diff options
Diffstat (limited to 'drivers/block')
-rw-r--r-- | drivers/block/aoe/aoeblk.c | 1 | ||||
-rw-r--r-- | drivers/block/drbd/drbd_bitmap.c | 4 | ||||
-rw-r--r-- | drivers/block/drbd/drbd_req.c | 27 | ||||
-rw-r--r-- | drivers/block/floppy.c | 466 | ||||
-rw-r--r-- | drivers/block/loop.c | 387 | ||||
-rw-r--r-- | drivers/block/null_blk_main.c | 35 | ||||
-rw-r--r-- | drivers/block/null_blk_zoned.c | 41 | ||||
-rw-r--r-- | drivers/block/paride/pcd.c | 2 | ||||
-rw-r--r-- | drivers/block/rsxx/dev.c | 19 | ||||
-rw-r--r-- | drivers/block/swim.c | 6 | ||||
-rw-r--r-- | drivers/block/virtio_blk.c | 86 | ||||
-rw-r--r-- | drivers/block/zram/zcomp.c | 44 | ||||
-rw-r--r-- | drivers/block/zram/zcomp.h | 5 | ||||
-rw-r--r-- | drivers/block/zram/zram_drv.c | 24 |
14 files changed, 656 insertions, 491 deletions
diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index a27804d71e12..5ca7216e9e01 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -407,7 +407,6 @@ aoeblk_gdalloc(void *vp) WARN_ON(d->gd); WARN_ON(d->flags & DEVFL_UP); blk_queue_max_hw_sectors(q, BLK_DEF_MAX_SECTORS); - q->backing_dev_info->name = "aoe"; q->backing_dev_info->ra_pages = READ_AHEAD / PAGE_SIZE; d->bufpool = mp; d->blkq = gd->queue = q; diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c index 15e99697234a..df53dca5d02c 100644 --- a/drivers/block/drbd/drbd_bitmap.c +++ b/drivers/block/drbd/drbd_bitmap.c @@ -396,9 +396,7 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want) bytes = sizeof(struct page *)*want; new_pages = kzalloc(bytes, GFP_NOIO | __GFP_NOWARN); if (!new_pages) { - new_pages = __vmalloc(bytes, - GFP_NOIO | __GFP_ZERO, - PAGE_KERNEL); + new_pages = __vmalloc(bytes, GFP_NOIO | __GFP_ZERO); if (!new_pages) return NULL; } diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index 840c3aef3c5c..c80a2f1c3c2a 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -21,24 +21,6 @@ static bool drbd_may_do_local_read(struct drbd_device *device, sector_t sector, int size); -/* Update disk stats at start of I/O request */ -static void _drbd_start_io_acct(struct drbd_device *device, struct drbd_request *req) -{ - struct request_queue *q = device->rq_queue; - - generic_start_io_acct(q, bio_op(req->master_bio), - req->i.size >> 9, &device->vdisk->part0); -} - -/* Update disk stats when completing request upwards */ -static void _drbd_end_io_acct(struct drbd_device *device, struct drbd_request *req) -{ - struct request_queue *q = device->rq_queue; - - generic_end_io_acct(q, bio_op(req->master_bio), - &device->vdisk->part0, req->start_jif); -} - static struct drbd_request *drbd_req_new(struct drbd_device *device, struct bio *bio_src) { struct drbd_request *req; @@ -263,7 +245,7 @@ void drbd_req_complete(struct drbd_request *req, struct bio_and_error *m) start_new_tl_epoch(first_peer_device(device)->connection); /* Update disk stats */ - _drbd_end_io_acct(device, req); + bio_end_io_acct(req->master_bio, req->start_jif); /* If READ failed, * have it be pushed back to the retry work queue, @@ -1222,16 +1204,15 @@ drbd_request_prepare(struct drbd_device *device, struct bio *bio, unsigned long bio_endio(bio); return ERR_PTR(-ENOMEM); } - req->start_jif = start_jif; + + /* Update disk stats */ + req->start_jif = bio_start_io_acct(req->master_bio); if (!get_ldev(device)) { bio_put(req->private_bio); req->private_bio = NULL; } - /* Update disk stats */ - _drbd_start_io_acct(device, req); - /* process discards always from our submitter thread */ if (bio_op(bio) == REQ_OP_WRITE_ZEROES || bio_op(bio) == REQ_OP_DISCARD) diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index c3daa64cb52c..3e9db22db2a8 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -337,8 +337,7 @@ static bool initialized; /* * globals used by 'result()' */ -#define MAX_REPLIES 16 -static unsigned char reply_buffer[MAX_REPLIES]; +static unsigned char reply_buffer[FD_RAW_REPLY_SIZE]; static int inr; /* size of reply buffer, when called from interrupt */ #define ST0 0 #define ST1 1 @@ -595,12 +594,12 @@ static unsigned char in_sector_offset; /* offset within physical sector, static inline unsigned char fdc_inb(int fdc, int reg) { - return fd_inb(fdc_state[fdc].address + reg); + return fd_inb(fdc_state[fdc].address, reg); } static inline void fdc_outb(unsigned char value, int fdc, int reg) { - fd_outb(value, fdc_state[fdc].address + reg); + fd_outb(value, fdc_state[fdc].address, reg); } static inline bool drive_no_geom(int drive) @@ -668,16 +667,12 @@ static struct output_log { static int output_log_pos; -#define current_reqD -1 #define MAXTIMEOUT -2 static void __reschedule_timeout(int drive, const char *message) { unsigned long delay; - if (drive == current_reqD) - drive = current_drive; - if (drive < 0 || drive >= N_DRIVE) { delay = 20UL * HZ; drive = 0; @@ -827,59 +822,70 @@ static int set_dor(int fdc, char mask, char data) return olddor; } -static void twaddle(void) +static void twaddle(int fdc, int drive) { - if (drive_params[current_drive].select_delay) + if (drive_params[drive].select_delay) return; - fdc_outb(fdc_state[current_fdc].dor & ~(0x10 << UNIT(current_drive)), - current_fdc, FD_DOR); - fdc_outb(fdc_state[current_fdc].dor, current_fdc, FD_DOR); - drive_state[current_drive].select_date = jiffies; + fdc_outb(fdc_state[fdc].dor & ~(0x10 << UNIT(drive)), + fdc, FD_DOR); + fdc_outb(fdc_state[fdc].dor, fdc, FD_DOR); + drive_state[drive].select_date = jiffies; } /* - * Reset all driver information about the current fdc. + * Reset all driver information about the specified fdc. * This is needed after a reset, and after a raw command. */ -static void reset_fdc_info(int mode) +static void reset_fdc_info(int fdc, int mode) { int drive; - fdc_state[current_fdc].spec1 = fdc_state[current_fdc].spec2 = -1; - fdc_state[current_fdc].need_configure = 1; - fdc_state[current_fdc].perp_mode = 1; - fdc_state[current_fdc].rawcmd = 0; + fdc_state[fdc].spec1 = fdc_state[fdc].spec2 = -1; + fdc_state[fdc].need_configure = 1; + fdc_state[fdc].perp_mode = 1; + fdc_state[fdc].rawcmd = 0; for (drive = 0; drive < N_DRIVE; drive++) - if (FDC(drive) == current_fdc && + if (FDC(drive) == fdc && (mode || drive_state[drive].track != NEED_1_RECAL)) drive_state[drive].track = NEED_2_RECAL; } -/* selects the fdc and drive, and enables the fdc's input/dma. */ +/* + * selects the fdc and drive, and enables the fdc's input/dma. + * Both current_drive and current_fdc are changed to match the new drive. + */ static void set_fdc(int drive) { - unsigned int new_fdc = current_fdc; + unsigned int fdc; - if (drive >= 0 && drive < N_DRIVE) { - new_fdc = FDC(drive); - current_drive = drive; + if (drive < 0 || drive >= N_DRIVE) { + pr_info("bad drive value %d\n", drive); + return; } - if (new_fdc >= N_FDC) { + + fdc = FDC(drive); + if (fdc >= N_FDC) { pr_info("bad fdc value\n"); return; } - current_fdc = new_fdc; - set_dor(current_fdc, ~0, 8); + + set_dor(fdc, ~0, 8); #if N_FDC > 1 - set_dor(1 - current_fdc, ~8, 0); + set_dor(1 - fdc, ~8, 0); #endif - if (fdc_state[current_fdc].rawcmd == 2) - reset_fdc_info(1); - if (fdc_inb(current_fdc, FD_STATUS) != STATUS_READY) - fdc_state[current_fdc].reset = 1; + if (fdc_state[fdc].rawcmd == 2) + reset_fdc_info(fdc, 1); + if (fdc_inb(fdc, FD_STATUS) != STATUS_READY) + fdc_state[fdc].reset = 1; + + current_drive = drive; + current_fdc = fdc; } -/* locks the driver */ +/* + * locks the driver. + * Both current_drive and current_fdc are changed to match the new drive. + */ static int lock_fdc(int drive) { if (WARN(atomic_read(&usage_count) == 0, @@ -1062,12 +1068,9 @@ static void setup_DMA(void) unsigned long f; if (raw_cmd->length == 0) { - int i; - - pr_info("zero dma transfer size:"); - for (i = 0; i < raw_cmd->cmd_count; i++) - pr_cont("%x,", raw_cmd->cmd[i]); - pr_cont("\n"); + print_hex_dump(KERN_INFO, "zero dma transfer size: ", + DUMP_PREFIX_NONE, 16, 1, + raw_cmd->fullcmd, raw_cmd->cmd_count, false); cont->done(0); fdc_state[current_fdc].reset = 1; return; @@ -1104,62 +1107,62 @@ static void setup_DMA(void) #endif } -static void show_floppy(void); +static void show_floppy(int fdc); /* waits until the fdc becomes ready */ -static int wait_til_ready(void) +static int wait_til_ready(int fdc) { int status; int counter; - if (fdc_state[current_fdc].reset) + if (fdc_state[fdc].reset) return -1; for (counter = 0; counter < 10000; counter++) { - status = fdc_inb(current_fdc, FD_STATUS); + status = fdc_inb(fdc, FD_STATUS); if (status & STATUS_READY) return status; } if (initialized) { - DPRINT("Getstatus times out (%x) on fdc %d\n", status, current_fdc); - show_floppy(); + DPRINT("Getstatus times out (%x) on