mac_esp, zorro_esp, esp_scsi: Various improvements
cover mac_esp, zorro_esp, esp_scsi: Various improvements
This series has fixes and cleanup for mac_esp, zorro_esp and the core esp_scsi driver.
The improvements include elimination of duplicated code temporarily introduced for zorro_esp.
Michael, would you please regression test this series on elgar, if convenient? So far, only mac_esp has been tested with this code.
Finn Thain (6):
zorro_esp: Limit DMA transfers to 65535 bytes
esp_scsi: Track residual for PIO transfers
esp_scsi: Grant disconnect privilege for untagged commands
esp_scsi: Eliminate ESP_FLAG_DOING_SLOWCMD
esp_scsi: De-duplicate PIO routines
esp_scsi: Optimize PIO loops
drivers/scsi/esp_scsi.c | 192 +++++++++++++++++++++++++++++-------- drivers/scsi/esp_scsi.h | 10 +- drivers/scsi/mac_esp.c | 171 ++------------------------------- drivers/scsi/zorro_esp.c | 240 +++++++---------------------------------------- 4 files changed, 201 insertions(+), 412 deletions(-)
Signature · 1 line
2.16.4
PATCH 1/6 zorro_esp: Limit DMA transfers to 65535 bytes
The core driver, esp_scsi, does not use the ESP_CONFIG2_FENAB bit, so the chip's Transfer Counter register is only 16 bits wide (not 24). A larger transfer cannot work and will theoretically result in a failed command and a "DMA length is zero" error.
--- drivers/scsi/zorro_esp.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-)
drivers/scsi/zorro_esp.c +1 -7
@@ -245,7 +245,7 @@ static int fastlane_esp_irq_pending(struct esp *esp)static u32 zorro_esp_dma_length_limit (struct esp *esp,u32 dma_addr,u32 dma_len) { -return dma_len >0xFFFFFF ?0xFFFFFF : dma_len; +return dma_len >0xFFFF ?0xFFFF : dma_len; }static void zorro_esp_reset_dma (struct esp *esp)
@@ -484,7 +484,6 @@ static void zorro_esp_send_blz1230_dma_cmd(struct esp *esp, u32 addr,scsi_esp_cmd (esp,ESP_CMD_DMA );zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );scsi_esp_cmd (esp, cmd); }
@@ -529,7 +528,6 @@ static void zorro_esp_send_blz1230II_dma_cmd(struct esp *esp, u32 addr,scsi_esp_cmd (esp,ESP_CMD_DMA );zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );scsi_esp_cmd (esp, cmd); }
@@ -574,7 +572,6 @@ static void zorro_esp_send_blz2060_dma_cmd(struct esp *esp, u32 addr,scsi_esp_cmd (esp,ESP_CMD_DMA );zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );scsi_esp_cmd (esp, cmd); }
@@ -599,7 +596,6 @@ static void zorro_esp_send_cyber_dma_cmd(struct esp *esp, u32 addr,zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );if (write) {/* DMA receive */
@@ -649,7 +645,6 @@ static void zorro_esp_send_cyberII_dma_cmd(struct esp *esp, u32 addr,zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );if (write) {/* DMA receive */
@@ -691,7 +686,6 @@ static void zorro_esp_send_fastlane_dma_cmd(struct esp *esp, u32 addr,zorro_esp_write8 (esp, (esp_count >>0 ) &0xff ,ESP_TCLOW );zorro_esp_write8 (esp, (esp_count >>8 ) &0xff ,ESP_TCMED ); -zorro_esp_write8 (esp, (esp_count >>16 ) &0xff ,ESP_TCHI );if (write) {/* DMA receive */
Signature · 1 line
2.16.4
PATCH 2/6 esp_scsi: Track residual for PIO transfers
In reply to Finn ThainIf a target disconnects during a PIO data transfer the command may fail when the target reconnects:
scsi host1: DMA length is zero!
scsi host1: cur adr[04380000] len[00000000]
The scsi bus is then reset. This happens because the residual reached zero before the transfer was completed.
The usual residual calculation relies on the Transfer Count registers which works for DMA transfers but not for PIO transfers. Fix the problem by storing the PIO transfer residual and using that to correctly calculate bytes_sent.
--- drivers/scsi/esp_scsi.c | 1 + drivers/scsi/esp_scsi.h | 2 ++ drivers/scsi/mac_esp.c | 2 ++ 3 files changed, 5 insertions(+)
drivers/scsi/esp_scsi.c +1 -0
@@ -1338,6 +1338,7 @@ static int esp_data_bytes_sent(struct esp *esp, struct esp_cmd_entry *ent, bytes_sent = esp->data_dma_len; bytes_sent -= ecount; + bytes_sent -= esp->send_cmd_residual;/* * The am53c974 has a DMA 'pecularity'. The doc states:
drivers/scsi/esp_scsi.h +2 -0
@@ -540,6 +540,8 @@ struct esp {void *dma;int dmarev; + +int send_cmd_residual; };/* A front-end driver for the ESP chip should do the following in
drivers/scsi/mac_esp.c +2 -0
@@ -427,6 +427,8 @@ static void mac_esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count,scsi_esp_cmd (esp,ESP_CMD_TI ); } } + + esp->send_cmd_residual = esp_count; }static int mac_esp_irq_pending (struct esp *esp)
Signature · 1 line
2.16.4
PATCH 3/6 esp_scsi: Grant disconnect privilege for untagged commands
In reply to Finn ThainA SCSI device is not granted disconnect privilege by an esp_scsi host unless that device has its simple_tags flag set. However, a device may support disconnect/reselect and not support command queueing. Allow these devices to disconnect and improve bus utilization.
