| Message ID | 20171004020048.26379-2-eblake@redhat.com |
|---|---|
| State | New |
| Headers | show |
| Series | make bdrv_get_block_status byte-based | expand |
Am 04.10.2017 um 04:00 hat Eric Blake geschrieben: > Not all callers care about which BDS owns the mapping for a given > range of the file. This patch merely simplifies the callers by > consolidating the logic in the common call point, while guaranteeing > a non-NULL file to all the driver callbacks, for no semantic change. > The only caller that does not care about pnum is bdrv_is_allocated, > as invoked by vvfat; we can likewise add assertions that the rest > of the stack does not have to worry about a NULL pnum. > > Furthermore, this will also set the stage for a future cleanup: when > a caller does not care about which BDS owns an offset, it would be > nice to allow the driver to optimize things to not have to return > BDRV_BLOCK_OFFSET_VALID in the first place. In the case of fragmented > allocation (for example, it's fairly easy to create a qcow2 image > where consecutive guest addresses are not at consecutive host > addresses), the current contract requires bdrv_get_block_status() > to clamp *pnum to the limit where host addresses are no longer > consecutive, but allowing a NULL file means that *pnum could be > set to the full length of known-allocated data. > > Signed-off-by: Eric Blake <eblake@redhat.com> > > --- > v5: use second label for cleaner exit logic [John], use local_pnum > v4: only context changes > v3: rebase to recent changes (qcow2_measure), dropped R-b > v2: use local variable and final transfer, rather than assignment > of parameter to local > [previously in different series]: > v2: new patch, https://lists.gnu.org/archive/html/qemu-devel/2017-05/msg05645.html > --- > include/block/block_int.h | 10 ++++---- > block/io.c | 58 ++++++++++++++++++++++++++--------------------- > block/mirror.c | 3 +-- > block/qcow2.c | 8 ++----- > qemu-img.c | 10 ++++---- > 5 files changed, 45 insertions(+), 44 deletions(-) > > diff --git a/include/block/block_int.h b/include/block/block_int.h > index 79366b94b5..3b4158f576 100644 > --- a/include/block/block_int.h > +++ b/include/block/block_int.h > @@ -202,10 +202,12 @@ struct BlockDriver { > int64_t offset, int bytes); > > /* > - * Building block for bdrv_block_status[_above]. The driver should > - * answer only according to the current layer, and should not > - * set BDRV_BLOCK_ALLOCATED, but may set BDRV_BLOCK_RAW. See block.h > - * for the meaning of _DATA, _ZERO, and _OFFSET_VALID. > + * Building block for bdrv_block_status[_above] and > + * bdrv_is_allocated[_above]. The driver should answer only > + * according to the current layer, and should not set > + * BDRV_BLOCK_ALLOCATED, but may set BDRV_BLOCK_RAW. See block.h > + * for the meaning of _DATA, _ZERO, and _OFFSET_VALID. The block > + * layer guarantees non-NULL pnum and file. > */ > int64_t coroutine_fn (*bdrv_co_get_block_status)(BlockDriverState *bs, > int64_t sector_num, int nb_sectors, int *pnum, > diff --git a/block/io.c b/block/io.c > index 1e246315a7..e5a6f63eea 100644 > --- a/block/io.c > +++ b/block/io.c > @@ -698,7 +698,6 @@ int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags) > { > int64_t target_sectors, ret, nb_sectors, sector_num = 0; > BlockDriverState *bs = child->bs; > - BlockDriverState *file; > int n; > > target_sectors = bdrv_nb_sectors(bs); > @@ -711,7 +710,7 @@ int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags) > if (nb_sectors <= 0) { > return 0; > } > - ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &n, &file); > + ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &n, NULL); > if (ret < 0) { > error_report("error getting block status at sector %" PRId64 ": %s", > sector_num, strerror(-ret)); > @@ -1800,8 +1799,9 @@ int64_t coroutine_fn