fdc %d\n", status, fdc); + show_floppy(fdc); } - fdc_state[current_fdc].reset = 1; + fdc_state[fdc].reset = 1; return -1; } /* sends a command byte to the fdc */ -static int output_byte(char byte) +static int output_byte(int fdc, char byte) { - int status = wait_til_ready(); + int status = wait_til_ready(fdc); if (status < 0) return -1; if (is_ready_state(status)) { - fdc_outb(byte, current_fdc, FD_DATA); + fdc_outb(byte, fdc, FD_DATA); output_log[output_log_pos].data = byte; output_log[output_log_pos].status = status; output_log[output_log_pos].jiffies = jiffies; output_log_pos = (output_log_pos + 1) % OLOGSIZE; return 0; } - fdc_state[current_fdc].reset = 1; + fdc_state[fdc].reset = 1; if (initialized) { DPRINT("Unable to send byte %x to FDC. Fdc=%x Status=%x\n", - byte, current_fdc, status); - show_floppy(); + byte, fdc, status); + show_floppy(fdc); } return -1; } /* gets the response from the fdc */ -static int result(void) +static int result(int fdc) { int i; int status = 0; - for (i = 0; i < MAX_REPLIES; i++) { - status = wait_til_ready(); + for (i = 0; i < FD_RAW_REPLY_SIZE; i++) { + status = wait_til_ready(fdc); if (status < 0) break; status &= STATUS_DIR | STATUS_READY | STATUS_BUSY | STATUS_DMA; @@ -1169,24 +1172,24 @@ static int result(void) return i; } if (status == (STATUS_DIR | STATUS_READY | STATUS_BUSY)) - reply_buffer[i] = fdc_inb(current_fdc, FD_DATA); + reply_buffer[i] = fdc_inb(fdc, FD_DATA); else break; } if (initialized) { DPRINT("get result error. Fdc=%d Last status=%x Read bytes=%d\n", - current_fdc, status, i); - show_floppy(); + fdc, status, i); + show_floppy(fdc); } - fdc_state[current_fdc].reset = 1; + fdc_state[fdc].reset = 1; return -1; } #define MORE_OUTPUT -2 /* does the fdc need more output? */ -static int need_more_output(void) +static int need_more_output(int fdc) { - int status = wait_til_ready(); + int status = wait_til_ready(fdc); if (status < 0) return -1; @@ -1194,13 +1197,13 @@ static int need_more_output(void) if (is_ready_state(status)) return MORE_OUTPUT; - return result(); + return result(fdc); } /* Set perpendicular mode as required, based on data rate, if supported. * 82077 Now tested. 1Mbps data rate only possible with 82077-1. */ -static void perpendicular_mode(void) +static void perpendicular_mode(int fdc) { unsigned char perp_mode; @@ -1215,7 +1218,7 @@ static void perpendicular_mode(void) default: DPRINT("Invalid data rate for perpendicular mode!\n"); cont->done(0); - fdc_state[current_fdc].reset = 1; + fdc_state[fdc].reset = 1; /* * convenient way to return to * redo without too much hassle @@ -1226,12 +1229,12 @@ static void perpendicular_mode(void) } else perp_mode = 0; - if (fdc_state[current_fdc].perp_mode == perp_mode) + if (fdc_state[fdc].perp_mode == perp_mode) return; - if (fdc_state[current_fdc].version >= FDC_82077_ORIG) { - output_byte(FD_PERPENDICULAR); - output_byte(perp_mode); - fdc_state[current_fdc].perp_mode = perp_mode; + if (fdc_state[fdc].version >= FDC_82077_ORIG) { + output_byte(fdc, FD_PERPENDICULAR); + output_byte(fdc, perp_mode); + fdc_state[fdc].perp_mode = perp_mode; } else if (perp_mode) { DPRINT("perpendicular mode not supported by this FDC.\n"); } @@ -1240,16 +1243,15 @@ static void perpendicular_mode(void) static int fifo_depth = 0xa; static int no_fifo; -static int fdc_configure(void) +static int fdc_configure(int fdc) { /* Turn on FIFO */ - output_byte(FD_CONFIGURE); - if (need_more_output() != MORE_OUTPUT) + output_byte(fdc, FD_CONFIGURE); + if (need_more_output(fdc) != MORE_OUTPUT) return 0; - output_byte(0); - output_byte(0x10 | (no_fifo & 0x20) | (fifo_depth & 0xf)); - output_byte(0); /* pre-compensation from track - 0 upwards */ + output_byte(fdc, 0); + output_byte(fdc, 0x10 | (no_fifo & 0x20) | (fifo_depth & 0xf)); + output_byte(fdc, 0); /* pre-compensation from track 0 upwards */ return 1; } @@ -1274,7 +1276,7 @@ static int fdc_configure(void) * * These values are rounded up to the next highest available delay time. */ -static void fdc_specify(void) +static void fdc_specify(int fdc, int drive) { unsigned char spec1; unsigned char spec2; @@ -1286,10 +1288,10 @@ static void fdc_specify(void) int hlt_max_code = 0x7f; int hut_max_code = 0xf; - if (fdc_state[current_fdc].need_configure && - fdc_state[current_fdc].version >= FDC_82072A) { - fdc_configure(); - fdc_state[current_fdc].need_configure = 0; + if (fdc_state[fdc].need_configure && + fdc_state[fdc].version >= FDC_82072A) { + fdc_configure(fdc); + fdc_state[fdc].need_configure = 0; } switch (raw_cmd->rate & 0x03) { @@ -1298,13 +1300,13 @@ static void fdc_specify(void) break; case 1: dtr = 300; - if (fdc_state[current_fdc].version >= FDC_82078) { + if (fdc_state[fdc].version >= FDC_82078) { /* chose the default rate table, not the one * where 1 = 2 Mbps */ - output_byte(FD_DRIVESPEC); - if (need_more_output() == MORE_OUTPUT) { - output_byte(UNIT(current_drive)); - output_byte(0xc0); + output_byte(fdc, FD_DRIVESPEC); + if (need_more_output(fdc) == MORE_OUTPUT) { + output_byte(fdc, UNIT(drive)); + output_byte(fdc, 0xc0); } } break; @@ -1313,14 +1315,14 @@ static void fdc_specify(void) break; } - if (fdc_state[current_fdc].version >= FDC_82072) { + if (fdc_state[fdc].version >= FDC_82072) { scale_dtr = dtr; hlt_max_code = 0x00; /* 0==256msec*dtr0/dtr (not linear!) */ hut_max_code = 0x0; /* 0==256msec*dtr0/dtr (not linear!) */ } /* Convert step rate from microseconds to milliseconds and 4 bits */ - srt = 16 - DIV_ROUND_UP(drive_params[current_drive].srt * scale_dtr / 1000, + srt = 16 - DIV_ROUND_UP(drive_params[drive].srt * scale_dtr / 1000, NOMINAL_DTR); if (slow_floppy) srt = srt / 4; @@ -1328,14 +1330,14 @@ static void fdc_specify(void) SUPBOUND(srt, 0xf); INFBOUND(srt, 0); - hlt = DIV_ROUND_UP(drive_params[current_drive].hlt * scale_dtr / 2, + hlt = DIV_ROUND_UP(drive_params[drive].hlt * scale_dtr / 2, NOMINAL_DTR); if (hlt < 0x01) hlt = 0x01; else if (hlt > 0x7f) hlt = hlt_max_code; - hut = DIV_ROUND_UP(drive_params[current_drive].hut * scale_dtr / 16, + hut = DIV_ROUND_UP(drive_params[drive].hut * scale_dtr / 16, NOMINAL_DTR); if (hut < 0x1) hut = 0x1; @@ -1346,12 +1348,12 @@ static void fdc_specify(void) spec2 = (hlt << 1) | (use_virtual_dma & 1); /* If these parameters did not change, just return with success */ - if (fdc_state[current_fdc].spec1 != spec1 || - fdc_state[current_fdc].spec2 != spec2) { + if (fdc_state[fdc].spec1 != spec1 || + fdc_state[fdc].spec2 != spec2) { /* Go ahead and set spec1 and spec2 */ - output_byte(FD_SPECIFY); - output_byte(fdc_state[current_fdc].spec1 = spec1); - output_byte(fdc_state[current_fdc].spec2 = spec2); + output_byte(fdc, FD_SPECIFY); + output_byte(fdc, fdc_state[fdc].spec1 = spec1); + output_byte(fdc, fdc_state[fdc].spec2 = spec2); } } /* fdc_specify */ @@ -1513,7 +1515,7 @@ static void setup_rw_floppy(void) r = 0; for (i = 0; i < raw_cmd->cmd_count; i++) - r |= output_byte(raw_cmd->cmd[i]); + r |= output_byte(current_fdc, raw_cmd->fullcmd[i]); debugt(__func__, "rw_command"); @@ -1524,7 +1526,7 @@ static void setup_rw_floppy(void) } if (!(flags & FD_RAW_INTR)) { - inr = result(); + inr = result(current_fdc); cont->interrupt(); } else if (flags & FD_RAW_NEED_DISK) fd_watchdog(); @@ -1562,29 +1564,29 @@ static void seek_interrupt(void) floppy_ready(); } -static void check_wp(void) +static void check_wp(int fdc, int drive) { - if (test_bit(FD_VERIFY_BIT, &drive_state[current_drive].flags)) { + if (test_bit(FD_VERIFY_BIT, &drive_state[drive].flags)) { /* check write protection */ - output_byte(FD_GETSTATUS); - output_byte(UNIT(current_drive)); - if (result() != 1) { - fdc_state[current_fdc].reset = 1; + output_byte(fdc, FD_GETSTATUS); + output_byte(fdc, UNIT(drive)); + if (result(fdc) != 1) { + fdc_state[fdc].reset = 1; return; } - clear_bit(FD_VERIFY_BIT, &drive_state[current_drive].flags); + clear_bit(FD_VERIFY_BIT, &drive_state[drive].flags); clear_bit(FD_NEED_TWADDLE_BIT, - &drive_state[current_drive].flags); - debug_dcl(drive_params[current_drive].flags, + &drive_state[drive].flags); + debug_dcl(drive_params[drive].flags, "checking whether disk is write protected\n"); - debug_dcl(drive_params[current_drive].flags, "wp=%x\n", + debug_dcl(drive_params[drive].flags, "wp=%x\n", reply_buffer[ST3] & 0x40); if (!