--- drivers/scsi/esp_scsi.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/scsi/esp_scsi.c +2 -6
@@ -800,13 +800,9 @@ static void esp_maybe_execute_command(struct esp *esp) build_identify:/* If we don't have a lun-data struct yet, we're probing -* so do not disconnect. Also, do not disconnect unless -* we have a tag on this command. +* so do not disconnect. */ -if (lp && (tp->flags &ESP_TGT_DISCONNECT ) && ent->tag[0 ]) - *p++ =IDENTIFY (1 , lun); -else - *p++ =IDENTIFY (0 , lun); + *p++ =IDENTIFY (lp && (tp->flags &ESP_TGT_DISCONNECT ), lun);if (ent->tag[0 ] && esp->rev ==ESP100 ) {/* ESP100 lacks select w/atn3 command, use select
Signature · 1 line
2.16.4
PATCH 4/6 esp_scsi: Eliminate ESP_FLAG_DOING_SLOWCMD
In reply to Finn ThainThe concept of a 'slow command' as it appears in esp_scsi is confusing because it could refer to an ESP command or a SCSI command. It turns out that it refers to a particular ESP select command which the driver also tracks as 'ESP_SELECT_MSGOUT'. For readability, it is better to use the terminology from the datasheets.
The global ESP_FLAG_DOING_SLOWCMD flag is redundant anyway, as it can be inferred from esp->select_state. Remove the ESP_FLAG_DOING_SLOWCMD cruft and just use a boolean local variable.
--- drivers/scsi/esp_scsi.c | 57 ++++++++++++++++++++----------------------------- drivers/scsi/esp_scsi.h | 3 +-- 2 files changed, 24 insertions(+), 36 deletions(-)
drivers/scsi/esp_scsi.c +23 -34
@@ -478,17 +478,6 @@ static void esp_restore_pointers(struct esp *esp, struct esp_cmd_entry *ent) spriv->tot_residue = ent->saved_tot_residue; } -static void esp_check_command_len (struct esp *esp,struct scsi_cmnd *cmd) -{ -if (cmd->cmd_len ==6 || - cmd->cmd_len ==10 || - cmd->cmd_len ==12 ) { - esp->flags &= ~ESP_FLAG_DOING_SLOWCMD ; - }else { - esp->flags |=ESP_FLAG_DOING_SLOWCMD ; - } -} -static void esp_write_tgt_config3 (struct esp *esp,int tgt) {if (esp->rev >ESP100A ) {
@@ -721,6 +710,7 @@ static void esp_maybe_execute_command(struct esp *esp)struct scsi_device *dev;struct scsi_cmnd *cmd;struct esp_cmd_entry *ent; +bool select_and_stop = false;int tgt, lun, i;u32 val, start_cmd;u8 *p;
@@ -752,7 +742,8 @@ static void esp_maybe_execute_command(struct esp *esp)esp_map_dma (esp, cmd);esp_save_pointers (esp, ent); -esp_check_command_len (esp, cmd); +if (!(cmd->cmd_len ==6 || cmd->cmd_len ==10 || cmd->cmd_len ==12 )) + select_and_stop = true; p = esp->command_block;
@@ -793,9 +784,9 @@ static void esp_maybe_execute_command(struct esp *esp) tp->flags &= ~ESP_TGT_CHECK_NEGO ; } -/* Process it like a slow command. */ -if (tp->flags & (ESP_TGT_NEGO_WIDE |ESP_TGT_NEGO_SYNC )) - esp->flags |=ESP_FLAG_DOING_SLOWCMD ; +/* If there are multiple message bytes, use Select and Stop */ +if (esp->msg_out_len) + select_and_stop = true; } build_identify:
@@ -808,23 +799,10 @@ static void esp_maybe_execute_command(struct esp *esp)/* ESP100 lacks select w/atn3 command, use select * and stop instead. */ - esp->flags |=ESP_FLAG_DOING_SLOWCMD ; + select_and_stop = true; } -if (!(esp->flags &ESP_FLAG_DOING_SLOWCMD )) { - start_cmd =ESP_CMD_SELA ; -if (ent->tag[0 ]) { - *p++ = ent->tag[0 ]; - *p++ = ent->tag[1 ]; - - start_cmd =ESP_CMD_SA3 ; - } - -for (i =0 ; i < cmd->cmd_len; i++) - *p++ = cmd->cmnd[i]; - - esp->select_state =ESP_SELECT_BASIC ; - }else { +if (select_and_stop) { esp->cmd_bytes_left = cmd->cmd_len; esp->cmd_bytes_ptr = &cmd->cmnd[0 ];
@@ -839,6 +817,19 @@ static void esp_maybe_execute_command(struct esp *esp) start_cmd =ESP_CMD_SELAS ; esp->select_state =ESP_SELECT_MSGOUT ; + }else { + start_cmd =ESP_CMD_SELA ; +if (ent->tag[0 ]) { + *p++ = ent->tag[0 ]; + *p++ = ent->tag[1 ]; + + start_cmd =ESP_CMD_SA3 ; + } + +for (i =0 ; i < cmd->cmd_len; i++) + *p++ = cmd->cmnd[i]; + + esp->select_state =ESP_SELECT_BASIC ; } val = tgt;if (esp->rev ==FASHME )
@@ -1248,7 +1239,6 @@ static int esp_finish_select(struct esp *esp)esp_unmap_dma (esp, cmd);esp_free_lun_tag (ent, cmd->device->hostdata); tp->flags &= ~(ESP_TGT_NEGO_SYNC |ESP_TGT_NEGO_WIDE ); - esp->flags &= ~ESP_FLAG_DOING_SLOWCMD ; esp->cmd_bytes_ptr =NULL ; esp->cmd_bytes_left =0 ; }else {
@@ -1299,9 +1289,8 @@ static int esp_finish_select(struct esp *esp)esp_flush_fifo (esp); } -/* If we are doing a slow command, negotiation, etc. -* we'll do the right thing as we transition to the -* next phase. +/* If we are doing a Select And Stop command, negotiation, etc. +* we'll do the right thing as we transition to the next phase. */ esp_event (esp,ESP_EVENT_CHECK_PHASE );return 0 ;