bdrv_co_get_block_status_from_backing(BlockDriverState *bs, > * beyond the end of the disk image it will be clamped; if 'pnum' is set to > * the end of the image, then the returned value will include BDRV_BLOCK_EOF. > * > - * If returned value is positive and BDRV_BLOCK_OFFSET_VALID bit is set, 'file' > - * points to the BDS which the sector range is allocated in. > + * If returned value is positive, BDRV_BLOCK_OFFSET_VALID bit is set, and > + * 'file' is non-NULL, then '*file' points to the BDS which the sector range > + * is allocated in. > */ > static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > int64_t sector_num, > @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > int64_t total_sectors; > int64_t n; > int64_t ret, ret2; > + BlockDriverState *local_file = NULL; > + int local_pnum = 0; I don't quite see what the point of local_pnum is if we assert anyway that the real pnum is non-NULL. > - *file = NULL; > + assert(pnum); > total_sectors = bdrv_nb_sectors(bs); > if (total_sectors < 0) { > - return total_sectors; > + ret = total_sectors; > + goto early_out; > } > > if (sector_num >= total_sectors || !nb_sectors) { > - *pnum = 0; > - return sector_num >= total_sectors ? BDRV_BLOCK_EOF : 0; > + ret = sector_num >= total_sectors ? BDRV_BLOCK_EOF : 0; > + goto early_out; > } > > n = total_sectors - sector_num; > @@ -1829,30 +1832,30 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > } > > if (!bs->drv->bdrv_co_get_block_status) { > - *pnum = nb_sectors; > + local_pnum = nb_sectors; > ret = BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED; > if (sector_num + nb_sectors == total_sectors) { > ret |= BDRV_BLOCK_EOF; > } > if (bs->drv->protocol_name) { > ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE); > - *file = bs; > + local_file = bs; > } > - return ret; > + goto early_out; > } > > bdrv_inc_in_flight(bs); > - ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, pnum, > - file); > + ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, > + &local_pnum, &local_file); > if (ret < 0) { > - *pnum = 0; > + local_pnum = 0; > goto out; > } > > if (ret & BDRV_BLOCK_RAW) { > - assert(ret & BDRV_BLOCK_OFFSET_VALID && *file); > - ret = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS, > - *pnum, pnum, file); > + assert(ret & BDRV_BLOCK_OFFSET_VALID && local_file); > + ret = bdrv_co_get_block_status(local_file, ret >> BDRV_SECTOR_BITS, > + local_pnum, &local_pnum, &local_file); > goto out; > } > > @@ -1870,14 +1873,13 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > } > } > > - if (*file && *file != bs && > + if (local_file && local_file != bs && > (ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) && > (ret & BDRV_BLOCK_OFFSET_VALID)) { > - BlockDriverState *file2; > int file_pnum; > > - ret2 = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS, > - *pnum, &file_pnum, &file2); > + ret2 = bdrv_co_get_block_status(local_file, ret >> BDRV_SECTOR_BITS, > + local_pnum, &file_pnum, NULL); > if (ret2 >= 0) { > /* Ignore errors. This is just providing extra information, it > * is useful but not necessary. > @@ -1892,17 +1894,22 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > ret |= BDRV_BLOCK_ZERO; > } else { > /* Limit request to the range reported by the protocol driver */ > - *pnum = file_pnum; > + local_pnum = file_pnum; > ret |= (ret2 & BDRV_BLOCK_ZERO); > } > } > } > > -out: > + out: git grep tells me that the old style (unindented label, as opposed to indented by a single space) is a lot more common both within the block layer and across qemu as a whole. > bdrv_dec_in_flight(bs); > - if (ret >= 0 && sector_num + *pnum == total_sectors) { > + if (ret >= 0 && sector_num + local_pnum == total_sectors) { > ret |= BDRV_BLOCK_EOF; > } > + early_out: So it's better to change this one to match the old out: than the other way round. > + *pnum = local_pnum; > + if (file) { > + *file = local_file; > + } > return ret; > } Kevin