(reply_buffer[ST3] & 0x40)) set_bit(FD_DISK_WRITABLE_BIT, - &drive_state[current_drive].flags); + &drive_state[drive].flags); else clear_bit(FD_DISK_WRITABLE_BIT, - &drive_state[current_drive].flags); + &drive_state[drive].flags); } } @@ -1628,7 +1630,7 @@ static void seek_floppy(void) track = 1; } } else { - check_wp(); + check_wp(current_fdc, current_drive); if (raw_cmd->track != drive_state[current_drive].track && (raw_cmd->flags & FD_RAW_NEED_SEEK)) track = raw_cmd->track; @@ -1639,9 +1641,9 @@ static void seek_floppy(void) } do_floppy = seek_interrupt; - output_byte(FD_SEEK); - output_byte(UNIT(current_drive)); - if (output_byte(track) < 0) { + output_byte(current_fdc, FD_SEEK); + output_byte(current_fdc, UNIT(current_drive)); + if (output_byte(current_fdc, track) < 0) { reset_fdc(); return; } @@ -1742,14 +1744,14 @@ irqreturn_t floppy_interrupt(int irq, void *dev_id) do_print = !handler && print_unex && initialized; - inr = result(); + inr = result(current_fdc); if (do_print) print_result("unexpected interrupt", inr); if (inr == 0) { int max_sensei = 4; do { - output_byte(FD_SENSEI); - inr = result(); + output_byte(current_fdc, FD_SENSEI); + inr = result(current_fdc); if (do_print) print_result("sensei", inr); max_sensei--; @@ -1771,8 +1773,8 @@ static void recalibrate_floppy(void) { debugt(__func__, ""); do_floppy = recal_interrupt; - output_byte(FD_RECALIBRATE); - if (output_byte(UNIT(current_drive)) < 0) + output_byte(current_fdc, FD_RECALIBRATE); + if (output_byte(current_fdc, UNIT(current_drive)) < 0) reset_fdc(); } @@ -1782,7 +1784,7 @@ static void recalibrate_floppy(void) static void reset_interrupt(void) { debugt(__func__, ""); - result(); /* get the status ready for set_fdc */ + result(current_fdc); /* get the status ready for set_fdc */ if (fdc_state[current_fdc].reset) { pr_info("reset set in interrupt, calling %ps\n", cont->error); cont->error(); /* a reset just after a reset. BAD! */ @@ -1792,7 +1794,9 @@ static void reset_interrupt(void) /* * reset is done by pulling bit 2 of DOR low for a while (old FDCs), - * or by setting the self clearing bit 7 of STATUS (newer FDCs) + * or by setting the self clearing bit 7 of STATUS (newer FDCs). + * This WILL trigger an interrupt, causing the handlers in the current + * cont's ->redo() to be called via reset_interrupt(). */ static void reset_fdc(void) { @@ -1800,7 +1804,7 @@ static void reset_fdc(void) do_floppy = reset_interrupt; fdc_state[current_fdc].reset = 0; - reset_fdc_info(0); + reset_fdc_info(current_fdc, 0); /* Pseudo-DMA may intercept 'reset finished' interrupt. */ /* Irrelevant for systems with true DMA (i386). */ @@ -1819,7 +1823,7 @@ static void reset_fdc(void) } } -static void show_floppy(void) +static void show_floppy(int fdc) { int i; @@ -1842,7 +1846,7 @@ static void show_floppy(void) print_hex_dump(KERN_INFO, "", DUMP_PREFIX_NONE, 16, 1, reply_buffer, resultsize, true); - pr_info("status=%x\n", fdc_inb(current_fdc, FD_STATUS)); + pr_info("status=%x\n", fdc_inb(fdc, FD_STATUS)); pr_info("fdc_busy=%lu\n", fdc_busy); if (do_floppy) pr_info("do_floppy=%ps\n", do_floppy); @@ -1868,7 +1872,7 @@ static void floppy_shutdown(struct work_struct *arg) unsigned long flags; if (initialized) - show_floppy(); + show_floppy(current_fdc); cancel_activity(); flags = claim_dma_lock(); @@ -1934,7 +1938,7 @@ static void floppy_ready(void) "calling disk change from floppy_ready\n"); if (!(raw_cmd->flags & FD_RAW_NO_MOTOR) && disk_change(current_drive) && !drive_params[current_drive].select_delay) - twaddle(); /* this clears the dcl on certain + twaddle(current_fdc, current_drive); /* this clears the dcl on certain * drive/controller combinations */ #ifdef fd_chose_dma_mode @@ -1946,20 +1950,20 @@ static void floppy_ready(void) #endif if (raw_cmd->flags & (FD_RAW_NEED_SEEK | FD_RAW_NEED_DISK)) { - perpendicular_mode(); - fdc_specify(); /* must be done here because of hut, hlt ... */ + perpendicular_mode(current_fdc); + fdc_specify(current_fdc, current_drive); /* must be done here because of hut, hlt ... */ seek_floppy(); } else { if ((raw_cmd->flags & FD_RAW_READ) || (raw_cmd->flags & FD_RAW_WRITE)) - fdc_specify(); + fdc_specify(current_fdc, current_drive); setup_rw_floppy(); } } static void floppy_start(void) { - reschedule_timeout(current_reqD, "floppy start"); + reschedule_timeout(current_drive, "floppy start"); scandrives(); debug_dcl(drive_params[current_drive].flags, @@ -2004,6 +2008,9 @@ static const struct cont_t intr_cont = { .done = (done_f)empty }; +/* schedules handler, waiting for completion. May be interrupted, will then + * return -EINTR, in which case the driver will automatically be unlocked. + */ static int wait_til_done(void (*handler)(void), bool interruptible) { int ret; @@ -2059,18 +2066,19 @@ static void success_and_wakeup(void) * ========================== */ -static int next_valid_format(void) +static int next_valid_format(int drive) { int probed_format; - probed_format = drive_state[current_drive].probed_format; + probed_format = drive_state[drive].probed_format; while (1) { - if (probed_format >= 8 || !drive_params[current_drive].autodetect[probed_format]) { - drive_state[current_drive].probed_format = 0; + if (probed_format >= FD_AUTODETECT_SIZE || + !drive_params[drive].autodetect[probed_format]) { + drive_state[drive].probed_format = 0; return 1; } - if (floppy_type[drive_params[current_drive].autodetect[probed_format]].sect) { - drive_state[current_drive].probed_format = probed_format; + if (floppy_type[drive_params[drive].autodetect[probed_format]].sect) { + drive_state[drive].probed_format = probed_format; return 0; } probed_format++; @@ -2083,7 +2091,7 @@ static void bad_flp_intr(void) if (probing) { drive_state[current_drive].probed_format++; - if (!next_valid_format()) + if (!next_valid_format(current_drive)) return; } err_count = ++(*errors); @@ -2843,6 +2851,9 @@ static int set_next_request(void) return current_req != NULL; } +/* Starts or continues processing request. Will automatically unlock the + * driver at end of request. + */ static void redo_fd_request(void) { int drive; @@ -2867,7 +2878,7 @@ do_request: } drive = (long)current_req->rq_disk->private_data; set_fdc(drive); - reschedule_timeout(current_reqD, "redo fd request"); + reschedule_timeout(current_drive, "redo fd request"); set_floppy(drive); raw_cmd = &default_raw_cmd; @@ -2885,7 +2896,7 @@ do_request: if (!_floppy) { /* Autodetection */ if (!probing) { drive_state[current_drive].probed_format = 0; - if (next_valid_format()) { + if (next_valid_format(current_drive)) { DPRINT("no autodetectable formats\n"); _floppy = NULL; request_done(0); @@ -2904,7 +2915,7 @@ do_request: } if (test_bit(FD_NEED_TWADDLE_BIT, &drive_state[current_drive].flags)) - twaddle(); + twaddle(current_fdc, current_drive); schedule_bh(floppy_start); debugt(__func__, "queue fd request"); return; @@ -2917,6 +2928,7 @@ static const struct cont_t rw_cont = { .done = request_done }; +/* schedule the request and automatically unlock the driver on completion */ static void process_fd_request(void) { cont = &rw_cont; @@ -2938,17 +2950,17 @@ static blk_status_t floppy_queue_rq(struct blk_mq_hw_ctx *hctx, (unsigned long long) current_req->cmd_flags)) return BLK_STS_IOERR; - spin_lock_irq(&floppy_lock); - list_add_tail(&bd->rq->queuelist, &floppy_reqs); - spin_unlock_irq(&floppy_lock); - if (test_and_set_bit(0, &fdc_busy)) { /* fdc busy, this new request will be treated when the current one is done */ is_alive(__func__, "old request running"); - return BLK_STS_OK; + return BLK_STS_RESOURCE; } + spin_lock_irq(&floppy_lock); + list_add_tail(&bd->rq->queuelist, &floppy_reqs); + spin_unlock_irq(&floppy_lock); + command_status = FD_COMMAND_NONE; __reschedule_timeout(MAXTIMEOUT, "fd_request"); set_fdc(0); @@ -2996,6 +3008,10 @@ static const struct cont_t reset_cont = { .done = generic_done }; +/* + * Resets the FDC connected to drive <drive>. + * Both current_drive and current_fdc are changed to match the new drive. + */ static int user_reset_fdc(int drive, int arg, bool interruptible) { int ret; @@ -3006,6 +3022,9 @@ static int user_reset_fdc(int drive, int arg, bool interruptible) if (arg == FD_RESET_ALWAYS) fdc_state[current_fdc].reset = 1; if (fdc_state[current_fdc].reset) { + /* note: reset_fdc will take care of unlocking the driver + * on completion. + */ cont = &reset_cont; ret = wait_til_done(reset_fdc, interruptible); if (ret == -EINTR) @@ -3059,7 +3078,7 @@ static void raw_cmd_done(int flag) raw_cmd->flags |= FD_RAW_HARDFAILURE; } else { raw_cmd->reply_count = inr; - if (raw_cmd->reply_count > MAX_REPLIES) + if (raw_cmd->reply_count > FD_RAW_REPLY_SIZE) raw_cmd->reply_count = 0; for (i = 0; i < raw_cmd->reply_count; i++) raw_cmd->reply[i] = reply_buffer[i]; @@ -3170,18 +3189,10 @@ loop: if (ret) return -EFAULT; param += sizeof(struct floppy_raw_cmd); - if (ptr->cmd_count > 33) - /* the command may now also take up the space - * initially intended for the reply & the - * reply count. Needed for long 82078 commands - * such as RESTORE, which takes ... 