[PULL 34/88] esp.c: update esp_do_dma() bypass if async_len is zero to include non-zero transfer check

In the PDMA case the last transfer from the device to the FIFO has occurred
(async_len is zero) but esp_do_dma() is still being called to drain the
remaining FIFO contents.

The additional non-zero transfer check ensures that we still defer the SCSI
layer in the case where we are waiting for data for a TI command or a DMA
enable signal.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Tested-by: Helge Deller <deller@gmx.de>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20240112125420.514425-35-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
---
 hw/scsi/esp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/scsi/esp.c b/hw/scsi/esp.c
index 14284ef54a..9647be4cc3 100644
--- a/hw/scsi/esp.c
+++ b/hw/scsi/esp.c
@@ -696,7 +696,7 @@ static void esp_do_dma(ESPState *s)
     if (!s->current_req) {
         return;
     }
-    if (s->async_len == 0) {
+    if (s->async_len == 0 && esp_get_tc(s) && s->ti_size) {
         /* Defer until data is available.  */
         return;
     }
-- 
2.39.2

• [PULL 01/88] esp.c: don't clear cmdfifo when esp_select() fails in get_cmd()
• [PULL 02/88] esp.c: move existing request cancel check into esp_select()
• [PULL 03/88] esp.c: add FIFO wraparound support to esp_fifo_pop_buf()
• [PULL 04/88] esp.c: remove FIFO clear from esp_select()
• [PULL 05/88] esp.c: move esp_select() to ESP selection commands from get_cmd()
• [PULL 06/88] esp.c: update esp_set_tc() to set STAT_TC flag
• [PULL 07/88] esp.c: start removal of manual STAT_TC setting when transfer counter reaches zero
• [PULL 08/88] esp.c: move command execution logic to new esp_run_cmd() function
• [PULL 09/88] esp.c: update TC check logic in do_dma_pdma_cb() to check for TC == 0
• [PULL 10/88] esp.c: move buffer and TC logic into separate to/from device paths in esp_do_dma()
• [PULL 11/88] esp.c: remove unused case from esp_pdma_read()
• [PULL 12/88] esp.c: don't accumulate directly into cmdfifo
• [PULL 13/88] esp.c: decrement the TC during MESSAGE OUT and COMMAND phases
• [PULL 14/88] esp.c: introduce esp_set_phase() helper function
• [PULL 15/88] esp.c: remove another set of manual STAT_TC updates
• [PULL 16/88] esp.c: remove MacOS TI workaround that pads FIFO transfers to ESP_FIFO_SZ
• [PULL 17/88] esp.c: don't reset the TC and ESP_RSEQ state when executing a SCSI command
• [PULL 18/88] esp.c: don't clear RFLAGS register when DMA is complete
• [PULL 19/88] esp.c: remove zero transfer size check from esp_do_dma()
• [PULL 20/88] esp.c: update condition for esp_dma_done() in esp_do_dma() from device path
• [PULL 21/88] esp.c: update condition for esp_dma_done() in esp_do_dma() to device path
• [PULL 22/88] esp.c: ensure that the PDMA callback is called for every device read
• [PULL 23/88] esp.c: don't immediately raise INTR_BS if SCSI data needed in esp_do_dma()
• [PULL 24/88] esp.c: remove TC adjustment in esp_do_dma() from device path
• [PULL 25/88] esp.c: remove unaligned adjustment in do_dma_pdma_cb() to device path
• [PULL 26/88] esp.c: remove unneeded if() check in esp_transfer_data()
• [PULL 27/88] esp.c: update end of transfer logic at the end of esp_transfer_data()
• [PULL 28/88] esp.c: consolidate async_len and TC == 0 checks in do_dma_pdma_cb() and esp_do_dma()
• [PULL 29/88] esp.c: fix premature end of phase logic esp_command_complete
• [PULL 30/88] esp.c: move TC and FIFO check logic into esp_dma_done()
• [PULL 31/88] esp.c: rename esp_dma_done() to esp_dma_ti_check()
• [PULL 32/88] esp.c: copy PDMA logic for transfers to device from do_dma_pdma_cb() to esp_do_dma()
• [PULL 33/88] esp.c: copy logic for do_cmd transfers from do_dma_pdma_cb() to esp_do_dma()
• [PULL 34/88] esp.c: update esp_do_dma() bypass if async_len is zero to include non-zero transfer check
• [PULL 35/88] esp.c: move end of SCSI transfer check after TC adjustment in do_dma_pdma_cb()
• [PULL 36/88] esp.c: remove s_without_satn_pdma_cb() PDMA callback