drivers/scsi/esp_scsi.h +1 -2
@@ -490,7 +490,6 @@ struct esp {u32 flags;#define ESP_FLAG_DIFFERENTIAL 0x00000001#define ESP_FLAG_RESETTING 0x00000002 -#define ESP_FLAG_DOING_SLOWCMD 0x00000004#define ESP_FLAG_WIDE_CAPABLE 0x00000008#define ESP_FLAG_QUICKIRQ_CHECK 0x00000010#define ESP_FLAG_DISABLE_SYNC 0x00000020
@@ -532,7 +531,7 @@ struct esp {u32 min_period;u32 radelay; -/* Slow command state. */ +/* ESP_CMD_SELAS command state */ u8 *cmd_bytes_ptr;int cmd_bytes_left;
Signature · 1 line
2.16.4
PATCH 5/6 esp_scsi: De-duplicate PIO routines
In reply to Finn ThainAs a temporary measure, the code to implement PIO transfers was duplicated in zorro_esp and mac_esp. Now that this code has stabilized, move it into the core driver to eliminate the duplication.
This replaces the inline assembler with more portable writesb() calls. Optimizing the m68k writesb() implementation is a separate patch.
--- drivers/scsi/esp_scsi.c | 126 +++++++++++++++++++++++++ drivers/scsi/esp_scsi.h | 5 + drivers/scsi/mac_esp.c | 173 ++--------------------------------- drivers/scsi/zorro_esp.c | 232 +++++++---------------------------------------- 4 files changed, 171 insertions(+), 365 deletions(-)
drivers/scsi/esp_scsi.c +126 -0
@@ -2776,3 +2776,129 @@ MODULE_PARM_DESC(esp_debug,module_init (esp_init);module_exit (esp_exit); + +#if IS_ENABLED (CONFIG_SCSI_MAC_ESP ) ||IS_ENABLED (CONFIG_SCSI_ZORRO_ESP ) +static inline unsigned int esp_wait_for_fifo (struct esp *esp) +{ +int i =500000 ; + +do { +unsigned int fbytes =esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ; + +if (fbytes) +return fbytes; + +udelay (2 ); + }while (--i); + +pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS )); +return 0 ; +} + +static inline int esp_wait_for_intr (struct esp *esp) +{ +int i =500000 ; + +do { + esp->sreg =esp_read8 (ESP_STATUS ); +if (esp->sreg &ESP_STAT_INTR ) +return 0 ; + +udelay (2 ); + }while (--i); + +pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg); +return 1 ; +} + +#define ESP_FIFO_SIZE 16 + +void esp_send_pio_cmd (struct esp *esp,u32 addr,u32 esp_count, +u32 dma_count,int write,u8 cmd) +{ +u8 phase = esp->sreg &ESP_STAT_PMASK ; + + cmd &= ~ESP_CMD_DMA ; + esp->send_cmd_error =0 ; + +if (write) { +u8 *dst = (u8 *)addr; +u8 mask = ~(phase ==ESP_MIP ?ESP_INTR_FDONE :ESP_INTR_BSERV ); + +scsi_esp_cmd (esp, cmd); + +while (1 ) { +if (!esp_wait_for_fifo (esp)) +break ; + + *dst++ =esp_read8 (ESP_FDATA ); + --esp_count; + +if (!esp_count) +break ; + +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & mask) { + esp->send_cmd_error =1 ; +break ; + } + +if (phase ==ESP_MIP ) +scsi_esp_cmd (esp,ESP_CMD_MOK ); + +scsi_esp_cmd (esp,ESP_CMD_TI ); + } + }else { +unsigned int n =ESP_FIFO_SIZE ; +u8 *src = (u8 *)addr; + +scsi_esp_cmd (esp,ESP_CMD_FLUSH ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n; + esp_count -= n; + +scsi_esp_cmd (esp, cmd); + +while (esp_count) { +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & ~ESP_INTR_BSERV ) { + esp->send_cmd_error =1 ; +break ; + } + + n =ESP_FIFO_SIZE - + (esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n; + esp_count -= n; + +scsi_esp_cmd (esp,ESP_CMD_TI ); + } + } + + esp->send_cmd_residual = esp_count; +} +EXPORT_SYMBOL (esp_send_pio_cmd); +#endif
drivers/scsi/esp_scsi.h +5 -0
@@ -431,6 +431,7 @@ struct esp_driver_ops {struct esp {void __iomem *regs;void __iomem *dma_regs; +u8 __iomem *fifo_reg;const struct esp_driver_ops *ops;
@@ -540,6 +541,7 @@ struct esp {void *dma;int dmarev; +int send_cmd_error;int send_cmd_residual; };
@@ -581,4 +583,7 @@ extern void scsi_esp_unregister(struct esp *);extern irqreturn_t scsi_esp_intr (int ,void *);extern void scsi_esp_cmd (struct esp *,u8 ); +extern void esp_send_pio_cmd (struct esp *esp,u32 dma_addr,u32 esp_count, +u32 dma_count,int write,u8 cmd); +#endif /* !(_ESP_SCSI_H) */
drivers/scsi/mac_esp.c +6 -167
225 lines not on this page · raw message
drivers/scsi/zorro_esp.c +34 -198
343 lines not on this page · raw message
Signature · 1 line
2.16.4
Re: [PATCH 5/6] esp_scsi: De-duplicate PIO routines
Quoted 1 line
+#if
IS_ENABLED(CONFIG_SCSI_MAC_ESP)||IS_ENABLED(CONFIG_SCSI_ZORRO_ESP)
Please add a new CONFIG_SCSI_ESP_PIO symbol that is selected by the mac and zorro drivers instead.