On 10/10/2017 08:59 AM, Kevin Wolf wrote: > Am 04.10.2017 um 04:00 hat Eric Blake geschrieben: >> Not all callers care about which BDS owns the mapping for a given >> range of the file. This patch merely simplifies the callers by >> consolidating the logic in the common call point, while guaranteeing >> a non-NULL file to all the driver callbacks, for no semantic change. >> The only caller that does not care about pnum is bdrv_is_allocated, >> as invoked by vvfat; we can likewise add assertions that the rest >> of the stack does not have to worry about a NULL pnum. >> >> Furthermore, this will also set the stage for a future cleanup: when >> a caller does not care about which BDS owns an offset, it would be >> nice to allow the driver to optimize things to not have to return >> BDRV_BLOCK_OFFSET_VALID in the first place. In the case of fragmented >> allocation (for example, it's fairly easy to create a qcow2 image >> where consecutive guest addresses are not at consecutive host >> addresses), the current contract requires bdrv_get_block_status() >> to clamp *pnum to the limit where host addresses are no longer >> consecutive, but allowing a NULL file means that *pnum could be >> set to the full length of known-allocated data. >> >> Signed-off-by: Eric Blake <eblake@redhat.com> >> >> --- >> v5: use second label for cleaner exit logic [John], use local_pnum >> @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, >> int64_t total_sectors; >> int64_t n; >> int64_t ret, ret2; >> + BlockDriverState *local_file = NULL; >> + int local_pnum = 0; > > I don't quite see what the point of local_pnum is if we assert anyway > that the real pnum is non-NULL. I did it in parallel with fallout from John's review on v4: https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg06958.html but since it wasn't specifically asked for, and is now getting questions, I'm fine with not having it in v6. >> >> -out: >> + out: > > git grep tells me that the old style (unindented label, as opposed to > indented by a single space) is a lot more common both within the block > layer and across qemu as a whole. emacs likes to add the space (so that grepping for content in column 1 sees only function declarations, not mid-function labels - which in turn makes 'diff -p' output nicer to read). But I can override emacs' insistence, and go with the label flush to the left.
On 10/10/2017 09:43 AM, Eric Blake wrote: >>> --- >>> v5: use second label for cleaner exit logic [John], use local_pnum > >>> @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, >>> int64_t total_sectors; >>> int64_t n; >>> int64_t ret, ret2; >>> + BlockDriverState *local_file = NULL; >>> + int local_pnum = 0; >> >> I don't quite see what the point of local_pnum is if we assert anyway >> that the real pnum is non-NULL. > > I did it in parallel with fallout from John's review on v4: > https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg06958.html > > but since it wasn't specifically asked for, and is now getting > questions, I'm fine with not having it in v6. Okay, I re-read v4, and here's the comment (on 21/23) that led to my experiment in v5 patch 1 with local_pnum: https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00293.html and I did argue: https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00311.html >> Is it asking for trouble to be updating pnum here before we undo our >> alignment corrections? For readability reasons and preventing an >> accidental context-based oopsy-daisy. > > As in, write the code to make all calculations in a temporary, and then > assign *pnum only at the end? I suppose I can tweak the code along > those lines, but I'm not sure it will make the end result any more legible.
