) {
>> + error_setg(errp, "backing file wrong size "
>> + "(%" PRId64 " != %" PRId64")", backing_len, device_len);
>> + return;
>> + }
>> + }
>>
>> memory_region_init_rom_device(
>> &pfl->mem, OBJECT(dev),
>>
>
> I have two suggestions:
> - backing_len and device_len are both uint64_t; we should print them
> with PRIu64
blk_getlength actually returns int64_t for some reason (do signed
lengths even make sense? maybe it's for error handling?). But sure I can
make it PRIu64
> - from a user POV, I find it more useful if the error message also shows
> which quantity is which, not just two inequal numbers.
How about:
"backing file size (%) not enough for whole device (%)"
?
>
> I don't feel too strongly about this, so up to you.
>
> Thanks,
> Laszlo
--
Alex Bennée
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
2019-02-14 22:38 ` Alex Bennée
@ 2019-02-15 6:23 ` Geert Stappers
2019-02-15 9:12 ` Markus Armbruster
2019-02-15 13:57 ` Laszlo Ersek
2 siblings, 0 replies; 6+ messages in thread
From: Geert Stappers @ 2019-02-15 6:23 UTC (permalink / raw)
To: Alex Bennée; +Cc: Laszlo Ersek, pkg-qemu-devel, qemu-devel, ard.biesheuvel
On Thu, Feb 14, 2019 at 10:38:35PM +0000, Alex Bennée wrote:
> Laszlo Ersek <lersek@redhat.com> writes:
> > On 02/14/19 16:57, Alex Bennée wrote:
> >> + /*
> >> + * Validate the backing store is the right size for pflash
> >> + * devices. It has to be padded to a multiple of the flash block
> >> + * size.
> >> + */
> >> + if (pfl->blk) {
> >> + uint64_t backing_len = blk_getlength(pfl->blk);
> >> + if (device_len != backing_len) {
> >> + error_setg(errp, "backing file wrong size "
> >> + "(%" PRId64 " != %" PRId64")", backing_len, device_len);
> >> + return;
> >> + }
> >> + }
> >>
> >> memory_region_init_rom_device(
> >> &pfl->mem, OBJECT(dev),
> >>
>
> > - from a user POV, I find it more useful if the error message also shows
> > which quantity is which, not just two inequal numbers.
>
> How about:
>
> "backing file size (%) not enough for whole device (%)"
>
> ?
>
from a user POV, I find it more useful if the error messsage
resembles the actual test in the source. Above is test !=
How about
"backing file size (%) not equal to whole device size (%)"
?
Groeten
Geert Stappers
--
Leven en laten leven
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
2019-02-14 22:38 ` Alex Bennée
2019-02-15 6:23 ` Geert Stappers
@ 2019-02-15 9:12 ` Markus Armbruster
2019-02-15 13:57 ` Laszlo Ersek
2 siblings, 0 replies; 6+ messages in thread
From: Markus Armbruster @ 2019-02-15 9:12 UTC (permalink / raw)
To: Alex Bennée; +Cc: Laszlo Ersek, pkg-qemu-devel, qemu-devel, ard.biesheuvel
Alex Bennée <alex.bennee@linaro.org> writes:
> Laszlo Ersek <lersek@redhat.com> writes:
>
>> On 02/14/19 16:57, Alex Bennée wrote:
>>> It looks like there was going to be code to check we had some sort of
>>> alignment so lets replace it with an actual check. This is a bit more
>>> useful than the enigmatic "failed to read the initial flash content"
>>> when we attempt to read the number of bytes the device should have.
>>>
>>> This is a potential confusing stumbling block when you move from using
>>> -bios to using -drive if=pflash,file=blob,format=raw,readonly for
>>> loading your firmware code.
>>>
>>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>>> ---
>>> hw/block/pflash_cfi01.c | 19 +++++++++++++------
>>> 1 file changed, 13 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
>>> index bffb4c40e7..f3251b236c 100644
>>> --- a/hw/block/pflash_cfi01.c
>>> +++ b/hw/block/pflash_cfi01.c
>>> @@ -722,12 +722,19 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
>>> }
>>> device_len = sector_len_per_device * blocks_per_device;
>>>
>>> - /* XXX: to be fixed */
>>> -#if 0
>>> - if (total_len != (8 * 1024 * 1024) && total_len != (16 * 1024 * 1024) &&
>>> - total_len != (32 * 1024 * 1024) && total_len != (64 * 1024 * 1024))
>>> - return NULL;
>>> -#endif
>>> + /*
>>> + * Validate the backing store is the right size for pflash
>>> + * devices. It has to be padded to a multiple of the flash block
>>> + * size.