drivers/scsi/esp_scsi.c MODULE_PARM_DESC(esp_debug, · line 2794
Quoted 1 line
drivers/scsi/esp_scsi.c +1
-0+pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));
This should probably use dev_err (at least with my series).
Re: [PATCH 5/6] esp_scsi: De-duplicate PIO routines
Quoted 5 lines · Christoph Hellwig
Quoted 1 line
+#if
IS_ENABLED(CONFIG_SCSI_MAC_ESP)||IS_ENABLED(CONFIG_SCSI_ZORRO_ESP)Please add a new
CONFIG_SCSI_ESP_PIOsymbol that is selected by the mac and zorro drivers instead.
OK.
Quoted 4 lines
drivers/scsi/esp_scsi.c MODULE_PARM_DESC(esp_debug, · line 2794
Quoted 1 line
drivers/scsi/esp_scsi.c +1
-0+pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));This should probably use
dev_err(at least with my series).
How about shost_printk()? That would be consistent with the rest of esp_scsi.c. It also means my series need not depend on yours which would complicate things.
Re: [PATCH 5/6] esp_scsi: De-duplicate PIO routines
Quoted 8 lines · Finn Thain
Quoted 4 lines
drivers/scsi/esp_scsi.c MODULE_PARM_DESC(esp_debug, · line 2794
Quoted 1 line
drivers/scsi/esp_scsi.c +1
-0+pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));This should probably use
dev_err(at least with my series).How about
shost_printk()? That would be consistent with the rest ofesp_scsi.c. It also means my series need not depend on yours which would complicate things.
Yes, that sounds even better.
Re: [PATCH 5/6] esp_scsi: De-duplicate PIO routines
In reply to Finn ThainQuoted 140 lines · Finn Thain
As a temporary measure, the code to implement PIO transfers was duplicated in
zorro_espandmac_esp. Now that this code has stabilized, move it into the core driver to eliminate the duplication.This replaces the inline assembler with more portable
writesb()calls. Optimizing the m68kwritesb()implementation is a separate patch.Tested-by Stan Johnson <[email protected]>Signed-off-by Finn Thain <[email protected]>--- drivers/scsi/esp_scsi.c | 126 +++++++++++++++++++++++++ drivers/scsi/esp_scsi.h | 5 + drivers/scsi/mac_esp.c | 173 ++--------------------------------- drivers/scsi/zorro_esp.c | 232 +++++++---------------------------------------- 4 files changed, 171 insertions(+), 365 deletions(-)drivers/scsi/esp_scsi.c +116
-0@@ -2776,3 +2776,129 @@ MODULE_PARM_DESC(esp_debug,module_init (esp_init);module_exit (esp_exit); + +#if IS_ENABLED (CONFIG_SCSI_MAC_ESP ) ||IS_ENABLED (CONFIG_SCSI_ZORRO_ESP ) +static inline unsigned int esp_wait_for_fifo (struct esp *esp) +{ +int i =500000 ; + +do { +unsigned int fbytes =esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ; + +if (fbytes) +return fbytes; + +udelay (2 ); + }while (--i); + +pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS )); +return 0 ; +} + +static inline int esp_wait_for_intr (struct esp *esp) +{ +int i =500000 ; + +do { + esp->sreg =esp_read8 (ESP_STATUS ); +if (esp->sreg &ESP_STAT_INTR ) +return 0 ; + +udelay (2 ); + }while (--i); + +pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg); +return 1 ; +} + +#define ESP_FIFO_SIZE 16 + +void esp_send_pio_cmd (struct esp *esp,u32 addr,u32 esp_count, +u32 dma_count,int write,u8 cmd) +{ +u8 phase = esp->sreg &ESP_STAT_PMASK ; + + cmd &= ~ESP_CMD_DMA ; + esp->send_cmd_error =0 ; + +if (write) { +u8 *dst = (u8 *)addr; +u8 mask = ~(phase ==ESP_MIP ?ESP_INTR_FDONE :ESP_INTR_BSERV ); + +scsi_esp_cmd (esp, cmd); + +while (1 ) { +if (!esp_wait_for_fifo (esp)) +break ; + + *dst++ =esp_read8 (ESP_FDATA ); + --esp_count; + +if (!esp_count) +break ; + +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & mask) { + esp->send_cmd_error =1 ; +break ; + } + +if (phase ==ESP_MIP ) +scsi_esp_cmd (esp,ESP_CMD_MOK ); + +scsi_esp_cmd (esp,ESP_CMD_TI ); + } + }else { +unsigned int n =ESP_FIFO_SIZE ; +u8 *src = (u8 *)addr; + +scsi_esp_cmd (esp,ESP_CMD_FLUSH ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n; + esp_count -= n; + +scsi_esp_cmd (esp, cmd); + +while (esp_count) { +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & ~ESP_INTR_BSERV ) { + esp->send_cmd_error =1 ; +break ; + } + + n =ESP_FIFO_SIZE - + (esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n;
