Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[patch\]\s+3c59x\.c\s+for\s+2\.2\.17\s*$/: 18 ]

Total 18 documents matching your query.

1. [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 21:20:49 +1000
- Andrey's tester reported that the driver was still oopsing every three days on vortex-specific code for a 3c905. So I've split the ISR into vortex_interrupt() and boomerang_interrupt() as per the
/archives/netdev/2000-06/msg00271.html (27,527 bytes)

2. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 13:09:41 +0100 (BST)
Ok This is a mess in a lot of Don Becker skeleton base drivers. Its much cleaner if the last ring buffer slot is about to be used and the new alloc fails to simply recycle the buffer and throw the r
/archives/netdev/2000-06/msg00272.html (8,932 bytes)

3. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 22:42:16 +1000
Agreed. Mark Hemment's fix basically did this. He reserved an 'emergency' skbuff and never gave it away. I was not attracted to it because it handles a _very_ rare case by adding complexity to a core
/archives/netdev/2000-06/msg00273.html (8,849 bytes)

4. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 16:10:30 +0200
Applied against pre6, on PowerMac 7200, ok. Stress testing follows. Andreas
/archives/netdev/2000-06/msg00274.html (8,514 bytes)

5. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 16:08:18 +0200 (CEST)
If I understand it right, you want that the driver never passes one of its preallocated Rx buffers to the upper levels, so that the Rx ring is always of size RX_RING_SIZE. On receiving, it tries to a
/archives/netdev/2000-06/msg00275.html (10,027 bytes)

6. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 16:24:48 +0100 (BST)
No. I want to do if(<copybreak) { skb=alloc_skb(RING_SIZE); if(skb==NULL) recycle original } else { alloc skb if(skb!=NULL) copy }
/archives/netdev/2000-06/msg00276.html (8,623 bytes)

7. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 12:45:50 -0400 (EDT)
The previous code would allow operation even if not all buffers could be allocated. This is a good thing. I suppose the correct operation is to refuse to start if no buffers at all can be allocated.
/archives/netdev/2000-06/msg00277.html (9,903 bytes)

8. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 18:48:51 +0200 (CEST)
With all due respect, I don't understand! What has RING_SIZE to do with the allocation of a buffer? Should it be "PKT_BUF_SZ" instead (="size of each temporary Rx buffer") ? What does "recycle" mean:
/archives/netdev/2000-06/msg00278.html (9,356 bytes)

9. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: xxxx>
Date: Sun, 25 Jun 2000 23:37:23 +0000
I agree. I think. This is fairly easy to do: partially fill the rx ring and leave the other slots with a NULL skb, and set cur_rx and dirty_rx appropriately. Then hope that boomerang_rx will eventual
/archives/netdev/2000-06/msg00280.html (8,986 bytes)

10. [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Andrew Morton <andrewm@xxxxxxxxxx>
Date: Sun, 25 Jun 2000 21:20:49 +1000
Hi, Alan. - Andrey's tester reported that the driver was still oopsing every three days on vortex-specific code for a 3c905. So I've split the ISR into vortex_interrupt() and boomerang_interrupt() as
/archives/netdev/2000-06/msg00570.html (27,572 bytes)

11. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>
Date: Sun, 25 Jun 2000 13:09:41 +0100 (BST)
Ok This is a mess in a lot of Don Becker skeleton base drivers. Its much cleaner if the last ring buffer slot is about to be used and the new alloc fails to simply recycle the buffer and throw the r
/archives/netdev/2000-06/msg00571.html (8,957 bytes)

12. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Andrew Morton <andrewm@xxxxxxxxxx>
Date: Sun, 25 Jun 2000 22:42:16 +1000
Agreed. Mark Hemment's fix basically did this. He reserved an 'emergency' skbuff and never gave it away. I was not attracted to it because it handles a _very_ rare case by adding complexity to a core
/archives/netdev/2000-06/msg00572.html (8,899 bytes)

13. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Andreas Tobler <toa@xxxxxxxxxxx>
Date: Sun, 25 Jun 2000 16:10:30 +0200
Applied against pre6, on PowerMac 7200, ok. Stress testing follows. Andreas
/archives/netdev/2000-06/msg00573.html (8,539 bytes)

14. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Bogdan Costescu <Bogdan.Costescu@xxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 25 Jun 2000 16:08:18 +0200 (CEST)
If I understand it right, you want that the driver never passes one of its preallocated Rx buffers to the upper levels, so that the Rx ring is always of size RX_RING_SIZE. On receiving, it tries to a
/archives/netdev/2000-06/msg00574.html (10,052 bytes)

15. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>
Date: Sun, 25 Jun 2000 16:24:48 +0100 (BST)
No. I want to do if(<copybreak) { skb=alloc_skb(RING_SIZE); if(skb==NULL) recycle original } else { alloc skb if(skb!=NULL) copy }
/archives/netdev/2000-06/msg00575.html (8,672 bytes)

16. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Donald Becker <becker@xxxxxxxxx>
Date: Sun, 25 Jun 2000 12:45:50 -0400 (EDT)
The previous code would allow operation even if not all buffers could be allocated. This is a good thing. I suppose the correct operation is to refuse to start if no buffers at all can be allocated.
/archives/netdev/2000-06/msg00576.html (9,943 bytes)

17. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Bogdan Costescu <Bogdan.Costescu@xxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 25 Jun 2000 18:48:51 +0200 (CEST)
With all due respect, I don't understand! What has RING_SIZE to do with the allocation of a buffer? Should it be "PKT_BUF_SZ" instead (="size of each temporary Rx buffer") ? What does "recycle" mean:
/archives/netdev/2000-06/msg00577.html (9,381 bytes)

18. Re: [patch] 3c59x.c for 2.2.17 (score: 1)
Author: Andrew Morton <andrewm@xxxxxxxxxx>
Date: Sun, 25 Jun 2000 23:37:23 +0000
I agree. I think. This is fairly easy to do: partially fill the rx ring and leave the other slots with a NULL skb, and set cur_rx and dirty_rx appropriately. Then hope that boomerang_rx will eventual
/archives/netdev/2000-06/msg00579.html (9,060 bytes)


This search system is powered by Namazu