>>> + */
>>> + if (pfl->blk) {
>>> + uint64_t backing_len = blk_getlength(pfl->blk);
>>> + if (device_len != backing_len) {
>>> + error_setg(errp, "backing file wrong size "
>>> + "(%" PRId64 " != %" PRId64")", backing_len, device_len);
>>> + return;
>>> + }
>>> + }
>>>
>>> memory_region_init_rom_device(
>>> &pfl->mem, OBJECT(dev),
>>>
>>
>> I have two suggestions:
>> - backing_len and device_len are both uint64_t; we should print them
>> with PRIu64
>
> blk_getlength actually returns int64_t for some reason (do signed
> lengths even make sense? maybe it's for error handling?). But sure I can
> make it PRIu64
Use of signed integers for file offsets is pervasive in the block layer.
It's convenient when we return either a non-negative offset or a
negative error code. It's admittedly sloppy anywhere else.
>> - from a user POV, I find it more useful if the error message also shows
>> which quantity is which, not just two inequal numbers.
>
> How about:
>
> "backing file size (%) not enough for whole device (%)"
>
> ?
Not bad. Another one, avoiding parentheses:
"device needs DDD bytes, backing file provides only BBB bytes"
>
>>
>> I don't feel too strongly about this, so up to you.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned
2019-02-14 22:38 ` Alex Bennée
2019-02-15 6:23 ` Geert Stappers
2019-02-15 9:12 ` Markus Armbruster
@ 2019-02-15 13:57 ` Laszlo Ersek
2 siblings, 0 replies; 6+ messages in thread
From: Laszlo Ersek @ 2019-02-15 13:57 UTC (permalink / raw)
To: Alex Bennée; +Cc: qemu-devel, pkg-qemu-devel, ard.biesheuvel
On 02/14/19 23:38, Alex Bennée wrote:
>
> Laszlo Ersek <lersek@redhat.com> writes:
>
>> On 02/14/19 16:57, Alex Bennée wrote:
>>> It looks like there was going to be code to check we had some sort of
>>> alignment so lets replace it with an actual check. This is a bit more
>>> useful than the enigmatic "failed to read the initial flash content"
>>> when we attempt to read the number of bytes the device should have.
>>>
>>> This is a potential confusing stumbling block when you move from using
>>> -bios to using -drive if=pflash,file=blob,format=raw,readonly for
>>> loading your firmware code.
>>>
>>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>>> ---
>>> hw/block/pflash_cfi01.c | 19 +++++++++++++------
>>> 1 file changed, 13 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
>>> index bffb4c40e7..f3251b236c 100644
>>> --- a/hw/block/pflash_cfi01.c
>>> +++ b/hw/block/pflash_cfi01.c
>>> @@ -722,12 +722,19 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
>>> }
>>> device_len = sector_len_per_device * blocks_per_device;
>>>
>>> - /* XXX: to be fixed */
>>> -#if 0
>>> - if (total_len != (8 * 1024 * 1024) && total_len != (16 * 1024 * 1024) &&
>>> - total_len != (32 * 1024 * 1024) && total_len != (64 * 1024 * 1024))
>>> - return NULL;
>>> -#endif
>>> + /*
>>> + * Validate the backing store is the right size for pflash
>>> + * devices. It has to be padded to a multiple of the flash block
>>> + * size.
>>> + */
>>> + if (pfl->blk) {
>>> + uint64_t backing_len = blk_getlength(pfl->blk);
>>> + if (device_len != backing_len) {
>>> + error_setg(errp, "backing file wrong size "
>>> + "(%" PRId64 " != %" PRId64")", backing_len, device_len);
>>> + return;
>>> + }
>>> + }
>>>
>>> memory_region_init_rom_device(
>>> &pfl->mem, OBJECT(dev),
>>>
>>
>> I have two suggestions:
>> - backing_len and device_len are both uint64_t; we should print them
>> with PRIu64
>
> blk_getlength actually returns int64_t for some reason (do signed
> lengths even make sense? maybe it's for error handling?).
Ah, sorry, I didn't realize. I guess we should cover negative values
then explicitly? Not sure.
> But sure I can
> make it PRIu64
>
>> - from a user POV, I find it more useful if the error message also shows
>> which quantity is which, not just two inequal numbers.
>
> How about:
>
> "backing file size (%) not enough for whole device (%)"
"not enough" means "<", but the C expression uses "!=". How about
"backing file size (...) does not match device size (...)"?
Or "does not equal", whichever sounds more palatable.
Thanks!
Laszlo
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-02-15 13:57 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-02-14 15:57 [Qemu-devel] [PATCH] hw/block: report when pflash backing file isn't aligned Alex Bennée
2019-02-14 19:54 ` Laszlo Ersek
2019-02-14 22:38 ` Alex Bennée
2019-02-15 6:23 ` Geert Stappers
2019-02-15 9:12 ` Markus Armbruster
2019-02-15 13:57 ` Laszlo Ersek
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).