On 10/10/2017 03:00 PM, Eric Blake wrote: > On 10/10/2017 09:43 AM, Eric Blake wrote: > >>>> --- >>>> v5: use second label for cleaner exit logic [John], use local_pnum >> >>>> @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, >>>> int64_t total_sectors; >>>> int64_t n; >>>> int64_t ret, ret2; >>>> + BlockDriverState *local_file = NULL; >>>> + int local_pnum = 0; >>> >>> I don't quite see what the point of local_pnum is if we assert anyway >>> that the real pnum is non-NULL. >> >> I did it in parallel with fallout from John's review on v4: >> https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg06958.html >> >> but since it wasn't specifically asked for, and is now getting >> questions, I'm fine with not having it in v6. > > Okay, I re-read v4, and here's the comment (on 21/23) that led to my > experiment in v5 patch 1 with local_pnum: > > https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00293.html > > and I did argue: > > https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00311.html > Well, Kevin's the boss :D >>> Is it asking for trouble to be updating pnum here before we undo our >>> alignment corrections? For readability reasons and preventing an >>> accidental context-based oopsy-daisy. >> >> As in, write the code to make all calculations in a temporary, and then >> assign *pnum only at the end? I suppose I can tweak the code along >> those lines, but I'm not sure it will make the end result any more legible. >
Am 10.10.2017 um 21:24 hat John Snow geschrieben: > > > On 10/10/2017 03:00 PM, Eric Blake wrote: > > On 10/10/2017 09:43 AM, Eric Blake wrote: > > > >>>> --- > >>>> v5: use second label for cleaner exit logic [John], use local_pnum > >> > >>>> @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, > >>>> int64_t total_sectors; > >>>> int64_t n; > >>>> int64_t ret, ret2; > >>>> + BlockDriverState *local_file = NULL; > >>>> + int local_pnum = 0; > >>> > >>> I don't quite see what the point of local_pnum is if we assert anyway > >>> that the real pnum is non-NULL. > >> > >> I did it in parallel with fallout from John's review on v4: > >> https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg06958.html > >> > >> but since it wasn't specifically asked for, and is now getting > >> questions, I'm fine with not having it in v6. > > > > Okay, I re-read v4, and here's the comment (on 21/23) that led to my > > experiment in v5 patch 1 with local_pnum: > > > > https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00293.html > > > > and I did argue: > > > > https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00311.html > > Well, Kevin's the boss :D I'm not sure how renaming *pnum into local_pnum addresses your concerns? We still update local_pnum before we undo our alignment corrections. Or are you talking about some other part of these mails? Kevin
On 10/11/2017 04:42 AM, Kevin Wolf wrote: > Am 10.10.2017 um 21:24 hat John Snow geschrieben: >> >> >> On 10/10/2017 03:00 PM, Eric Blake wrote: >>> On 10/10/2017 09:43 AM, Eric Blake wrote: >>> >>>>>> --- >>>>>> v5: use second label for cleaner exit logic [John], use local_pnum >>>> >>>>>> @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, >>>>>> int64_t total_sectors; >>>>>> int64_t n; >>>>>> int64_t ret, ret2; >>>>>> + BlockDriverState *local_file = NULL; >>>>>> + int local_pnum = 0; >>>>> >>>>> I don't quite see what the point of local_pnum is if we assert anyway >>>>> that the real pnum is non-NULL. >>>> >>>> I did it in parallel with fallout from John's review on v4: >>>> https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg06958.html >>>> >>>> but since it wasn't specifically asked for, and is now getting >>>> questions, I'm fine with not having it in v6. >>> >>> Okay, I re-read v4, and here's the comment (on 21/23) that