drivers/scsi/esp_scsi.c MODULE_PARM_DESC(esp_debug, · line 2904
Quoted 10 lines
drivers/scsi/esp_scsi.c +10
-0+ esp_count -= n; + + scsi_esp_cmd(esp, ESP_CMD_TI); + } + } + + esp->send_cmd_residual = esp_count; +} +EXPORT_SYMBOL (esp_send_pio_cmd); +#endif
These function are conditional, but
Quoted 21 lines
drivers/scsi/esp_scsi.h +2
-0@@ -431,6 +431,7 @@ struct esp_driver_ops {struct esp {void __iomem *regs;void __iomem *dma_regs; +u8 __iomem *fifo_reg;const struct esp_driver_ops *ops;@@ -540,6 +541,7 @@ struct esp {void *dma;int dmarev; +int send_cmd_error;int send_cmd_residual; };
drivers/scsi/esp_scsi.h extern void scsi_esp_unregister(struct esp *); · line 589
Quoted 7 lines
drivers/scsi/esp_scsi.h +3
-0@@ -581,4 +583,7 @@ extern void scsi_esp_unregister(struct esp *);extern irqreturn_t scsi_esp_intr (int ,void *);extern void scsi_esp_cmd (struct esp *,u8 ); +extern void esp_send_pio_cmd (struct esp *esp,u32 dma_addr,u32 esp_count, +u32 dma_count,int write,u8 cmd); +#endif /* !(_ESP_SCSI_H) */
These are not.
Which will result in building errors when the said configuration is not enabled. Please fix by either making everything conditional or remove the conditional completely.
Cheers,
Hannes
Signature · 5 lines
Dr. Hannes Reinecke Teamlead Storage & Networking [email protected] +49 911 74053 688 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton HRB 21284 (AG Nürnberg)
Re: [PATCH 5/6] esp_scsi: De-duplicate PIO routines
Quoted 178 lines · Hannes Reinecke
Quoted 140 lines · Finn Thain
As a temporary measure, the code to implement PIO transfers was duplicated in
zorro_espandmac_esp. Now that this code has stabilized, move it into the core driver to eliminate the duplication.This replaces the inline assembler with more portable
writesb()calls. Optimizing the m68kwritesb()implementation is a separate patch.Tested-by Stan Johnson <[email protected]>Signed-off-by Finn Thain <[email protected]>--- drivers/scsi/esp_scsi.c | 126 +++++++++++++++++++++++++ drivers/scsi/esp_scsi.h | 5 + drivers/scsi/mac_esp.c | 173 ++--------------------------------- drivers/scsi/zorro_esp.c | 232 +++++++---------------------------------------- 4 files changed, 171 insertions(+), 365 deletions(-)drivers/scsi/esp_scsi.c +116
-0@@ -2776,3 +2776,129 @@ MODULE_PARM_DESC(esp_debug,module_init (esp_init);module_exit (esp_exit); + +#if IS_ENABLED (CONFIG_SCSI_MAC_ESP ) ||IS_ENABLED (CONFIG_SCSI_ZORRO_ESP ) +static inline unsigned int esp_wait_for_fifo (struct esp *esp) +{ +int i =500000 ; + +do { +unsigned int fbytes =esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ; + +if (fbytes) +return fbytes; + +udelay (2 ); + }while (--i); + +pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS )); +return 0 ; +} + +static inline int esp_wait_for_intr (struct esp *esp) +{ +int i =500000 ; + +do { + esp->sreg =esp_read8 (ESP_STATUS ); +if (esp->sreg &ESP_STAT_INTR ) +return 0 ; + +udelay (2 ); + }while (--i); + +pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg); +return 1 ; +} + +#define ESP_FIFO_SIZE 16 + +void esp_send_pio_cmd (struct esp *esp,u32 addr,u32 esp_count, +u32 dma_count,int write,u8 cmd) +{ +u8 phase = esp->sreg &ESP_STAT_PMASK ; + + cmd &= ~ESP_CMD_DMA ; + esp->send_cmd_error =0 ; + +if (write) { +u8 *dst = (u8 *)addr; +u8 mask = ~(phase ==ESP_MIP ?ESP_INTR_FDONE :ESP_INTR_BSERV ); + +scsi_esp_cmd (esp, cmd); + +while (1 ) { +if (!esp_wait_for_fifo (esp)) +break ; + + *dst++ =esp_read8 (ESP_FDATA ); + --esp_count; + +if (!esp_count) +break ; + +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & mask) { + esp->send_cmd_error =1 ; +break ; + } + +if (phase ==ESP_MIP ) +scsi_esp_cmd (esp,ESP_CMD_MOK ); + +scsi_esp_cmd (esp,ESP_CMD_TI ); + } + }else { +unsigned int n =ESP_FIFO_SIZE ; +u8 *src = (u8 *)addr; + +scsi_esp_cmd (esp,ESP_CMD_FLUSH ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n; + esp_count -= n; + +scsi_esp_cmd (esp, cmd); + +while (esp_count) { +if (esp_wait_for_intr (esp)) { + esp->send_cmd_error =1 ; +break ; + } + +if ((esp->sreg &ESP_STAT_PMASK ) != phase) +break ; + + esp->ireg =esp_read8 (ESP_INTRPT ); +if (esp->ireg & ~ESP_INTR_BSERV ) { + esp->send_cmd_error =1 ; +break ; + } + + n =ESP_FIFO_SIZE - + (esp_read8 (ESP_FFLAGS ) &ESP_FF_FBYTES ); + +if (n > esp_count) + n = esp_count; +writesb (esp->fifo_reg, src, n); + src += n;drivers/scsi/esp_scsi.c MODULE_PARM_DESC(esp_debug, · line 2904
