Message ID | 20190226193408.23862-9-armbru@redhat.com |
---|---|
State | New |
Headers | show |
Series | pflash: Fixes and cleanups | expand |
On 2/26/19 8:34 PM, Markus Armbruster wrote: > pflash_cfi02_register() takes a size in bytes, a block size in bytes > and a number of blocks. r2d_init() passes FLASH_SIZE, 16 * KiB, > FLASH_SIZE >> 16. Does not compute: size doesn't match block size * > number of blocks. The latter happens to win. I tried to find > documentation on the physical hardware, no luck. > > For now, adjust the byte size passed to match the actual size created, > and add a FIXME comment. > > Cc: Magnus Damm <magnus.damm@gmail.com> > Signed-off-by: Markus Armbruster <armbru@redhat.com> > --- > hw/sh4/r2d.c | 9 ++++++++- > 1 file changed, 8 insertions(+), 1 deletion(-) > > diff --git a/hw/sh4/r2d.c b/hw/sh4/r2d.c > index dcdb3728cb..ed18d1f351 100644 > --- a/hw/sh4/r2d.c > +++ b/hw/sh4/r2d.c > @@ -290,7 +290,14 @@ static void r2d_init(MachineState *machine) > > /* onboard flash memory */ > dinfo = drive_get(IF_PFLASH, 0, 0); > - pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE, > + /* > + * FIXME The code is confused about the size of the flash. It > + * used to pass FLASH_SIZE bytes, in FLASH_SIZE >> 16 blocks of > + * 16KiB each, which does not compute, but creates one of > + * FLASH_SIZE / 4 bytes anyway. The current code does so too, but > + * whether it's the right size is anybody's guess. > + */ > + pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE / 4, NAck, please see suggestion on v1: https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg01206.html > dinfo ? blk_by_legacy_dinfo(dinfo) : NULL, > 16 * KiB, FLASH_SIZE >> 16, > 1, 4, 0x0000, 0x0000, 0x0000, 0x0000, >
Philippe Mathieu-Daudé <philmd@redhat.com> writes: > On 2/26/19 8:34 PM, Markus Armbruster wrote: >> pflash_cfi02_register() takes a size in bytes, a block size in bytes >> and a number of blocks. r2d_init() passes FLASH_SIZE, 16 * KiB, >> FLASH_SIZE >> 16. Does not compute: size doesn't match block size * >> number of blocks. The latter happens to win. I tried to find >> documentation on the physical hardware, no luck. >> >> For now, adjust the byte size passed to match the actual size created, >> and add a FIXME comment. >> >> Cc: Magnus Damm <magnus.damm@gmail.com> >> Signed-off-by: Markus Armbruster <armbru@redhat.com> >> --- >> hw/sh4/r2d.c | 9 ++++++++- >> 1 file changed, 8 insertions(+), 1 deletion(-) >> >> diff --git a/hw/sh4/r2d.c b/hw/sh4/r2d.c >> index dcdb3728cb..ed18d1f351 100644 >> --- a/hw/sh4/r2d.c >> +++ b/hw/sh4/r2d.c >> @@ -290,7 +290,14 @@ static void r2d_init(MachineState *machine) >> >> /* onboard flash memory */ >> dinfo = drive_get(IF_PFLASH, 0, 0); >> - pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE, >> + /* >> + * FIXME The code is confused about the size of the flash. It >> + * used to pass FLASH_SIZE bytes, in FLASH_SIZE >> 16 blocks of >> + * 16KiB each, which does not compute, but creates one of >> + * FLASH_SIZE / 4 bytes anyway. The current code does so too, but >> + * whether it's the right size is anybody's guess. >> + */ >> + pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE / 4, > > NAck, please see suggestion on v1: > https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg01206.html I'll use that to replace this patch in v3. Thanks! [...]
diff --git a/hw/sh4/r2d.c b/hw/sh4/r2d.c index dcdb3728cb..ed18d1f351 100644 --- a/hw/sh4/r2d.c +++ b/hw/sh4/r2d.c @@ -290,7 +290,14 @@ static void r2d_init(MachineState *machine) /* onboard flash memory */ dinfo = drive_get(IF_PFLASH, 0, 0); - pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE, + /* + * FIXME The code is confused about the size of the flash. It + * used to pass FLASH_SIZE bytes, in FLASH_SIZE >> 16 blocks of + * 16KiB each, which does not compute, but creates one of + * FLASH_SIZE / 4 bytes anyway. The current code does so too, but + * whether it's the right size is anybody's guess. + */ + pflash_cfi02_register(0x0, NULL, "r2d.flash", FLASH_SIZE / 4, dinfo ? blk_by_legacy_dinfo(dinfo) : NULL, 16 * KiB, FLASH_SIZE >> 16, 1, 4, 0x0000, 0x0000, 0x0000, 0x0000,
pflash_cfi02_register() takes a size in bytes, a block size in bytes and a number of blocks. r2d_init() passes FLASH_SIZE, 16 * KiB, FLASH_SIZE >> 16. Does not compute: size doesn't match block size * number of blocks. The latter happens to win. I tried to find documentation on the physical hardware, no luck. For now, adjust the byte size passed to match the actual size created, and add a FIXME comment. Cc: Magnus Damm <magnus.damm@gmail.com> Signed-off-by: Markus Armbruster <armbru@redhat.com> --- hw/sh4/r2d.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)