led to my >>> experiment in v5 patch 1 with local_pnum: >>> >>> https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00293.html >>> >>> and I did argue: >>> >>> https://lists.gnu.org/archive/html/qemu-devel/2017-10/msg00311.html >> >> Well, Kevin's the boss :D > > I'm not sure how renaming *pnum into local_pnum addresses your concerns? > We still update local_pnum before we undo our alignment corrections. Or > are you talking about some other part of these mails? > > Kevin > I made only a stylistic comment and it looked to me at a glance like it had resulted in conflicting feedback from you (since I all but asked for a local pnum) -- but my feedback was only minor stylistic stuff, so I'm deferring to you. --js
diff --git a/include/block/block_int.h b/include/block/block_int.h index 79366b94b5..3b4158f576 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -202,10 +202,12 @@ struct BlockDriver { int64_t offset, int bytes); /* - * Building block for bdrv_block_status[_above]. The driver should - * answer only according to the current layer, and should not - * set BDRV_BLOCK_ALLOCATED, but may set BDRV_BLOCK_RAW. See block.h - * for the meaning of _DATA, _ZERO, and _OFFSET_VALID. + * Building block for bdrv_block_status[_above] and + * bdrv_is_allocated[_above]. The driver should answer only + * according to the current layer, and should not set + * BDRV_BLOCK_ALLOCATED, but may set BDRV_BLOCK_RAW. See block.h + * for the meaning of _DATA, _ZERO, and _OFFSET_VALID. The block + * layer guarantees non-NULL pnum and file. */ int64_t coroutine_fn (*bdrv_co_get_block_status)(BlockDriverState *bs, int64_t sector_num, int nb_sectors, int *pnum, diff --git a/block/io.c b/block/io.c index 1e246315a7..e5a6f63eea 100644 --- a/block/io.c +++ b/block/io.c @@ -698,7 +698,6 @@ int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags) { int64_t target_sectors, ret, nb_sectors, sector_num = 0; BlockDriverState *bs = child->bs; - BlockDriverState *file; int n; target_sectors = bdrv_nb_sectors(bs); @@ -711,7 +710,7 @@ int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags) if (nb_sectors <= 0) { return 0; } - ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &n, &file); + ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &n, NULL); if (ret < 0) { error_report("error getting block status at sector %" PRId64 ": %s", sector_num, strerror(-ret)); @@ -1800,8 +1799,9 @@ int64_t coroutine_fn bdrv_co_get_block_status_from_backing(BlockDriverState *bs, * beyond the end of the disk image it will be clamped; if 'pnum' is set to * the end of the image, then the returned value will include BDRV_BLOCK_EOF. * - * If returned value is positive and BDRV_BLOCK_OFFSET_VALID bit is set, 'file' - * points to the BDS which the sector range is allocated in. + * If returned value is positive, BDRV_BLOCK_OFFSET_VALID bit is set, and + * 'file' is non-NULL, then '*file' points to the BDS which the sector range + * is allocated in. */ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, int64_t sector_num, @@ -1811,16 +1811,19 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, int64_t total_sectors; int64_t n; int64_t ret, ret2; + BlockDriverState *local_file = NULL; + int local_pnum = 0; - *file = NULL; + assert(pnum); total_sectors = bdrv_nb_sectors(bs); if (total_sectors < 0) { - return total_sectors; + ret = total_sectors; + goto early_out; } if (sector_num >= total_sectors || !nb_sectors) { - *pnum = 0; - return sector_num >= total_sectors ? BDRV_BLOCK_EOF : 0; + ret = sector_num >= total_sectors ? BDRV_BLOCK_EOF : 0; + goto early_out; } n = total_sectors - sector_num; @@ -1829,30 +1832,30 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, } if (!bs->drv->bdrv_co_get_block_status) { - *pnum = nb_sectors; + local_pnum = nb_sectors; ret = BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED; if (sector_num + nb_sectors == total_sectors) { ret |= BDRV_BLOCK_EOF; } if (bs->drv->protocol_name) { ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE); - *file = bs; + local_file = bs; } - return ret; + goto early_out; } bdrv_inc_in_flight(bs); - ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, pnum, - file); + ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, + &local_pnum, &local_file); if (ret < 0) { - *pnum = 0; + local_pnum = 0; goto out; } if (ret & BDRV_BLOCK_RAW) { - assert(ret & BDRV_BLOCK_OFFSET_VALID && *file); - ret = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS, - *pnum, pnum, file); + assert(ret & BDRV_BLOCK_OFFSET_VALID && local_file); + ret = bdrv_co_get_block_status(local_file, ret >> BDRV_SECTOR_BITS, + local_pnum, &local_pnum, &local_file); goto out; } @@ -1870,14 +1873,13 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, } } - if (*file && *file != bs && + if (local_file && local_file != bs && (ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) && (ret & BDRV_BLOCK_OFFSET_VALID)) { - BlockDriverState *file2; int file_pnum; - ret2 = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS, - *pnum, &file_pnum, &file2); + ret2 = bdrv_co_get_block_status(local_file, ret >> BDRV_SECTOR_BITS, + local_pnum, &file_pnum, NULL); if (ret2 >= 0) { /* Ignore errors. This is just providing extra information, it * is useful but not necessary. @@ -1892,17 +1894,22 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, ret |= BDRV_BLOCK_ZERO; } else { /* Limit request to the range reported by the protocol driver */ - *pnum = file_pnum; + local_pnum = file_pnum; ret |= (ret2 & BDRV_BLOCK_ZERO); } } } -out: + out: bdrv_dec_in_flight(bs); - if (ret >= 0 && sector_num + *pnum == total_sectors) { + if (ret >= 0 && sector_num + local_pnum == total_sectors) { ret |= BDRV_BLOCK_EOF; } + early_out: + *pnum = local_pnum; + if (file) { + *file = local_file; + } return ret; } @@ -2002,7 +2009,6 @@ int64_t bdrv_get_block_status(BlockDriverState *bs, int coroutine_fn bdrv_is_allocated(BlockDriverState *bs, int64_t offset, int64_t bytes, int64_t *pnum) { - BlockDriverState *file; int64_t sector_num = offset >> BDRV_SECTOR_BITS; int nb_sectors = bytes >> BDRV_SECTOR_BITS; int64_t ret; @@ -2011,7 +2017,7 @@ int coroutine_fn bdrv_is_allocated(BlockDriverState *bs, int64_t offset, assert(QEMU_IS_ALIGNED(offset, BDRV_SECTOR_SIZE)); assert(QEMU_IS_ALIGNED(bytes, BDRV_SECTOR_SIZE) && bytes < INT_MAX); ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &psectors, - &file); + NULL); if (ret < 0) { return ret; } diff --git a/block/mirror.c b/block/mirror.c index 459b80f8f3..e664a5dc5a 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -390,7 +390,6 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) int io_sectors; unsigned int io_bytes; int64_t io_bytes_acct; - BlockDriverState *file; enum MirrorMethod { MIRROR_METHOD_COPY, MIRROR_METHOD_ZERO, @@ -401,7 +400,7 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) ret = bdrv_get_block_status_above(source, NULL, offset >> BDRV_SECTOR_BITS, nb_chunks * sectors_per_chunk, - &io_sectors, &file); + &io_sectors, NULL); io_bytes = io_sectors * BDRV_SECTOR_SIZE; if (ret < 0) { io_bytes = MIN(nb_chunks * s->granularity, max_io_bytes); diff --git a/block/qcow2.c b/block/qcow2.c index b8da8ca105..bcd5c4a34c 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -2976,7 +2976,6 @@ static bool is_zero_sectors(BlockDriverState *bs, int64_t start, uint32_t count) { int nr; - BlockDriverState *file; int64_t res; if (start + count > bs->total_sectors) { @@ -2986,8 +2985,7 @@ static bool is_zero_sectors(BlockDriverState *bs, int64_t start, if (!count) { return true; } - res = bdrv_get_block_status_above(bs, NULL, start, count, - &nr, &file); + res = bdrv_get_block_status_above(bs, NULL, start, count, &nr, NULL); return res >= 0 && (res & BDRV_BLOCK_ZERO) && nr == count; } @@ -3680,13 +3678,11 @@ static BlockMeasureInfo *qcow2_measure(QemuOpts *opts, BlockDriverState *in_bs, offset += pnum * BDRV_SECTOR_SIZE) { int nb_sectors = MIN(ssize - offset, BDRV_REQUEST_MAX_BYTES) / BDRV_SECTOR_SIZE; - BlockDriverState *file; int64_t ret; ret = bdrv_get_block_status_above(in_bs, NULL, offset >> BDRV_SECTOR_BITS, - nb_sectors, - &pnum, &file); + nb_sectors, &pnum, NULL); if (ret < 0) { error_setg_errno(&local_err, -ret, "Unable to get block status"); diff --git a/qemu-img.c b/qemu-img.c index d6007b2a6d..e9c7b30c91 100644 --- a/qemu-img.c +++ b/qemu-img.c @@ -1375,7 +1375,6 @@ static int img_compare(int argc, char **argv) for (;;) { int64_t status1, status2; - BlockDriverState *file; nb_sectors = sectors_to_process(total_sectors, sector_num); if (nb_sectors <= 0) { @@ -1383,7 +1382,7 @@ static int img_compare(int argc, char **argv) } status1 = bdrv_get_block_status_above(bs1, NULL, sector_num, total_sectors1 - sector_num, - &pnum1, &file); + &pnum1, NULL); if (status1 < 0) { ret = 3; error_report("Sector allocation test failed for %s", filename1); @@ -1393,7 +1392,7 @@ static int img_compare(int argc, char **argv) status2 = bdrv_get_block_status_above(bs2, NULL, sector_num, total_sectors2 - sector_num, - &pnum2, &file); + &pnum2, NULL); if (status2 < 0) { ret = 3; error_report("Sector allocation test failed for %s", filename2); @@ -1599,15 +1598,14 @@ static int convert_iteration_sectors(ImgConvertState *s, int64_t sector_num) n = MIN(s->total_sectors - sector_num, BDRV_REQUEST_MAX_SECTORS); if (s->sector_next_status <= sector_num) { - BlockDriverState *file; if (s->target_has_backing) { ret = bdrv_get_block_status(blk_bs(s->src[src_cur]), sector_num - src_cur_offset, - n, &n, &file); + n, &n, NULL); } else { ret = bdrv_get_block_status_above(blk_bs(s->src[src_cur]), NULL, sector_num - src_cur_offset, - n, &n, &file); + n, &n, NULL); } if (ret < 0) { return ret;
Not all callers care about which BDS owns the mapping for a given range of the file. This patch merely simplifies the callers by consolidating the logic in the common call point, while guaranteeing a non-NULL file to all the driver callbacks, for no semantic change. The only caller that does not care about pnum is bdrv_is_allocated, as invoked by vvfat; we can likewise add assertions that the rest of the stack does not have to worry about a NULL pnum. Furthermore, this will also set the stage for a future cleanup: when a caller does not care about which BDS owns an offset, it would be nice to allow the driver to optimize things to not have to return BDRV_BLOCK_OFFSET_VALID in the first place. In the case of fragmented allocation (for example, it's fairly easy to create a qcow2 image where consecutive guest addresses are not at consecutive host addresses), the current contract requires bdrv_get_block_status() to clamp *pnum to the limit where host addresses are no longer consecutive, but allowing a NULL file means that *pnum could be set to the full length of known-allocated data. Signed-off-by: Eric Blake <eblake@redhat.com> --- v5: use second label for cleaner exit logic [John], use local_pnum v4: only context changes v3: rebase to recent changes (qcow2_measure), dropped R-b v2: use local variable and final transfer, rather than assignment of parameter to local [previously in different series]: v2: new patch, https://lists.gnu.org/archive/html/qemu-devel/2017-05/msg05645.html --- include/block/block_int.h | 10 ++++---- block/io.c | 58 ++++++++++++++++++++++++++--------------------- block/mirror.c | 3 +-- block/qcow2.c | 8 ++----- qemu-img.c | 10 ++++---- 5 files changed, 45 insertions(+), 44 deletions(-)