17 command - * bytes. Murphy's law #137: When you reserve - * 16 bytes for a structure, you'll one day - * discover that you really need 17... - */ + if (ptr->cmd_count > FD_RAW_CMD_FULLSIZE) return -EINVAL; - for (i = 0; i < 16; i++) + for (i = 0; i < FD_RAW_REPLY_SIZE; i++) ptr->reply[i] = 0; ptr->resultcode = 0; @@ -3423,13 +3434,13 @@ static int fd_getgeo(struct block_device *bdev, struct hd_geometry *geo) return 0; } -static bool valid_floppy_drive_params(const short autodetect[8], +static bool valid_floppy_drive_params(const short autodetect[FD_AUTODETECT_SIZE], int native_format) { size_t floppy_type_size = ARRAY_SIZE(floppy_type); size_t i = 0; - for (i = 0; i < 8; ++i) { + for (i = 0; i < FD_AUTODETECT_SIZE; ++i) { if (autodetect[i] < 0 || autodetect[i] >= floppy_type_size) return false; @@ -3610,7 +3621,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int case FDTWADDLE: if (lock_fdc(drive)) return -EINTR; - twaddle(); + twaddle(current_fdc, current_drive); process_fd_request(); return 0; default: @@ -3654,7 +3665,7 @@ struct compat_floppy_drive_params { struct floppy_max_errors max_errors; char flags; char read_track; - short autodetect[8]; + short autodetect[FD_AUTODETECT_SIZE]; compat_int_t checkfreq; compat_int_t native_format; }; @@ -4298,79 +4309,79 @@ static const struct block_device_operations floppy_fops = { /* Determine the floppy disk controller type */ /* This routine was written by David C. Niemi */ -static char __init get_fdc_version(void) +static char __init get_fdc_version(int fdc) { int r; - output_byte(FD_DUMPREGS); /* 82072 and better know DUMPREGS */ - if (fdc_state[current_fdc].reset) + output_byte(fdc, FD_DUMPREGS); /* 82072 and better know DUMPREGS */ + if (fdc_state[fdc].reset) return FDC_NONE; - r = result(); + r = result(fdc); if (r <= 0x00) return FDC_NONE; /* No FDC present ??? */ if ((r == 1) && (reply_buffer[0] == 0x80)) { - pr_info("FDC %d is an 8272A\n", current_fdc); + pr_info("FDC %d is an 8272A\n", fdc); return FDC_8272A; /* 8272a/765 don't know DUMPREGS */ } if (r != 10) { pr_info("FDC %d init: DUMPREGS: unexpected return of %d bytes.\n", - current_fdc, r); + fdc, r); return FDC_UNKNOWN; } - if (!fdc_configure()) { - pr_info("FDC %d is an 82072\n", current_fdc); + if (!fdc_configure(fdc)) { + pr_info("FDC %d is an 82072\n", fdc); return FDC_82072; /* 82072 doesn't know CONFIGURE */ } - output_byte(FD_PERPENDICULAR); - if (need_more_output() == MORE_OUTPUT) { - output_byte(0); + output_byte(fdc, FD_PERPENDICULAR); + if (need_more_output(fdc) == MORE_OUTPUT) { + output_byte(fdc, 0); } else { - pr_info("FDC %d is an 82072A\n", current_fdc); + pr_info("FDC %d is an 82072A\n", fdc); return FDC_82072A; /* 82072A as found on Sparcs. */ } - output_byte(FD_UNLOCK); - r = result(); + output_byte(fdc, FD_UNLOCK); + r = result(fdc); if ((r == 1) && (reply_buffer[0] == 0x80)) { - pr_info("FDC %d is a pre-1991 82077\n", current_fdc); + pr_info("FDC %d is a pre-1991 82077\n", fdc); return FDC_82077_ORIG; /* Pre-1991 82077, doesn't know * LOCK/UNLOCK */ } if ((r != 1) || (reply_buffer[0] != 0x00)) { pr_info("FDC %d init: UNLOCK: unexpected return of %d bytes.\n", - current_fdc, r); + fdc, r); return FDC_UNKNOWN; } - output_byte(FD_PARTID); - r = result(); + output_byte(fdc, FD_PARTID); + r = result(fdc); if (r != 1) { pr_info("FDC %d init: PARTID: unexpected return of %d bytes.\n", - current_fdc, r); + fdc, r); return FDC_UNKNOWN; } if (reply_buffer[0] == 0x80) { - pr_info("FDC %d is a post-1991 82077\n", current_fdc); + pr_info("FDC %d is a post-1991 82077\n", fdc); return FDC_82077; /* Revised 82077AA passes all the tests */ } switch (reply_buffer[0] >> 5) { case 0x0: /* Either a 82078-1 or a 82078SL running at 5Volt */ - pr_info("FDC %d is an 82078.\n", current_fdc); + pr_info("FDC %d is an 82078.\n", fdc); return FDC_82078; case 0x1: - pr_info("FDC %d is a 44pin 82078\n", current_fdc); + pr_info("FDC %d is a 44pin 82078\n", fdc); return FDC_82078; case 0x2: - pr_info("FDC %d is a S82078B\n", current_fdc); + pr_info("FDC %d is a S82078B\n", fdc); return FDC_S82078B; case 0x3: - pr_info("FDC %d is a National Semiconductor PC87306\n", current_fdc); + pr_info("FDC %d is a National Semiconductor PC87306\n", fdc); return FDC_87306; default: pr_info("FDC %d init: 82078 variant with unknown PARTID=%d.\n", - current_fdc, reply_buffer[0] >> 5); + fdc, reply_buffer[0] >> 5); return FDC_82078_UNKN; } } /* get_fdc_version */ @@ -4534,11 +4545,13 @@ static void floppy_device_release(struct device *dev) static int floppy_resume(struct device *dev) { int fdc; + int saved_drive; + saved_drive = current_drive; for (fdc = 0; fdc < N_FDC; fdc++) if (fdc_state[fdc].address != -1) - user_reset_fdc(-1, FD_RESET_ALWAYS, false); - + user_reset_fdc(REVDRIVE(fdc, 0), FD_RESET_ALWAYS, false); + set_fdc(saved_drive); return 0; } @@ -4646,16 +4659,15 @@ static int __init do_floppy_init(void) config_types(); for (i = 0; i < N_FDC; i++) { - current_fdc = i; - memset(&fdc_state[current_fdc], 0, sizeof(*fdc_state)); - fdc_state[current_fdc].dtr = -1; - fdc_state[current_fdc].dor = 0x4; + memset(&fdc_state[i], 0, sizeof(*fdc_state)); + fdc_state[i].dtr = -1; + fdc_state[i].dor = 0x4; #if defined(__sparc__) || defined(__mc68000__) /*sparcs/sun3x don't have a DOR reset which we can fall back on to */ #ifdef __mc68000__ if (MACH_IS_SUN3X) #endif - fdc_state[current_fdc].version = FDC_82072A; + fdc_state[i].version = FDC_82072A; #endif } @@ -4697,30 +4709,29 @@ static int __init do_floppy_init(void) msleep(10); for (i = 0; i < N_FDC; i++) { - current_fdc = i; - fdc_state[current_fdc].driver_version = FD_DRIVER_VERSION; + fdc_state[i].driver_version = FD_DRIVER_VERSION; for (unit = 0; unit < 4; unit++) - fdc_state[current_fdc].track[unit] = 0; - if (fdc_state[current_fdc].address == -1) + fdc_state[i].track[unit] = 0; + if (fdc_state[i].address == -1) continue; - fdc_state[current_fdc].rawcmd = 2; - if (user_reset_fdc(-1, FD_RESET_ALWAYS, false)) { + fdc_state[i].rawcmd = 2; + if (user_reset_fdc(REVDRIVE(i, 0), FD_RESET_ALWAYS, false)) { /* free ioports reserved by floppy_grab_irq_and_dma() */ - floppy_release_regions(current_fdc); - fdc_state[current_fdc].address = -1; - fdc_state[current_fdc].version = FDC_NONE; + floppy_release_regions(i); + fdc_state[i].address = -1; + fdc_state[i].version = FDC_NONE; continue; } /* Try to determine the floppy controller type */ - fdc_state[current_fdc].version = get_fdc_version(); - if (fdc_state[current_fdc].version == FDC_NONE) { + fdc_state[i].version = get_fdc_version(i); + if (fdc_state[i].version == FDC_NONE) { /* free ioports reserved by floppy_grab_irq_and_dma() */ - floppy_release_regions(current_fdc); - fdc_state[current_fdc].address = -1; + floppy_release_regions(i); + fdc_state[i].address = -1; continue; } if (can_use_virtual_dma == 2 && - fdc_state[current_fdc].version < FDC_82072A) + fdc_state[i].version < FDC_82072A) can_use_virtual_dma = 0; have_no_fdc = 0; @@ -4728,7 +4739,7 @@ static int __init do_floppy_init(void) * properly, so force a reset for the standard FDC clones, * to avoid interrupt garbage. */ - user_reset_fdc(-1, FD_RESET_ALWAYS, false); + user_reset_fdc(REVDRIVE(i, 0), FD_RESET_ALWAYS, false); } current_fdc = 0; cancel_delayed_work(&fd_timeout); @@ -4855,6 +4866,8 @@ static void floppy_release_regions(int fdc) static int floppy_grab_irq_and_dma(void) { + int fdc; + if (atomic_inc_return(&usage_count) > 1) return 0; @@ -4882,24 +4895,24 @@ static int floppy_grab_irq_and_dma(void) } } - for (current_fdc = 0; current_fdc < N_FDC; current_fdc++) { - if (fdc_state[current_fdc].address != -1) { - if (floppy_request_regions(current_fdc)) + for (fdc = 0; fdc < N_FDC; fdc++) { + if (fdc_state[fdc].address != -1) { + if (floppy_request_regions(fdc)) goto cleanup; } } - for (current_fdc = 0; current_fdc < N_FDC; current_fdc++) { - if (fdc_state[current_fdc].address != -1) { - reset_fdc_info(1); - fdc_outb(fdc_state[current_fdc].dor, current_fdc, FD_DOR); + for (fdc = 0; fdc < N_FDC; fdc++) { + if (fdc_state[fdc].address != -1) { + reset_fdc_info(fdc, 1); + fdc_outb(fdc_state[fdc].dor, fdc, FD_DOR); } } - current_fdc = 0; + set_dor(0, ~0, 8); /* avoid immediate interrupt */ - for (current_fdc = 0; current_fdc < N_FDC; current_fdc++) - if (fdc_state[current_fdc].address != -1) - fdc_outb(fdc_state[current_fdc].dor, current_fdc, FD_DOR); + for (fdc = 0; fdc < N_FDC; fdc++) + if (fdc_state[fdc].address != -1) + fdc_outb(fdc_state[fdc].dor, fdc, FD_DOR); /* * The driver will try and free resources and relies on us * to know if they were allocated