• [PULL 37/88] esp.c: introduce esp_get_phase() function
• [PULL 38/88] esp.c: convert esp_do_dma() to switch statement based upon SCSI phase
• [PULL 39/88] esp.c: convert do_dma_pdma_db() to switch statement based upon SCSI phase
• [PULL 40/88] esp.c: convert esp_do_nodma() to switch statement based upon SCSI phase
• [PULL 41/88] esp.c: convert esp_do_dma() do_cmd path to check for SCSI phase instead
• [PULL 42/88] esp.c: convert do_dma_pdma_cb() do_cmd path to check for SCSI phase instead
• [PULL 43/88] esp.c: convert esp_do_nodma() do_cmd path to check for SCSI phase instead
• [PULL 44/88] esp.c: convert esp_reg_write() do_cmd path to check for SCSI phase instead
• [PULL 45/88] esp.c: remove do_cmd from ESPState
• [PULL 46/88] esp.c: untangle MESSAGE OUT and COMMAND phase logic in esp_do_dma()
• [PULL 47/88] esp.c: untangle MESSAGE OUT and COMMAND phase logic in do_dma_pdma_cb()
• [PULL 48/88] esp.c: untangle MESSAGE OUT and COMMAND phase logic in esp_do_nodma()
• [PULL 49/88] esp.c: move CMD_SELATN end of message phase detection to esp_do_dma() and do_dma_pdma_cb()
• [PULL 50/88] esp.c: move CMD_TI end of message phase detection to esp_do_dma() and do_dma_pdma_cb()
• [PULL 51/88] esp.c: don't use get_cmd() for CMD_SEL DMA commands
• [PULL 52/88] esp.c: move CMD_SELATNS end of command logic to esp_do_dma() and do_dma_pdma_cb()
• [PULL 53/88] esp.c: replace do_dma_pdma_cb() with esp_do_dma()
• [PULL 54/88] esp.c: move CMD_ICCS command logic to esp_do_dma()
• [PULL 55/88] esp.c: always use esp_do_dma() in pdma_cb()
• [PULL 56/88] esp.c: remove unused PDMA callback implementation
• [PULL 57/88] esp.c: rename data_in_ready to to data_ready
• [PULL 58/88] esp.c: separate logic based upon ESP command in esp_command_complete()
• [PULL 59/88] esp.c: separate logic based upon ESP command in esp_transfer_data()
• [PULL 60/88] esp.c: use deferred interrupts for both DATA IN and DATA OUT phases
• [PULL 61/88] esp.c: remove DATA IN phase logic when reading from FIFO
• [PULL 62/88] esp.c: zero command register when TI command terminates due to phase change
• [PULL 63/88] esp.c: remove unneeded ti_cmd field
• [PULL 64/88] esp.c: don't raise INTR_BS interrupt in DATA IN phase until TI command issued
• [PULL 65/88] esp.c: move non-DMA TI logic to separate esp_nodma_ti_dataout() function
• [PULL 66/88] esp.c: process non-DMA FIFO writes in esp_do_nodma()
• [PULL 67/88] esp.c: replace get_cmd() with esp_do_nodma()
• [PULL 68/88] esp.c: move write_response() non-DMA logic to esp_do_nodma()
• [PULL 69/88] esp.c: consolidate end of command sequence after ICCS command
• [PULL 70/88] esp.c: ensure that STAT_INT is cleared when reading ESP_RINTR
• [PULL 71/88] esp.c: don't clear the SCSI phase when reading ESP_RINTR
• [PULL 72/88] esp.c: handle TC underflow for DMA SCSI requests
• [PULL 73/88] esp.c: remove restriction on FIFO read access when DMA memory routines defined
• [PULL 74/88] esp.c: handle non-DMA FIFO writes used to terminate DMA commands
• [PULL 75/88] esp.c: improve ESP_RSEQ logic consolidation
• [PULL 76/88] esp.c: only transfer non-DMA COMMAND phase data for specific commands
• [PULL 77/88] esp.c: only transfer non-DMA MESSAGE OUT phase data for specific commands
• [PULL 78/88] esp.c: consolidate DMA and PDMA logic in DATA OUT phase
• [PULL 79/88] esp.c: consolidate DMA and PDMA logic in DATA IN phase
• [PULL 80/88] esp.c: consolidate DMA and PDMA logic in MESSAGE OUT phase
• [PULL 81/88] esp.c: remove redundant n variable in PDMA COMMAND phase
• [PULL 82/88] esp.c: consolidate DMA and PDMA logic in STATUS and MESSAGE IN phases
• [PULL 83/88] esp.c: replace n variable with len in esp_do_nodma()
• [PULL 84/88] esp.c: implement DMA Transfer Pad command for DATA phases
• [PULL 85/88] esp.c: rename irq_data IRQ to drq_irq
• [PULL 86/88] esp.c: keep track of the DRQ state during DMA
• [PULL 87/88] esp.c: switch TypeInfo registration to use DEFINE_TYPES() macro
• [PULL 88/88] esp.c: add my copyright to the file