Quoted 10 lines
drivers/scsi/esp_scsi.c +10
-0+ esp_count -= n; + + scsi_esp_cmd(esp, ESP_CMD_TI); + } + } + + esp->send_cmd_residual = esp_count; +} +EXPORT_SYMBOL (esp_send_pio_cmd); +#endif These function are conditional, but
Quoted 24 lines
drivers/scsi/esp_scsi.h +2
-0@@ -431,6 +431,7 @@ struct esp_driver_ops {struct esp {void __iomem *regs;void __iomem *dma_regs; +u8 __iomem *fifo_reg;const struct esp_driver_ops *ops;@@ -540,6 +541,7 @@ struct esp {void *dma;int dmarev; +int send_cmd_error;int send_cmd_residual; };@@ -581,4 +583,7 @@ extern void scsi_esp_unregister(struct esp *);extern irqreturn_t scsi_esp_intr (int ,void *);extern void scsi_esp_cmd (struct esp *,u8 );
Quoted 8 lines
drivers/scsi/esp_scsi.h extern void scsi_esp_unregister(struct esp *); · line 589
Quoted 4 lines
drivers/scsi/esp_scsi.h +3
-0+extern void esp_send_pio_cmd (struct esp *esp,u32 dma_addr,u32 esp_count, +u32 dma_count,int write,u8 cmd); +#endif /* !(_ESP_SCSI_H) */ These are not.
Which will result in building errors when the said configuration is not enabled.
I didn't see errors or new warnings when build-testing with CONFIG_SCSI_AM53C974=y with gcc-5.4 or gcc-7.3, nor when build-testing CONFIG_SUN3X_ESP=y with gcc-4.4.
Are you perhaps thinking of "declared static but never defined"? That doesn't apply here because esp_send_pio_cmd() isn't static. Likewise "static declaration of 'foo' follows non-static declaration"...
Quoted 10 lines
Cheers,
Hannes
Signature · 5 lines
Dr. Hannes Reinecke Teamlead Storage & Networking [email protected] +49 911 74053 688 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg GF: F. Imend?rffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton HRB 21284 (AG N?rnberg)
PATCH 6/6 esp_scsi: Optimize PIO loops
In reply to Finn ThainAvoid function calls in the inner PIO loops. On a Centris 660av this improves throughput for sequential read transfers by about 40% and sequential write by about 10%.
Unfortunately it is not possible to have method calls like esp_write8() placed inline so this is always going to be slow (even with LTO).
--- drivers/scsi/esp_scsi.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/scsi/esp_scsi.c +7 -7
@@ -2788,7 +2788,7 @@ static inline unsigned int esp_wait_for_fifo(struct esp *esp)if (fbytes)return fbytes; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));
@@ -2804,7 +2804,7 @@ static inline int esp_wait_for_intr(struct esp *esp)if (esp->sreg &ESP_STAT_INTR )return 0 ; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg);
@@ -2831,7 +2831,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count,if (!esp_wait_for_fifo (esp))break ; - *dst++ =esp_read8 (ESP_FDATA ); + *dst++ =readb (esp->fifo_reg); --esp_count;if (!esp_count)
@@ -2852,15 +2852,15 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, }if (phase ==ESP_MIP ) -scsi_esp_cmd (esp,ESP_CMD_MOK ); +esp_write8 (ESP_CMD_MOK ,ESP_CMD ); -scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD ); } }else {unsigned int n =ESP_FIFO_SIZE ;u8 *src = (u8 *)addr; -scsi_esp_cmd (esp,ESP_CMD_FLUSH ); +esp_write8 (ESP_CMD_FLUSH ,ESP_CMD );if (n > esp_count) n = esp_count;
@@ -2894,7 +2894,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, src += n; esp_count -= n; -scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD ); } }
Signature · 1 line
2.16.4
Re: [PATCH 6/6] esp_scsi: Optimize PIO loops
Hi Finn,
Am 13.10.2018 um 13:51 schrieb Finn Thain:
Quoted 45 lines
Avoid function calls in the inner PIO loops. On a Centris 660av this improves throughput for sequential read transfers by about 40% and sequential write by about 10%.