or not. @@ -4910,15 +4923,16 @@ static int floppy_grab_irq_and_dma(void) cleanup: fd_free_irq(); fd_free_dma(); - while (--current_fdc >= 0) - floppy_release_regions(current_fdc); + while (--fdc >= 0) + floppy_release_regions(fdc); + current_fdc = 0; atomic_dec(&usage_count); return -1; } static void floppy_release_irq_and_dma(void) { - int old_fdc; + int fdc; #ifndef __sparc__ int drive; #endif @@ -4959,11 +4973,9 @@ static void floppy_release_irq_and_dma(void) pr_info("auxiliary floppy timer still active\n"); if (work_pending(&floppy_work)) pr_info("work still pending\n"); - old_fdc = current_fdc; - for (current_fdc = 0; current_fdc < N_FDC; current_fdc++) - if (fdc_state[current_fdc].address != -1) - floppy_release_regions(current_fdc); - current_fdc = old_fdc; + for (fdc = 0; fdc < N_FDC; fdc++) + if (fdc_state[fdc].address != -1) + floppy_release_regions(fdc); } #ifdef MODULE diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 14372df0f354..2e96d8b8758b 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -228,26 +228,36 @@ static void __loop_update_dio(struct loop_device *lo, bool dio) blk_mq_unfreeze_queue(lo->lo_queue); } +/** + * loop_validate_block_size() - validates the passed in block size + * @bsize: size to validate + */ static int -figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit) +loop_validate_block_size(unsigned short bsize) { - loff_t size = get_size(offset, sizelimit, lo->lo_backing_file); - sector_t x = (sector_t)size; - struct block_device *bdev = lo->lo_device; + if (bsize < 512 || bsize > PAGE_SIZE || !is_power_of_2(bsize)) + return -EINVAL; - if (unlikely((loff_t)x != size)) - return -EFBIG; - if (lo->lo_offset != offset) - lo->lo_offset = offset; - if (lo->lo_sizelimit != sizelimit) - lo->lo_sizelimit = sizelimit; - set_capacity(lo->lo_disk, x); - bd_set_size(bdev, (loff_t)get_capacity(bdev->bd_disk) << 9); - /* let user-space know about the new size */ - kobject_uevent(&disk_to_dev(bdev->bd_disk)->kobj, KOBJ_CHANGE); return 0; } +/** + * loop_set_size() - sets device size and notifies userspace + * @lo: struct loop_device to set the size for + * @size: new size of the loop device + * + * Callers must validate that the size passed into this function fits into + * a sector_t, eg using loop_validate_size() + */ +static void loop_set_size(struct loop_device *lo, loff_t size) +{ + struct block_device *bdev = lo->lo_device; + + bd_set_size(bdev, size << SECTOR_SHIFT); + + set_capacity_revalidate_and_notify(lo->lo_disk, size, false); +} + static inline int lo_do_transfer(struct loop_device *lo, int cmd, struct page *rpage, unsigned roffs, @@ -919,7 +929,7 @@ static void loop_unprepare_queue(struct loop_device *lo) static int loop_kthread_worker_fn(void *worker_ptr) { - current->flags |= PF_LESS_THROTTLE | PF_MEMALLOC_NOIO; + current->flags |= PF_LOCAL_THROTTLE | PF_MEMALLOC_NOIO; return kthread_worker_fn(worker_ptr); } @@ -952,23 +962,125 @@ static void loop_update_rotational(struct loop_device *lo) blk_queue_flag_clear(QUEUE_FLAG_NONROT, q); } -static int loop_set_fd(struct loop_device *lo, fmode_t mode, - struct block_device *bdev, unsigned int arg) +static int +loop_release_xfer(struct loop_device *lo) +{ + int err = 0; + struct loop_func_table *xfer = lo->lo_encryption; + + if (xfer) { + if (xfer->release) + err = xfer->release(lo); + lo->transfer = NULL; + lo->lo_encryption = NULL; + module_put(xfer->owner); + } + return err; +} + +static int +loop_init_xfer(struct loop_device *lo, struct loop_func_table *xfer, + const struct loop_info64 *i) +{ + int err = 0; + + if (xfer) { + struct module *owner = xfer->owner; + + if (!try_module_get(owner)) + return -EINVAL; + if (xfer->init) + err = xfer->init(lo, i); + if (err) + module_put(owner); + else + lo->lo_encryption = xfer; + } + return err; +} + +/** + * loop_set_status_from_info - configure device from loop_info + * @lo: struct loop_device to configure + * @info: struct loop_info64 to configure the device with + * + * Configures the loop device parameters according to the passed + * in loop_info64 configuration. + */ +static int +loop_set_status_from_info(struct loop_device *lo, + const struct loop_info64 *info) +{ + int err; + struct loop_func_table *xfer; + kuid_t uid = current_uid(); + + if ((unsigned int) info->lo_encrypt_key_size > LO_KEY_SIZE) + return -EINVAL; + + err = loop_release_xfer(lo); + if (err) + return err; + + if (info->lo_encrypt_type) { + unsigned int type = info->lo_encrypt_type; + + if (type >= MAX_LO_CRYPT) + return -EINVAL; + xfer = xfer_funcs[type]; + if (xfer == NULL) + return -EINVAL; + } else + xfer = NULL; + + err = loop_init_xfer(lo, xfer, info); + if (err) + return err; + + lo->lo_offset = info->lo_offset; + lo->lo_sizelimit = info->lo_sizelimit; + memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE); + memcpy(lo->lo_crypt_name, info->lo_crypt_name, LO_NAME_SIZE); + lo->lo_file_name[LO_NAME_SIZE-1] = 0; + lo->lo_crypt_name[LO_NAME_SIZE-1] = 0; + + if (!xfer) + xfer = &none_funcs; + lo->transfer = xfer->transfer; + lo->ioctl = xfer->ioctl; + + lo->lo_flags = info->lo_flags; + + lo->lo_encrypt_key_size = info->lo_encrypt_key_size; + lo->lo_init[0] = info->lo_init[0]; + lo->lo_init[1] = info->lo_init[1]; + if (info->lo_encrypt_key_size) { + memcpy(lo->lo_encrypt_key, info->lo_encrypt_key, + info->lo_encrypt_key_size); + lo->lo_key_owner = uid; + } + + return 0; +} + +static int loop_configure(struct loop_device *lo, fmode_t mode, + struct block_device *bdev, + const struct loop_config *config) { struct file *file; struct inode *inode; struct address_space *mapping; struct block_device *claimed_bdev = NULL; - int lo_flags = 0; int error; loff_t size; bool partscan; + unsigned short bsize; /* This is safe, since we have a reference from open(). */ __module_get(THIS_MODULE); error = -EBADF; - file = fget(arg); + file = fget(config->fd); if (!file) goto out; @@ -977,7 +1089,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, * here to avoid changing device under exclusive owner. */ if (!(mode & FMODE_EXCL)) { - claimed_bdev = bd_start_claiming(bdev, loop_set_fd); + claimed_bdev = bd_start_claiming(bdev, loop_configure); if (IS_ERR(claimed_bdev)) { error = PTR_ERR(claimed_bdev); goto out_putf; @@ -999,52 +1111,58 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, mapping = file->f_mapping; inode = mapping->host; + size = get_loop_size(lo, file); + + if ((config->info.lo_flags & ~LOOP_CONFIGURE_SETTABLE_FLAGS) != 0) { + error = -EINVAL; + goto out_unlock; + } + + if (config->block_size) { + error = loop_validate_block_size(config->block_size); + if (error) + goto out_unlock; + } + + error = loop_set_status_from_info(lo, &config->info); + if (error) + goto out_unlock; + if (!(file->f_mode & FMODE_WRITE) || !(mode & FMODE_WRITE) || !file->f_op->write_iter) - lo_flags |= LO_FLAGS_READ_ONLY; + lo->lo_flags |= LO_FLAGS_READ_ONLY; - error = -EFBIG; - size = get_loop_size(lo, file); - if ((loff_t)(sector_t)size != size) - goto out_unlock; error = loop_prepare_queue(lo); if (error) goto out_unlock; - error = 0; - - set_device_ro(bdev, (lo_flags & LO_FLAGS_READ_ONLY) != 0); + set_device_ro(bdev, (lo->lo_flags & LO_FLAGS_READ_ONLY) != 0); - lo->use_dio = false; + lo->use_dio = lo->lo_flags & LO_FLAGS_DIRECT_IO; lo->lo_device = bdev; - lo->lo_flags = lo_flags; lo->lo_backing_file = file; - lo->transfer = NULL; - lo->ioctl = NULL; - lo->lo_sizelimit = 0; lo->old_gfp_mask = mapping_gfp_mask(mapping); mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS)); - if (!(lo_flags & LO_FLAGS_READ_ONLY) && file->f_op->fsync) + if (!(lo->lo_flags & LO_FLAGS_READ_ONLY) && file->f_op->fsync) blk_queue_write_cache(lo->lo_queue, true, false); - if ((lo->lo_backing_file->f_flags & O_DIRECT) && inode->i_sb->s_bdev) { + if (config->block_size) + bsize = config->block_size; + else if ((lo->lo_backing_file->f_flags & O_DIRECT) && inode->i_sb->s_bdev) /* In case of direct I/O, match underlying block size */ - unsigned short bsize = bdev_logical_block_size( - inode->i_sb->s_bdev); + bsize = bdev_logical_block_size(inode->i_sb->s_bdev); + else + bsize = 512; - blk_queue_logical_block_size(lo->lo_queue, bsize); - blk_queue_physical_block_size(lo->lo_queue, bsize); - blk_queue_io_min(lo->lo_queue, bsize); - } + blk_queue_logical_block_size(lo->lo_queue, bsize); + blk_queue_physical_block_size(lo->lo_queue, bsize); + blk_queue_io_min(lo->lo_queue, bsize); loop_update_rotational(lo); loop_update_dio(lo); - set_capacity(lo->lo_disk, size); - bd_set_size(bdev, size << 9); loop_sysfs_init(lo); - /* let user-space know about the new size */ - kobject_uevent(&disk_to_dev(bdev->bd_disk)->kobj, KOBJ_CHANGE); + loop_set_size(lo, size); set_blocksize(bdev, S_ISBLK(inode->i_mode) ? block_size(inode->i_bdev) : PAGE_SIZE); @@ -1062,14 +1180,14 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, if (partscan) loop_reread_partitions(lo, bdev); if (claimed_bdev) - bd_abort_claiming(bdev, claimed_bdev, loop_set_fd); + bd_abort_claiming(bdev, claimed_bdev, loop_configure); return 0; out_unlock: mutex_unlock(&loop_ctl_mutex); out_bdev: if (claimed_bdev) - bd_abort_claiming(bdev, claimed_bdev, loop_set_fd); + bd_abort_claiming(bdev, claimed_bdev, loop_configure); out_putf: fput(file); out: @@ -1078,43 +1196,6 @@ out: return error; } -static int -loop_release_xfer(struct loop_device *lo) -{ - int err = 0; - struct loop_func_table *xfer = lo->lo_encryption; - - if (xfer) { - if (xfer->release) - err = xfer->release(lo); - lo->transfer = NULL; - lo->lo_encryption = NULL; - module_put(xfer->owner); - } - return err; -} - -static int -loop_init_xfer(struct loop_device *lo, struct loop_func_table *xfer, - const struct loop_info64 *i) -{ - int err = 0; - - if (xfer) { - struct module *owner = xfer->owner; - - if (!try_module_get(owner)) - return -EINVAL; - if (xfer->init) - err = xfer->init(lo, i); - if (err) - module_put(owner); - else - lo->lo_encryption = xfer; - } - return err; -} - static int __loop_clr_fd(struct loop_device *lo, bool release) { struct file *filp = NULL; @@ -1263,10 +1344,11 @@ static int loop_set_status(struct loop_device *lo, const struct loop_info64 *info) { int err; - struct loop_func_table *xfer; - kuid_t uid = current_uid(); struct block_device *bdev; + kuid_t uid = current_uid(); + int prev_lo_flags; bool partscan = false; + bool size_changed = false; err = mutex_lock_killable(&loop_ctl_mutex); if (err) @@ -1281,13 +1363,10 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) err = -ENXIO; goto out_unlock; } - if ((unsigned int) info->lo_encrypt_key_size > LO_KEY_SIZE) { - err = -EINVAL; - goto out_unlock; - } if (lo->lo_offset != info->lo_offset || lo->lo_sizelimit != info->lo_sizelimit) { + size_changed = true; sync_blockdev(lo->lo_device); kill_bdev(lo->lo_device); } @@ -1295,79 +1374,44 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) /* I/O need to be drained during transfer transition */ blk_mq_freeze_queue(lo->lo_queue); - err = loop_release_xfer(lo); - if (err) + if (size_changed && lo->lo_device->bd_inode->i_mapping->nrpages) { + /* If any pages were dirtied after kill_bdev(), try again */ + err = -EAGAIN; + pr_warn("%s: loop%d (%s) has still dirty pages (nrpages=%lu)\n", + __func__, lo->lo_number, lo->lo_file_name, + lo->lo_device->bd_inode->i_mapping->nrpages); goto out_unfreeze; + } - if (info->lo_encrypt_type) { - unsigned int type = info->lo_encrypt_type; - - if (type >= MAX_LO_CRYPT) { - err = -EINVAL; - goto out_unfreeze; - } - xfer = xfer_funcs[type]; - if (xfer == NULL) { - err = -EINVAL; - goto out_unfreeze; - } - } else - xfer = NULL; + prev_lo_flags = lo->lo_flags; - err = loop_init_xfer(lo, xfer, info); + err = loop_set_status_from_info(lo, info); if (err) goto out_unfreeze; - if (lo->lo_offset != info->lo_offset || - lo->lo_sizelimit != info->lo_sizelimit) { - /* kill_bdev should have truncated all the pages */ - if (lo->lo_device->bd_inode->i_mapping->nrpages) { - err = -EAGAIN; - pr_warn("%s: loop%d (%s) has still dirty pages (nrpages=%lu)\n", - __func__, lo->lo_number, lo->lo_file_name, - lo->lo_device->bd_inode->i_mapping->nrpages); - goto out_unfreeze; - } - if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit)) { - err = -EFBIG; - goto out_unfreeze; - } + /* Mask out flags that can't be set using LOOP_SET_STATUS. */ + lo->lo_flags &= ~LOOP_SET_STATUS_SETTABLE_FLAGS; + /* For those flags, use the previous values instead */ + lo->lo_flags |= prev_lo_flags & ~LOOP_SET_STATUS_SETTABLE_FLAGS; + /* For flags that can't be cleared, use previous values too */ + lo->lo_flags |= prev_lo_flags & ~LOOP_SET_STATUS_CLEARABLE_FLAGS; + + if (size_changed) { + loff_t new_size = get_size(lo->lo_offset, lo->lo_sizelimit, + lo->lo_backing_file); + loop_set_size(lo, new_size); } loop_config_discard(lo); - memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE); - memcpy(lo->lo_crypt_name, info->lo_crypt_name, LO_NAME_SIZE); - lo->lo_file_name[LO_NAME_SIZE-1] = 0; - lo->lo_crypt_name[LO_NAME_SIZE-1] = 0; - - if (!xfer) - xfer = &none_funcs; - lo->transfer = xfer->transfer; - lo->ioctl = xfer->ioctl; - - if ((lo->lo_flags & LO_FLAGS_AUTOCLEAR) != - (info->lo_flags & LO_FLAGS_AUTOCLEAR)) - lo->lo_flags ^= LO_FLAGS_AUTOCLEAR; - - lo->lo_encrypt_key_size = info->lo_encrypt_key_size; - lo->lo_init[0] = info->lo_init[0]; - lo->lo_init[1] = info->lo_init[1]; - if (info->lo_encrypt_key_size) { - memcpy(lo->lo_encrypt_key, info->lo_encrypt_key, - info->lo_encrypt_key_size); - lo->lo_key_owner = uid; - } - /* update dio if lo_offset or transfer is changed */ __loop_update_dio(lo, lo->use_dio); out_unfreeze: blk_mq_unfreeze_queue(lo->lo_queue); - if (!err && (info->lo_flags & LO_FLAGS_PARTSCAN) && - !(lo->lo_flags & LO_FLAGS_PARTSCAN)) { - lo->lo_flags |= LO_FLAGS_PARTSCAN; + if (!err && (lo->lo_flags & LO_FLAGS_PARTSCAN) && + !(prev_lo_flags & LO_FLAGS_PARTSCAN)) { lo->lo_disk->flags &= ~GENHD_FL_NO_PART_SCAN; bdev = lo->lo_device; partscan = true; @@ -1531,10 +1575,15 @@ loop_get_status64(struct loop_device *lo, struct loop_info64 __user *arg) { static int loop_set_capacity(struct loop_device *lo) { + loff_t size; + if (unlikely(lo->lo_state != Lo_bound)) return -ENXIO; - return figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit); + size = get_loop_size(lo, lo->lo_backing_file); + loop_set_size(lo, size); + + return 0; } static int loop_set_dio(struct loop_device *lo, unsigned long arg) @@ -1558,8 +1607,9 @@ static int loop_set_block_size(struct loop_device *lo, unsigned long arg) if (lo->lo_state != Lo_bound) return -ENXIO; - if (arg < 512 || arg > PAGE_SIZE || !is_power_of_2(arg)) - return -EINVAL; + err = loop_validate_block_size(arg); + if (err) + return err; if (lo->lo_queue->limits.logical_block_size == arg) return 0; @@ -1617,11 +1667,31 @@ static int lo_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd, unsigned long arg) { struct loop_device *lo = bdev->bd_disk->private_data; + void __user *argp = (void __user *) arg; int err; switch (cmd) { - case LOOP_SET_FD: - return loop_set_fd(lo, mode, bdev, arg); + case LOOP_SET_FD: { + /* + * Legacy case - pass in a zeroed out struct loop_config with + * only the file descriptor set , which corresponds with the + * default parameters we'd have used otherwise. + */ + struct loop_config config; + + memset(&config, 0, sizeof(config)); + config.fd = arg; + + return loop_configure(lo, mode, bdev, &config); + } + case LOOP_CONFIGURE: { + struct loop_config config; + + if (copy_from_user(&config, argp, sizeof(config))) + return -EFAULT; + + return loop_configure(lo, mode, bdev, &config); + } case LOOP_CHANGE_FD: return loop_change_fd(lo, bdev, arg); case LOOP_CLR_FD: @@ -1629,21 +1699,19 @@ static int lo_ioctl(struct block_device *bdev, fmode_t mode, case LOOP_SET_STATUS: err = -EPERM; if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN)) { - err = loop_set_status_old(lo, - (struct loop_info __user *)arg); + err = loop_set_status_old(lo, argp); } break; case LOOP_GET_STATUS: - return loop_get_status_old(lo, (struct loop_info __user *) arg); + return loop_get_status_old(lo, argp); case LOOP_SET_STATUS64: err = -EPERM; if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN)) { - err = loop_set_status64(lo, - (struct loop_info64 __user *) arg); + err = loop_set_status64(lo, argp); } break; case LOOP_GET_STATUS64: - return loop_get_status64(lo, (struct loop_info64 __user *) arg); + return loop_get_status64(lo, argp); case LOOP_SET_CAPACITY: case LOOP_SET_DIRECT_IO: case LOOP_SET_BLOCK_SIZE: @@ -1795,6 +1863,7 @@ static int lo_compat_ioctl(struct block_device *bdev, fmode_t mode, case LOOP_CLR_FD: case LOOP_GET_STATUS64: case LOOP_SET_STATUS64: + case LOOP_CONFIGURE: arg = (unsigned long) compat_ptr(arg); /* fall through */ case LOOP_SET_FD: @@ -2037,7 +2106,7 @@ static int loop_add(struct loop_device **l, int i) lo->tag_set.queue_depth = 128; lo->tag_set.numa_node = NUMA_NO_NODE; lo->tag_set.cmd_size = sizeof(struct loop_cmd); - lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE; + lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING; lo->tag_set.driver_data = lo; err = blk_mq_alloc_tag_set(&lo->tag_set); diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c index 8efd8778e209..87b31f9ca362 100644 --- a/drivers/block/null_blk_main.c +++ b/drivers/block/null_blk_main.c @@ -1250,8 +1250,34 @@ static inline blk_status_t null_handle_memory_backed(struct nullb_cmd *cmd, return errno_to_blk_status(err); } +static void nullb_zero_read_cmd_buffer(struct nullb_cmd *cmd) +{ + struct nullb_device *dev = cmd->nq->dev; + struct bio *bio; + + if (dev->memory_backed) + return; + + if (dev->queue_mode == NULL_Q_BIO && bio_op(cmd->bio) == REQ_OP_READ) { + zero_fill_bio(cmd->bio); + } else if (req_op(cmd->rq) == REQ_OP_READ) { + __rq_for_each_bio(bio, cmd->rq) + zero_fill_bio(bio); + } +} + static inline void nullb_complete_cmd(struct nullb_cmd *cmd) { + /* + * Since root privileges are required to configure the null_blk + * driver, it is fine that this driver does not initialize the + * data buffers of read commands. Zero-initialize these buffers + * anyway if KMSAN is enabled to prevent that KMSAN complains + * about null_blk not initializing read data buffers. + */ + if (IS_ENABLED(CONFIG_KMSAN)) + nullb_zero_read_cmd_buffer(cmd); + /* Complete IO by inline, softirq or timer */ switch (cmd->nq->dev->irqmode) { case NULL_IRQ_SOFTIRQ: @@ -1397,7 +1423,7 @@ static bool should_requeue_request(struct request *rq) static enum blk_eh_timer_return null_timeout_rq(struct request *rq, bool res) { pr_info("rq %p timed out\n", rq); - blk_mq_complete_request(rq); + blk_mq_force_complete_rq(rq); return BLK_EH_DONE; } @@ -1535,6 +1561,13 @@ static void null_config_discard(struct nullb *nullb) { if (nullb->dev->discard == false) return; + + if (nullb->dev->zoned) { + nullb->dev->discard = false; + pr_info("discard option is ignored in zoned mode\n"); + return; + } + nullb->q->limits.discard_granularity = nullb->dev->blocksize; nullb->q->limits.discard_alignment = nullb->dev->blocksize; blk_queue_max_discard_sectors(nullb->q, UINT_MAX >> 9); diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c index 9e4bcdad1a80..cc47606d8ffe 100644 --- a/drivers/block/null_blk_zoned.c +++ b/drivers/block/null_blk_zoned.c @@ -23,6 +23,10 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q) pr_err("zone_size must be power-of-two\n"); return -EINVAL; } + if (dev->zone_size > dev->size) { + pr_err("Zone size larger than device capacity\n"); + return -EINVAL; + } dev->zone_size_sects = dev->zone_size << ZONE_SIZE_SHIFT; dev->nr_zones = dev_size >> @@ -70,13 +74,20 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q) int null_register_zoned_dev(struct nullb *nullb) { + struct nullb_device *dev = nullb->dev; struct request_queue *q = nullb->q; - if (queue_is_mq(q)) - return blk_revalidate_disk_zones(nullb->disk); + if (queue_is_mq(q)) { + int ret = blk_revalidate_disk_zones(nullb->disk, NULL); - blk_queue_chunk_sectors(q, nullb->dev->zone_size_sects); - q->nr_zones = blkdev_nr_zones(nullb->disk); + if (ret) + return ret; + } else { + blk_queue_chunk_sectors(q, dev->zone_size_sects); + q->nr_zones = blkdev_nr_zones(nullb->disk); + } + + blk_queue_max_zone_append_sectors(q, dev->zone_size_sects); return 0; } @@ -138,7 +149,7 @@ size_t null_zone_valid_read_len(struct nullb *nullb, } static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, - unsigned int nr_sectors) + unsigned int nr_sectors, bool append) { struct nullb_device *dev = cmd->nq->dev; unsigned int zno = null_zone_no(dev, sector); @@ -158,9 +169,21 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, case BLK_ZONE_COND_IMP_OPEN: case BLK_ZONE_COND_EXP_OPEN: case BLK_ZONE_COND_CLOSED: - /* Writes must be at the write pointer position */ - if (sector != zone->wp) + /* + * Regular writes must be at the write pointer position. + * Zone append writes are automatically issued at the write + * pointer and the position returned using the request or BIO + * sector. + */ + if (append) { + sector = zone->wp; + if (cmd->bio) + cmd->bio->bi_iter.bi_sector = sector; + else + cmd->rq->__sector = sector; + } else if (sector != zone->wp) { return BLK_STS_IOERR; + } if (zone->cond != BLK_ZONE_COND_EXP_OPEN) zone->cond = BLK_ZONE_COND_IMP_OPEN; @@ -242,7 +265,9 @@ blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op, { switch (op) { case REQ_OP_WRITE: - return null_zone_write(cmd, sector, nr_sectors); + return null_zone_write(cmd, sector, nr_sectors, false); + case REQ_OP_ZONE_APPEND: + return null_zone_write(cmd, sector, nr_sectors, true); case REQ_OP_ZONE_RESET: case REQ_OP_ZONE_RESET_ALL: case REQ_OP_ZONE_OPEN: diff --git a/drivers/block/paride/pcd.c b/drivers/block/paride/pcd.c index cda5cf917e9a..5124eca90e83 100644 --- a/drivers/block/paride/pcd.c +++ b/drivers/block/paride/pcd.c @@ -1032,7 +1032,7 @@ static int __init pcd_init(void) for (unit = 0, cd = pcd; unit < PCD_UNITS; unit++, cd++) { if (cd->present) { - register_cdrom(&cd->info); + register_cdrom(cd->disk, &cd->info); cd->disk->private_data = cd; add_disk(cd->disk); } diff --git a/drivers/block/rsxx/dev.c b/drivers/block/rsxx/dev.c index 8ffa8260dcaf..3ba07ab30c84 100644 --- a/drivers/block/rsxx/dev.c +++ b/drivers/block/rsxx/dev.c @@ -96,20 +96,6 @@ static const struct block_device_operations rsxx_fops = { .ioctl = rsxx_blkdev_ioctl, }; -static void disk_stats_start(struct rsxx_cardinfo *card, struct bio *bio) -{ - generic_start_io_acct(card->queue, bio_op(bio), bio_sectors(bio), - &card->gendisk->part0); -} - -static void disk_stats_complete(struct rsxx_cardinfo *card, - struct bio *bio, - unsigned long start_time) -{ - generic_end_io_acct(card->queue, bio_op(bio), - &card->gendisk->part0, start_time); -} - static void bio_dma_done_cb(struct rsxx_cardinfo *card, void *cb_data, unsigned int error) @@ -121,7 +107,7 @@ static void bio_dma_done_cb(struct rsxx_cardinfo *card, if (atomic_dec_and_test(&meta->pending_dmas)) { if (!card->eeh_state && card->gendisk) - disk_stats_complete(card, meta->bio, meta->start_time); + bio_end_io_acct(meta->bio, meta->start_time); if (atomic_read(&meta->error)) bio_io_error(meta->bio); @@ -167,10 +153,9 @@ static blk_qc_t rsxx_make_request(struct request_queue *q, struct bio *bio) bio_meta->bio = bio; atomic_set(&bio_meta->error, 0); atomic_set(&bio_meta->pending_dmas, 0); - bio_meta->start_time = jiffies; if (!unlikely(card->halt)) - disk_stats_start(card, bio); + bio_meta->start_time = bio_start_io_acct(bio); dev_dbg(CARD_TO_DEV(card), "BIO[%c]: meta: %p addr8: x%llx size: %d\n", bio_data_dir(bio) ? 'W' : 'R', bio_meta, diff --git a/drivers/block/swim.c b/drivers/block/swim.c index 4c297f69171d..dd34504382e5 100644 --- a/drivers/block/swim.c +++ b/drivers/block/swim.c @@ -327,7 +327,7 @@ static inline void swim_motor(struct swim __iomem *base, swim_select(base, RELAX); if (swim_readbit(base, MOTOR_ON)) break; - current->state = TASK_INTERRUPTIBLE; + set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(1); } } else if (action == OFF) { @@ -346,7 +346,7 @@ static inline void swim_eject(struct swim __iomem *base) swim_select(base, RELAX); if (!swim_readbit(base, DISK_IN)) break; - current->state = TASK_INTERRUPTIBLE; + set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(1); } swim_select(base, RELAX); @@ -370,7 +370,7 @@ static inline int swim_step(struct swim __iomem *base) for (wait = 0; wait < HZ; wait++) { - current->state = TASK_INTERRUPTIBLE; + set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(1); swim_select(base, RELAX); diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 93468b7c6701..9d21bf0f155e 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -33,6 +33,15 @@ struct virtio_blk_vq { } ____cacheline_aligned_in_smp; struct virtio_blk { + /* + * This mutex must be held by anything that may run after + * virtblk_remove() sets vblk->vdev to NULL. + * + * blk-mq, virtqueue processing, and sysfs attribute code paths are + * shut down before vblk->vdev is set to NULL and therefore do not need + * to hold this mutex. + */ + struct mutex vdev_mutex; struct virtio_device *vdev; /* The disk structure for the kernel. */ @@ -44,6 +53,13 @@ struct virtio_blk { /* Process context for config space updates */ struct work_struct config_work; + /* + * Tracks references from block_device_operations open/release and + * virtio_driver probe/remove so this object can be freed once no + * longer in use. + */ + refcount_t refs; + /* What host tells us, plus 2 for header & tailer. */ unsigned int sg_elems; @@ -295,10 +311,55 @@ out: return err; } +static void virtblk_get(struct virtio_blk *vblk) +{ + refcount_inc(&vblk->refs); +} + +static void virtblk_put(struct virtio_blk *vblk) +{ + if (refcount_dec_and_test(&vblk->refs)) { + ida_simple_remove(&vd_index_ida, vblk->index); + mutex_destroy(&vblk->vdev_mutex); + kfree(vblk); + } +} + +static int virtblk_open(struct block_device *bd, fmode_t mode) +{ + struct virtio_blk *vblk = bd->bd_disk->private_data; + int ret = 0; + + mutex_lock(&vblk->vdev_mutex); + + if (vblk->vdev) + virtblk_get(vblk); + else + ret = -ENXIO; + + mutex_unlock(&vblk->vdev_mutex); + return ret; +} + +static void virtblk_release(struct gendisk *disk, fmode_t mode) +{ + struct virtio_blk *vblk = disk->private_data; + + virtblk_put(vblk); +} + /* We provide getgeo only to please some old bootloader/partitioning tools */ static int virtblk_getgeo(struct block_device *bd, struct hd_geometry *geo) { struct virtio_blk *vblk = bd->bd_disk->private_data; + int ret = 0; + + mutex_lock(&vblk->vdev_mutex); + + if (!vblk->vdev) { + ret = -ENXIO; + goto out; + } /* see if the host passed in geometry config */ if (virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_GEOMETRY)) { @@ -314,11 +375,15 @@ static int virtblk_getgeo(struct block_device *bd, struct hd_geometry *geo) geo->sectors = 1 << 5; geo->cylinders = get_capacity(bd->bd_disk) >> 11; } - return 0; +out: + mutex_unlock(&vblk->vdev_mutex); + return ret; } static const struct block_device_operations virtblk_fops = { .owner = THIS_MODULE, + .open = virtblk_open, + .release = virtblk_release, .getgeo = virtblk_getgeo, }; @@ -655,6 +720,10 @@ static int virtblk_probe(struct virtio_device *vdev) goto out_free_index; } + /* This reference is dropped in virtblk_remove(). */ + refcount_set(&vblk->refs, 1); + mutex_init(&vblk->vdev_mutex); + vblk->vdev = vdev; vblk->sg_elems = sg_elems; @@ -820,8 +889,6 @@ out: static void virtblk_remove(struct virtio_device *vdev) { struct virtio_blk *vblk = vdev->priv; - int index = vblk->index; - int refc; /* Make sure no work handler is accessing the device. */ flush_work(&vblk->config_work); @@ -831,18 +898,21 @@ static void virtblk_remove(struct virtio_device *vdev) blk_mq_free_tag_set(&vblk->tag_set); + mutex_lock(&vblk->vdev_mutex); + /* Stop all the virtqueues. */ vdev->config->reset(vdev); - refc = kref_read(&disk_to_dev(vblk->disk)->kobj.kref); + /* Virtqueues are stopped, nothing can use vblk->vdev anymore. */ + vblk->vdev = NULL; + put_disk(vblk->disk); vdev->config->del_vqs(vdev); kfree(vblk->vqs); - kfree(vblk); - /* Only free device id if we don't have any users */ - if (refc == 1) - ida_simple_remove(&vd_index_ida, index); + mutex_unlock(&vblk->vdev_mutex); + + virtblk_put(vblk); } #ifdef CONFIG_PM_SLEEP diff --git a/drivers/block/zram/zcomp.c b/drivers/block/zram/zcomp.c index 1a8564a79d8d..5ee8e3fae551 100644 --- a/drivers/block/zram/zcomp.c +++ b/drivers/block/zram/zcomp.c @@ -37,19 +37,16 @@ static void zcomp_strm_free(struct zcomp_strm *zstrm) if (!IS_ERR_OR_NULL(zstrm->tfm)) crypto_free_comp(zstrm->tfm); free_pages((unsigned long)zstrm->buffer, 1); - kfree(zstrm); + zstrm->tfm = NULL; + zstrm->buffer = NULL; } /* - * allocate new zcomp_strm structure with ->tfm initialized by - * backend, return NULL on error + * Initialize zcomp_strm structure with ->tfm initialized by backend, and + * ->buffer. Return a negative value on error. */ -static struct zcomp_strm *zcomp_strm_alloc(struct zcomp *comp) +static int zcomp_strm_init(struct zcomp_strm *zstrm, struct zcomp *comp) { - struct zcomp_strm *zstrm = kmalloc(sizeof(*zstrm), GFP_KERNEL); - if (!zstrm) - return NULL; - zstrm->tfm = crypto_alloc_comp(comp->name, 0, 0); /* * allocate 2 pages. 1 for compressed data, plus 1 extra for the @@ -58,9 +55,9 @@ static struct zcomp_strm *zcomp_strm_alloc(struct zcomp *comp) zstrm->buffer = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, 1); if (IS_ERR_OR_NULL(zstrm->tfm) || !zstrm->buffer) { zcomp_strm_free(zstrm); - zstrm = NULL; + return -ENOMEM; } - return zstrm; + return 0; } bool zcomp_available_algorithm(const char *comp) @@ -113,12 +110,13 @@ ssize_t zcomp_available_show(const char *comp, char *buf) struct zcomp_strm *zcomp_stream_get(struct zcomp *comp) { - return *get_cpu_ptr(comp->stream); + local_lock(&comp->stream->lock); + return this_cpu_ptr(comp->stream); } void zcomp_stream_put(struct zcomp *comp) { - put_cpu_ptr(comp->stream); + local_unlock(&comp->stream->lock); } int zcomp_compress(struct zcomp_strm *zstrm, @@ -159,17 +157,15 @@ int zcomp_cpu_up_prepare(unsigned int cpu, struct hlist_node *node) { struct zcomp *comp = hlist_entry(node, struct zcomp, node); struct zcomp_strm *zstrm; + int ret; - if (WARN_ON(*per_cpu_ptr(comp->stream, cpu))) - return 0; + zstrm = per_cpu_ptr(comp->stream, cpu); + local_lock_init(&zstrm->lock); - zstrm = zcomp_strm_alloc(comp); - if (IS_ERR_OR_NULL(zstrm)) { + ret = zcomp_strm_init(zstrm, comp); + if (ret) pr_err("Can't allocate a compression stream\n"); - return -ENOMEM; - } - *per_cpu_ptr(comp->stream, cpu) = zstrm; - return 0; + return ret; } int zcomp_cpu_dead(unsigned int cpu, struct hlist_node *node) @@ -177,10 +173,8 @@ int zcomp_cpu_dead(unsigned int cpu, struct hlist_node *node) struct zcomp *comp = hlist_entry(node, struct zcomp, node); struct zcomp_strm *zstrm; - zstrm = *per_cpu_ptr(comp->stream, cpu); - if (!IS_ERR_OR_NULL(zstrm)) - zcomp_strm_free(zstrm); - *per_cpu_ptr(comp->stream, cpu) = NULL; + zstrm = per_cpu_ptr(comp->stream, cpu); + zcomp_strm_free(zstrm); return 0; } @@ -188,7 +182,7 @@ static int zcomp_init(struct zcomp *comp) { int ret; - comp->stream = alloc_percpu(struct zcomp_strm *); + comp->stream = alloc_percpu(struct zcomp_strm); if (!comp->stream) return -ENOMEM; diff --git a/drivers/block/zram/zcomp.h b/drivers/block/zram/zcomp.h index 1806475b919d..40f6420f4b2e 100644 --- a/drivers/block/zram/zcomp.h +++ b/drivers/block/zram/zcomp.h @@ -5,8 +5,11 @@ #ifndef _ZCOMP_H_ #define _ZCOMP_H_ +#include <linux/local_lock.h> struct zcomp_strm { + /* The members ->buffer and ->tfm are protected by ->lock. */ + local_lock_t lock; /* compression/decompression buffer */ void *buffer; struct crypto_comp *tfm; @@ -14,7 +17,7 @@ struct zcomp_strm { /* dynamic per-device compression frontend */ struct zcomp { - struct zcomp_strm * __percpu *stream; + struct zcomp_strm __percpu *stream; const char *name; struct hlist_node node; }; diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index ebb234f36909..6e2ad90b17a3 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -1510,13 +1510,8 @@ static void zram_bio_discard(struct zram *zram, u32 index, static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, int offset, unsigned int op, struct bio *bio) { - unsigned long start_time = jiffies; - struct request_queue *q = zram->disk->queue; int ret; - generic_start_io_acct(q, op, bvec->bv_len >> SECTOR_SHIFT, - &zram->disk->part0); - if (!op_is_write(op)) { atomic64_inc(&zram->stats.num_reads); ret = zram_bvec_read(zram, bvec, index, offset, bio); @@ -1526,8 +1521,6 @@ static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, ret = zram_bvec_write(zram, bvec, index, offset, bio); } - generic_end_io_acct(q, op, &zram->disk->part0, start_time); - zram_slot_lock(zram, index); zram_accessed(zram, index); zram_slot_unlock(zram, index); @@ -1548,6 +1541,7 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) u32 index; struct bio_vec bvec; struct bvec_iter iter; + unsigned long start_time; index = bio->bi_iter.bi_sector >> SECTORS_PER_PAGE_SHIFT; offset = (bio->bi_iter.bi_sector & @@ -1563,6 +1557,7 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) break; } + start_time = bio_start_io_acct(bio); bio_for_each_segment(bvec, bio, iter) { struct bio_vec bv = bvec; unsigned int unwritten = bvec.bv_len; @@ -1571,8 +1566,10 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) bv.bv_len = min_t(unsigned int, PAGE_SIZE - offset, unwritten); if (zram_bvec_rw(zram, &bv, index, offset, - bio_op(bio), bio) < 0) - goto out; + bio_op(bio), bio) < 0) { + bio->bi_status = BLK_STS_IOERR; + break; + } bv.bv_offset += bv.bv_len; unwritten -= bv.bv_len; @@ -1580,12 +1577,8 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) update_position(&index, &offset, &bv); } while (unwritten); } - + bio_end_io_acct(bio, start_time); bio_endio(bio); - return; - -out: - bio_io_error(bio); } /* @@ -1633,6 +1626,7 @@ static int zram_rw_page(struct block_device *bdev, sector_t sector, u32 index; struct zram *zram; struct bio_vec bv; + unsigned long start_time; if (PageTransHuge(page)) return -ENOTSUPP; @@ -1651,7 +1645,9 @@ static int zram_rw_page(struct block_device *bdev, sector_t sector, bv.bv_len = PAGE_SIZE; bv.bv_offset = 0; + start_time = disk_start_io_acct(bdev->bd_disk, SECTORS_PER_PAGE, op); ret = zram_bvec_rw(zram, &bv, index, offset, op, NULL); + disk_end_io_acct(bdev->bd_disk, op, start_time); out: /* * If I/O fails, just return error(ie, non-zero) without |