Unfortunately it is not possible to have method calls like
esp_write8()placed inline so this is always going to be slow (even with LTO).Tested-by Stan Johnson <[email protected]>Signed-off-by Finn Thain <[email protected]>--- drivers/scsi/esp_scsi.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)drivers/scsi/esp_scsi.c +3
-3@@ -2788,7 +2788,7 @@ static inline unsigned int esp_wait_for_fifo(struct esp *esp)if (fbytes)return fbytes; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));@@ -2804,7 +2804,7 @@ static inline int esp_wait_for_intr(struct esp *esp)if (esp->sreg &ESP_STAT_INTR )return 0 ; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg);@@ -2831,7 +2831,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count,if (!esp_wait_for_fifo (esp))break ; - *dst++ =esp_read8 (ESP_FDATA ); + *dst++ =readb (esp->fifo_reg); --esp_count;if (!esp_count)
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2855
Quoted 5 lines
drivers/scsi/esp_scsi.c +1
-1@@ -2852,15 +2852,15 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, }if (phase ==ESP_MIP ) -scsi_esp_cmd (esp,ESP_CMD_MOK ); +esp_write8 (ESP_CMD_MOK ,ESP_CMD );
You're no longer logging this command with this patch. (That'll be the reason for the speedup you saw ...)
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2856
Quoted 3 lines
drivers/scsi/esp_scsi.c +1
-1-scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );
Same here..
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2863
Quoted 7 lines
drivers/scsi/esp_scsi.c +1
-1} }else {unsigned int n =ESP_FIFO_SIZE ;u8 *src = (u8 *)addr; -scsi_esp_cmd (esp,ESP_CMD_FLUSH ); +esp_write8 (ESP_CMD_FLUSH ,ESP_CMD );
here..
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2897
Quoted 9 lines
drivers/scsi/esp_scsi.c +1
-1if (n > esp_count) n = esp_count;@@ -2894,7 +2894,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, src += n; esp_count -= n; -scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );
and here.
The burst of ESP_CMD_TI's in the log was quite useful to spot what went wrong during PIO. Maybe mention in the changelog that commands during PIO are no longer logged? Or introduce a new ESP_EVENT_PIO and log that at the start of PIO?
Cheers,
Michael
Quoted 4 lines
drivers/scsi/esp_scsi.c +0
-0} }
Re: [PATCH 6/6] esp_scsi: Optimize PIO loops
Quoted 84 lines · Michael Schmitz
Hi Finn,
Am 13.10.2018 um 13:51 schrieb Finn Thain:
Quoted 47 lines
Avoid function calls in the inner PIO loops. On a Centris 660av this improves throughput for sequential read transfers by about 40% and sequential write by about 10%.
Unfortunately it is not possible to have method calls like
esp_write8()placed inline so this is always going to be slow (even with LTO).Tested-by Stan Johnson <[email protected]>Signed-off-by Finn Thain <[email protected]>--- drivers/scsi/esp_scsi.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)drivers/scsi/esp_scsi.c +1
-1@@ -2788,7 +2788,7 @@ static inline unsigned int esp_wait_for_fifo(struct esp *esp)if (fbytes)return fbytes; -udelay (2 ); +udelay (1 ); }while (--i);drivers/scsi/esp_scsi.c +2
-2pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));@@ -2804,7 +2804,7 @@ static inline int esp_wait_for_intr(struct esp *esp)if (esp->sreg &ESP_STAT_INTR )return 0 ; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg);@@ -2831,7 +2831,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, if (!esp_wait_for_fifo(esp)) break; - *dst++ =esp_read8 (ESP_FDATA ); + *dst++ =readb (esp->fifo_reg); --esp_count;drivers/scsi/esp_scsi.c +0
-0if (!esp_count)drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2856
Quoted 6 lines
drivers/scsi/esp_scsi.c +1
-1@@ -2852,15 +2852,15 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, }if (phase ==ESP_MIP ) -scsi_esp_cmd (esp,ESP_CMD_MOK ); +esp_write8 (ESP_CMD_MOK ,ESP_CMD );You're no longer logging this command with this patch. (That'll be the reason for the speedup you saw ...)
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2857
Quoted 3 lines
drivers/scsi/esp_scsi.c +1
-1-scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );Same here..
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2863
Quoted 7 lines
drivers/scsi/esp_scsi.c +1
-1} }else {unsigned int n =ESP_FIFO_SIZE ;u8 *src = (u8 *)addr; -scsi_esp_cmd (esp,ESP_CMD_FLUSH ); +esp_write8 (ESP_CMD_FLUSH ,ESP_CMD );here..
Quoted 8 lines
drivers/scsi/esp_scsi.c +0
-0if (n > esp_count) n = esp_count;@@ -2894,7 +2894,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, src += n; esp_count -= n;
Quoted 5 lines
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2898
Quoted 2 lines
drivers/scsi/esp_scsi.c +1
-1-scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );and here.
Yes, it's deliberate.
Quoted 2 lines
The burst of ESP_CMD_TI's in the log was quite useful to spot what went wrong during PIO.
I don't think it's as useful as you seem to think. Compare mac_esp_send_pdma_cmd().
Quoted 4 lines
Maybe mention in the changelog that commands during PIO are no longer logged? Or introduce a new
ESP_EVENT_PIOand log that at the start of PIO?
Yes, and I did leave a scsi_esp_cmd(esp, cmd) call at the start of PIO.
That should be sufficient, right?
Quoted 11 lines
Cheers,
Michael
Quoted 4 lines
drivers/scsi/esp_scsi.c +0
-0} }
Re: [PATCH 6/6] esp_scsi: Optimize PIO loops
Hi Finn,
Am 13.10.2018 um 17:09 schrieb Finn Thain:
Quoted 86 lines · Michael Schmitz
Quoted 85 lines
Hi Finn,
Am 13.10.2018 um 13:51 schrieb Finn Thain:
Quoted 47 lines
Avoid function calls in the inner PIO loops. On a Centris 660av this improves throughput for sequential read transfers by about 40% and sequential write by about 10%.
Unfortunately it is not possible to have method calls like
esp_write8()placed inline so this is always going to be slow (even with LTO).Tested-by Stan Johnson <[email protected]>Signed-off-by Finn Thain <[email protected]>--- drivers/scsi/esp_scsi.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)drivers/scsi/esp_scsi.c +1
-1@@ -2788,7 +2788,7 @@ static inline unsigned int esp_wait_for_fifo(struct esp *esp)if (fbytes)return fbytes; -udelay (2 ); +udelay (1 ); }while (--i);drivers/scsi/esp_scsi.c +2
-2pr_err ("FIFO is empty (sreg %02x)\n" ,esp_read8 (ESP_STATUS ));@@ -2804,7 +2804,7 @@ static inline int esp_wait_for_intr(struct esp *esp)if (esp->sreg &ESP_STAT_INTR )return 0 ; -udelay (2 ); +udelay (1 ); }while (--i);pr_err ("IRQ timeout (sreg %02x)\n" , esp->sreg);@@ -2831,7 +2831,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, if (!esp_wait_for_fifo(esp)) break; - *dst++ =esp_read8 (ESP_FDATA ); + *dst++ =readb (esp->fifo_reg); --esp_count;drivers/scsi/esp_scsi.c +0
-0if (!esp_count)drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2856
Quoted 6 lines
drivers/scsi/esp_scsi.c +1
-1@@ -2852,15 +2852,15 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, }if (phase ==ESP_MIP ) -scsi_esp_cmd (esp,ESP_CMD_MOK ); +esp_write8 (ESP_CMD_MOK ,ESP_CMD );You're no longer logging this command with this patch. (That'll be the reason for the speedup you saw ...)
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2857
Quoted 3 lines
drivers/scsi/esp_scsi.c +1
-1-scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );Same here..
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, · line 2863
Quoted 7 lines
drivers/scsi/esp_scsi.c +1
-1} }else {unsigned int n =ESP_FIFO_SIZE ;u8 *src = (u8 *)addr; -scsi_esp_cmd (esp,ESP_CMD_FLUSH ); +esp_write8 (ESP_CMD_FLUSH ,ESP_CMD );here..
Quoted 8 lines
drivers/scsi/esp_scsi.c +0
-0if (n > esp_count) n = esp_count;@@ -2894,7 +2894,7 @@ void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 esp_count, src += n; esp_count -= n;
Quoted 7 lines
Quoted 5 lines
drivers/scsi/esp_scsi.c void esp_send_pio_cmd(struct esp *esp, u32 addr, u32 · line 2898
Quoted 2 lines
drivers/scsi/esp_scsi.c +1
-1-scsi_esp_cmd (esp,ESP_CMD_TI ); +esp_write8 (ESP_CMD_TI ,ESP_CMD );and here.
Yes, it's deliberate.
I'm sure it was... and I wasn't objecting to that.
Quoted 6 lines
Quoted 2 lines
The burst of ESP_CMD_TI's in the log was quite useful to spot what went wrong during PIO.
I don't think it's as useful as you seem to think. Compare
mac_esp_send_pdma_cmd().
Quoted 6 lines
Quoted 4 lines
Maybe mention in the changelog that commands during PIO are no longer logged? Or introduce a new
ESP_EVENT_PIOand log that at the start of PIO?Yes, and I did leave a
scsi_esp_cmd(esp, cmd)call at the start of PIO.
Which I missed from just looking at the patch, sorry.
Quoted 1 line
That should be sufficient, right?
It would indeed. Thanks for clarifying.
Cheers,
Michael
Hi Finn,
Am 13.10.2018 um 13:51 schrieb Finn Thain:
Quoted 18 lines
This series has fixes and cleanup for
mac_esp,zorro_espand the coreesp_scsidriver.The improvements include elimination of duplicated code temporarily introduced for
zorro_esp.Michael, would you please regression test this series on elgar, if convenient? So far, only
mac_esphas been tested with this code.Finn Thain (6):
zorro_esp: Limit DMA transfers to 65535 bytes
esp_scsi: Track residual for PIO transfers
esp_scsi: Grant disconnect privilege for untagged commands
esp_scsi: EliminateESP_FLAG_DOING_SLOWCMD
esp_scsi: De-duplicate PIO routines
esp_scsi: Optimize PIO loops
Quoted 6 lines
drivers/scsi/esp_scsi.c | 192 +++++++++++++++++++++++++++++-------- drivers/scsi/esp_scsi.h | 10 +- drivers/scsi/mac_esp.c | 171 ++------------------------------- drivers/scsi/zorro_esp.c | 240 +++++++---------------------------------------- 4 files changed, 201 insertions(+), 412 deletions(-)
For Amiga zorro-esp: